Sutton Kyra Leigh

PARENTHOOD AND ORGANIZATIONAL NETWORKS:

A RELATIONAL VIEW OF THE CAREER MOBILITY OF WORKING

PARENTS

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the Degree of Doctor of

Philosophy in the Graduate School of The Ohio State University

By

Kyra Leigh Sutton

**************

The Ohio State University

2006

Dissertation Committee:

Professor, Raymond A. Noe, Adviser Approved by

Professor, David Greenberger __________________________

Professor, Howard J. Klein Adviser Graduate Program in Labor and Human Resources

Copyrighted Kyra Leigh Sutton

ABSTRACT

This dissertation examined how parental responsibilities impacted three

organizational networks characteristics, network size, network ties, and network content

across employees. The study was based on theory and research from sociology (i.e. social

networks), careers, and the work-family literature. The study was designed to understand

the potential moderators of the relationship between parental status and the three network

characteristics. This dissertation sought to understand the relationship between the three

network characteristics of interest and two career outcomes including career success and

career management perceptions. The career success measures included in this study were

salary, salary growth, promotions, and career satisfaction. The career management

perception measures included in this study were career planning, career tactics, and

career mobility preparedness. This study utilized network analysis and investigated the

organizational networks of working adults with children in comparison to the networks of

working adults without children. The goal of this dissertation was to understand the

following research questions:

Research Question: First, how do networks differ after the birth of a child for males vs females? Secondly, how do networks differ between working adults with and without children? Thirdly, what constraints produce those differences?

ii

The study results suggest that network characteristic, network content, may vary

across parental status, where working parents reported a higher percentage of their

network content (i.e. topics of conversation) to be non-work and kin relevant topics.

There were no significant interactions between parental status and the four moderators of

interest, including gender, family involvement, job involvement, and role segmentation.

However, significant main effects were found for both job involvement and role

segmentation on network ties and network content. Finally, the results suggest that

network size has a significant main effects on salary growth and career mobility

preparedness, individual and peer-related career satisfaction. Network content was also

significantly related to career tactics.

ii

DEDICATION Dedicated to my three favorite people, Mom (Wanda G. Sutton), Dad (James C. Sutton),

and Grandma (Gloria I Gibson)

and in Memory Of My Favorite Little Buddy, “Sweetie Pie Sutton” (7/11/89 – 6/04/04)

iii

ACKNOWLEDGMENTS First, I must thank God, the higher power that has sustained me throughout my life, this

program, and in all future endeavors.

In reflecting on this dissertation process, I have many people to thank. I will start by

thanking my undergraduate thesis committee at Spelman College including Dr. Romie

Tribble, Dr. Ann Hornsby, and Dr. Jack Stone. Each of you saw my potential to go to

graduate school and complete a doctoral program, and I thank for your guidance,

encouragement, and confidence that I would begin and successfully complete my

program. I also want to thank my current dissertation committee members including Dr.

Raymond A. Noe, Dr. Howard Klein, and Dr. David Greenberger. First, Ray, as my

advisor and committee chair, I thank you for your time, wonderful feedback, and

willingness to push to get the best out of me. I also thank Ray for being the first faculty

member to really introduce me to the work-family area and for continuously giving me

the encouragement to do my best throughout the program, and really being the first

professor to give me a chance in the program. Howard, I thank you for many things

including your willingness to help me by answering as I progressed through this

dissertation process and other research projects. I appreciate your willingness to include

myself and my cohort during our first year in a research project, and I thank you for

always being accessible. David, I thank you for being a person that always has a open-

door policy and for making sure that as doctoral students we are taking care of ourselves

as well as our minds. I also thank you for challenging me to look through a “social

identity lens”, dissertation not excluded. Finally, in working on my dissertation I also had

the help of two very special individuals including Dr. James (Jim) Moody (Duke

University) and Eddie Willett. Jim, I thank you for giving my first taste of the social

iv

networking! This is a fascinating field and I am so appreciative to have had your

guidance and knowledge as I worked on my dissertation. I also appreciate you for

encouraging me to take ownership of the dissertation and encouraging me to really think

about the data and questions I wanted to analyze. Finally, I was very fortunate to have

what I consider to be the best experience with a computer programmer, that is, Eddie.

Eddie, I thank you for your time and patience and willingness to respond so quickly as

we worked on this project.

In addition, I would also like to thank my family and friends. First, Mom, I absolutely

needed your support in this program and I thank you forever. You are always available to

listen to me, support me, give a hug when I needed one, and make me laugh when I

needed to do that. Also, you are my perfect role model, beautiful, intelligent, generous

and thoughtful. Dad, thank you for always reminding me of the bigger goal, and being

able to see the bigger picture. Also Dad, thank you for always checking on me; I always

appreciated your calls just to “see how your young lady is doing”, it always gave me

great comfort to talk to you. Grandma, thank you for just being you. It was so comforting

to talk to you on the phone and just hear your words of encouragement and praise. You

have and always will make such a difference in my life. Knowing that I have Mom, Dad,

and Grandma in my life, always let me know that everything will be more than okay, it’ll

be great!

Also, I want to thank all of my friends, and two very special friends, Dr. Michele Harvey

and Justina Richards. Michele and Justina, as an only child I could not have been so

lucky to be blessed with two beautiful people who I both consider my sisters. I thank you

both for your support and encouragement, and I thank you for just always being available

to talk and listen, even when I couldn’t always do the same. Throughout the program I

developed many colleagues and friends some of whom include

v

Marie-elene Roberge, Dr. Monica Wang, Dr. Ed Tomlinson, Dr. Boyce Watkins,

Dr. Chris O.L. H. Porter, Janice Molloy, Dr. Hyondong Kim, Aden Heuser, and many,

many others. I thank you for your collegiality and friendship now and in the future.

Lastly, I wanted to thank the organization that allowed me to collect data, although the

name of the organization will be withheld. I also want to thank Shari-Mickey Boggs with

whom I worked very closely to collect data for the focus group portion of my

dissertation. Shari you are great to work with and I look forward to working with you in

the future. I also want to thank Heidi Dugger. Heidi, you are one in a million and the

Management and Human Resources Department at Fisher is fortunate to have you.

In closing, I am thankful for the financial support I received for my dissertation project

from two grants including the Coca-Cola Critical Difference for Women Graduate

Studies Grants for Research on Women, Gender, and Gender Equity (OSU) and the Ohio

State University, Graduate School’s Alumni Grants for Graduate Research and

Scholarship (AGGRS) Fund.

vi

VITA

September 25, 1976………………………… .....Born- USA 1998……………………………………………..B.A. Economics, Spelman College 1998-2000……………………………………….Business Analyst, AT Kearney

Consulting Firm 2000-2001…………………………………….....Senior Corporate Forecasting

Analyst, Delta Airlines 2001-present…………………………………….Graduate Research and Teaching Associate, The Ohio State University

PUBLICATIONS Sutton, K.L. & Noe, R.A. (2004). Family Friendly Programs and Work-Life Integration: More Myth Than Magic. In Kossek, E. E. & Lambert, S. (Eds.) Work And Life Integration: Organizational, Cultural and Psychological Perspectives. Mahwah, N.J.: Lawrence Erlbaum Associates. Referred Conference Publications Klein, H.J., Heuser, A. E., & Sutton, K.L. (April, 2006). “The Dimensions and Levels of Socialization Content. Paper accepted for presentation at the Annual Conference of the Society for Industrial and Organizational Psychology. Dallas, TX. Sutton, K.L. & Dunn-Jensen, L. (August, 2005). “Managing Work-Family Balance in the 21st Century: Do Informal Work Practices Help or Hinder Employees". Organizer, Co-chair and Presenter. Symposium accepted for presentation at the annual meeting of the Academy of Management, Honolulu, HW.

**Symposium nominated for Best Symposium Award, Careers Division** Sutton, K.L. & Noe, R.A. (2004). Work Family Practices: A Pragmatic Perspective: Do We Really Know How These Practices Work? Organizer and co-chair. Symposium accepted for presentation at the annual meeting of the Academy of Management, New Orleans, LA. Wang, C. and Sutton, K.L. (2004). "Nodding Along or Fighting for 'Us': Do Conflict Management Style and Propensity to Initiate Negotiations Influence Group Identification and Effectiveness?" Accepted for presentation at the International Association of Conflict Management, Pittsburgh, PA.

vii

Sutton, K.L., Klein, H., & Barnard, J.,& Noe, R. A. (2003). Distance Learning and Learning Preferences: Does Gender Matter? Presented at a poster session during the annual meeting of the Academy of Management, Seattle, WA. Technical Publications –Organizational Use Only Ellingson, J.E., Reichers, A., Molloy, J. & Sutton, K. (2005). Retaining Female Tenure-Track Assistant Professors. A Descriptive Evaluation of the Faculty Cohort Project Conducted at The Ohio State University. Department of Management and Human Resources. Fisher College of Business, The Ohio State University.

FIELDS OF STUDY Major Field: Labor and Human Relations

viii

TABLES OF CONTENTS

ABSTRACT................................................................................................................. ii DEDICATION............................................................................................................ iii ACKNOWLEDGMENTS ......................................................................................... iv VITA.......................................................................................................................... viii LIST OF TABLES…………………………………………………………………..xi LIST OF FIGURES………………………………………………………………....xv Chapters: 1. INTRODUCTION............................................................................................... 1 Problem Statement.................................................................................................... 11 Contributions of this dissertation............................................................................ 15 2. LITERATURE REVIEW ................................................................................ 18 Social Capital and The Relational Approach to Career Development ................ 27 Network Measures and Characteristics.................................................................. 33 The Importance of Organizational Networks ........................................................ 41 Gender and organizational networks..................................................................... 43 Parental Status and Organizational Networks....................................................... 52 Careers: An Overview of Various Approaches...................................................... 57 Careers and Organizational Networks ................................................................... 61 Careers and Gender.................................................................................................. 67 Careers and Parental Status .................................................................................... 70 Careers and Work-Family Concerns ...................................................................... 74 Job and Family Involvement.................................................................................... 81 Weak Tie Theory....................................................................................................... 84 Boundary Theory...................................................................................................... 88 3. CONCEPTUAL MODEL AND HYPOTHESES DEVELOPMENT .......... 95 Gender Constraint .................................................................................................. 102 Family Involvement Constraint............................................................................. 104

ix

Role Segmentation Constraint ............................................................................... 107 Job Involvement Constraint................................................................................... 110 Perceptions of Career Success ............................................................................... 113 Perceptions of Career Self-Management .............................................................. 116 4. METHOD ........................................................................................................ 119 Focus Group Study ................................................................................................. 120 Pilot Testing............................................................................................................. 127 Field Study............................................................................................................... 130 Survey Response Rate............................................................................................. 140 Measures .................................................................................................................. 145 Plan for Data Analysis ............................................................................................ 175 5. RESULTS ........................................................................................................ 177 Preliminary Analysis .............................................................................................. 177 Data Checks and Cleaning ..................................................................................... 186 Scale Reliability Analysis ....................................................................................... 192 Tests of Hypotheses................................................................................................. 210 6. DISCUSSION .................................................................................................. 281 Overview of Findings.............................................................................................. 281 Theoretical Implications......................................................................................... 309 Practical Implications............................................................................................. 314 Future Research Questions .................................................................................... 319 Study Limitations.................................................................................................... 321 Works Cited............................................................................................................ .330 Appendix A: Focus Group Interview Survey ............................................... .333 Appendix B: Wave 1 E-mail: Invitation to Participate in The Study ....... 346Appendix C, Wave 2 E-mail: Invitation to Participate in The Study ....... 348 Appendix D: Field Survey Wave 1 ........................................................................ 352 Appendix E: Organizational Networks and Careers Survey- Wave 2Error! Bookmark not defined.

x

LIST OF TABLES Table 2.1: Frequently Measured Items Related to Network Ties ....................... 34 Table 2.2: Typical Social Network Measures Used to Describe Entire Networks36 Table 4.1: Non-Response Bias Analysis ................................................................ 144 Table 4.2: Calculation of Role segmentation ‘Actual’ Measure ......................... 157 Table 5.1: Descriptive Statistics for Demographic Variables ............................. 182 Table 5.2: Descriptive Statistics and Correlations on Career Success and Career Management……………………………………………………………………….180 Table 5.3: Family Involvement Scale .................................................................... 193 Table 5.4: Role segmentation Scale A ................................................................... 197 Table 5.5: Role segmentation Scale B ................................................................... 198 Table 5.6: Role segmentation Scale C ................................................................... 199 Table 5.7: Job Involvement Scale ......................................................................... 201 Table 5.8: 3-Factor Solution, Career Management ............................................. 206 Table 5.9: Career Success-Individual Career Satisfaction Scale........................ 208 Table 5.10: Career Success-Peer/Related Career Satisfaction Scale ................. 209 Table 5.11: One-Way ANOVAs between Parental Status and Ego Network Characteristics......................................................................................................... 214 Table 5.12: Regression Results of the Relationship between Parental Status and Network Constraint (Gender) on Network Size, Hypothesis 1a......................... 218 Table 5.13 Regression Results of the Relationship between Parental Status and Network Constraint (Gender) on Network Ties .................................................. 219

Table 5.14 Regression Results of the Relationship between Parental Status and Network Constraint (Gender) on Network Content............................................ 220 xi

Table 5.15: Regression Results of the Relationship between Parental Status and Network Constraint (Family Involvement) on Network Size ............................. 223 Table 5.16 Regression Results of the Relationship between Parental Status and Network Constraint (Family Involvement) on Network Ties ............................. 224 Table 5.17 Regression Results of the Relationship between Parental Status and Network Constraint (Family Involvement) on Network Content ...................... 225 Table 5.18 Regression Results of the Relationship between Parental Status and Role Segmentation (Perceptual) on Network Size, ....................................................... 230 Table 5.19 Regression Results of the Relationship between Parental Status and Role Segmentation (Perceptual) on Network Ties........................................................ 231 Table 5.20 Regression Results of the Relationship between Parental Status and Role Segmentation (Perceptual) on Network Content ................................................. 232 Table 5.21 Regression Results of the Relationship between Parental Status and Role Segmentation (Actual) on Network Size ............................................................... 233 Table 5.22 Regression Results of the Relationship between Parental Status and Role Segmentation (Actual) on Network Ties ............................................................... 234 Table 5.23 Regression Results of the Relationship between Parental Status and Role Segmentation (Actual) on Network Content ........................................................ 235 Table 5.24 Regression Results of the Relationship between Parental Status and Network Constraint (Actual) on Network Size .................................................... 238 Table 5.25 Regression Results of the Relationship between Parental Status and Network Constraint (Actual) on Network Ties .................................................... 239 Table 5.26 Regression Results of the Relationship between Parental Status and Network Constraint (Actual) on Network Content ............................................. 240 Table 5.27 Regression Results of the Relationship of Network Size on Salary . 243 Table 5.28 Regression Results of the Relationship of Network Size on Salary Growth................................................................................................................................... 244 xii

Table 5.29 Regression Results of the Relationship of Network Size on Promotions................................................................................................................................... 245 Table 5.30 Regression Results of the Relationship of Network Size on Individual Career Satisfaction.................................................................................................. 246 Table 5.31 Regression Results of the Relationship of Network Size on Peer-Related Career Satisfaction.................................................................................................. 247 Table 5.32 Regression Results of the Relationship of Network Ties on Salary. 250 Table 5.33 Regression Results of the Relationship of Network Ties on Salary. 251 Table 5.34 Regression Results of the Relationship of Network Ties on Promotions................................................................................................................................... 252 Table 5.35 Regression Results of the Relationship of Network Ties on Individual Career Satisfaction.................................................................................................. 253 Table 5.36 Regression Results of the Relationship of Network Ties on Peer-Related Career Satisfaction.................................................................................................. 254 Table 5.37 Regression Results of the Relationship of Network Content on Salary257 Table 5.38 Regression Results of the Relationship of Network Content on Salary Growth ..................................................................................................................... 258 Table 5.39 Regression Results of the Relationship of Network Content on Promotions............................................................................................................... 259 Table 5.40 Regression Results of the Relationship of Network Content on Individual Career Satisfaction.................................................................................................. 260 Table 5.41 Regression Results of the Relationship of Network Content on Peer-Related Career Satisfaction ................................................................................... 261 Table 5.42 Regression Results of the Relationship of Network Size on Career Planning ................................................................................................................... 264 Table 5.43 Regression Results of the Relationship of Network Size on Career 265 xiii

Table 5.44 Regression Results of the Relationship of Network Size on Career Mobility Preparedness............................................................................................ 266 Table 5.45: Regression Results of the Relationship of Network Ties on Career Planning ................................................................................................................... 268 Table 5.46: Regression Results of The Relationship of Network Ties on Career Tactics ...................................................................................................................... 269 Table 5.47 Regression Results of the Relationship of Network Ties on Career Mobility Preparedness............................................................................................ 270 Table 5.48 Regression Results of the Relationship of Network Content on Career Planning ................................................................................................................... 272 Table 5.49 Regression Results of the Relationship of Network Content on Career Tactics ...................................................................................................................... 273 Table 5.50 Regression Results of the Relationship of Network Content on Career Mobility Preparedness............................................................................................ 274 Table 5.51: Summary of Study Hypotheses and Findings .................................. 275 Table 6.1: Summary Results of Network Size on Career Objective Career Success Indicators ................................................................................................................. 292 Table 6.2: Summary Results of Network Ties on Career Objective Career Success Indicators ................................................................................................................. 297 Table 6.3: Summary Results of Network Content on Career Objective Career Success Indicators ................................................................................................... 300 Table 6.4: Summary Results of Network Size on Career Management Indicators302 Table 6.5: Summary Results of Network Ties on Career Management Indicators304 Table 6.6: Summary Results of Network Content on Career Management Indicators................................................................................................................................... 307

xiv

LIST OF FIGURES

Figure 1: Conceptual Model of The Relationship Between Family Status Differences, Network Constraints, Job Involvement, Career Success, and Career Self-Management .................................................................................................... 100

xv

CHAPTER 1

INTRODUCTION

One of the recent pressures employees face within organizations includes the

responsibility of managing their own careers. Hall (2002) describes careers as “the

individually perceived sequence of attitudes and behaviors associated with work-related

experiences and activities over the span of the person’s life” (p.4). Recently, there have

been many changes in the concept of careers. Thus includes the trend that more

individuals are engaging in career mobility or frequent job changes (Hall, 2002).

Frequent job changes are often motivated by individuals “taking advantage of better job

opportunities and searching for a better match between job characteristics and personal

interests” (Hall, 2002). Also, frequent job changes are characteristic of what Hall (1976,

1996, 2002) describes as the “protean career”. The protean career is:

A process which the person, not the organization is managing… a process characterized by frequent change and self-invention, autonomy, and self-direction…The protean person’s own personal choices and search for self-fulfillment are the unifying or integrating elements in their life.. The criterion of success is internal (psychological success). (pp 36, 38, 43)

One outcome of the “protean career” is decreased job stability and increased job

mobility, both within and between organizations (Hall, 2002).

1

This mobility across jobs is also characteristic of the “permeability” of organizational

boundaries, that is, intra-organizational mobility (“movement between levels, projects,

products, functions, and locations”) and inter-organizational mobility has become more

acceptable, frequent and essential (Hall, 2002). To increase mobility, an individual must

have access to information regarding job and career opportunities, even when that

individual chooses to remain with the same organization. Therefore, researchers must

consider how individuals gain information about job and career opportunities, which is

useful for them moving either within or across organizational boundaries.

Drawing from a social network framework, researchers can study networks within

organizations to understand how information is exchanged between individuals (or

actors) within the network (Higgins & Kram, 2001; Wasserman and Faust, 1994).

Although understanding the role of organizational networks as it relates to acquisition of

job and career-related information is important, researchers have often overlooked the

importance of (informal) networks on the career mobility of their employees (Podolny &

Baron, 1997). This is an important matter to consider, as now the burden (or

responsibility) for an employee’s career development has shifted from the organization to

the individual (Forret & Dougherty, 2004).

A few distinctions should be made clear about the terminology that is used in this

dissertation. First, the phrase social network refers to the set of actors (individuals) and

the ties among them (Wasserman and Faust, 1994). It should be noted that a single social

networking theory does not exist; rather a social networking perspective is used to

understand mostly patterns of behavior and interaction (Brass, 1995). Networking is a

specific behavior, that is, it’s a behavior where an individuals attempt to develop and

2

maintain relationships with others who have the potential to assist them in their work or

career (Forret & Dougherty, 2004). Network capital is a form of social capital that

makes resources available through interpersonal ties (Wellman & Frank, 2001).

Organizational networks are the set of nodes and ties representing some relationship or

the lack of a relationship between nodes, where nodes are actors (individuals, work units,

or organizations) (Brass et al, 2004). Finally, (individual) social capital can be defined

as investments into social relations which allow individuals to gain access to embedded

resources to enhance expected returns of instrumental or expressive actions; that is, social

capital is “how individuals access and use resources embedded in social networks to gain

returns in instrumental actions (e.g. finding better jobs)… individuals are said to invest in

social capital in hopes of some return to them” (Lin 2001, p.19).

The social networking perspective provides researchers with a framework to

understand both the (networking) behavior of individuals (actors) and how these

individuals (actors) exchange information within organizational networks. This

dissertation will use the social networking perspective to understand if specific network

characteristics, that is, network size, ties, and content influence career outcomes, career

success and career satisfaction. Also, the social networking perspective is used to

understand if there are specific moderators that cause differences in the network

characteristics between working adults without parental responsibility, and working

adults with parental responsibility.

3

The career development literature discusses the importance of networking

behaviors, and the importance of organizational networks for employees’ careers. The

careers literature investigates the importance of the relationships that one can develop

through their participation in organizational networks. For example, in his 1996 book,

The Career is Dead-Long Live the Career. A Relational Approach to Careers, Hall

suggested that the primary resources for career development, which refers to how people

“grow” in their careers, include relationships with other people, and work challenges.

Although limited, empirical support confirms the notion that relationships at work

are positively related to career outcomes. For example, Higgins and Thomas (2001)

found that as the number of developmental (networking) relationships increased, the

individuals were more likely to remain with the organization and to experience higher

work satisfaction. Also, career advancement was positively related to multiple

hierarchical relationships, rather than just the development of one primary relationship

(Higgins and Thomas, 2001). That is, career advancement was related to the

development of multiple relationships in comparison to the development of a single,

primary relationship such as the one developed with a mentor. Seibert et al (2001) found

interpersonal relationships led to various career benefits including access to information,

access to material and financial resources, and visibility and sponsorship within the

organization. Bozionelos (2003) found network resources (e.g. relationship ties who’s

function is to promote the career interests of others) were associated with intra-

organizational career success, over and above human capital, demographics and

mentoring received. Podolny & Baron (1997) found results similar to Burt’s (1992)

finding. That is, informal social ties impacted advancement within organizations. Ties are

4

defined as a relationship between two sets of nodes or actors within a network (Brass,

1995). Specifically, large information ties that lack indirect ties (or relationships)

promote upward mobility within organizations.

The relational view of careers links to the social network perspective of

relationships. That is, a social networking framework can be used to study the influence

of relationships (and how information is exchanged within these relationships) on various

career outcomes. Specifically, by analyzing patterns of independent relationships

researchers can gain an understanding of how information is exchanged across multiple

relationships (Wasserman & Galaskiewicz, 1994). This dissertation will use a

networking framework to explore how individual differences (i.e. parental status and

gender) impact the relationships people develop within their organizational networks.

Researchers have been successful in using networks to study individual

satisfaction, performance, job turnover, group structure and performance, and

organizational innovation and survival (Brass et al., 2004). Networks can also be used to

study how individuals leverage existing relationships while searching for new jobs.

Specifically, Granovetter’s (1973) weak tie argument suggests that people find jobs

through their acquaintances or weak ties. This phenomena occurs because an individual’s

weak ties or associates are not likely to know each other. Individuals benefit from weak

ties because they provide an individual with diverse and nonredundant information

(Granovetter, 1973; Brass et al., 2004). In addition to finding new jobs, organizational

networks also help individuals mobilize their career within their current organizations.

Described as social capital, that is, “who you know”, most managers owe a significant

5

amount of their career success to their connections with others (this is especially true at

mid-level management) (Brass et al., 2004).

Burt’s (1992) structural hole argument suggests that individuals must not only be

concerned with developing weak ties (Granovetter, 1973), but they must be concerned

about the diversity of their contacts. Specifically, Burt (1992) argues that people should

have a network of abundant structural holes. Structural holes exist when there is an

absence of a link between two actors who are both linked to a common actor (Burt, 1992;

Brass et al., 2004). As a result of this structural hole, individuals are less likely to have

access to the same kind of information. In simpler terms, an individual that is interested

in maximizing the amount of job or career-related information should develop a large

number of weak ties, and they should make certain that the weak ties are not in contact

with each other. This can be achieved by an individual joining multiple organizations in

which the there is little overlap between the members across organizations. Consistent

with this notion, Podolny and Baron (1997) found that individuals whom have a large,

sparse informal network with many structural holes experienced enhanced career

mobility (i.e. higher amount of movement between jobs throughout their career).

Given the importance placed on the role of networks, (where networks are

operationalized as the development of multiple, nonredundant relationships) in

mobilizing an employee’s career, it also makes sense to identify specific network

properties that are likely to be related to career outcomes. The network properties of

interest to this study include network size, network ties, and network topics of

conversation. Network size refers to the number of individuals within a network (Brass,

1995). Network tie, is defined by the relationship between a specific tie and the ego in the

6

network (e.g. friendship tie, coworker tie, family tie). Finally, network topics of

conversation, refers to the topics an ego discusses with their network members. This

study will operationalize network topics of conversation by asking individuals to identify

a set of conversation topics discussed with each member within their network and ask

them to identify which of those conversation topics they discuss most frequently from the

set of conversation topics they have identified.

This study suggests that four moderators may impact the three network properties

included in this study (network size, network ties, and network topics of conversation).

The four moderators this dissertation proposes will influence network properties include

gender, parenthood, family involvement, job involvement, and the extent to which

individuals want their home and work life integrated or segmented. Next, a brief

description of each of the factors thought to impact the development of multiple,

nonredundant relationships are provided. Chapter 3 provides a detailed discussion of how

each of these four factors will influence each of the three network properties.

Of the four factors mentioned, research has empirically tested the impact of

gender on an individual’s network and the advancement of one’s careers (e.g. Ibarra and

Smith-Lovin, 1997). A case can be made that men have better networks, that is broader

networks (i.e. weak ties) and more powerful contacts within their networks, hence

resulting in men having better access to job and career related information (e.g. Ragins &

Sundstrom, 1989) than women. This provides evidence that an individual’s gender may

impact at least two network characteristics network size and ties, which results in men

having better access to job and career-related information. However, the other variables

included in this dissertation research, that is parental status, family involvement, and role

7

segmentation, are not gender specific, and are likely additional variables that impose

constraints on the networking behaviors of employees. Specifically, as traditional family

roles evolve and men become more involved with parental responsibilities, (e.g. dual-

earner families, an influx of single dads raising children), it is no longer sufficient to

primarily focus on gender differences within networks. Instead, it is important to consider

additional individual differences (e.g. parental status, family involvement) that impact the

networks within organizations. Thus this dissertation suggests that the work-family,

careers, and networking fields must evolve from merely focusing on gender differences

within networks to identifying additional individual characteristics (e.g. family

involvement, job involvement) that contributes to differences in the networking

experiences of employees.

Previous research findings regarding gender are consistent, that is, gender matters

in terms of explaining differences within organizational networks. Now the question

becomes, in addition to gender, what else matters? It is likely that, parental status, family

involvement, and role segmentation have an effect on networks. However, research has

not yet investigated how those factors influence an individual’s network and its

relationship to career management and career success.

A number of studies have investigated the relationship between gender,

parenthood, networks, and career outcomes. For example Valcour and Tolbert (2003)

found women’s intra-organizational career mobility was negatively related to the number

of children. Also, Smith-Lovin and McPhearson (1993) proposed that the networks of

unmarried career women and men are mostly similar, however, the birth of a child tends

to produce dramatic changes in the gender and kin composition of career women’s

8

networks; changes which reduce the “career-related value” of the network (e.g. career

mobility). For example, women, after the birth of a child, have a tendency to use their

networks to gain and share information about family and households (Smith-Lovin and

McPhearson, 1993). In comparison, men seem to use their networks to exchange

information related to career, money and recreation pursuits (Smith-Lovin and

McPhearson, 1993). To date, this notion of changes within a woman’s network after the

birth of a child, has not been empirically tested.

Family involvement refers to the psychological involvement with and importance

of the family to a person (Parasuraman, Purohit, Godshalk & Beutell, 1996). Family

involvement is an important variable to consider because when an individual is highly

involved with the family, they are likely to devote more time to their family demands (in

comparison to their work demands). Most of the work-family research measures the

relationship between family involvement and work-family conflict and a negative

relationship is usually found between the two variables (e.g. Parasuraman et, al., 1996).

That is, the more an individual is involved with their family, the more likely they are to

experience work-family conflict (e.g. Eby et al., 2005). Work-family conflict occurs

when involvement in a work-related activity, interferes with participation in a competing

family activity (Greenhaus and Powell, 2002). Family involvement and parental status

have been tested separately in previous empirical studies. However, previous studies have

not investigated how both parental status and family involvement impacts an individual’s

network. This study will make a unique contribution to the field by investigating the

interaction between parental status and family involvement, and how that interaction

impacts network size, ties, and content.

9

Similar to family involvement, job involvement also refers the psychological

involvement with and importance of the job to the individual (Lodahl & Kejner, 1965).

Job involvement is an important variable to consider as the more an individual is

involved in their job, the less time and involvement that individual will have with things

that are not specific to their job (e.g. family). While research has investigated the

relationship between job involvement and work-family conflict, little is know about how

job involvement many influence the development of multiple, nonredundant

relationships.

Finally, the extent to which an individual wants their family and work life

segmented or integrated is drawn from Ashforth, Kreiner, & Fugate ‘s boundary theory

(2000). This study will argue that individuals that segment their role and family roles will

have organizational networks with different characteristics compared to individuals that

integrate their work and family roles. For example, those individuals that segment their

work and family roles would be less likely to prefer to bring their families to social

networking activities such as company picnics. In comparison, individuals that choose to

integrate their work and family roles, would seek out opportunities to bring their families

to work-sponsored events. It is important to note these differences, as information about

jobs and careers can be exchanged in multiple informal environments including social

gatherings where both employees and their families may be present.

For example in a recent empirical study, Scott (2001) describes how family, specifically

children and spouses, act as social capital. Taken from an in-depth semi structure

interview, one of the respondents reports:

10

“There are summer conferences that bring together state legislative officials, so at those conferences, you can get to meet the state senator as well as his wife and three kids. That has been helpful in terms of a network at the state level, where you can’t get to all these state capitals, and not being able to go as often as I used to. At least I get to see people once a year, and they know my husband. I don’t have children- I bring one of my nieces. So there’s a social element to it, very family oriented” (pp 23).

This remark “is an indication of the significance of family status, particularly parenthood,

for establishing work connections (relationships)” (Scott, 2001, pp 23).

In short, integrating work and family, specifically children and spouses seems to make a

difference in helping an employee further develop their work relationships. Therefore, an

employee whom prefers to segment their work and family lives, may be not fully

utilizing their family ties as social capital to help them establish relationships at work.

Individuals that do not leverage their family in the development of social ties may

experience a reduction in the overall size of their network, which according to

Granovetter’s strength of weak ties argument (1973), leads to individuals attaining less

job and career related information. This study contributes to the field, by investigating if

role segmentation will interact with parental status and contribute to differences in

network size, ties, and content.

Problem Statement

This study assumes one of the key functions of organizational networks is for

employees to leverage these networks to share information which will be helpful for

career mobility, that is, acquiring information related to job openings and career

opportunities within the organization. The purpose of this study is to understand if there

is evidence to support the notion that parental responsibility results in several key

changes within an employee’s organizational network.

11

Three major research questions will be addressed in this study include: First, how do

networks differ after the birth of a child for males vs females? Secondly, how do

networks differ between working adults with and without children? Thirdly, what

constraints contribute to differences in those networks?

This study contributes to the work-family literature in the following ways.

First, very little research has investigated the impact of parenthood on the networking

behaviors of employees. Most networking literature has looked at various factors

including gender and it’s impact on organizational networks (e.g. Ibarra and Smith-

Lovin, 1997). Also, most previous work-family research has focused on the relationship

between work-family concerns and career satisfaction (e.g. Sturges and Guest, 2004;

Almer et al., 2004), work-life and career issues specific to women (e.g. Harris, 2004; Ng

& Posh, 2004), the issues of role conflict among dual-career couples (e.g. Hall, 2002;

Elloy & Smith, 2003; Burke, 2000), and concerns related to protean careers and dual

career relationships (e.g. Hall, 2002). Some studies have also addressed the

relationships between intra and inter- organizational mobility and work-family concerns

(e.g. Valcour and Tolbert, 2003) and have focused on work-family challenges from the

life course perspective (e.g. Moen and Sweet, 2004). The life-course perspective views an

individual’s career as a series of stages characterized by the changing patterns of

developmental tasks, career concerns, activities, values, and needs which emerge as the

individual passes through various age ranges (Hall, 2002). Work-family researchers have

not yet addressed the various ways that developmental networks influence career choice

and decision making of parents (Hall, 2002). A recent search in Business Source Premier

yielded only 7 conference abstracts and zero research articles when the key words

12

“networks, career and work-family” were entered as search terms. Sullivan (1999)

adequately summarized the potential contribution of networking theory for studying

careers:

“social networking theory may provide an effective framework by which to conduct future research on the careers… networking theory may be useful in understanding the interactions of short-term and long-term careerists… and researchers should understand “the effect of gender, race, age and personal characteristics on the development of large, non-redundant networks” (pp 469).

This study investigates how parental responsibilities impacts employee

organizational networks, by comparing the network characteristics of working parents to

the experiences of working adults without parental responsibility. This study suggests

that parental responsibilities will impact three aspects of an employee’s organizational

network including, the size of their networks, the members (i.e. ties) within their

organizational networks, and what employees discuss amongst members of their

organizational networks (i.e. network content). Previous research suggests that the size

of one’s network will change after the birth of a child, that is, an individual’s network

will become smaller (e.g. Smith-Lovin & McPhearson, 1993). This change is thought to

be related to employee gender. That is, women will experience a greater change in their

network size than men. Therefore, this study expects to find that female parents have

smaller networks then male parents. According to Smith-Lovin & McPhearson (1993)

women have more childcare responsibilities, therefore resulting in smaller networks.

However, while gender may play a role in a change in network size, this study also

suggests that family involvement may impact the change in networks. Thus, in addition

to gender, the employee with the highest amount of family involvement may experience

the largest change in their organizational network.

13

In addition to a change in size of organizational network, this study proposes that

the membership of an employee’s network will be different for individuals that are

parents. Specifically, the members of the employee’s network with who they are in

frequent contact will be different from employees with no parental responsibilities.

Usually frequency of contact is a function of social closeness (intimate, active, latent),

spatial closeness (same neighborhood, metropolitan area), and kinship closeness (that is,

immediate vs extended kin) (Wellman, 1992). A network framework can be used to

identify the members of the network with whom the focal individual (ego) has active

relationships (e.g. kin ties, friend ties, work ties). Specifically, this study suggests,

parental responsibilities will increase the frequency of contact with family and friends for

working parents. Most people have contact at least once a week with their active network

members (Wellman, 1992). This dissertation will test the hypothesis that working

parents will have a higher proportion of kin ties in their network, than working adults

without parental responsibility.

In addition, this study also will investigate how network content differ across

parental status. Specifically this study suggests that working adults with parental

responsibility are likely to have a higher proportion of non-work/kin network content.

That is, working adults without parental responsibility will be more likely to discuss

topics related to children and other non-work issues among members of their networks.

This study assumes that people will discuss important matters among people within their

networks and we operationalize information exchanged by assessing the conversation

topics discuss among the ego and members within their network.

14

Finally, this study suggests that parental responsibilities will impose a large

constraint on employees who desire to segment their work and family lives. Boundary

theory suggests that employees differ in the degree to which they desire to integrate their

home and family life. Individuals whom segment their work and family lives want a clear

temporal and physical boundary established between their work and family roles. For

examples, these individuals prefer not do their work from home, most of their work will

be completed at the office. In addition, these individuals are reluctant to integrate their

family members in work activities, including work-related social activities (e.g. baseball

outings) that may occur during or after work hours. As a result of clearly establishing

boundaries between their work and home life, individuals who prefer to segment their

work and home lives are likely to have less time to dedicate to the maintenance of their

organizational networks. As a result, this dissertation expects role segmentation to

explain differences in the organizational networks of working adults with children and

working adults without children. For those individuals whom segment their work and

family life, it is expected that there will be differences in their network size, membership

and content exchanged within their network. This study makes a contribution using role

theory to understand the differences in network characteristics between working parents

with children and working parents without children.

Contributions of this dissertation

This study will make the following contributions to the field of human resources

management and organizational behavior. First, this study will utilize social networking

theory (e.g. weak ties argument) and social networking methods (e.g. ego (individual)

network size, ties, content) and apply it to research on careers, and work-family concerns

15

(i.e. working parents). Second, by considering the impact of family involvement and role

segmentation, in addition to gender and the onset of parenthood, this dissertation

examines additional factors that influence networks and the advancement of one’s career.

Third, this study contributes to the literature on careers, by taking a relational perspective,

to understand how an individual’s network influences their access to information and

subsequent perceptions of career success and attitudes toward network resources. In

addition, this study will also contribute to the research on networks and careers. That is,

most research on the relationship between networks and careers, investigates outcomes

such as career mobility or career advancement (e.g. Podolny & Baron, 1997; Granovetter,

1973; Burt, 1992). This study will use attitudinal measures (e.g. career satisfaction) to

understand the extent to which parents are constrained by various networking properties

(e.g. size, ties, and content) in utilizing their networks to gain career and job related

information. Finally, this research addresses a need identified by Higgins and Kram

(2001) to better understand individual’s developmental relationships.

Of note, social network analysis can be studied at various levels (Brass, 1995).

Two examples include network research at the individual level (i.e. ego networks) or

studying the entire systems of networks. This dissertation will study networks at the ego

or individual level. Studying network characteristics at the ego-level allows researches to

gather an individual’s perception of their network (and specific network characteristics,

inclduing network size) (Brass, 1995). Specifically this dissertation studies the relational

aspects of networks at the ego (i.e. individual) level. The relational aspect of social

networks is focused on the social characteristics (e.g. gender) of these network members

and of their ties (Wellman, 1992). The relational aspect of social networks can be also

16

used to study characteristics of the relationships between ties in a network (e.g. frequency

of contact, intimacy).

Chapter 2 presents the literature review that provides theoretical and empirical

support for the hypotheses tested in this dissertation. Chapter 3 introduces the theoretical

model used in the study and concludes by presenting hypotheses based on the model. The

methods used in this study are discussed in Chapter 4. Chapter 5 discusses the results of

the study. Finally, the practical and research implications, theoretical contributions, and

directions for future research resulting from this study will be discussed in Chapter 6.

17

CHAPTER 2

LITERATURE REVIEW

The relational approach to careers suggest that people will be responsible for their

own careers and can use various relationships to share information and acquire

information from others about various job and career opportunities. Also, research related

to the relational approach of careers helps provide insight into an individual’s entire

social environment (Hall, 1996). One common example of the study of relationships and

career development, is that of the mentoring relationship. Mentoring serves two

functions including providing support for career development (which contributes to the

mentee’s advancement in the organization; and “psychosocial support” which contributes

to the mentee’s personal and professional growth (Kidd et al., 2003). Mentoring has been

linked to many career outcomes including, enhanced career development, career

progress, higher rates of promotion, and career satisfaction (Higgins and Kram, 2001).

The literature on mentoring has generally described it as a single relationship, or

focused on primary mentoring relationships, where primary mentoring relationships are

those in which the mentor holds a higher-level or more senior level position (compared to

their mentee) (Higgins & Thomas, 2001). That is, the mentor-mentee relationship is

characterized by organizational members of unequal status, and it does not generally

focus on peer-peer mentoring relationships (Kram, 1988). Thus previous research has

restricted the study of the mentoring relationship, to a single relationship with a more 18

senior individual in the mentee’s organization (Higgins and Thomas, 2001). The

nature of these mentoring relationships is usually dyadic, where only the relationship

between the primary mentor and mentee are studied. Research typically does not study

concurrent, mentoring relationships.

More recently, researchers (e.g. Higgins and Kram, 2001; Kram, 1988) have

begun to investigate relationships beyond the single relationships or dyadic relationships

that were prevalent in the mentoring literature. Thus current research acknowledges that

individuals receive mentoring types of support from a set or “constellation” of

developmental relationships including peers, subordinates, friends, families and bosses,

that is, individuals receive support from more than one person (Kram, 1988). As

previously suggested, the relational model of career development is focused on

understanding how an employee’s interaction with people (both within and outside of the

organization) helps influence their career development. The relational perspective of

mentoring is also described by some researchers as a developmental constellation

(Higgins and Thomas, 2001; Kram, 1988). The developmental constellation is “the set of

relationships an individual has with people who take an active interest in the

advancement of an individual’s career by assisting with his/her personal and professional

development” (Higgins and Thomas, 2001). Constellations are operationalized as a set of

“multiple concurrent relationships”, that is, relationships an individual identifies at a

single point in time as having been important to their career development (Higgins and

Thomas, 2001). Previous research that has investigated multiple relationships generally

considers a series of relationships over time, instead of a set of relationships at one

specific time period (Higgins and Thomas, 2001). Therefore, this study will investigate a

19

set of relationships at one specific time, that is, relationships that are included in the

individual (or ego’s) network.

Higgins (1998) demonstrated that relationships impact an individual’s career by

influencing the extent to which one seeks information from nonredundant sources (hence

providing a variety of different information) and the extent to which the individual’s

sources are heterogeneous and provide exposure to different points of view. In addition,

the most effective organizations take a relational approach of the development of their

employees. That is, “the employer will provide opportunities and flexibility and

resources, particularly people resources, to enable the employee to develop identity and

adaptability and thus be in charge of their own career” (pp 40) (Hall, 2002). If

organizations are moving towards a relational approach of the development of their

employee’s careers, it makes sense that researchers understand more about the role of

relational networks in individual career experiences and their impact at the organizational

level (Hall, 2002). In fact, one can argue that the major source of power of networks lies

precisely in the information exchanges and collective learning that they can promote”

(Hall, 2002).

In a recent study, Parker, Arthur, & Inkson (2004) also demonstrate the value of

relationships in supporting an individual’s career. Parker et al (2004) draw from several

approaches (e.g. the relational approach to careers, the subjective career (i.e. an

individual’s perspective of their career) to describe career communities. Of note, the

concept of career communities differs slightly from the relational view of careers. When

career communities are studied, researchers investigate the career support that members

find within identifiable communities (Parker et al., 2004,). In comparison, the relational

20

view of careers investigates the career support individuals receive through their

interpersonal relationships. Career communities include three factors (knowing-why,

knowing-how, knowing-whom) through which individuals draw career support (Parker et

al., 2004). The three factors, knowing-why, knowing-how, and knowing whom describe

the investments that individuals make to their career over time. For example, knowing-

why describes the values (and various motivational factors) that impact an individual’s

career choice, especially their career adaptability and commitment (Parker et al., 2004).

Knowing-why is also related to an individual’s perception of non-work issues that impact

an individual’s motivation and values about their careers including personal interests and

changes in family status. Knowing-how is related to an individual’s career-related

expertise or skills, that is, those skills that are applied to an individual’s work. Of note, in

the traditional view of careers, an individual’s knowing-how skills were considered one

of the most important predictors of career success (Parker, et al., 2004). Finally knowing-

whom describes the relationships that support an individual’s career. This aspect of the

career communities, knowing whom, is most directly related to the relational aspect of

careers. Taken together, the knowing-why, knowing-how, and knowing-whom comprise

an individual’s career community and all three aspects represent investments an

individual makes in their career in hopes of achieving a successful career.

In the most recent Parket et al (2004) paper, the ideas of the career communities

were tested empirically. In this study, Parker et al (2004) selected three organizations in

which they expected to find a specific type of career community. That is, the authors

expected to be able to distinguish between a knowing why community, a knowing how

community and a knowing whom community. The results of the study suggest that the

21

three community types do not emerge as three pure types of communities (Parket et al.,

2004). Rather, the career communities described in each organization seems to suggest a

hybrid perspective of career communities. The study suggests that an investment into all

three areas of an individual’s career will be important, across different organizational

settings. In addition, the investment an individual makes into each of the three career

communities will increase their perception of perceived subjective career success.

Organizational networks

Networking, as defined by Forret and Dougherty (2004) is one’s ability to

develop and maintain relationships with others who have the potential to assist them in

their work or career. That is, networking is concerned with the specific behaviors an

individuals uses to obtain information. Further networking behaviors are developmental

relationships used to help one improve both their personal life and their professional

lives. Networking also plays a key role in regulating access to jobs, providing mentoring

and sponsorship, channeling the flow of information and increasing the likelihood and

speed of promotions (Ibarra & Smith-Lovin, 1997).

Previous research has found that networking behavior was more beneficial for the

career progress of males in comparison to females (Forret and Dougherty, 2004). For

example, increased internal visibility was related to the number of promotions and total

compensation for men, but not women. Forret and Dougherty (2004) found networking

behaviors were not as advantageous for women as for men. Men tended to engage more

frequently in networking and socializing activities. They suggest that men may engage in

networking behavior more frequently than women, as a result of men having more hours

to allocate to networking; specifically participating in networking events after work

22

hours. The researchers posit that women may have had less time to participate in after-

hours networking activities, as a result of their childcare responsibilities, which require

them to go directly home after work. Research needs to be conducted to understand why

networking behaviors are not as advantageous for men as they are for women. Further,

there has not been a comprehensive study on networking and career mobility for women

(Ibarra & Smith-Lovin, 1997).

Networking enables employees to effectively manage their career because as

suggested by Burt (1992), networks provide employees with at least two advantages.

First, employees whom are embedded in networks have access to information and people.

In addition, networks offer employees “just-in-time” information. Stated differently, Burt

(1992) suggests,

“Timing is a significant feature of information received by networks. Beyond making sure that you are well informed, personal contacts can make sure you are one of the people that is informed early. Therefore, personal contacts get your name mentioned at the right time in the right place so that opportunities are presented to you” (p 14).

Networking or informal interactions remain important because this is often

described as one strategy women can use to break through the glass ceiling (Forret &

Dougherty, 2004). However, research suggests that women are often challenged by

exclusion from or limited access to informal dialogues which occur within organizational

networks. Research also indicates that the “absence of women from informal networks

and the limitation of publicity of job openings may lead to restriction primarily to men of

information about high-level openings in organizations” (Ragins and Sundstrom, 1989).

Previous research also suggests that organizations may inadvertently or deliberately

23

exclude women from powerful positions through recruitment (Ragins and Sundstrom,

1989 and Plake et al., 1987).

Higgins and Kram (2001) also discuss the importance of networking on the career

mobility of employees. The researchers describe the developmental network as the set of

people an individual identifies as taking an active interest in and action to advancing their

career by providing developmental assistance (Higgins and Kram, 2001). Within a

developmental network, at least two types of support are meaningful, (1) career support

(such as exposure and visibility, sponsorship and protection), and, (2) psychosocial

support (e.g. friendship, counseling, acceptance, confirmation and sharing beyond work)

(Higgins and Kram, 2001). The developmental network differs from other organizational

networks (e.g friendship or social networks) in that it does not consists of all the members

of an individual’s networks, rather it consists of the people whom an individual identifies

at a particular point in time as being important to his/her career development.

In the article, Higgins and Kram (2001) describe a typology of 4 developmental

networks, entrepreneurial, opportunistic, traditional, and receptive. The two

developmental networks that seemed to be most relevant to this study are the

entrepreneurial and opportunistic developmental networks. The entrepreneurial network

was described as one where individuals have the “best of both worlds”. That is,

individuals have a network with strong ties and members in their networks that provide

access to a wide array of information. Strong ties are defined as relationships between an

individual and their closest friends and family members, and the individuals usually all

know each other or are somehow connected; this makes for a highly dense network

(Granovetter, 1973). The opportunistic networks differ from the entrepreneurial network,

24

in terms of the strength and breathe of ties that exist within an individual’s network. In

the entrepreneurial network, individuals have multiple ties or relationships with

individuals and very strong (or intimate) ties. In comparison, the strength of the ties or

relationships developed within the opportunistic network, tend to be weak. That is, within

the opportunistic network, an individual must cultivate, that is, communicate frequently

and actively seek help from members within their networks. If they do not, then the

individual may only receive help from members in their network when it is offered.

Specifically, they will not utilize their networks effectively to gain information, because

they are not willing to invest the time in maintaining these relationships. As a result, the

ties or relationships within an entrepreneurial network are likely to be strong. That is,

individuals can depend on relationships in the entrepreneurial network for specific

guidance and information. The ties within the entrepreneurial network are strong as a

result of the ego within that network making certain to have frequent contact with

individuals within that network that can help them.

Working parents are likely unable to have frequent contact with individuals

within their network, due to their parental responsibilities and subsequent time

constraints. Hence, this study suggests that working adults with no children are likely to

have more frequent contact with members within their networks, which will lead to them

have better access to job and career related information. In comparison, working adults

with parental responsibility may be more likely to have less frequent contact with

members in their networks, due to their responsibilities at home.

25

Specifically, they may be more likely to have less frequent contact with individuals that

would be able to provide them with career-related information. Therefore, working

parents may be more likely to have an opportunistic network.

Although the entrepreneurial or the opportunistic networks are not directly related

to this dissertation, they are worth mentioning for at least two reasons. The opportunistic

developmental network described by Higgins and Kram (2001) can lead one to conclude

that working parents are not able to use their networks to gain information because they

do not cultivate the relationships within their networks particularly after the birth of a

child. In comparison, working individuals without children may have a better chance of

developing an entrepreneurial developmental network. In comparison to parents, whom

have less time to cultivate relationships in their network, working individuals without

children could have more time to develop stronger relationships within their networks.

Also the networks of working adults without parental responsibility may be larger, and

individuals within their network will know them well enough (as a result of more

frequent contact) to speak on their behalf (Higgins and Kram, 2001), when they are

looking for new job or career opportunities. Consistent with this view, Higgins and Kram

(2001) propose that individuals with entrepreneurial developmental networks are more

likely to experience change in their careers than individuals who have opportunistic

networks. The entrepreneurial networks tend to be larger (i.e. more ties) and individuals

tend to have more frequent contact with the members in their networks. Therefore,

consistent with Burt’s (1992) and Granovetter’s (1973) arguments, entrepreneurial

networks will lead to larger networks where larger amounts of nonredundant information

will be shared between the ego and the members of their network.

26

Social Capital and The Relational Approach to Career Development.

Several different components of networks have been studied from a social capital

perspective. Similar to the relational perspective of career development, social capital is

described as the investment in social relations with expected return (Lin, 2001). Social

capital is suggested to provide a return on relationships for the following reasons: (1)

social capital facilitates the flow of information, specifically, social ties located in certain

strategic locations and/or hierarchical positions can provide an individual with useful

information about opportunities and choices which otherwise may not available, (2)

social capital often exerts influence on the agents (e.g. recruiters or supervisors of the

organization) who play a critical role in decisions (e.g. hiring or promotion) involving the

actor, (3) perceptually, an individual’s social capital can be viewed as their social

credentials or an indication of whom supports the individual within (and across)

organizations, that is, an indicator of their “network”, and, (4) social capital reinforces

identity and recognition, that is, it makes a member feel like they are part of the

organization and reinforces their efforts will be recognized by others within the

organization (Lin, 2001).

The notion of social capital can be studied at both a group and individual level of

analysis. Consistent with this study, researchers use social capital framework at the

individual level to understand how individuals gain access to and use the resources

embedded in their social networks to gain returns in instrumental actions (e.g. finding

better jobs) (Lin, 2001). In comparison, researchers studying social capital at the group

level may be interested in investigating how certain groups develop and maintain more or

less social capital as a collective asset (Lin, 2001).

27

Social capital is considered a resource for an individual’s career, and people

utilize their interactions with others in an organizational setting to share information

(Palgi and Moore, 2004). Two widely used mechanisms which foster the development of

social networks include mentors and organizational networks. Social capital is beneficial

to employees and employers mutually. For example, employers value employees with

social capital, because as a result of hiring these employees, the organization will

enhance its social capital by utilizing the new employees’ contact to reach various

organizational goals (Erickson, 2001). An example of how an organization leverages

their employee’s social capital is when the organization relies on an employee’s contacts

to develop new relationships with suppliers (Erickson, 2001). Of note, social capital is

usually a requirement/asset for employees at higher levels within the organization; in fact

employees with greater social contact get better higher-level jobs whether they were hired

through personal contacts or not. Lower level employees are not typically responsible for

recruiting new clients (although in the insurance industry this is an inaccurate

assumption) or making deals with clients (Erickson, 2001).

In short, social capital seems to be important within organizations for at least four

reasons: (1) social capital facilitates the flow of information (by providing an individual

with useful information about opportunities and choices not otherwise available), (2) the

social ties one develops, may be able to influence certain agents (e.g. recruiters or

supervisors) who play an important role in making critical decisions within organization,

(3) people social ties, that is their acknowledged relationships with others in an

organization, may help highlight an individual’s social credentials, and (4) social capital

should help reinforce an individual’s identity within an organization (Lin, 2001).

28

Taken together, both the social capital framework and the relational perspective of

career development demonstrate the importance of relationships between members in an

organization. Both social capital and the relational perspective of career development

suggest that employees (1) need access to information which will enable them to learn

about opportunities within the organization, and (2) employees need to invest time in

social relations that will enable them to gain access to information, and provide them

social support. Social network analysis then is a set of measurements that are used to

capture specific aspects of an individual’s network (e.g. network size, density, cohesion

and closeness within social networks). Specifically, at the network level of analysis,

researchers look at the composition of the networks (e.g. network size, network

heterogeneity, mean frequency of contact) and the structure of these networks; thus

network analysis is used to understand how the properties of networks affect what

happens in them (and to them) (Wellman and Frank, 2001).

Social network analysis can also be used to study various relationships within

organizations. For example, social network analysis can be used to study supporting

partnerships and alliances. Executives are increasingly employing cross-organizational

initiatives such as alliances or other forms of strategic partnerships to leverage the firm’s

unique capabilities. Thus social network analysis can illuminate the effectiveness of

information flow, knowledge transfer and decision making. In short, organizational

networks provide companies a diagnostic tool which can be used to assess specific

processes and flows of information (including information related to career mobility)

(Cross, Parker& Cross, 2004). Network analysis can also be used to investigate the extent

to which hierarchy conditions impact the flow of information across organizational levels

29

(Cross, Parker& Cross, 2004). Specific to this context, future research could investigate if

hierarchy amongst organizational networks impacts the flow of information within the

network. That is, are those more central in an organizational network more informed

about career opportunities and if so, what are the implications? In general, work-family

research has lacked a networking perspective; however suggestions have been made that

work-family research should begin to understand the relationship between networking

and career mobility (e.g. women’s work-related networks may be subject to disruption as

a result of participation in work-family programs).

The Formation of Networks and Network Data Collection Techniques.

Sociologists first begin thinking about networks by looking at Konigsberg’s

bridges of Prussia (Wasserman &Faust, 1994). This graph/map contained links and

nodes. Networks are formed through social links, and these links form as people interact

with each other. In a graphical representation, people are “nodes” and each encounter

with a new actor is represented by a social “link”. Networks are often represented

graphically on a histogram, where a series of links are drawn between various nodes.

Networks start from a small nucleus and expand with the addition of new nodes. Some of

the fundamental characteristics of networks include the following (Wasserman & Faust,

1994):

o Actor – Actors are discrete individual, corporate, or collective social units. Examples of actors are people in a group, departments within a corporation, public services agents in a city, or nation-states in the world system.

o Relational Tie- Actors are linked to one another by social ties. The defining feature of a tie is that it establishes a linkage between a pair of actors.

o Dyad- A dyad consists of a pair of actors and the (possible) ties between them. Dyadic analyses focus on the properties of pairwise relationships,

30

such as whether ties are reciprocated or not, or whether specific types of multiple relationships tend to occur together.

o Triad- Relationships among larger subsets of actors may also be studied. Many important social network methods and models focus on the triad: a sublet of three actors and the possible tie(s) among them.

o Subgroup- A set of actors that and all ties among them. o Group- The collection of all actors on which ties are to be measured. o Relation- The collection of ties of a specific kind among members of a

group is called relation. o Ego-Centric Network- An ego-centered network consists of a focal actor,

termed ego, as set of alters who have ties to the ego, and measurements on the ties among these alters. Ego-centered networks are also used quite often in the study of social support. The term “social support” has been used to refer to social relationships that aid the health or well-being of an individual.

Social networking data can be collected at multiple levels of analysis. The

multiple levels include different levels: the individual actor, the pair of actors or dyad,

the triple of actors or triad, or the network as a whole. The level of analysis is called

the modeling unit (Wasserman & Faust, 1994). Most social network data is collected

by observing, interviewing or questioning individual actors about the ties from these

actors to other actors in the set (Wasserman & Faust, 1994). The questionnaire

method of collecting data is most often used when the actors are people. The

questionnaire usually contains questions about the respondent’s ties to the other

actors. Questionnaires are most useful when the actors are people, and the relations

being studied are ones that the respondent can report (Wasserman & Faust, 1994).

The three types of questionnaire formats that can be used include (Wasserman &

Faust, 1994) :

o Roster vs Recall- Roster indicates actors are presented complete list of people in the set. If the researcher does not have a complete roster they can use the recall method which ask respondents to name those people with whom you (fill in specific tie)… The respondents generating a list a names is called free recall.

31

o Free vs Fixed Choice – If actors are told how many people to nominate on a questionnaire, then each person has a fixed number of choices. In a fixed design, each actor has a fixed max number of ties to the other actors in the set. If actors are not given any such constraints on how many nominations to make, the data are free choice.

o Ratings vs complete rankings – In some cases, actors are asked to rank or rate order all the other actors in the set for each measured relation. Ratings ask an actor to assign a value or rating to each tie.

There are at least three sampling techniques used to collect network data (1) the

saturation sampling technique, (2) the ego-networking technique and (3) the position-

generator technique (Lin, 2001). The ego-networking technique will be used in this

dissertation, but a brief description of each techniques has been included. The saturation

sampling technique is used when it is possible to map and define the boundaries of an

entire network. This technique is most often used within a single organization, or a small

network among organizations (Lin, 2001). The position name-generator technique is used

when a sample of positions within an organization are identified (e.g. marketing

representative, financial analyst), and the individual is asked to indicate if s/he know

anyone holding that position within the organization; based on these responses the

researchers is able to identify various network characteristics including the range of the

individual’s network (that is the distance between the individual with the highest and

lowest position within a network) or the heterogeneity of a given network. The last

technique is the ego-network technique, and the one that will be used in this dissertation.

The ego-network is the sampling method generally used when a researcher is

investigating a large or less definable network (e.g. the researcher is interested in

studying the networks of people across several organizations). The ego-network

technique uses a name-generator survey in which the ego identifies a list of alters within

32

their network and they are able to answer questions related to various characteristics of

their alters (frequency of contact, physical proximity of alters, etc). One common

shortcoming in using the ego-network technique is it usually elicits strong (people with

whom the ego is close too) rather than weak ties, that is, the ego usually names

individuals with whom they have more intimate relationships with or people with whom

they have more frequent contact (Lin, 2001). However, this bias may be beneficial if the

researchers are studying issues related to quality of life or social support, or various

perceptual or psychological outcomes (Lin, 2001).

Network Measures and Characteristics

There are many components of a network that can be measured and used to

describe an individual’s (ego) network. Below, are two tables of typical measures used to

describe networks among individuals.

33

Measure Definition Example

• indirect links Path between two actors is mediated by one or more others

A is linked to B, B is linked to C, thus A is indirectly linked to C through B

• frequency How many times, or how often the link occurs A talks to B 10 times per week

• stability Existence of link over time A has been friends with B for 5 years

• multiplexity Extent to which two actors are linked together by more than one relationship

A and B are friends, they seek out each other for advice, and work together

• strength Amount of time, emotional intensity, intimacy, or reciprocal services (frequency or multiplexity often used as measure of strength of tie)

A and B are close friends, or spend much time together

• direction Extent to which link is from one actor to another

Work flows from A to B, but not from B to A

• symmetry (reciprocity)

Extent to which relationship is bi-directional A asks for B for advice, and B asks A for advice

From D Brass (1995)

Table 2.1: Frequently Measured Items Related to Network Ties

34

The measures listed in Table 2.1 are usually used to describe a link between at least two

nodes (e.g. actors, groups). Of note, a network consists of nodes, and a set of relations

linking these points (Smith-Lovin and McPherson, 1993). Examples of relations that the

measures in the table could be used to study include Person A “gives orders to” (Smith-

Lovin & McPherson, 1993). This would be an example of a directional measures. In

comparison, Table 2.2 (which appears on the proceeding page) describes measures that

are used to capture an entire network, whereas the measures in Table 2.1 are used to

describe a specific link between nodes within a network.

35

Measure Definition

• Size Number of actors in the network

• Inclusiveness Total number of actors in a network minus the number of isolated actors (not connected to any other actors). Also measured as the ratio of connected actors to the total number of actors.

• Component Largest connected subset of network nodes and links. All nodes in the component are connected (either direct or indirect links) and no nodes have links to nodes outside the component.

• Connectivity (Reachability)

Extent to which actors in the network are linked to one another by direct or indirect ties. Sometimes measured by the maximum, or average, path distance between any two actors in the network.

• Connectedness Ratio of pairs of nodes that are mutually reachable to total number of pairs of nodes

• Density Ratio of the number of actual links to the number of possible links in the network.

• Centralization Difference between the centrality scores of the most central actor and those of other actors in a network is calculated, and used to form ratio of the actual sum of the differences to the maximum sum of the differences

• Symmetry Ratio of number of symmetric to asymmetric links (or to total number of links) in a network.

• Transitivity Three actors(A, B, C) are transitive if whenever A is linked to B and B is linked to C, then C is linked to A. Transitivity is the number of transitive triples divided by the number of potential transitive triples (number of paths of length 2).

From D Brass (1995) Table 2.2: Typical Social Network Measures Used to Describe Entire Networks

36

As demonstrated by Brass (1995), several characteristics of an individual’s network can

be studied. Of the network characteristics that were described in the Brass (1995) table,

two will be included in this dissertation. The two measures include network size (the

number of ties in the ego’s network) and network ties/frequency (the number of times the

ego has contact with a specific tie in their network, and the frequency of contact with

those ties). These measures were selected because they are the measures most relevant to

the study of work-family concerns, networks, and careers.

Specifically, previous research has suggested that the characteristics that change

the most in an individual’s network after the birth of a child, include network size and

network ties (e.g. Bost, Cox, & Payne, 2002; Belsky & Rovine, 1984). Network ties, refer

to the individuals with whom the ego has contact. The ties are usually described in terms

of a relational measure, that is, family or kin ties, friendship ties, coworker ties, neighbor

ties, etc. Kin ties include, but are not limited to immediate kin (e.g. parents, adult children

and siblings and in-laws) and extended kin (e.g. aunts, uncles, cousins). If individuals

have networks that are mostly composed of kin, the disadvantage of this network includes

kin are less likely to obtain job and career related information (Wellman, 1992).

However, working parents, specifically, those with children under six years of age are

likely to have a high number of kin in their intimate network. Immediate kin are likely to

be very close to working parents as they are more likely to trust and receive help

(especially with childcare responsibilities) from the kin ties in their network. In fact,

individuals, especially those with parental responsibilities, may want stronger ties.

According to Wellman (1992), strong, intimate ties provide more emotional support and

companionship, in comparison to weaker ties. Interestingly, aside from immediate kin,

37

most of the active ties (that is, those with whom the ego is in frequent contact with)

include friend ties. In fact, friends and neighbors make up nearly half of the most active

and intimate networks (Wellman, 1992).

A distinction is made in the literature between assessing social networks at two

different organizational levels, the system or organizational level and the personal

network level (Ibarra and Smith-Lovin, 1997). The organizational level of social

networking research is concerned with identifying networking characteristics among a

group of individuals who co-exist within defined boundaries (e.g. a specific organization

or a specific workgroup). This dissertation is concerned with measuring social network at

the personal or ego-centric level. That is, the “egocentric or personal networks” are

concerned with assessing the network at the individual level and are more interested in a

specific individual’s contacts.

The total number of actors (i.e. ties) in an ego’s network is a measure of network

size. Network size is an important, commonly measured variable included in studies

interested in investigating career-related outcomes. For example, Carroll and Teo (1996)

studied differences in the organizational networks of managers and non-mangers,

including overall network size. The researchers compared the organizational networks of

managers and non-managers. The study found several differences in the network

characteristics of mangers and non-managers including managers tended to have more

external network ties than non-managers (e.g. more membership in clubs and societies).

38

Also, managers seemed to have more nonredundant ties within their networks (i.e. people

that were total strangers within their network) and have larger networks. The results

from the Carroll and Teo (1996) study were not very surprising, as one would expect the

networks of managers, or people with higher level jobs to be larger, and networks with

nonredundant ties.

In addition, to the size and type/frequency of contact with ties, the third measure

that will be used to describe the ego’s network characteristics includes topics of

conversation or content of conversations within networks. The topics of conversation

within informal studies have been studied previously (e.g. Kidd et al. 2003). The findings

from previous studies suggest individuals experienced positive career-related outcomes

when employees engaged in informal conversations, such as those one would have within

their networks. Some of the positive outcomes reported included “future direction, self-

insight, awareness of opportunities and feel-good” (Kidd et al., 2003). In the Kidd et

al,.2003 study: “future directions” was described as identification or exploration of

particular career options relevant to self and it described career decision or clearer

direction in terms of a career path within a given organization, “self-insight” was greater

awareness of ambitions, lifestyle, values, skills, and strengths, “awareness of

opportunities” included knowledge of a range of career opportunities and general

knowledge of external opportunities, and “feel-good” was described as individuals who

reported feeling better or more reassured about their jobs and careers. In addition,

about one-third of the respondents reported that informal conversations with people led to

a job move and about one quarter of the conversations led to “developmental

opportunities, on-going dialogue, greater political awareness about internal processes, or

39

improved career skills” (Kidd et al., 2003). The Kidd et al (2003) study made a

contribution to the literature because it measured outcomes related to informal

conversations; most previous literature measured outcomes using measures such as

learning and career outcomes. Kidd et al (2003) also studied the following topics during

their semi-structured interviews: who the respondent had discussions with, the setting the

discussion took place, how the parties involved behaved during the discussion, what the

outcomes of the discussion were, and how the receiver felt after the discussion.

Therefore, this study, similar to the Kidd et al (2003) study, will contribute to the

field by measuring the topics discussed within the ego’s network and outcomes related to

career management perceptions and indicators of objective/subjective career success. In

addition, this dissertation will go beyond the Kidd et al (2003) study because it will

compare topics of conversations discussed across four groups, working mothers, working

fathers, working women with no parental responsibilities, and working men with no

parental responsibilities. Further, this study will analyze what topics were discussed

among the people that the respondents have identified as important to their professional

careers. In the Kidd et al (2003) study, the respondents were asked to identify anyone

with whom they had an effective career-oriented discussion. In comparison, this

dissertation will analyze the members within a network that are most important for

individuals to talk to, rather than just analyzing career-oriented conversations with

individuals that may or may not be included in their network. Further, the Kidd et al

(2003) study assessed only the career outcomes that resulted from these discussions.

40

This study will assess career-related outcomes and other important topics (e.g. family,

childcare) to gain an understanding of whether there are differences in conversations

topics across gender and parental status. The Kidd et al (2003) study does not compare

the conversation topics discussed across gender or family status; rather the study reports

12 overall categories of conversation, without accounting for individual differences (e.g.

gender) in topics/outcomes discussed.

In summary, it is important to consider the topics of conversation that are

discussed within networks. For example, this study suggests that women may not realize

the full advantages of networks in comparison to men for the following reasons. Men,

talk about work-related or career related topics when they initiate informal conversations

within their networks or in conversations with their immediate supervisors. In

comparison, it may be the case that women initiate dialogue about their work-family

concerns, in an effort to begin dialogues within their networks and during informal

conversations. The concern is if women initiate these informal work and family dialogues

within their networks, are they able to successfully shift their conversations within in

their networks from “family talk” to more relevant talk about jobs and careers?

The Importance of Organizational Networks

As mentioned previously, organizational networks are a key factor to securing

advancement within an organization. Networks offer many benefits including

information, career guidance, advocacy for promotion, and “having a relationship with

influential people within one’s organization provides entry into social networks that are

inaccessible through formal communications” (King, 2003, p120). Further, networking

41

or the development of informal relationships is important, because the burden (or

responsibility) for an employee’s career development has shifted from the organization to

the individual (Forret & Dougherty, 2004). In addition to networking being beneficial to

an employee’s career development, other benefits of organizational networks include

formal and informal information exchange, career planning, professional support and

encouragement, greater visibility with senior management, and personal and career

development, as reported by a sample of women in a recent empirical study (Linehan,

2001). In addition, networks were thought to be advantageous especially for women who

did not have a mentor throughout their career (Linehan, 2001).

Another advantage of organizational networks is the development of social

capital. There are two expected returns from investment into social capital, (1) returns to

instrumental action and (2) returns to expressive action (Lin, 2001). Returns to

instrumental capital includes elements such as career development opportunities,

inclusion in powerful organizational networks, access to information, access to higher-

level and higher paying jobs, and increases in power within organizations. Returns on

expressive action, refer to an individual’s effort to gain access to others whom share

common interests and control (access to) specific resources. For example, Lin (2001)

describes three types of return on expressive action that individuals usually seek

including physical health, mental health and life satisfaction. Of interest to work-family

researchers, one outcome related to social capital is the return related to expressive

action, that is, life satisfaction. Life satisfaction suggests optimism and satisfaction with

various life domains such as family, marriage, work and community and neighborhood

42

environments (Lin, 2001). Also, life satisfaction is an outcome variable that is also

commonly measured in the work-family literature.

Gender and Organizational Networks

Research using network analysis suggest that similar people tend to interact with

each other, where similarity can be defined as age, gender, prestige, tenure and

occupation (Brass et al,. 2004). This line of reasoning may explain why women have

been excluded from organizational networks, that is, the idea that the old boys network

where men generally congregate with each other. Of note, similarity is a relational

phenomena such that people are similar with each other only when they are dissimilar to

another group or person (Brass et al. 2004).

Within the literature, there have been some key differences noted in the

networking behaviors of men and women. For example, some argue that men tend to

have broader informal networks that often included top executives and outside contacts

(Ragins and Sundstrom, 1989). In addition, men in comparison to women generally have

higher position power. That is, power associated with the control over resources, rewards,

and punishments, information, the work environment, and work procedures (Ragins and

Sundstrom, 1989). As a result, men may be better informed and have a broader

networks, which helps them facilitate their career mobility. Broader networks are those in

which the ties within the network are at various levels within the organization, spanning

from entry-level to executive levels. Men generally have more fully developed

organizational networks, and men tend to utilize their organizational networks to bid for

jobs, in comparison to women (Cannings & Montmarquette, 1991).

43

Research has also suggested that women tend to benefit less than men from

participation in organizational networks (Davidson and Cooper, 1992; Broadbridge,

2004). Brass (1985) provides three explanations for why women benefit less from

networks than men including: (1) they are unaware of informal networks, (2) they interact

with people whom are like them (in respect to values, attitudes and experiences) and, (3)

they are intentionally excluded from informal networks by men. There has been little

support found for the later rationale, that is, women are intentionally excluded from

networks. Empirical evidence has found that women are included in informal networks,

however many women choose to participate in gender-based networks, that is they prefer

networking with other women (Brass, 1985).

One of the factors that may lead to homophily within groups, is the argument of

weak and strong ties. Homophily refers to interaction with similar others (Brass, 1995).

Similarity can be operationalized on many dimensions including age, gender, education,

social class, tenure, and occupation. Research suggests that in general, cross-gender and

cross-racial networks are perceived to be weaker than homophilous relationships (i.e.

interaction with similar others). This perception (and attraction) to the network is mostly

based on the perceived ability and power of a specific network group (Ibarra, 1993). In

addition, homophilous groups are often formed naturally, particularly at higher levels of

the organization. In senior manager and executive level jobs, people forms groups with

others within their functional groups. Thus homophilous groups sustain themselves in

organizations because organizational members have a disincentive to join cross-gender or

cross racial groups; thus providing minorities and women with less access to more

powerful networks within their organization (Ibarra, 1993). Both Brass (1985) and

44

Ibarra (1993) also found that many informal groups are also “male-only” groups. It is

important to consider the role of homophilous groups because the relationship between

two actors (individuals) within networks can vary due to actor similarity. For example, if

there are two individuals within a network and they are dissimilar (i.e. different ages,

gender, education) they are likely to have less frequent communication, there relationship

will be less stable over time, and they are more likely to have a weak in comparison to a

strong tie (Brass, 1995).

Thus, within organizations, informal networks are often segregated by gender,

leaving women at a disadvantage. However, one must take into account how male-only

groups are formed. These groups may not always intentionally exclude women. Previous

studies have indicated that there are fewer minorities and women in senior level

positions. Moreover, the number of employees at the very senior levels of management

within organizations is limited. That is, there are a higher number of female and minority

employees who hold entry and middle level positions in organizations. Therefore, people

within this relatively small executive group, that is people at the top levels of

organizations, naturally form a network amongst each other, without purposely excluding

women and minorities. Women and minorities can not be intentionally excluded if there

are none employed in a specific functional area, or in senior management levels within

organizations.

An additional factor that makes it difficult for women to join organizational

network groups is the topic of conversations during exchanges within organizational

groups. Conversation topics are often “male dominated” including sports or computer-

oriented conversations (Broadbridge, 2004). Further, other studies have suggested men

45

talk more in formal task-oriented contexts (i.e. workplace settings), while women are

more likely to talk in informal settings (James and Drakich, 1993). This leads one to

question to what extent are women disadvantaged within their organizational networks.

Thus, even if women are included in organizational networks, the James and Drakich

(1993) study suggests that women may not feel comfortable talking in these formal

settings. This provides additional evidence to suggest that men and women may have

different experiences within their organizational networks.

Some research has investigated acceptable and unacceptable conversations at

work. For example, Singh et al (2002) found that women deliberately avoid talking about

their families at work. Most women thought they should only speak about work-related

topics while they were on the job. In fact in the study some women perceived only a

work-focused individual would be promoted to the next level. This leads one to question

if women and working parents in general feel comfortable discussing their family

responsibilities within their informal networks, especially if their networks include

individuals who do not openly discuss family-oriented issues. This raises a question as to

whether or not working parents can comfortably discuss work-family topics at work.

Specifically, if working parents can only talk about work-family concerns in specific

forums, do they purposely exclude themselves from larger organizational networks,

where other job-oriented topics are discussed (e.g. career progression) and participate in

smaller (or non-work based) networks where they feel comfortable discussing work-

family issues (this may be especially true for working moms). In addition to discussing

work-family issues within networks, some studies suggest women, especially working

mothers do not have time to participate in organizational networks (e.g. Linehan, 2001).

46

For example, in a recent qualitative study a working female shared the following

comment about organizational networks:

“To be quite honest, I think women have less time than men for networking. Networking has to take place to a great extent after work and on top of your job. If you are a woman with a family, you have less time. Men have more time for networking. Working women are very busy”(Linehan, 2001, pp. 826).

The women that were found to benefit the most from informal networks were

those whom participated in integrated informal networks, where the majority of the

members were men (Brass, 1985). In general, the informal networks which include

mostly men seemed to be perceived by the organization as most influential. Personal

contacts have been found to have a major impact on hiring decisions especially for top-

level and upper managerial job openings. In short, various organizational barriers have

limited the career mobility of women. This study suggests that informal exchanges are

one of the key organizational barriers (i.e. insufficient knowledge about job openings)

that have limited the career mobility of women. This explanation provides an alternative

view to the notion that women are not promoted as a result of a supervisor’s perceptions

that women will be unable to fulfill job obligations (due to the role conflict they

experience in attempting to balance their home and family lives).

In general, although network size does not generally vary between men and

women, the key network members differ. For example, men seem to network the most

with coworkers and volunteer groups, while women tend to network with relatives and

neighbors (Mardsen, 1990).

47

Arguably, this trend is changing as more women enter the workforce. Another important

finding is there is an overall tendency of gender homophily, that is most networks are

same-gender, rather than cross-gender networks (Ridgeway & Smith-Lovin, 1999).

Therefore, it is presumable that women and men participate in organizational networks.

Ragins and Sundstrom (1989) suggest that women with children have less time to

devote to development of networks. Research has also found that the (generally

speaking) smaller representation of women in networks appeared to result from

exclusionary pressures in comparison to women’s preferences for female friends (Mehra,

et al., 1998). In addition to gender, differences within organizational networks also seem

to vary across levels. For example, Carroll and Teo (1996) found in comparing managers

and nonmanagers within the same sample, managers tend to participate in wider

organizational networks, that is, they belong to more clubs and societies. Further, this

study found managers tended to have networks with a greater number of people and more

“weak ties” or pairs of strangers included in their network in comparison to nonmanagers

(Carroll and Teo, 1996). The authors also found these differences in sub-groups

(managers vs non-managers) even when controlling for differences in education, age,

gender and ethnicity. The findings of these studies led Carroll and Teo (1996) to

conclude that the differences in organizational network patterns of managers and non-

managers may be a result of “the selective entry and retention of people with particular

types of networks into management” (p.437) or “the behavioral transformation of people

placed in the intense social structures surrounding managers” (p.437). In other words,

consistent with previous studies, one manner in which employees can enhance their

chances of career mobility is through attaining social capital, through the development of

48

multiple relationships (e.g. managers were found to belong to more clubs and societies in

comparison to non-managers in thee Carroll and Teo (1996) study). Therefore,

employees whom have successfully developed their social capital, by enhancing

networks, may be more likely to be promoted into management positions, than employees

with less social capital.

Burt suggests that building a network around the immediate supervisor is

unproductive. Instead Burt (1992) suggests that the greatest benefit of a network is with

people completely removed from the immediate workgroup. Burt also makes a distinction

between task and opportunity networks. Burt suggests that task networks are ones in

which contacts include your current workgroup. Task oriented networks suggest you

communicate only within your workgroups and are prominent among employees that

were recently promoted. Comparatively, opportunity networks are ones in which people

invest a lot of time outside of their immediate workgroup and they spend very little time

with people within their direct workgroups. When deciding between the two networks,

Burt (1992) suggests that the choice of networks to develop may differ by gender. For

example, high ranking men seem to benefit from opportunity networks, which will lead

to faster promotions within the organization. In comparison women (and entry-level men)

seem to benefit from a task-oriented network built around a significant organizational

member, besides their immediate boss. Although differences measured across

organizational level will not be included in this study, this is a factor that should be

investigated in future studies.

Ibarra and Smith-Lovin (1997) describe at least two contextual factors that

contribute to differences in men and women’s networks. The first factor is the

49

stratification of women and men in terms of their careers and occupations. In the past,

women tended to be concentrated in industries dominated by women. Alternatively

women held jobs in lower positions within male-dominated organizations. Also, women

have been challenged by individuals in organizations making attributions that they are

less powerful and have been excluded from participation in various networks. Another

contextual factor that contributes to the differences in men and women’s networks

include homophily, that is, the idea that individuals tend to network with others that are

similar to them. Gender has been one of the key elements that has contributed to

homophily, which is the tendency for men to network with men and women to network

with women. A distinction is made in the literature between induced homophily and

choice homophily. Induced homophily suggests that similar members are included in

organizational networks as a result of lack of variance amongst employees (e.g. auto

repair tends to be a male-dominated field). In comparison, choice homophily is the result

of preference for interaction with others (Ibarra and Smith-Lovin, 1997). Therefore, in

some organizations were there is a strong presence of both male and female employees,

some individuals may actively choose to network with others of their same gender,

thereby leading to choice homophily. Based on anecdotal evidence, this often happens in

such industries as banking (e.g. investment banking) and entertainment.

Research has also suggested that women tend to benefit less than men from

participation in organizational networks, that is, the impact or organizational networks on

career advancement creates a significant advantage for men; this also results in men’s

greater influence and centrality in (elite) networks (e.g. Brass, 1985; Cannings and

Montmarquette, 1991; Moore, 1988). In fact, as a result of women relying on informal

50

networks specifically for career promotion opportunities, women are often confronted by

an invisible ceiling that only can only be removed by social relationships being

restructured within organizations (i.e. the development of network or affinity groups

within organizations that are organized around common themes, for example, an Asian-

American affinity group that meets to discuss the concerns of Asian-American employees

across gender) (Cannings and Montmarquette, 1991).

Recently, studies have suggested new models of career success must include

situational and personal factors (Eddleston, Baldridge, & Veiga 2004). This study

suggests that a key organizational contextual factor worth reexamination is the influence

of networking behaviors across men and women and its influence on career mobility.

Eddleston, et al. (2004) found exposure to powerful (organizational) networks was

significantly related to number of promotions offered (number of promotions is a

commonly used measure of career mobility). Also, efficacy of mentoring, that is, an

individual’s belief that mentoring influenced their career progression, was also positively

related to both exposure to networks and number of promotions offered.

Specific to gender differences (i.e. a personal factor), Eddleston, et al. (2004)

found a significant relationship between exposure to powerful networks and promotions

offered for men, but not women (Eddleston, et. al 2004). The findings from this study

seem to suggest that interpersonal relationship, specifically exposure to powerful

networks, play a more important role in shaping the career progression of men in

comparison to women (Eddleston, et al, 2004). This finding suggests that women do not

have the same access to networks as men, and it suggests that women do not use their

networks as effectively as men (Eddleston, et al, 2004). Eddleston, et al. (2004)

51

identified several areas for future research including (1) examining why exposure to

powerful networks benefits men’s careers, but not women’s careers, or (2) examining

why women do not utilize their connections (network) to top management to aid their

careers (e.g. are there subtle forms of discrimination that keep women from benefiting

from exposure to powerful networks) (Eddleston et al., 2004). The proposed dissertation

research will help to address the research needs by investigating differences in the

network experiences of men compared to women.

Parental Status and Organizational Networks

The birth of a child and family involvement impacts the organizational networks

of employees. The time and energy formally allotted for the development of relationships

within networks is now allocated to the tasks and duties of childbearing (Munch et, al.,

1997). Empirical evidence also suggests that the influence of the birth of a child may

differ by gender. For example, Campbell (1988) suggests that having young children at

home decreases women’s but not men’s job-related contacts. Munch et al (1997) found

the age of the child impacted network size, that is, women whose youngest child is 3 or 4

years old displayed significantly smaller networks than do their counterparts with adult

children. Further, having a young child did not produce a significant effect on men’s

network size and in fact, the men’s networks remain at a relatively constant size over the

childbearing years (Munch et al., 1997). Another interesting finding from this study was

women’s networks tend to be smallest when children are infants, and through the time

that their children reach the age of four (Munch et al, 1997). After children reach the age

of four, the network size of women tends to rebound (Munch et al, 1997). This finding

may be consistent with the age that children begin going to pre-school, that is, parents

52

may have more time to develop their networks once their children begin attending school.

In addition, at the age of three, children begin to develop motor skills that require

increasing amounts of parent’s attention and time (e.g. when children begin to walk) in

comparison to an infant child that may spend a fair amount of time sleeping (Munch et

al., 1997). In short, parental status seems to be directly related to at least once change in

networks, that is, a reduction in network size (where size indicates the number of ties

within an individual’s network). Future research needs to investigate if the loss of

contacts (i.e. a reduction in network size) leads to a failure of parents to establish new

contacts, especially after the child passes the age of 3.

One of the reasons that the network size of a working parent may be smaller is

related to the multiplexity argument. The multiplexity argument suggests that individuals

that have strong ties may have smaller networks (Wellman, 1992). Specifically,

multiplexity occurs when an individual’s network members have multiple role relations

to that individual. For example, if a working parent hires their sibling to provide childcare

provision, this sibling is fulfilling multiple roles for the working parent, that is, the role of

sibling and the role of childcare provider. Research suggests that those individuals that

have multiplex ties, that is, when a person within your network fulfills at least two roles,

that these types of ties are stronger and are more supportive because these ties have a

detailed knowledge of each other’s needs (Wellman, 1992). As a result, working parents

may have a tendency to have stronger, but fewer ties, because they may have a tendency

towards spending time with individuals that are able to fulfill multiple roles for them (e.g.

when they are seeking childcare providers). The constraint that parents may face is,

although stronger ties are likely to be more supportive, they are also less likely to provide

53

working parents with new information. That is, the weak tie argument, to be discussed in

detail later in this chapter, suggests that individuals should maximize the number of weak

ties, especially when it comes to planning their jobs or careers, as they want to ensure that

they maximize the amount of nonredundant information they receive, which is best

accomplished, when an individual seeks information from weak vs. strong ties.

In addition, many women leave the labor force at least temporarily when their

first child is born (Waite, et al., 1986). As a result, working parents are likely to

experience an interruption to their organizational networks. In addition, in an effort to

allow work flexibility, organizations now are offering options that allow working parents

to continue their career with the organization by providing teleworking, part-time job

opportunities and condensed work weeks. However, organizations and researchers have

given little thought to the impact these alternative works arrangements have on an

working parent’s organizational network (participation). In a study where the researchers

determined whether new parents differ (in terms of career aspirations) from adults who

forgo or delay the birth of a child, Waite et al., (1986) found the following: (1) work

expectations for women decrease at the onset of pregnancy and remain well below

expectations for the next two years, (2) the work orientation for fathers increased after the

birth of a child and, (3) the total earnings per week is higher for men whom are

expecting children in comparison to those who are not (Waite, et al., 1986). Similarly,

another study found that the average hours worked differed between employed women

and employed mothers, where employed mothers worked five hours less per week on

average (Kaufman & Uhlenberg, 2000). In addition, the effect of parental status on the

number of hours worked was most significant when the number of children increases

54

(less hours are worked) and the age of the child increases (as age of the child increase,

moms work more hours) (Kaufman & Uhlenberg, 2000). Interestingly, the relationship

between hours worked and parental status differed between women and men in this study.

As the number of children increased, men tend to work more hours, but the amount of

hours works seems to vary by attitudes toward desirable childcare arrangements

(Kaufman & Uhlenberg, 2000). That is, the more satisfied men were with the childcare

arrangements, the more hours they were likely to work. If the men in the study were not

satisfied by the childcare arrangements, they were likely to work less hours.

The impact of the birth of a child on networks is important because it has been

suggested that time spent with family is thought to constrain the networks of working

employees, especially women. Thus time spent with family is viewed by social scientists

as limiting women’s success in the workplace (Scott, 2001). If women or working

parents in general are dedicating more of their time to their home responsibilities, they

are taking time away from their opportunities to participate in organizational activities.

Also researchers suggest the social ties formed at home are not as beneficial to women as

the ties formed in organizations (Scott, 2001).

In a recent article, Scott (2001) examined the gender differences in family

involvement and responsibilities and explores how these differences impact a women’s

ability to participate in organizational networks. This study was conducted among

workers in the public affairs, specifically, governmental relations divisions. Networking

with corporate executives, employees and the public is an important component for those

working in governmental affairs. Therefore this study tries to understand to what extent

marriage and family responsibilities create obstacles to interacting with government

55

clients and with key people in business (e.g. are married women, or women with children,

limited in the particular kinds of networks in which they engage?) This dissertation

addresses a key research need identified in previous work; “researchers have examined

the relationship between gender and work networks, but few have included the effects of

families in their analysis” (Scott, 2001, p9).

There were several interesting findings reported in the Scott (2001) study. The

factor that seemed to impact a woman’s ability to network was not parental status, instead

it was marital status. Married women were found to be at a slight disadvantage when it

comes to socializing in non-work settings (e.g. concerts, theatre, sporting events). This

finding was inconsistent with the notion that women with children are at the greatest

disadvantage in terms of participating in organizational networks. Based on this finding,

The findings from Scott’s (2001) research may suggest that that once ‘boundaries’ of

marriage are formed and solidified, the presence of children produces little additive effect

in terms of negatively impacting a woman’s opportunity to socializing and fostering

relationships within their networks.

One factor that is believed to help working parents continue to participate in

organizational networks was the availability and quality of childcare. A comment made

by one of the individuals interviewed in Scott’s (2001) study suggested that women, after

the birth of child, will have a hard time maintaining their professional relationships unless

they have some kind of help at home (in terms of childcare). Also, some women may not

directly maintain their own professional relationships after the birth of a child. Often

women began attending the non-work events of their husbands, and do not attend the

non-work events of their own organizations. As a result women may loose several of

56

their professional ties, especially if they are attending events where members of their

personal professional networks are not present.

Careers: An Overview of Various Approaches

In order to understand the role of careers within organizations, it is useful to know

how careers are defined. Hall (2002) describes career as “the individually perceived

sequence of attitudes and behaviors associated with work-related experiences and

activities over the span of one’s life”. The relational approach to careers, as described

previously, is concerned with how employees develop and leverage interpersonal

relationships to assist them obtaining information about new job and career opportunities.

Although interest in the relational approach to careers appears to be increasing, that is,

there is an increase in empirical and theoretical work (e.g. Higgins and Kram, 2001;

Seibert et al, 2001), the relational approach is not the only perspective used to study the

career behaviors of employees. The previous perspectives of careers, typically examined

the developmental aspect of careers (Kram, 1996). Therefore, if one knew a person’s age,

tenure, personality, values, and/or learning style one could accurately predict what the

person’s salient career concerns and developmental tasks might be (Kram, 1996).

One example of a developmental perspective of careers includes Hall and

Nougaim’s (1968) three stage model of organizational career development. In the first

stage, the establishment stage, the employee is at the beginning of their career and the

individual is mostly concerned with defining their position and feeling comfortable in this

position, that is, the individual is looking for a way to integrate themselves into the

organization (Hall and Nougaim, 1968). In the next stage, advancement, the individual is

no longer concerned about fitting into the organization, instead they are interested in

57

looking for opportunities to advance their career. Further in this stage, individuals show a

higher need for achievement (Hall and Nougaim, 1968). In the final stage, maintenance,

successful individuals may be concerned with helping or mentoring others. In

comparison, for those individuals who are not successful, they may try to derail the career

progression of individuals newer to the organization, in an effort to maintain their

position in the organization, and to reduce competition for their specific job. Further, this

last stage represents a career plateau (Hall, 2001). Employees meet career plateaus when

they are stuck in their current jobs with little likelihood of promotion or with very few

opportunities for increased promotion (Greenhaus et al, 2000).

Another example of a developmental model of careers includes Super (1957) and

Levinson (1978). Using a life-span approach, Super’s (1957) model suggested that

individuals develop their self-concept through their choices in vocations. Described in

four stages, this model describes differences and behaviors across various career stages.

For example, the exploration stage (1) includes a period of self-examination, formal

education and the exploration of various career options (Super, 1990; Sullivan, 1999).

The second stage, establishment, is the period of finding employment and developing a

specific niche, and the third stage, maintenance, describes an individual upholding their

job and developing skills necessary to function in that job (Super, 1990; Sullivan, 1999).

Disengagement, the final stage, marks the retirement period for an employee (Super et al.

1988; Sullivan, 1999). Levinson’s (1978) model, in comparison, suggests a punctuated

equilibrium model, in which after a certain period, an individual will reassess their goals.

Specifically, the Levinson (1978) model suggests there are four eras of the human life

cycle: pre-adulthood, early adulthood, middle adulthood and late adulthood. The

58

fundamental argument Levinson’s (1978) model made was an individual goes through

both stable and transitional period during their lives, which are closely related to the

individual’s age. Levinson (1978) further argues that the typical transition period about

five years and transitional periods last about five to seven years. During the periods of

stability, individuals focus on non-work issues, develop job-related skills and mentally

prepare for themselves for transition periods (Levinson, 1978, 1986).

One criticism of the both the Super (1990) and Levinson (1986) models is these

models may not apply to women. Very little empirical work has been done to tests these

models on women. Further it has been suggested that the traditional (developmental)

career stage models were developed to explain the careers of men and were tested

primarily with male samples (Sullivan, 1999). Therefore, researchers are calling for

future research to take a broader approach to and consider the interaction of multiple

factors, including the timing of parenthood, family responsibilities (e.g. childcare) the

career stage of the woman’s partner, and organizational support (e.g. flexible schedules).

The relational model, which suggests the value of relationships in the career development

of employees, is not specific to any gender and therefore may provide a more

comprehensive view of the factors that impact the careers of men and women, and

parents.

In addition, to the potential gender biased introduced by the use of stage models,

it has been suggested that the developmental models are not consistent with the average

tenure of employees. The career developmental models were developed at a time when an

employee, for the most part, stayed with one organization (Sullivan, 1999). Given the

trend of employees today, that is to move between organizations, on average, every four

59

to five years, the developmental models are no longer a means of understanding the

employee’s career behaviors or attitudes.

The most current perspective on careers is the protean career. The protean career

is defined as a self-directed career “in the pursuit of psychological success in one’s work”

(Hall, 2002), and a career that is “self-determined, driven by personal values rather than

organizational rewards, and serving the whole person, family, and life purpose” (Hall,

2004, p2). Some of the characteristics of the new protean career include: (1) the career is

managed by the employee, not the organization, (2) the career consists of a lifelong series

of experiences, skills, transitions, and identity changes, (3) career development is marked

by continuous, self-directed learning and the development of relationships, (4) career

development does not necessarily include formal training and upward mobility, and (5)

the employee expects their organization will provide challenging assignments, and

informational and other developmental resources (Hall, 2002; Hall and Mirvis, 1996). In

a recent article, Higgins and Kram (2001) suggested the way to make protean careers

work is through the development of career networks. That is, if an individual is going to

be successful in managing their own career and pursuing opportunities that meet their

subjective measures of career success, individuals need to have specific tools to manage a

protean career. Drawing from Higgins and Kram (2001) work, this dissertation suggests

that gender, family involvement, job involvement, and role segmentation will moderate

the relationship between parental status and the network characteristics size, ties, and

content. As a result, when compared to working adults without children, this dissertation

argues that working parents face several disadvantages that make pursuing a self-

managed or protean career difficult. Because of the constraints faced by working parents,

60

they have a more homogenous set of relationships within their career networks. Although

not studied in this dissertation, future studies could also assess the role of the

organization in helping individuals develop relationships that would be useful in pursuing

a protean career. Some of the contributions of organizations in supporting employees

develop relationships may include encouraging employees to engage in career dialogues

with a manager, peer or career coach (Hall, 2004).

Careers and Organizational Networks

Previous research has examined the relationship between organizational network

and various career outcomes. Some of the outcome measures include career success,

career mobility, and career progression. A brief description of each of the career

outcomes will be included in this section, followed by examples of empirical findings

related to career outcomes. Career mobility is usually measured in terms of the number

of promotions an individual receives in a specific time period, the salary progression of

the individual (including salary, commission, and bonus), and the movement of

employees in various organizational roles (e.g. Vardi, 1980; Martin & Strauss, 1956). In

some empirical studies, career mobility is also called career progression (Turban and

Dougherty, 1994). In prior literature, some studies measured either salary progression (a

measure across years of salary and bonuses received) or promotions (number of new jobs

received in a specified amount of time) individually. That is, salary progression and

promotions are treated as two separate dependent variables (e.g. Wallace, 2001). In

comparison, when studies use both salary progression and promotions, these two

variables are combined into one larger dependent variable category, that is, career

mobility or career progression (e.g. Stroh, Brett, & Reily, 1992). Other measures of

61

career mobility include intra-organizational mobility (i.e. the total number of job changes

with the same employer) and inter-organizational career mobility (i.e. the total number of

job changes that involved moving from one employer to another) (e.g. Valcour &

Tolbert, 2003). For example, Polodny and Baron (1997) studied the relationship between

networks and career mobility. They found having a large sparse informal network (that is

a large number of nonredundant ties) led to upward career mobility among individuals

studied.

Burt (1997) also examined career mobility. He studied the relationship between

network size and career mobility (e.g. promotability and salary increases). Burt (1997)

examined a group of managers whom had a large network of nonredundant ties. He found

the managers with structural holes, that is networks where the alters of the network do not

know each other (i.e. the alters are various ties or individuals within the ego’s network),

were promoted more quickly and received larger bonuses than individuals with networks

composed of strong ties (i.e. strong, but very few ties within their networks). In

addition, Saxenian (1996) studied the impact of networks on career mobility among

professionals in Silicon Valley. Saxenian (1996) found that those interested in inter-

organizational mobility (the Silicon Valley has a norm for high inter-organizational

mobility) and those individuals whose ties included members of professional

organizations (another example of nonredundant ties) were able to move easier between

organizations, than an individual whom had only organizational ties. Finally, other

studies have investigated the relationship between career mobility and a number of

attitudinal measures including career satisfaction and organizational commitment. For

example, Murrell et. al (1996) conducted a study to address the overall affect of job

62

changes on career outcomes. The authors found frequent moves within an organization

had a negative impact on work-related attitudes, specifically overall job satisfaction and

commitment.

In addition to career mobility measures, several studies have used attitudinal or

perceptual measures, for example career satisfaction and career self-management. Career

success is based on psychological success. That is, career success is a person’s

perceptions of how successful their careers are, regardless of observable indicators such

as salary or hierarchical attainment (e.g. Valcour & Tolbert, 2003; Turban & Dougherty,

1994). Vardi (1980) characterized the psychological approach to careers as any work that

is based on understanding the:

“individual’s perspective, career attitudes, perceptions, expectations and behaviors; “thus a psychological approach to careers looks at both the antecedents of the career behavior (e.g. personality, ability, choices) and the consequences of the outcome of this behavior” (e.g. career success, career satisfaction) “ (p.344).

Consistent with this view, career success is based on individual perceptions.

Career success can be defined as “the accomplishment of desirable work-related

outcomes at any one point in a person’s work experiences over time” (Arthur, Khapova,

& Wilderom, 2005, p. 179). Therefore, career success is a perceptual measure and it can

be measured by items such as “I am in a position to do work that I like”,” I am respected

by my colleagues” or “I am pleased with the promotions I have received so far” (Gattiker

and Larwood, 1986). A common distinction made in the literature is between objective

and subjective career success. Objective career success is success that is directly

observable, measurable, and verifiable by a third party (Hughes, 1937; Hughes, 1958).

Examples of career success may include pay increases or promotion to a higher position

63

(e.g. from bank teller to loan officer within a commercial banking setting). In fact, the

most widely used measures of career success include “salary, salary growth, and

promotions” (Heslin, 2005, p.115). Consistent with this view, Arthur and Rousseau

(1996) found that more than 75 percent of empirical studies published between 1980 and

1994, used an objective career success measure. The large presence of objective career

success in published articles is a result of several methodological advantages; specifically

when objective success career measures are collected from a 3rd party (e.g. employee

records) there are no issues related to common method variance or self-serving bias

(Heslin, 2005).

Despite the prevalence of objective career success measures in the literature, some

studies (e.g. Greenhaus, 2003; Hall, 2002) have begun to investigate the antecedents of

subjective career success (Heslin, 2005). Subjective career success alludes more to an

internal satisfaction an individual experiences. Specifically, subjective career success is

defined by an individuals reactions (across any dimensions that are important to that

individual) to their career experiences and it is usually operationalized by a career

satisfaction or a job satisfaction measure (Hughes, 1937, 1958; Heslin, 2005; Arthur et

al., 2005). Career success is usually measured by items that assess global career success.

It can also be measured with items such as satisfaction with pay, promotions and the

development of specific skills (Turban & Dougherty, 1994). It is important to consider

multiple measures of career success (e.g. intrinsic and extrinsic career success) in the era

of the protean or boundaryless careers, because careers do not follow a template in the

same way traditional careers do (Valcour & Tolbert, 2003).

64

An example of a study that measured the relationship between network and career

success was conducted by Bozionelos (2003). In this study, she found network resources

(e.g. a perceptual measure of the availability of intra-organizational network resources)

were associated with career success and networks were found to make an additive

contribution to career success beyond any mentoring received. In the Bozionelos (2003)

study, both extrinsic and intrinsic career success were measured. Extrinsic career success

was measured as current organizational grade. Intrinsic career success was measured

using ten items from the Gattiker and Larwood (1985) scale. The ten items include

measures of job success (e.g. “ I am in a position to do work I really like”), inter-personal

success (e.g. “I am well liked by my colleagues) and hierarchical success (e.g. “I am

pleased with the promotions I received so far”).

In another study measuring career success, Seibert et al (2001) found weak ties

and structural holes, that is factors to contributing to larger networks and networks with

nonredundant information, related to social resources (e.g. information), which led to

promotions and career satisfaction among the sample studied. In addition, Valcour

&Tolbert (2003) measured career success and the relationship between intra-and inter-

career mobility. They found when people move between jobs (inter-organizational

mobility), they tend to experience a decline in their pay, but this does not affect how

successful people feel in their jobs (subjective career success). In comparison intra-

career mobility, that is taking a new job in the same organization, resulted in an increase

in earning, but negative effect on perceived career success.

65

A perceptual measure, career self-management is best described as performing

activities such as setting-career related goals and devising strategies to achieve them

(Noe, 1996). That is,

“career management is the process by which individuals collect information about values, interests, and skill strengths and weaknesses, identify a career goal, and engage in career strategies that increase the likelihood the goal will be achieved” (Noe, 1996, p.119).

One study that used a career-self management measure recently was conducted by

Sturges, Guest, & Davey (2000). In this study career self-management was measured by

asking the research participants the extent to which they had practiced or intended to

practice a set of career self-management behaviors (e.g. developing built contacts with

people in other areas that they would like to work). Further this study measured the

relationship between career self-management techniques and organizational commitment.

The findings of the study suggested that recent graduates perceived career-self-

management techniques should be used in addition to participation in organizational

career management techniques (e.g. training programs, formal mentoring programs).

Specifically, participation in organizational career management techniques, that is,

participation in more formal organizational activities was related to higher organizational

commitment, especially when the employees first join the organization. In comparison,

the employees that had been with the organization at least 8 years were more concerned

about participating in informal organizational practices (e.g. being introduced or

networking with people).

The Sturges et al (2000) study suggests that individuals need various kinds of

career self-management guidance depending on the length of tenure with the

organization. Perhaps the individuals in Sturges et al (2000) study desire to build their

66

human capital skills (e.g. participation in training) when they initially enter the company

and develop their social capital (e.g. participation in networks) after becoming adjusted to

their work environment and job responsibilities.

Careers and Gender

For some time, researchers have studied the influence of gender on promotions

and career mobility. Career mobility, that is, a movement between jobs results in an

individual acquiring additional skills and broader job experience. Previous studies have

addressed why women and men experience variance in their career mobility patterns. For

example, research suggests that men are advanced at a faster rate than women (Ragins

and Sundstrom, 1989). Although differences related to gender have been noted, previous

research has not been able to clearly determine why those gender differences persist, nor

why gender (in comparison to tenure) explains a large amount of variance in promotion

rates. Thus, the interesting question becomes, why does this occur and what is the role of

networking in career mobility?

The two elements that appear to be important to en employee’s career

advancement include human and social capital. Human capital is the accumulation of

knowledge and skills over time (Becker, 1993). Examples of human capital variables

that are important to careers (i.e. career success) include mental ability and educational

attainment (Melamed, 1996). Mental ability is said to positively impact the pace at which

job knowledge is acquired and educational attainment enables individuals to access

higher paying and higher status jobs, thereby leading to career success (at least objective

career success, that is, when salary is used as an indicator of objective career success. In

comparison, social capital can be defined as the ability of individuals to reap the benefits

67

of membership in social networks or other social structures (Portes, 1998). Networking

within organizations is one example of enhancing one’s social capital. Human capital (an

employee’s knowledge and skills) is expected to help employees gain entry into

organization, while social capital is expected to help employees to middle and senior

management (Metz and Tharenou 2001). Of interest, Metz and Tharenou (2001) found

little support for the notion that networks facilitated the career development of women.

Instead, Metz and Tharenou (2001) found internal networks were negatively related to

the career advancement of women at the middle and senior level management positions.

Consistent with this view, additional studies have found that women, specifically in

senior-level management, were excluded from informal groups, thereby having limited

access to information that was being exchanged (Davis-Netzley, 1998; Moore, 1988;

Swiss, 1996). Also, Lyness and Thompson (2000) found exclusion from informal

networks was a perceived barrier of success for female executives in comparison to

males. When asked to identify the factors necessary for success in executive level

positions, women reported that networking with men in powerful positions was the most

significant requirement for success in executive level positions (Davies-Netzley, 1998).

Researchers have suggested that the influence of organizational context on the

persistent of the glass ceiling, is a key area for future research (Chernesky, 2003). Glass

ceilings are one example of career barriers often described by researchers and

practioners. The glass ceiling is described as a non-job related barrier that individuals

face (London, 1993). Specifically, the glass ceiling occurs when people are denied access

to a career goal or negative decisions made about them on the basis of their race or

gender. Therefore reaching the glass ceiling suggests a career barrier of not being

68

promoted beyond a certain point and it is a common barrier faced by women (London,

1993). In addition to the glass ceiling phenomena, another example of a career barrier

faced by women and parents is the multirole conflict barrier. The multirole conflict

barrier refers to problems or demands from one’s family (spouse or children) or friends

that may distract the individual’s attention from working toward a career goal (London,

1993). It is important for researchers to address various career barriers (e.g. the glass

ceiling phenomena), as research suggest individuals will have an emotional reaction to

career barriers. That is, strong emotions interrupt an individual’s thought process and

behaviors; this influences the extent to which people can rationally interpret a career

barrier and determine a constructive course of action (London, 1993). Thus we need to

develop a more comprehensive model of career barriers that considers challenges specific

to women. Although outside of the scope of this article, emotions should be included in a

more comprehensive model of career barriers (e.g. how to balance work-family demands

while also meeting individual career goals).

This dissertation seeks to understand the attitudes of working parents specific to

their ability to use organizational networks to learn specific job or career-relevant

information. An example of an attitudinal measure that will be used in this dissertation

draws from Parker’s et al. (2004) study. In this study, Parker measured individual’s

perceptions or attitudes towards members in their organizational networks and the extent

to which these members influence their careers. The measure used was called knowing

whom and it was used to assessed the relational aspect of the individual’s career

situation. Additional detail will be provided in Chapter 4, regarding all measures that will

be used to assess the attitudes of individuals toward their networks and their careers.

69

Careers and Parental Status

Previous research has examined the impact of parental and marital status on

career advancement. In a recent literature review it was reported that work-family

conflict may be negatively associated with less career satisfaction and lower perceived

career success (Eby et al., 2005). Empirical evidence also seems to suggest career

outcomes may vary across parental status. For example, Tharenou (1995) found childless

singles experienced less career advancement than married employees with or without

children. Tharenou (1995) also found that single men advanced less than married men.

This finding was consistent with the commonly held notion that single men has less

financial need than married men (especially if married men are a part of a single-income

household). Thus, since single men have less financial need than married men, they may

experience less career advancement. In short, the key finding from the study indicated

that married men and women, regardless of parental status, advanced more in their

careers than childless singles. This finding contradicts some of the prior literature that

suggests that employees with children have less time to devote to their careers, will

therefore experience less career advancement, in comparison to employees without

children. Of note, Tharenou’s (1995) sample was limited to early and middle level

managers and professional employees.

Tharenou’s (1995) findings are consistent with other studies that have found that

career advancement, measured in terms of salary and salary progression, favor married

men (e.g. Schneer & Reitman, 2002). That is, married men tend to earn more than both

single men and women. Married men may earn more than single men and women,

because their family status (e.g. husband, father) signals to organizations that they are

70

more stable (i.e. unlikely to leave the organization due to family responsibilities) and

demonstrate a higher financial need, especially if their spouse is unemployed.

Some studies have also measured the impact of children on salary progression,

that is, the career progression of individuals with children. For example, Miree & Frone

(1999) investigated the impact of children on career outcomes of MBA graduates. The

study found an effect for age. That is, working parents, regardless of gender, made higher

salaries (where salary is a commonly used measure for salary progression) when the

individuals had older children only (the study did not specify the age distribution of older

vs younger children). Therefore, working parents with younger children made less than

the employees with older children. One factor that may have contributed to the

differences in salaries between those workers with younger in comparison to older

children, was the age of the employee. It may be the case that the employees with older

children were older and had more work experience (hence higher pay) than the

employees with younger children. The interesting fact from this study was that the age of

the child matters. That is, parents seem to experience more challenges with younger

children in comparison to older children regarding career progression (i.e. salary

progression). As a result, this dissertation will focus on the challenges working parents

face with younger children as far as career outcomes are concerned.

Metz and Tharenou (2001) found that family responsibilities hindered the career

advancement of women more at lower levels of the organization in comparison to higher

levels within the organization. In other words, women at higher levels within

organizations do not perceive family responsibilities to burden their career advancement.

71

This may give some insight into women that decide to wait until later in their careers to

have children, after obtaining the higher managerial position they desire.

Valcour and Tolbert (2003) studied the impact of family and gender on careers,

specifically intra-organizational and inter-organizational mobility. Their results suggested

both gender and family characteristics impact mobility. For example, men were more

likely to move to new careers within the same organization (intra-organizational

mobility), while women were more likely to move across organizations (inter-

organizational mobility). Also, men’s intra-organizational mobility was positively

influenced by number of children, while women’s intra-organizational mobility was

negatively influenced by children (and positively influenced by marriage). Inter-

organizational mobility was unaffected by the presence of children for men, while

children positively influenced the inter-organizational mobility for women. It was not

surprising that children positively influenced inter-organizational mobility for women, as

women sometimes move between organizations after the birth of child, in order to find

work-arrangements (e.g. part-time work) that help meet their needs to balance their work

and family demands. In conclusion, the Valcour and Tolbert (2003) study provides

evidence that gender and parental status matter in terms of career mobility. This

difference seems to be most prevalent for women, regardless if they are moving within or

between organizations. In fact the women who obtained the greatest career success, in

terms of mobility within a single employer, were those who had previously been

divorced, and who were either childless or had fewer children.

The empirical results related to the impact of parenthood on networks are

inconsistent. For example, in an empirical study conducted in the public sector, Scott

72

(2001) found women with children were not significantly disadvantaged relative to their

colleagues in terms of networks. They were as likely as women without children to talk

with staff members and other government relations managers and top executives. Of

interest, one of the employees interviewed in the study reported that their organization

pays for baby-sitting services which allow the employees to attend social events after

traditional working hours (Scott, 2001). One possible interpretation of the Scott (2001)

study is that women are not disadvantaged by their parental status and in some cases

parental status may help working parents gain social capital, through their interaction

with other members of the organization who also have children.

Previous studies have shown that a change in family status does impact an

individual’s social network. For example, Kearns and Leonard (2004) found the social

networks of men and women were reshaped by marriage. Specifically, the networks of

married couples become more interdependent, as indicated by the increased overlap

between husbands’ and wives’ friends and family network. Arguably, this change in

networks, may lead to a reduction of nonredundant information shared within networks.

For example, if the husband and wife begin to interact primarily with an overlapping

group of individuals, this may suggest that they lose contact with individuals with whom

they communicated prior to marriage. This suggests that similar to a change in parental

status, a change in marital status may also reduce the number of direct ties within the

ego’s network. Consider a scenario where the wife loses contact with her college friends,

as a result of moving with her husband to a new city to take a job. The wife is no longer

in touch with her former college friends as a result of physical proximity. Further, from a

networking perspective it can be argued that wife has experienced a reduction in her ego

73

network, thereby leading to a reduction in nonredundant information related to job or

career opportunities.

Careers and Work-Family Concerns

Previous research has examined the relationship between family and career issues.

Most of the previous research has focused on the relationship between dual-career

couples and role conflict and the protean career. (Hall, 2002). Previous research has also

investigated the relationship between work-family and family-work conflict on (career)

withdrawal behavior, and the moderating effect of family involvement and career

involvement (e.g. Greenhaus et al., 2001). Of note, work-to-family conflict occurs when

involvement in a work-related activity, interferes with participation in a competing family

activity (Greenhaus and Powell, 2002). In comparison, family-work-conflict occurs when

involvement in a family activity interferes with participating in a work activity

(Greenhaus and Powell, 2002). A recent study found when work-family conflict is high,

leaving the professions that interferes with family life, is likely to reduce the stress of that

individual (Greenhaus et al. 2001). In comparison, an individual experiencing family-to-

work stress is not likely to withdraw from that profession because the stressor is related at

home, not the workplace (Greenhaus et al., 2001). Therefore, an individual’s intention to

leave a specific profession is related to the direction of interference between work and

family roles (Greenhaus, et al., 2001).

In addition, the Greenhaus et al. (2001) study found career involvement impacted

the individual’s decision to leave the organization, but family involvement had no impact

on the decision to leave. Therefore, when the employees were highly involved with their

careers, they were not disturbed, greatly, by the interference of work with family life and

74

they are willing to tolerate the interference for the sake of their careers (Greenhaus, et al.

2001). In contrast, the study did not find an interaction between family involvement and

intentions to leave the profession (Greenhaus et al., 2001). This may suggest that

employees who are highly involved with their families are concerned about maintaining

their position with the organization, providing them with the ability to financially support

their family.

Very little research has investigated the relationship between family status and a

relational approach to careers (as suggested in this dissertation). Most of the studies have

investigated the antecedents (and consequences) of work-family or family-work conflict.

For example, Evans and Bartolome (1981, 1984) found when work roles and home roles

come into conflict, work usually wins.

Some studies have looked at the relationship between family involvement and

career involvement. For example, Hall and Hall (1979) developed a typology based on

the four possible combinations of career and family involvement. This typology suggests

couples will experience conflict based upon two factors, career involvement and home

involvement. The group expected to experience the most stress are the couples that want

high involvement careers and high involvement family lives (referred to by Hall & Hall

(1979) as acrobats). Hall and Hall (1979) suggested that the typology is not static, that is,

a couple can move between various stages based on career and life stages development.

The Hall and Hall (1979) typology for dual career couple is based on role theory. Role

theory is often used in work-family studies to describe the conflict individuals experience

when they try to manage competing demands from both home and work. Thus, while

previous research has suggested the antecedents and consequences of role conflict among

75

dual-career couples, future research needs to investigate the relational aspects of careers,

that is, the intersection (and tradeoffs) between family and the role of relational networks

in careers.

Other studies have looked at the relationship between family structures and career

satisfaction. For example, Schneer and Reitman (1993) investigated the relationship

between family structure (e.g. marital status, parental status, and spousal employment

status), career and income satisfaction, and gender. This study was conducted on a

sample of MBA students and self-report data was used to measure multiple variables of

interest including employment status, income, career satisfaction, parental status, marital

status, etc. Schneer and Reitman (1993) found that families where both spouses were

employed with children, earned more than families with only one spouse working. This

result is not surprising as the number of dual income earners have increased, and both

men and women are working full-time and balancing parental responsibilities. Further,

Schneer and Reitman (1993) found that families in which both adults were working and

children were present in the home, were more satisfied with their careers than families

with only one parent working and children are present in the home. According to Schneer

and Reitman (1993) this finding was surprising as these individuals, that is families with

both parents working, depart from the traditional successful manager model (i.e. where

one parents stays at home to raise the children and the other parent works to support the

family financially). The authors suggest that the career success of dual-earner families

with children may be attributed to individuals being able to fill multiple roles (i.e. spouse,

worker and parent), which leads them to be more satisfied with their careers.

76

Another set of studies have examined the relationship between the birth of a child

and occupational mobility. These types of studies are important, because the birth of a

child is one of the major career interruptions for women (Brannen, 1989). That is, these

studies investigate if women experience a career interruption if they leave the workforce

temporarily after having children. For example, Brannen (1989) compared the careers of

two groups of women, those that resumed their full-time jobs after maternity leave with

their pre-birth employers to those women that moved to new employers after the birth of

a child. The study was a longitudinal study conducted over five years. Brannen (1989)

found that the women that remained with the same employer (even if they reduced their

work hours) were at “much less risk of downward mobility (Brannen, 1989, p186) and

were much more likely to be upwardly- mobile, that is, receiving a job promotion. In

comparison, for women that left their original (or pre-birth) employer and moved to a

new employer after returning from maternity leave, were more likely to experience lower

pay, than the women who returned to their original employer after returning from

maternity leave. Remaining with the same employer (i.e. pre-birth employer) after

returning from maternity leave appears to offer women an advantage (i.e. higher pay and

higher likelihood of an upward promotion). In addition to increasing the likelihood of an

upward promotion, the women who remained with their pre-birth employer after

returning from maternity leave were more likely to have more job security and a higher

number of paid holidays, than the women that took a job with a new employer after

retuning from maternity leave (Brannen, 1989). Based on the findings in the Brannen

(1989) study, it seems that the birth of a child may not impose as large of an interruption

on women’s careers as once thought. Instead the findings from the Brannen (1989) study

77

seemed to suggest that women can be successful in their careers (e.g. experience an

upward promotion) if they remain with the same pre-birth employer. Given this finding,

it would be interesting to identify the factors that contribute to the career mobility of

women after returning from maternity leave. Although outside of the scope of this

dissertation, it would be interesting to study the role of interpersonal relationships in

facilitating the career mobility of women after they return from maternity leave.

Presumably, if women return to the same pre-birth employers they are likely to have a

large number of existing interpersonal relationships (i.e. a networks with numerous co-

worker ties), depending on the length of tenure with their pre-birth employer. However,

the women that take jobs at new employers after returning from maternity leave, would

have to develop new interpersonal relationships. Therefore, they would have fewer

interpersonal ties (and less access to job and career-related information) than the women

that are returning to their pre-birth employers who are more likely to have a larger

number of existing ties (and thereby more access to job and career-relevant information

and a higher chance of experiencing career mobility).

Based on the empirical studies, literature reviews, and popular press articles, it is

fair to suggest that working parents face many challenges, which may deter their career

progression. For example, women are challenged by career interruptions, or time taken

off from work to stay at home with children. Broadbridge (2003) offered insight into the

challenges faced by employees in organizations, specifically working mothers. In

general organizations lack both formal and informal policies that will help working moms

address the challenges they face in trying to advance their careers and engage in their

family responsibilities. Formal policies (e.g. lack of childcare facilities, lack of flex-time,

78

long “after work” social hours) are held responsible for the perceived reasons why there

are so few women in senior management positions. However, organizational culture and

informal policies are also important in helping or hindering career development. In short,

work-family researchers often study to what extent formal policies impact an employee’s

ability to balance work and family responsibilities, without considering the role of

informal practices (e.g. developing or maintaining social capital), specifically related to

career development.

Moreover, it has been suggested that organizations and researchers may be

overlooking the significance of informal policies, specifically in understanding why the

myth of the glass ceiling phenomenon has persisted for women (Broadbridge, 2004). This

dissertation suggests that work-family researchers in the past may not have ignored the

challenges working parents face in terms of their ability to adhere to organizational

expectations set forth by informal policies (e.g. the importance of “face time” in

organizations, the importance of developing social networks in order to advance your

career). In short, employees’ use of work-family practices will hinder their career

advancement and success (e.g. taking an external parental leave or setting limits on the

number of hours worked) (Blair-Loy & Wharton, 2002). Also, many employees fear that

starting a family will also hinder their career.

This dissertation does not propose that all working parents participate in work-

family programs (e.g. condensed work weeks). Rather, it is likely that whether

employees participate in work-family programs or not, working parents will have less

time to allocate towards the development and socialization of organizational networks as

a result of career interruptions (taking time off from work soon after the children are

79

born), will spend less time at work (especially networking functions that take place after

work, conferences, etc), and when engaging in their organizational networks may spend

more time discussing their family-life (in comparison to career-related information). It

has been suggested in the existing literature that perhaps professional and managerial

employees with fewer family responsibilities can focus more single-mindedly on their

careers, spend more energy cultivating critical networks, and thus ascend higher or more

quickly through organizational ranks (Blair-Loy and Wharton, 2002). As such,

organizations and work-family researchers must address both the importance and barriers

employees face as a result of not complying with both formal and informal organizational

policies. This topic requires further investigation. For example, one may study if the

ability to develop or participate in organizational networking activities is determined by

the type of work-family program in which the employee participates. For example, are

those employees who participate in teleworking less likely to have an opportunity to

participate in networking activities than an employee whom works a condensed work

week?

Previous work-family research has been concerned with the utilization of work-

family practices and the negative consequences utilizing these practices may have on the

career development of working parents. For example, Lewis and Taylor (2000) found

some of the women interviewed in their study raised concerns that their participation in

alternative working patterns would negatively impact their career progression,

specifically in organizations that measure productivity in terms of time spent at the office

(i.e. being seen at work). One activity employees may engage in while at work is

interacting within organizational network groups. The opportunity to network as

80

previously stated allows employees to gain information, and it allows employees to be

seen as engaged in work activities by both peers and upper-level management. Also, if

working parents have less time to participate in organizational networks, they may have

less information readily available to them, providing that individuals leverage

organizational networks to learn about job/career opportunities. It is important for w-f

researchers to consider the impact of gender on the career advancement of working

parents. Empirical evidence suggests that working moms are penalized by slower

advancement and reduced wages in comparison to women without children, while

working fathers experience more advanced career and salary progression, then men

without children (Tharenou et al, 1994). In contrast, Tharenou (1999) found, in a

longitudinal assessment, that working moms experienced more career advancement than

women without children. It appears that the empirical evidence is inconclusive in terms

of the impact of children on the career advancement of employees. Therefore, the role of

organizational networks in advancing the careers of employees may differ by both gender

and parental status.

Job and Family Involvement

Previous research has investigated the work-family conflict individuals

experience as a result of participating in dual roles in their lives. Role involvement is

thought to lead to conflict among individuals because (1) high levels of involvement in

that role may lead to increased amount of time spent in that role, therefore allowing less

time to be allocated to the second role (Greenhaus & Beutell, 1985). In addition, high role

involvement in one role may lead to individuals being mentally involved with the most

prominent role in their life, even when one is physically trying to fulfill the demands of

81

the second role in their lives (Greenhaus & Beutell, 1985). Two measures commonly

used to assess the importance of a specific roles in an individual’s life include job

involvement and family involvement (e.g. Lodahl and Kejner, 1965; Kanungo 1982). Job

involvement is a measure of how important a job is to an individual’s life. That is, a job

involved person is “one that views work as an important or central part of their lives”

(Lodahl & Kejner, 1965). In addition, family involvement is similar to the job

involvement measure, that is, it is a measure of how important an individual sees family

as important to their life. That is, job involvement is generally operationalized as the

extent to which one indicates job-related activities or the job itself to be of central and

unique importance in their lives, and a key source of personal identity (Reeve and Smith,

2001).

There are many factors that are thought to contribute to job involvement including

opportunities for promotion, financial reward, job security and supportive relationships

with coworkers and supervisors (Lodahl and Kejner, 1965; Lambert, 1991). Related to

job involvement, at least two themes have been identified in the literature. The first is the

notion of whether job involvement differs by gender, that is, are males or females more

likely to be involved with their job. Examples of this research include Golembiewski

(1977) who argued that work is perceived as more central by males in comparison to

females. This ascertain would suggest that males are more involved with their jobs than

females. A study by Lorence (1987) aimed to understand if gender differences exist by

gender. This research was carried out in two waves and was part of a measure, The

Quality of Employees Survey. Findings from the study included, (1) marital status and

the number of children did not affect job involvement among males or females, but (2)

82

overall women were less involved with their jobs than men. Contrary to what the authors

expected, the gender differences found in job involvement did not have any relationship

with family responsibilities. Instead the job involvement differences by gender were

related to women having less autonomy in their jobs or poor working climate.

Another theme that is related to gender differences includes differences between

gender by marital status. That is, are married men more/less likely to be involved with

their jobs in comparison to single men; and are married women more/less likely to be

involved with their jobs in comparison to single women? Consistent with this view,

studies have found married men are more involved with their jobs than single men (e.g.

Agassi, 1982). However, other research suggests that women, especially married women,

are more committed to their jobs (especially the longer they work) (Haller & Rosenmayr,

1971). Consistent with this view, married women were found to be more committed to

their jobs than single women (Agassi, 1982). Taken together, these studies suggest that

although gender differences may exist, there may be additional factors that influence the

job involvement of individuals, marital status, notwithstanding.

Previous work-family studies have investigated the relationship between role

involvement and work-family conflict. For example, Frone and Rice (1987) investigated

whether family involvement moderates the relationship between work-family conflict and

job involvement. The authors found that job involvement was positively related to job-

parent conflict, regardless of parental involvement. In addition, in a recent literature

review Eby, Casper, Lockwood, Bordeaux & Brinley (2005) reported several studies that

measured job involvement and various career-related outcomes. For example, Gould and

Werbel (1983) found job involvement (and organizational identification) to be lower

83

among males with working spouses in comparison to males without working spouses.

Further, for the individuals who had both working spouses and children, job involvement

was higher. Also, a study by Lambert (1991) also investigated the impact of several

extrinsic factors (e.g. opportunities for promotion) and intrinsic factors (e.g.

meaningfulness of work) on job involvement. The findings from the study suggest there

is very little difference between the job involvement of men and women. However certain

factors such as career-related rewards (e.g. opportunities for promotion) were more

important to men, but factors related to social rewards (e.g. having a good relationship

with co-workers) were more important to the job involvement of women.

These findings suggest (1) others factors besides gender are related to job

involvement and (2) job involvement is related to interpersonal relationships at work.

That is, it appears that one of the factors that contributes to high job involvement is an

individual’s relationships within their organizations. For the purposes of this dissertation,

job involvement will be measured as a moderator between the ego’s network

characteristics and parental status.

Weak Tie Theory

Granovetter’s (1973) weak tie argument is one of the most highly cited papers

regarding how individuals can best mobilize their networks. Granovetter’s (1973)

argument suggests individuals should maximize the number of nonredundant ties within

their network, as this will lead to the individual learning more information about future

job and career opportunities. Granovetter argues weak ties are sources of better

information, because they offer new information. In short, this study uses Graonovetter’s

(1973) paper as a basis to argue that the birth of a child will result in employees spending

84

more time with kin and less time with organizational members, and their overall size of

their social networks will decrease. A reduction in the size of an ego’s networks will lead

to redundant information related to job and career opportunities being shared within

networks. That is, as the ego’s network size reduces, they will have less members in their

network. Granovetter argues that individuals should maximize the size of their networks,

to increase the likelihood that they will learn non-redundant information from the

members within their networks.

Although Granovetter’s recommendation suggests that an individual should

maintain multiple nonredundant weak ties, there is a challenge in maintaining weak ties

that should be addressed. Higgins and Kram (2001) suggest that an individual must

communicate or reach out frequently to their weak ties. If individuals do not reach out

frequently to their weak ties then they may only receive help from members in their

network when it is offered and they do not utilize their network effectively to gain

information (some people are simply not willing to dedicate the time it requires to

maintain contact with multiple individuals within their network). This is especially true

for parents whom have less time to invest in maintaining those multiple contacts.

Working parents are not likely to have time to dedicate to maintaining multiple weak ties

(meaning they are in touch with these individuals at least once a week).

In addition, to Granovetter’s weak ties argument, Burt (1992) describes the

benefits of networks through his structural holes argument. Burt argues individuals

should focus on the pattern of relationships within their network. Specifically, Burt

(1992) suggests that people create an efficient network by maximizing their structural

holes, or the distance between two alters (individuals) within a network. Burt (1992)

85

defines structural holes as absence of connection between among those in the network,

suggesting that the more the ego (individual) is surrounded by structural holes, the more

likely they are to experience mobility. Burt (1992) similar to Granovetter argues that the

larger the network, that is, the larger number of direct, nonredundant ties, should lead to

an individual experiencing more upward career mobility. Therefore, both Granovetter

and Burt suggest that the ego should maximize the size and nonredundancy of their

network.

Burt argues that networks provide employees with at least two advantages. First,

employees who are embedded in networks have access to people and information. Also,

networks provide employees with just-in-time information. Stated differently, Burt

(1992) argues that timing is a significant feature of information received by networks.

Beyond making sure that you are well informed, personal contacts can make sure you are

one of the people that are informed early. Therefore, personal contacts get your name

mentioned at the right time in the right place so that opportunities are presented to you.

Burt’s theory also suggests that individuals should not build a network around an

immediate supervisor. Instead the greatest benefit of a network is with people completely

removed from the immediate group.

Burt (1992) argues that structural holes exist when the alters of a network do not

know each other. Thus, Burt’s structural holes argument suggests that structural holes

create a situation in which information is additive rather than overlapping. If you have a

number of weak ties but they all know the same information; then you will have

redundant information being shared rather than new information.

86

Redundancy comes in two forms: (1) cohesive contacts – contacts strongly

connected to each other are likely to have the same information and (2) Structurally

equivalent contacts (contacts linked the same third party). In summary, the key tenets of

Burt’s Structural Hole theory are:

All individuals have potential to develop social capital

Focus on patterns of relationships within their network

Efficient networks maximizes nonredundant contacts

Structural hole exists when two alters within a network are not

connected with each other.

Individual should be connected to many alters who themselves are

not to connected to other alters in the ego’s network. Leads to

additive rather than redundant information

Previous studies have used the weak tie theory as a basis for understanding the

role of social capital and career outcomes. For example, a study by Seibert et al (2001),

examined a conceptual model where two measures of social network structure, weak ties

and structural holes, were thought to be related to two social resources, the number of

contacts in other functional areas and the number of contacts in at higher organizational

levels. This model used social network structure and social resources to predict career

success (i.e. current salary, the number of promotions received over the career and career

satisfaction). The findings from this study suggested that social capital is important to

career success. Specifically, there was a strong relationship between weak ties and a

greater number of organizational contacts at higher levels (r=.44). This suggests that

individuals with weak ties (i.e. multiple, nonredundant ties) are likely to have multiple

87

contacts at higher levels. Further, social resources were positively related to current

salary, number of promotions over the career, and career satisfaction. Of note, none of the

paths from structural holes to network benefits (e.g. access to information, access to

resources, and career sponsorship) were significant. However, two of the three paths

from weak ties to structural holes (access to information and career sponsorship) were

significant. Therefore, this suggests that the weak ties argument supports the relationship

between social capital and various career outcomes, that is, the weak ties measure

appears to have a stronger and more robust effect on social resources (Seibert et al.,

2001). This finding is consistent with the use of the weak tie framework as a basis for this

dissertation. That is, this study demonstrates that the relationships between an

individual’s network and various career outcomes is significant and should continued to

be analyzed in future studies (hence this dissertation).

Boundary Theory

Boundary theory is useful for understanding how individuals move between their

work and family roles. Roles are specific forms of behavior associated with given

positions, and they develop from task requirements (Katz & Kahn, 1978). There are at

least three core characteristics of boundaries including they are flexible, permeable, and

directional (Hecht et al., 2004). Specifically, the flexibility characteristic of boundary

theory suggests that physical time and location markers can change (e.g. working hours).

The permeability characteristic of boundary theory describes the extent to which a person

that is physically present in one domain, is psychologically concerned with another

domain in their lives (e.g. a parent at work that is concerned about a sick child that is at

home with a babysitter). Finally, the directionality characteristic of boundary theory

88

describes the extent to which an individual defines the strengths of their boundaries, that

is, there is variance in the extent to which an individual allows work to cross over to

home, in comparison to the extent to which an individual allows home to cross over to

work (Hecht et. al., 2004). For example, a person may have a strong boundary, where

they do not allow their personal life to enter their work environment. At the same time, an

individual may have a weak boundary between work and home, where they allow the

office to call them at work after regular business hours. In general, an individual with

strong boundaries, generally prefers to segment their work and family lives, and the

boundaries between work and home are usually inflexible and impermeable (Hecht et al.,

2004). Meanwhile, an individual that enjoys weak boundaries, that is, a person that

prefers to integrate their home and work life, establish boundaries between home and

work that are both permeable and flexible (Hecht et. al., 2004).

Boundary theory is a theory that demonstrates that individuals have preferences

for the extent to which they desire two roles in their life, work and home, to be integrated

or segmented. A individual’s preference for integration or segmentation are along what

researchers describe as the integration- segmentation continuum. Segmentation refers to

the separation of work and home roles, while integration refers to inclusion of work and

home roles. Examples of segmentation include not displaying pictures of families in

offices, while a person that desires to integrate their work and family roles would display

pictures of family in their offices. Also, segmentors would no likely take work home,

instead they would rather finish it in the office. Meanwhile, integrators, are ore likely to

take work home (Rothbard et al., 2005). Integration and segmentation lie on the same

continuum, and instances of complete segmentation or complete integration are rare

89

(Rothbard et al, 2005). The primary reason individuals choose to integrate or segment

their work and home roles is to minimize the difficulty of participating in both roles

(Rothbard et al, 2005). That is, individuals are looking for a coping mechanism that

allows them to manage the difficulty of holding multiple roles in their lives.

The relevance of boundary theory to work and family research is to understand

how individuals engage in daily role transitions, and the psychological movement

between roles, from role exit to role entry (e.g. leave work and coming home to parenting

role) (Sutton and Noe, 2004). Role behavior is the recurring actions of an individual,

appropriately interrelated with the actions of others, and the best criterion for

understanding role behaviors is to study the expectations of a specific role (Katz & Kahn,

1978). Boundaries are physical and temporal limits that help individuals conceptualize

two entities, work and family, as separate from one another. Role boundaries specifically

describe how individuals make the distinction between various roles in which they are

engaged (i.e. employee, parent). Finally, role identity describes a social construction

where individuals use various cues (i.e. goals, values, beliefs) to identify their occupancy

in a particular role. Ashforth et al (2000) made four key assumptions regarding their

model of role transition. The first assumption was that assumed roles are relatively stable.

Second, there is variance among individuals in terms of the actual number of roles they

prefer to enact. Third, individuals vary in their preference for role segmentation or role

integration. Finally, people seek to minimize the difficulty associated with role

transitions.

Role segmentation suggests that there are large difference in the roles experienced

by individuals at work and home. Further, given the large discrepancies in roles, it is

90

unlikely that the various roles will influence one another; a “mental fence is drawn

around each identity” (Ashforth et al, 2000; Clark, 2000). In integrated roles, there is

hardly any difference between the roles of individuals at home and in their workplace.

For example, an individual who enjoys having their work and family roles integrated

would likely enjoy working from home.

According to boundary theory, individuals have preferences for the extent to

which they want their work and home life integrated or segmented. For example, Kossek

et al. (1999) theorized that the preferences individuals have for the extent to which they

want their work and home roles segmented or integrated may vary as a result of gender or

family status. Examples of segmentation include not displaying pictures of families in

offices, while a person that desires to integrate their work and family roles would display

pictures of family in their offices. Also, individuals that segment their work and homes

lives, would no likely take work home, instead they would rather finish it in the office.

Meanwhile, integrators, are ore likely to take work home (Rothbard et al., 2005).

Integration and segmentation lie on the same continuum, and instances of complete

segmentation or complete integration are rare (Rothbard et al, 2005). The primary reason

individuals choose to integrate or segment their work and home roles is to minimize the

difficulty of participating in both roles (Ashforth et al. (2000) & Rothbard et al. (2005)).

That is, individuals are looking for a coping mechanism that allows them to manage the

difficulty of holding multiple roles in their lives. For example, individuals that desire

segmentation between their work and home roles may want to avoid the spillover of

negative emotions (e.g. tension at work) from one domain (e.g. work) to another domain

(e.g. home) (Rothbard et al, 2005). In addition to individual preferences, it should be

91

noted that the current policies organizations offer to help employees manage their home

and work responsibilities also fall along the segmentation-integration continuum

(Rothbard et al, 2005). For example, an employer that offers on-site daycare is an

example of an organizational policy that allows for integration of work and home.

An interesting question that this study raises is whether an individual’s

preference for integration and segmentation of work and family roles, contribute to the

degree to which individuals share work-family concerns within their networks. This

dissertation draws on boundary theory to suggest that working parents may vary in the

extent to which they discuss family-related issues within their organizational networks.

As suggested previously, one of the elements within organizational networks that are

expected to change after the onset of parenthood, is the content of the discussion between

working parents and members within their networks.

This dissertation argues that the change in content of discussions of working

parents within their networks may be moderated by the extent to which the parents

choose to integrate or segment their work and family lives. This idea is consistent with

Ashforth et al’s (2000) proposition which stated that the greater the role segmentation,

the less difficult it tends to be to create and maintain role boundaries but the more

difficult it tends to be to cross boundaries. Consistent with boundary theory, for those

working parents that desire to integrate their work and home lives, the content of their

conversations within their organizational networks should change and reflect more

discussion related to family issues. In comparison, for those working parents whom

prefer to segment their work and home lives, the content of the discussion within their

organizational networks is not expected to change as a result of the birth of a child.

92

Although, not directly related to working parents, this next illustration provides an

example of two individuals that have different levels of desirability to segment their work

and home lives. This illustration speaks to the point that there is variance in the degrees

to which individuals want their family and work lives to overlap. Further, this dissertation

will argue that these individual differences to segment work and family roles may be

more severe for working parents.

An example of the differences in conversation topics and the extent to which an

individual wants to segment their work and family lives is seen in this next illustrative

example found in the Nippert-Eng (1996) article.

93

In this scenario, John clearly intermingles home and work dimensions, while Ed does not.

“John was married to his scientific collaborator. So not only did he freely do and discuss his work with his wife but also he shared much of what others might consider to be his private life with his coworkers. Also, John frequently could be found in his laboratory any day of the week, …and at other times he could be found working at home, reading work materials, puzzling out a problem, planning strategies, and discussing his work with his wife. His work days were also riddled with quick, random phone calls or in-person conversations with his wife about domestic issues and plans” (Nippert, 1996, pp 565-566). “Ed dislikes talking about work with her (his wife), seeing no reason to expose her to much of what goes on there (at work) or the boring details of work. When he gets home, he wants to forget about work, not drag it into the living room behind him. Ed also steadfastly refuses to divulge personal information to his coworkers. He firmly believes that the less other workers know about his life outside of the Lab, the less than can use this information against him. Ed, clocks in and out of work at the precise times mandated by his contract and vehemently insist on his right to evenings and weekends with his family. He never thinks (or talks) about work while he is at home and literally never socializes with colleagues, mush less invite them to his house. His wife only calls work and during breaks in the day, what Ed calls ‘my’ rather than the ‘Lab’s’ time (Nippert, 1996, pp 565-566).

94

CHAPTER 3

CONCEPTUAL MODEL AND HYPOTHESES DEVELOPMENT

Drawing from the relational view of careers, individuals are largely responsible

for managing their own careers. Moreover the relationship(s) they develop, that is their

network, will be a key component of career self-management (Hall, 1996). Further,

Higgins and Kram (2001) argue that individuals must be focused on the development

(and maintenance) of multiple, concurrent relationships. That is, the more people or ties

an individual has, the better they will be able to manage their careers. Relationships are

important to an individual’s career because they are a key component by which an

individual becomes aware of job and career-related information. To maximize the amount

of information an individual has about job and career-related information, they must

develop weak ties, that is, a large number of individuals that provide then with

nonredundant job and career-related information (Higgins & Kram, 2001; Granovetter,

1973). Social networks enable an individual to develop multiple, concurrent

relationships, and they allow individuals to exchange job and career-related information

(Forret& Doughterty, 2001).

This dissertation argues that networks are key to an individual acquiring career

and job-related information that will assist them in managing their career. This same idea

is captured by the social capital argument, that is, individuals making an investment into

multiple relationships for the purpose of gaining information (Lin, 2001). To better 95

understand the relationship between networks and career mobility Sullivan (1999)

suggested using a social networks framework (e.g. Burt’s Structural Holes or

Granovetter’s Weak Ties theories) conduct future research on careers, specifically the

relational aspects of career mobility. This dissertation also speaks to a research need

addressed by Sullivan (1999) in which it was suggested that individuals characteristics,

such as gender, age, and race, need to be investigated in terms of the overall

development of large non-redundant networks. Further, Sullivan (1999) suggests that

researchers need a better understanding of the factors that contribute to an individual’s

success in managing a boundaryless career; this dissertation suggests networks (i.e. large

and nonredundant networks) are a key factor to helping an individual successfully

manage their careers.

One way to describe an individual’s network is to identify key measures that are

used to describe the network. The specific network characteristics of interest to this

dissertation include network size, network ties, and network content (i.e. the topics

discussed between the ego and important people in their network). Specifically, network

size is a measure of the absolute numbers of ties within an individual’s network. Network

ties, is a measure of the type of ties individuals have in their network (e.g. co worker-ties,

kin ties, friend ties). Finally, network content describes the conversations individuals

have with important people within their network (e.g. from “work talk- to “family-talk”).

In describing specific network characteristics, it also makes sense to discuss

various factors that may impact specific network characteristics, specifically network

characteristics that are important to an individual’s career. When careers are studied, one

of the most common network characteristics studied is network size, that is, the number

96

of ties an individual identifies within their network. Several studies have demonstrated

that certain factors impact an individual’s network size. One of the most common factors

thought to impact network size is gender. For example, gender seems to impact the

composition of ties that comprise an individual’s networks. Mardsen (1990) found the

networks of men include mostly coworkers and volunteer ties, while the networks of

women tend to include more kin and neighbors ties. Thus, it can be concluded that

gender will interact with network size, which usually results in men having larger

networks than women.

In addition to gender, parental status also seems to impact specific networking

characteristics, especially network ties. Network ties, is the type of relationship (e,g, kin,

parent) that an ego has with the individuals they identify as part of the network. In

comparison to adults without parental responsibility, adults with parental responsibility

often have a higher proportion of kin ties in their network. For example, Bost, Cox,

Burchinal & Payne (2002) found that after the birth of a child, parents reported a decline

in the number of friend ties in comparison to the number of family ties within an

individual’s network. Consistent with this notion, as the number of kin ties (in

comparison to friend ties) increased, the respondents reported having more contact with

their kin ties in comparison to the friend ties. In short, after the birth of a child, the

respondents reported spending less time with their friends, in comparison to their family

members.

The examples provided above, suggests that network characteristics, such as

network size, tend to vary by individual differences including gender and parental status.

The first purpose of this dissertation is to understand the factors that cause the

97

relationship between parental status and the three network characteristics size, ties, and

content to vary. To accomplish this, this dissertation will identify and test four

moderators believed to influence the relationship between parental status and the three

network characteristics. The four moderating variables that will be studied include

gender, family involvement, role segmentation, and job involvement. Secondly, this

dissertation seeks to examine the relationship between the three network characteristics

and career outcomes. That is, this dissertation seeks to understand if differences in the

network characteristics leads to differences in the career management and career success

indicators included in this study.

This study begins by identifying the variables that are thought to moderate the

relationship between parental status and the three network characteristics, network size,

network ties, and network content. As mentioned previously, those variables include

gender, family involvement, role segmentation, and job involvement. However, in order

to begin the discussion of the antecedents that will moderate the relationship between

parental status and network characteristics, it makes sense to briefly describe why the

three network characteristics (size, ties, and content) were included in this study.

Following that discussion, the next section will describe the four moderators thought to

impact the relationship between parental status and network characteristics.

To identify the network characteristics that should be included in this study, a

literature search was conducted to identify the network characteristics in previous

literature that varied across parental status. Most notably, the network characteristics that

appear to differ across parental status include network size and network ties. Specifically,

research has demonstrated that the network size of adults after the birth of a child

98

decreases (e.g. Smith-Lovin). That is, the network size of adults with children, tends to be

smaller than the network size of adults without children. In addition, to network size,

another network characteristic that appears to differ across parental status is network ties.

Specifically, previous research also supports the notion that after the birth of a child,

contact with family increases and the percentage of kin ties within a network also

increases (e.g. Belsky & Rovine, 1984). Also, Munch et al., (1997) found that kin ties

accounted for 70% of the network ties (i.e. composition) within the ego’s network during

the first 4 years after the birth of a child. This increase in kin ties after the birth of a child

is usually consistent with a parent’s need, for additional emotional support from family

members. Thus, the proportion of kin ties is likely to be higher in the network of a

parent, in comparison to the network of an adult without parental responsibility.

In addition, to the two network characteristics network size and network ties

previous research has demonstrated are likely to differ across parental status, this

dissertation also investigates differences across parental status in network content; where

network content is a variable that has yet to be tested in empirical research. Network

content are the topics of conversation that individuals are discussing with the people they

have identified as part of the network (Bearman & Palgi, 2004). There is no previous

evidence to suggest that network content differs across parental status. Rather, network

content was included as an exploratory variable, as anecdotal evidence suggests that

parents tend to discuss family and children-related topics more than adults without

parental responsibility. Including network content into this dissertation, makes a

contribution to the field by introducing a new variable that is thought to vary across

parental status. Although the relationship between network content and parental status

99

has not been studied, previous research does demonstrate that network content is

important. Specifically, network content is an important measure, as this dissertation

assumes that the topics of conversations an individual discusses with members of their

network, is an indicator of the type of information exchanged between the ego and the

members within their network, and it is an indictaor of the matters that individuals find

important. That is, individuals talk about matters that are important to them with

members in their network (Bearman & Parigi, 2004). Social capital theory (Lin, 2001)

tells us that there is a direct link between the type of information shared amongst a

network, and specific career outcomes (e.g. career success or career advancement).

Specifically, individuals that leverage their networks to gain job and career-relevant

information are likely to experience career success.

If the assumption is made that network characteristics will vary across parental

status, the real question to address is, what variables cause the differences between

parental status and the three network characteristics that have been identified in this

dissertation? To address this issue, a conceptual model was developed, see Figure 1. The

conceptual model developed for this dissertation, demonstrates the following

relationships:

100

H2 H3

H4

H6

Family Status Differences Ego’s Org Network Characteristics

Size Ties Content

Role Segmentation

Gender Family Involvement

Career Management

Job Involvement

“Network Constraints”

H5

Career Success

Figure 1: Conceptual Model of The Relationship Between Family Status Differences, Network Constraints, Job Involvement, Career Success, and Career Self-Management

First, this model shows the relationship between parental status and network

characteristics (e.g. network size, ties, and content). This relationship suggests that

parental status will causes differences in the network characteristics included in the

model. That is, network size, network ties, and network content will differ for working

adults with parental responsibility in comparison to working adults without parental

responsibility. The model then shows the antecedents that are expected to moderate the

relationship between parental status and network characteristics. These antecedents

include gender, family involvement, role segmentation, and job involvement. The

hypotheses tested in this study assume that each of the antecedents will have a separate

relationship with each of the network characteristics, size, ties, and content. Thus, for

each of the four antecedents, there will be three hypotheses (e.g. 1a, 1b, and 1c) tested.

Further, the hypotheses tested in the study assume that parental status will interact with

101

each of the antecedents, and impact the relationship that parental status has on each of the

network characteristics. Next, a discussion of each of the moderators and the expected

interaction with parental status is discussed. Following this discussion, the last section of

this chapter will describe the expected impact that each of the network characteristics of

interest, that is, network size, content, and ties, will have on the career success indicators

and career management perceptions included in the model.

Gender Constraint

Gender has received much attention in the career, social networking, and work-

family literatures (e.g. Brass, 1985; Ibarar, 1997; Burt, 1992). Some of the gender-related

conclusions presented earlier in the literature review, include the following observations.

First, networking behaviors tend to vary by gender. Men usually have larger, broader

networks and have more ties to top executives and people in higher positions (Ragins &

Sundstrom, 1989). Thus, men are more likely to leverage networks to meet specific

career goals (e.g. career mobility) (Ragins &Sundstrom, 1989; Cannings &

Montmarquetter, 1991). In comparison, women tend to benefit less than men from

participation in organizational networks (Brass, 1985; Davidson & Cooper, 1992); that is

networks do not seem to consistently offer women advantages in terms of career mobility

or advancement. Informal networks are often segregated by gender, leaving women at a

disadvantage.

Also, important topics of conversation within network groups, especially those

network groups formed within organizations, are often male-dominated topics (e.g.

sports). This leaves women feeling uncomfortable discussing topics related to family and

other female-gendered topics (Broadbridge, 2004). Women avoid talking about their

102

families at work, as they believe topics unrelated to work, will make others perceive they

are not focused on their jobs. Finally, women are more likely to have (relational) ties with

relatives and neighbors, while men are more likely to have (relational ties) with to

coworkers and volunteer groups (Mardsen, 1990).

In this study, gender is expected to interact with parental status, and influence the

three network characteristics (network size, ties, and content) included in this study.

There is evidence in the literature that gender is related to network characteristics, and the

relationship is moderated by parental status. For example, previous work (e.g. Ibara &

Smih-Lovin 1997; Munch, et al., 1997) has demonstrated that women experience more

significant changes in their networks than men after childbirth. Also, Munch et al. (1997)

found having children impacts a woman’s network size negatively, but it does not impact

the network size for men. Many of the differences in network characteristics related to

gender, found in previous studies, were thought to have occurred because of women

being the primary childcare provider. Because gender has been shown in previous

research to impact networking behavior, it is expected that differences in organizational

networks will be more salient for women than men. Therefore, it is reasonable to expect

that the networking characteristics of female parents will differ from the networking

characteristics of male parents. That is, the relationship between family status and ego

involvement is moderated by gender. The following hypotheses will be tested.

Hypothesis 1a: Working adults with parental responsibility will have a smaller network

(i.e. network size) than working adults without parental status; also among working

parents, working mothers will have a smaller network than working fathers (i.e. the

103

interaction between gender and parental status will result in a negative relationship with

network size).

Hypothesis 1b: Working adults with parental responsibility will have a higher proportion

of kin ties within their network than working adults without parental responsibility; also

among working parents, working mothers will have a higher proportion of kin ties within

their network than working fathers (i.e. the interaction between gender and parental status

will result in a negative relationship with network ties).

Hypothesis 1c: Working adults with parental responsibility will have a higher proportion

of non-work network content, than working adults without parental responsibility; also

among working parents, working mothers will have a higher proportion of non-work

content than working fathers (i.e. the interaction between gender and parental status will

result in a negative relationship with network ties).

Family Involvement Constraint

Recently, there has been an increase in the number of female workers, an influx of

dual-earner families, single-parent households, and employees managing the care of both

elder members and children. Also, the total numbers of hours worked by employees has

continuously increased over the last twenty years (Saltzstein, et al, 2001). As a result,

both women and men are participating in household responsibilities, including childcare.

Although gender has been investigated in relationship to differences in networking

characteristics (e.g. women have smaller networks than men, women have more ties to

family and men have more ties to coworkers), family involvement is an additional

104

variable that is likely to influence an individual’s network characteristics. That is, family

involvement is expected to moderate the relationship between parental status and network

characteristics.

Family involvement is a measure of how important an individual sees family as a

prominent aspect of their lives (Lodahl & Kejner, 1965; Parasuraman, et al., 1996).

Family involvement is an important measure to consider because it is expected that the

more individuals become involved with their families, they will have less time and

energy to allocate to other roles in their lives (i.e. work roles). Parenthood or family

involvement is likely to greatly influence the amount of time an individual, regardless of

gender, devote to their family role, and they are likely to dedicate less time to their work

role (Parasuraman, et al., 1996); and subsequently less time to maintaining work-based

network ties. Also, research supports the notion that characteristics of an individual’s

network change after the birth of a child. For example, Belsky & Rovine (1984) found

that contact with family increases when a child is added to the family; this will likely lead

to an increased frequency of contact with kin ties.

Thus, in consideration of the changing nature of the workforce (e.g. an influx of

dual-earner careers, more women entering the workforce), the conceptual model shows

that gender is likely not to be the only factor that contributes to differences in the

characteristics in networks of working parents. Rather, it is hypothesized that family

involvement will help explain differences in the characteristics in the networks of

working parents in comparison to working adults without parental responsibility. That is,

the conceptual model suggests that the role of childcare provider no longer includes just

women. Instead, the childcare provider role is being shared by both men and women.

105

Therefore, this model suggests that regardless of gender, any working parent highly

involved with their family will experiences differences in their network characteristics

(e.g. size, ties, and content of conversations).

Therefore, it is expected that the networking characteristics of individuals

involved with their families will differ from the networking characteristics of individuals

less involved with their families. That is, the relationship between family status and ego

involvement is moderated by family involvement. The following hypotheses will be

tested:

Hypothesis 2a: The parental status-network size relationship will be moderated by family

involvement. That is, family involvement will interact with parental status, such that

parents that are highly involved with their families will have smaller networks compared

to parents that are less involved with their families. (i.e. When family involvement and

parental status interact, there will be a negative relationship with network size).

Hypothesis 2b: The parental status-network ties relationship will be moderated by family

involvement. That is, family involvement will interact with parental status, such that

parents that are highly involved with their families will have a higher proportion of kin

ties in their network compared to parents that are less involved with their families (i.e.

when family involvement and parental status interact, there will be a positive relationship

with network ties).

Hypothesis 2c: The parental status-network content relationship will be moderated by

family involvement. That is, family involvement will interact with parental status, such

106

that parents that are highly involved in their families will have a higher proportion of

kin/non-work network content compared to parents that are less involved with their

families (i.e. when family involvement and parental status interact, there will be a

positive relationship with network content).

Role Segmentation Constraint

Another variable expected to influence an individual’s networking characteristics

is role segmentation. Boundary theory suggests that individuals differ in the extent to

which they prefer to integrate or segment their work and family roles (Ashforth et al.,

2000). When work and home are fully integrated, this suggests that there are no

boundaries which separate the contents or meaning between the two, and all time is

allocated to multipurpose between these two very salient roles (Nippert-Eng, 1996). In

comparison when home and work are conceptually viewed as two separate roles, there is

no overlap between theses two roles and each task a person is responsible for completing

(e.g. maintaining the members within their network) is clearly designated to a home or

work category, that is, two categories that are mutually exclusive (Nippert-Eng, 1996).

There are four key assumptions related to boundary theory (Ashforth et al., 2000). The

first assumption is that roles are relatively stable. Second, there is variance among

individuals in terms of the actual number of roles they prefer to enact. Third, individuals

vary in their preference for role segmentation or role integration. Finally, people seek to

minimize the difficulty associated with role transitions. Thus, individuals who like to

segment their work and family roles, enjoy roles that are not only highly differentiated

but are tied to specific settings and permit few interruptions across roles. That is, those

107

individuals that want their work and family roles segmented desire clear proximal and

psychological separation between the two roles.

Thus, drawing from role theory, the conceptual model suggests that individuals

vary on the extent to which they want their work and family roles segmented. The model

suggests that those individuals that desire to segment their work and family roles, will be

more likely to report the following about each of their network characteristics, compared

to those who do not desire to segment their work and family roles. First, individuals that

prefer to segment their work and family roles are likely to have smaller network sizes.

Also, individuals that prefer to segment their work and family roles, will have a higher

proportion of kin ties within their network. Lastly, individuals that prefer to segment their

work and family roles will have a higher proportion of non-work/kin-related conversation

topics (i.e. network content).

As suggested by the example provided in Chapter 2, (i.e., the example comparing

Ed and John), two individuals that vary on the degree to which they want their home and

work lives segmented, it’s clear that an individual’s desire to segment their work and

family lives will impact many of their daily practices. For example, in the comparison

made between Ed and John, several characteristics varied between them including their

topics of conversation among people. This illustration provides further evidence to

support the idea that an individual’s desire to segment their work and family roles will

impact various areas of their lives, conversation topics not withstanding. This dissertation

also argues that in addition to conversation topics that both an individual’s network size

and ties will also be impacted by their desire to segment their work and family roles.

108

That is, the relationship between family status and ego network will be moderated by role

segmentation. The following hypotheses will be used to test this proposition.

Hypothesis 3a: The parental status-network size relationship will be moderated by role

segmentation. That is, role segmentation will interact with parental status, such that

parents that clearly segment their work and family roles will have smaller networks

compared to parents that do not clearly segment their work and family roles. (i.e. when

role segmentation and parental status interact, there will be a negative relationship with

network size).

Hypothesis 3b: The parental status-network ties relationship will be moderated by role

segmentation. That is, role segmentation will interact with parental status, such that

parents that clearly segment their work and family roles will have a higher proportion of

kin network ties compared to parents that do not clearly segment their work and family

roles. (i.e. when role segmentation and parental status interact, there will be a positive

relationship with network ties).

Hypothesis 3c: The parental status-network content relationship will be moderated by

role segmentation. That is, role segmentation will interact with parental status, such that

parents that clearly segment their work and family roles will have a higher proportion of

kin/non-work related network content compared to parents that do not clearly segment

their work and family roles. (i.e. when role segmentation and parental status interact,

there will be a positive relationship with network content).

109

Job Involvement Constraint

As stated previously, the conceptual model assumes first a main effect for

parental status. That is, the model expects that the network characteristics, size, ties, and

content will differ across parental status. Secondly, this model assumes that several

moderating variables will contribute to the differences in network characteristics across

parental status. One of the variables identified by the model that is thought to moderate

the relationship between parental status and network characteristics is job involvement.

Job involvement is defined as a measure of how important a job is to an individual’s life.

That is, a job involved person is one that views work as an important or central part of

their lives (Lodahl & Kejner, 1965). As mentioned previously, job involvement is

generally operationalized as the extent to which one indicates job-related activities or the

job itself to be of central and unique importance in their lives, and a key source of

personal identity (Reeve and Smith, 2001).

Some of the factors identified in the literature that are related to high levels of job

involvement include, opportunities for promotion, financial reward, job security, and

supportive relationships with coworkers and supervisors (Lodahl and Kejner, 1965;

Lambert, 1991). Job involvement has been shown to be unrelated to gender (Lorence,

1987). However, job involvement has been shown to vary by marital status, but the

findings in previous studies have been inconsistent. That is, some studies have suggested

that married people were found to be more involved with their jobs than unmarried

individuals (Haller & Rosenmayr, 1971; Agassi, 1982). In comparison, Singh et al (1981)

examined workers in two separate agencies. Singh et al (1981) found that married

110

individuals were less involved with their jobs than unmarried individuals. Therefore the

relationship between marital status and job involvement is still unknown.

It is expected that individuals married or with children are less likely to be

involved in their jobs, due to an increase in their family responsibilities. When

individuals are less involved in their job, network size, ties, and content will each be

impacted. Specifically, individuals that are less involved in their job are less likely to

engage in networking behaviors that would result in developing a larger network. That is,

individuals are not likely to show fewer proactive behavior that would lead to increases

within their network size. Further, individuals that are less involved in their jobs are

likely to have a higher proportion of non-work ties that are included in their network.

Finally, individuals that are less involved with their jobs are less likely to discuss their

jobs with members in their network. Therefore, they are likely to have a larger proportion

of non-work network content (i.e. they will have a smaller frequency of conversation

topics related to work when they speak to members within their network). As a result it

is hypothesized:

Hypothesis 4a: The parental status-network size relationship will be moderated by job

involvement. That is, job involvement will interact with parental status, such that parents

that are not highly involved with their jobs will have smaller networks compared to

parents that are highly involved with their jobs. (i.e. when job involvement and parental

status interact, there will be a negative relationship with network size).

Hypothesis 4b: The parental status-network ties relationship will be moderated by job


that are not highly involved with their jobs will have a higher proportion of kin ties

111

within their network compared to parents that are highly involved with their jobs. (i.e.

when job involvement and parental status interact, there will be a negative relationship

with network ties).

Hypothesis 4c: The parental status-network content relationship will be moderated by job


that are not highly involved with their jobs will have a higher proportion of kin/non-work

network content compared to parents that are highly involved with their jobs. (i.e. when

job involvement and parental status interact, there will be a negative relationship with

network content).

In summary, this study suggests that three specific network characteristics will

differ between working adults without children and working adults with children. The

three network characteristics that will differ include size, type of ties, and topics of

conversations within networks. This study argues that four moderating variables will

contribute to the differences in the networks of working adults with children and working

adults without children. The moderating variables that contribute to the differences in the

networking characteristics of working adults with children and working adults without

children include gender, family involvement, role segmentation and job involvement.

Overall, the conceptual model the hypotheses were based on, suggests that in comparison

to working adults without children, working parents will have a smaller network size,

their networks will have a higher proportion of kin ties, and the content of the

conversations within their networks will include topics related to family matters (e.g.

112

childcare). As a result of the differences in the network characteristics of working

parents (i.e. size, ties, and content), this study suggests that working adults with children

and working adults without children will perceive the following career outcomes

differently; those career outcomes include perceptions of career success, career

satisfaction, and career self-management.

Perceptions of Career Success

Career success is defined as the accomplishment of desirable work-related

outcomes at any point in an individual’s work experiences over time (Forrett

&Doughterty, 2004). In previous studies, objective career success, that is, success that is

directly observable, measurable, and verifiable by a third party, is the measure most often

used to operationalize objective career success. Typical measures of objective career

success include salary, salary growth, and promotions. In comparison, subjective career

success lends itself to more of an internal satisfaction that an individual experiences

(Heslin, 2003). Specifically, subjective career success is defined by an individual’s

reaction to their career experiences and it is usually operationalized by a career

satisfaction or job satisfaction measure. Examples of measures used to capture subjective

career success include global career success or satisfaction with pay and promotions.

This study is interested in understanding the relationship between the network

characteristics (network size, ties, and content of conversation), and career success. This

is consistent with the relational approach of studying careers which is concerned with

how employees develop and leverage interpersonal networks, and to what extent

individuals uses the networks to assist them in obtaining information about new and job

and career opportunities. Specific to this study, career success is one of the two outcome

113

variables identified in the conceptual model. Previous research has demonstrated that

there is a relationship between specific network characteristics and career success. Seibert

et al. (2001) found weak ties and structural holes led to promotions and career

satisfaction. Other studies have found similar relationships between career success and

network ties (e.g. Granovetter, 1973; Burt, 1992). Consistent with previous research, this

study suggests that individuals with multiple work ties should have more career success;

resulting from the person having a greater number of resources from whom they can gain

information to help make decisions or job changes that will result in career success In

addition to network size, this study predicts that network ties and network content will

also have a relationship with career success. For example, if one considers the impact of

network content on career success, it is likely individuals whom speak with the members

of their network about career and job related topics will experience more career success.

However, if an individual speaks with members of their network about non-work or non-

career –oriented topics, these types of conversations are not likely to contribute to an

individual’s success.

Similarly, this model suggests that it is important to consider both what you’re

talking about with members in your network, and who you’re talking to within your

network. Stated differently, this model suggests that individuals that spend a lot of time

talking to non-work/kin ties, may have differences in the level of career success that is

achieved, when compared to an individual that spends time talking to members within

their network that are work ties, and when an individual is also talking about work-

oriented topics. This model suggests that parents are more likely than working adults

without parental responsibility to talk have a larger proportion of kin ties within their

114

network. This model also suggests that working parents are more likely to talk about

non-work topics with the members within their network, and working parents are likely

to have smaller networks. Therefore, the following network characteristics, size, ties, and

content will have relationship with career success (objective and subjective). It is

hypothesized that:

Hypothesis 5a: Network size will negatively influence career success, such that as

network size decreases, the objective indicators of career success (i.e. salary, salary

growth, and promotions) will also decrease. Further, there will also be a negative

relationship between network size and career satisfaction; where if network size

decreases, career satisfaction will also decrease.

Hypothesis 5b: There will be a negative relationship between network ties and objective

career success. That is, as the proportion of kin ties increases, objective indicators of

career success (i.e. salary, salary growth, and promotions) will decrease. Also, there will

be a negative relationship between network ties and subjective indicators of career

success. That is, as the proportion of kin ties increases, subjective indicators of career

success (i.e. individual career satisfaction and peer-related career satisfaction) will

decrease.

Hypothesis 5c: There will be a negative relationship between network content and

objective career success. That is, as the proportion of non-work content (i.e. types of

conversations) increases, objective indicators of career success (i.e. salary, salary growth,

115

and promotions) will decrease. Also, there will be a negative relationship between

network content and subjective indicators of career success. That is, as the proportion of

non-work content (i.e. types of conversations) increases, subjective indicators of career

success (i.e. individual career satisfaction and peer-related career satisfaction) will

decrease.

Perceptions of Career Self-Management

As shown in the conceptual model, a relationship is expected between the

network characteristics, and career self-management techniques (i.e. career planning,

career tactics, and career mobility preparedness). First, career self-management is a

process where employees gather information to help them make key decisions about their

careers; employees usually engage in career self-management to achieve one of two

behaviors, gaining developmental feedback or learning about career mobility

opportunities (Kossek et al., 1998). Developmental feedback involves individuals seeking

information about their performance and their career developmental needs. In order to

gather appropriate career-related information, as individual must first plan the type of

information that will be of help to their own career development (e.g. what are their

current career goals, what are their strengths and what kinds of jobs would leverage those

strengths), this aspect of career management is called career planning (Hall, 1990). In

addition to planning the type of information that an individual should gather, an

individual must next decide which specific tactics they will use to gather career-relevant

information and seek developmental feedback. This stage of career management is called

career tactics (Hall, 1990). While one career tactic that is commonly used is networking,

116

there are many other tactics an individual can use to gather career-related information

and/or seek developmental feedback including making themselves visible to people at

higher levels within their organization.

The second career self-management behavior, learning about career mobility

opportunities, is achieved through informal networking with members inside and outside

of the organization. Specifically, job mobility preparedness is “the degree to which an

individual prepares his or herself to be ready to act on internal and external career

opportunities” (Kossek et al., 1998, pp 939). Therefore, one way of gathering job related

information is through participation in networks, that is, the development of relationships

that will help individuals learn new career-related information; the possession of new

information helps individuals prepare to move either laterally or horizontally and assists

them in moving between or within organizations (Kossek, et al, 1998). As a result, it

important to evaluate, perceptually, the extent to which individuals feel they have control

over their career self-management. Also, it is important to understand how their network

characteristics (i.e. network size, ties, and content) are related to career management

perceptions.

This study suggests that the there is a relationship between the three network

characteristics proposed in the hypothesized model, and perceptions of career self-

management. That is, the model suggest that an individual with a larger network, that is a

larger network work with a higher proportion of work ties, will have better perceptions of

their ability to manage their career. In addition to network size, this model also expects a

relationship to exist between network content and network kin, and career management

perceptions. Specifically, this model suggests that individuals with high proportion of kin

117

ties in their network will be less likely to experience career satisfaction that an individual

with a smaller portion of kin ties within their network. This study expects that an

individual with a large percentage of kin ties in their network, are likely to engage in

activities that are non-work related, and spend a lot of time with their kin ties. As a result,

they will have less time to allocate to managing their careers, thereby reducing their

perceptions of their ability to manage their careers. Similarly, if an individual is

frequently discussing non-work topics amongst members of their networks, they are less

likely to gain career- or job-relevant information, which would be helpful in managing

their careers. Therefore, the following network characteristics, size, ties, and content

will have relationship with career management perceptions. It is hypothesized:

Hypothesis 6a: There will be a negative relationship between network size and career

management perceptions. That is, as network size decreases, an individual’s perceptions

of their ability to manage their career will also increase.

Hypothesis 6b: There will be a negative relationship between network content and career

management perceptions. That is, as the proportion of non-work/kin ties increases,

perceptions of career management will decrease.

Hypothesis 6c: There will be a negative relationship between network content and career

management perceptions. That is, as the proportion of non-work content (i.e. topics of

conversation) increases, perceptions of career management will decrease.

118

CHAPTER 4

METHOD

This chapter describes the methods that were used to tests the hypotheses

discussed in Chapter 3. As a result, this chapter begins with a description of the focus

group study that was conducted prior to the field study. The purpose of the focus group

was to collect data that would help refine and develop the measures used in the field

study. The field study involved administering two separate web-based surveys to two

distinct groups, working adults with parental responsibility and working adults without

parental responsibility. The measures collected during the field study included

demographic variables (e.g. gender, parental status), attitudinal data related to feelings of

segmentation between work and family roles, job involvement, and family involvement.

Finally, measures of career success (e.g. salary, promotions, career success indicators)

and measures of career management perceptions (e.g. career mobility preparedness) were

collected during the field study. This chapter discusses how the variables included in the

conceptualized model were operationalized. Also, the chapter provides a brief description

of the pilot study that was conducted to refine some of the measures that would be

collected during the field survey. Finally, the chapter concludes by briefly describing the

analyses that will be used to test the study hypotheses.

119

Focus Group Study

A focus group study, the first step of this dissertation, was conducted in the

summer of 2005. The purpose of the focus group was to collect data that would help

further define the final measures used in the field study. The specific measures collected

during the focus group study included network size, network ties, and network topics of

conversation. In addition, the individuals who participated in the focus groups were asked

to define career success. The responses to this question were used to verify that both

subjective and objective measures of career success should be included in the study. That

is, the notes from the focus group study were reviewed, and it was determined that when

individuals were asked to define career success, the criteria they described related to both

objective (e.g. salary, opportunities for promotion) and subjective indicators (e.g. job

satisfaction, flexibility to work from home) of career success. Next, a brief overview of

the focus group study is provided.

Focus group interviews are defined as a research technique in which data is

collected through group interaction on a topic determined by the researcher (Morgan,

1996). The aim of focus group research is to draw conclusions about the participants’

views, ideas or experiences (Hydon & Bulow, 2003). There are three criteria that

describe focus group interviews including (1) they have to be conducted in formal

settings, (2) the interviews use directive interviewing, and (3) the interviews use

structured question formats. Focus group interviews are often paired with other methods

including surveys. Further, studies that include both focus group data and surveys, are

one of the primary ways to combine qualitative and quantitative methods (Morgan,

1996). A final advantage of using focus group interviews is, the amount of interaction

120

(and ideas shared) between the focus group participants. Finally, focus group interviews

are considered both generalizable and valid measures. The generalizability of the content

that is produced during focus interviews can be assessed through subsequent research

designed to test the implications learned from the focus groups, when tested using a

quantitative method. For this study, many of the components included in the focus group

interviews (e.g. number of ties, types of ties) will also be assessed through the field study.

To assess the importance of networking, specifically network characteristics and

the role they play in career mobility, the focus group interviews were conducted at a

Midwestern university. The focus of these interviews was to (1) gather information to

help identify the correct items that should be included in survey measures in future

empirical studies (e.g. measurement of the perceived usefulness of networks and its

relationship to career mobility) , (2) develop further the research hypotheses that will be

utilized in a future empirical study, and (3) collect information of the network-career-

management relationship based on gender and parental status. Focus group participanjts

were asked two types of questions Participants completed a survey at the beginning of

the focus group that was used to assess the three network measures including network

size, network ties, and network topics of conversation. After completing the survey

participants spent the remaining time in the focus group answering structured, open-

ended questions. These questions were used to determine participants opinions of

organizational networks. Additionally, several questions were asked about career success

and career management perceptions.

121

Sample

The Work-Life Initiative Office, a division of The Office of Human Resources,

and the Office of Human Resources were responsible for providing the names of the

individuals that were invited by the researcher to participate in the focus groups.

Specifically, the individuals identified by the Work-Life Initiative Office were the

working parents that were invited to participate in the study. In comparison, the

individuals identified by the Office of Human Resources were the individuals that were

working adults without parental responsibilities that were invited to participate in the

study. A brief description of the sampling procedure for each of the two groups, which is

the working adults without parental responsibility and working adults with parental

responsibility, is provided.

The individuals identified by the Work-Life Initiatives office were those working

adults who had a child, and had subsequently taken a parental leave within the last year.

In total, just over 300 individuals were identified through the Work-Life Initiative Office,

and invited to participate in the focus group interviews. E-mails were sent to the 300

individuals who’d participate in the parental leave program, and the individuals were

invited to participate in the focus group studies.

The individuals identified by the Office of Human Resources were working adults

who were employed by the university on a full-time basis. A random sample of 500

individuals employed by the University, were sent an e-mail inviting them to participate

in the focus groups and share their experiences of networking and their careers. That is,

the e-mails sent to the 500 individuals were drawn from a random sample of names

supplied by the Office of Human Resources at the university.

122

Four subsets of focus groups were created, and these focus groups were stratified

by gender and parental status. Those four groups included working fathers, working

mothers, working women without parental responsibilities and working men without

parental responsibilities. The stratification of focus group participants is consistent with

methods used in many fields (e.g. marketing and sociology) when the topic of interest

(e.g. gender and parental status and networks) is expected to vary by particular categories

(Morgan, 1996). Of note, focus group research studies are stratified across four to six

focus groups (Morgan, 1996).

Consistent with previous research, guidelines were followed such that twelve

participants were the maximum number allowed in each focus groups Previous research

suggests that focus groups should consist of 8 to 12 (homogenous) members (Fern,

1982). The criteria for inclusion in this study were the following: (a) on the job no less

than 30 working hours a week for a minimum of 9 months of the year (full-time

employee), and, (b) if parent, child less than the age of 6 years old.

In all, there were 10 focus group sessions offered. This included two sessions

each for working women without parental responsibility and working men without

parental responsibility. A total of 24 working women and men without parental

responsibility participated in the focus groups, conducted for individuals without parental

responsibility. Also, there were 40 working women and men with parental responsibility

participated in the focus groups. Total sample size for all participants in the 10 focus

group sessions was N=64.

123

Procedure

The participants were contacted via e-mail and asked to participate in the focus

groups. At that time the participants were given a description of the purpose of the focus

groups and their role in the focus groups should they decide to participate. The

participants were also made aware that all focus groups were scheduled for 90 minutes.

After agreeing to participate, on a voluntary basis, the focus group participants were

asked to sign-up for a focus group session that best suited their schedule (there were

asked to choose from 2 or 3 time periods).

Once the participants arrived at the focus group sessions, the participants were

given an opportunity to read and sign the consent form, and the purpose and procedure of

the focus groups was reviewed with the participants. Next, the participants were asked to

complete a survey which included several measures related to network ties, network

content of conversation, and network size. Specifically the participants were asked to

name (and write down) up to the 10 most important people in their professional lives.

After writing down the initials (the initials will ensure confidentiality during this

process), the individuals were also asked to indicate their frequency of contact, and the

relationship or tie they had with the individuals named. Of note, the name generator

question has been used in previous career literature (e.g. Gersick, et al. 2000) and this

method is commonly used in the networking literature (e.g. Burt, 1992). Typically, the

empirical studies in the networking literature use the GSS survey. The GSS survey is one

in which individuals are asked to name the important people within their networks.

Therefore, the methods used in this phase of this focus group are consistent with those

124

used in previous literature. After identifying up to the 10 most important people in their

professional lives, the third phase of the focus group study was initiated. During this

stage, the focus group interviews were still completing the surveys. Specifically, the

individuals were asked to look at the ‘name generator’ sheet they just completed on the

survey, and they were asked to identify the four most important topics they discuss with

these individuals they’d identified as part of their network. The participants were then

asked to name the topics they discuss with these individuals and provide an example of

each topic they discuss. The purpose of this step was to generate a list of themes the

individuals discuss with the people they have identified, and to understand the frequency

of the topics discussed.

Of note, the focus group participants were asked to identify four topics of

conversation in an attempt to simply the theme-oriented coding system that was used to

organize the topics of conversation. Examples some of the themes that emerged from the

focus group participant’s surveys included: health, family, work, personal finances,

sports, art/music/literature, politics. Examples of job-related topics mentioned on the

respondent’s surveys included: organizational politics, managers, employee

retention/loyalty (given the local job market). The analysis of the theme content and

frequency was conducted after all focus group interviews have been conducted.

The fourth and last phase of the focus group session began after the participants

completed the sections of the survey described previously. The purpose of this phase was

to allow participants to share their experiences with networks and career practices.

During the fourth phase of the focus group sessions, respondents were asked to respond

to a series of open-ended questions. Of note, the open-ended questions and the entire

125

interview guide can be found in Appendix A of this dissertation. Of note, questions 2-3

(Topic: Networks) were asked in the focus groups that included parents. The remainder

of the questions (Topic Networks: Q1and Topics: Careers Q 4-7) were asked of all focus

group participants. In total, the focus group interviews lasted 90 minutes. Lunch was

provided as an incentive for individuals to participate.

Finally, there were several outcomes analyzed from the focus group interviews.

First, the average number of ties was assessed by gender, parental status, and overall

group. The average number of ties is the indicator of network size, and is derived by

counting the number of ties each participant selects. In order to calculate average network

size, focus group participants were randomly selected where one-half of the participants

were asked to name up to 100 names of people that were important to them and the other

one-half were asked to name up to 10 names of people that were important to them.

Therefore an average and range (min and max) number of ties were developed. This

resulted in the respondents in the field survey being asked to identify up to 20 names of

people within their network. That is, 20 people were the average number of network ties

identified during the focus group study.

In addition, to size, this study also generated the type of ties that are included in

an individual’s network. Specifically, the participants were asked to describe the

relationship they had with the people whom they identified as part of their network (e.g.

co-worker). From those relationships identified, 10 categories of relationship types were

created and used in the field study. Also, frequency with each tie was assessed amongst

the focus group participants.

126

Finally, this study generated important topics of conversation discussed within

networks. Specifically, the interviews were transcribed; the written text from the

interviews was reviewed to identify common themes from the questions included in the

interview. The interview transcriptions were coded by two individuals to ensure

interrater reliability. The themes derived from the transcript were exploratory and

therefore did not need to be identified a priori. Nine categories of conversations topics

were identified and used in the field study.

Pilot Testing

Prior to the launch of Surveys 1 and 2, a pilot study was conducted on the

test instruments. Approximately fifteen individuals were contacted and asked to

participate in the pilot testing. Those fifteen individuals included a group which consisted

of university professors, graduate students, and working professionals. The pilot tests

were conducted to achieve two main purposes. First, the pilot tests were conducted to

ensure the readability of the survey instruments. That is, the individuals that participated

in the pilot testing were given specific instructions to read the Welcome Page, Consent

Form, and the Test Instrument (that includes both the survey and the directions for each

section of the survey) and provide feedback regarding the clarity and readability of all

sections mentioned previously. Secondly, the respondents were asked, to take the entire

survey, that is reading all directions and answering all questions, and provide feedback

about the total amount of time it took to complete the survey.

While reviewing the survey, pilot study participants were asked to comment on

the face validity of the survey measures, specifically the scales used to measure family

involvement, job involvement, role segmentation, career management perceptions, as

127

well as the career success measures. Acceptable face validity is apparent when

individuals review the operationalization of a measure, and determines the measure is a

good translation of the construct that is being measured (Schmidt & Hunter, 1998). To

assess the face validity of an instrument, researchers will look at the specific items used

to measure the construct, and determine if the items appear to measure the construct of

interest.

Face validity differs from content validity, because when content validity is

assessed, the items used to measure a construct are assessed against a detailed definition

of the construct (Schmidt & Hunter, 1998). For example, if an individual was interested

in determining the content validity of an instrument written to measure career success, an

individual would need to compare the items written in the instrument, to a list of criteria

that are used to define career success. An individual would then look at the list of criteria

that define career success, and determine which items tap those specific criteria. Any

items in the measure that did not tap the criteria listed would be eliminated from the

career success measure.

In comparison, if an individual was interested in determining the face validity of a

career success instrument, they would simply look at each of the items and determine if

they items seemed to reasonably measure career success. Face validity is typically the

least reliable way to assess the validity of the measure (as there is no criteria which is

compared when the measure of interest is investigated). However, one way to enhance

the face validity of a measure is to have experts who are familiar with the construct the

measures is trying to assess, review the measure.

128

That is, to ensure the accuracy of the face validity assessment of a measure, researchers

would use subject-matter experts to review their measure. For example, to conduct an

accurate assessment of face validity of a career success measure, you would want an

individual that specializes in career research or an individual that works as a career

counselor to verify the face validity of your instrument.

The feedback provided from the respondents was reviewed by the researcher and

changes were made (e.g. clarifying directions) to the test instruments, if deemed

appropriate. Of note, while minor changes were made to the survey instruments and/or

welcome page or consent forms, there were no changes made to the actual questions in

terms of removing or adding additional items. Rather any changes made were specific to

the questions wording or directions.

Finally, the average time to complete the survey was measured. The average tine

to complete the first survey was 20 -30 minutes .and 10-15 minutes to complete survey 2.

This information was used in the field study. The average time to complete the survey

was communicated to field study participants on Welcome Page, which was the first page

of survey that participants saw after they selected the link to the survey (see Appendix D:

Field Survey Wave 1 Welcome Page, Consent Form, and Wave 1 Survey Questions).

Finally, since this study was a web-based survey, the pilot testing allowed the researcher

to ensure that all data was being stored properly prior to the survey being launched in the

field.

129

Field Study

Sample

The sample used for this dissertation included a pool of respondents who were

either working adults, or working adults with children. The primary data collection

source for this survey was a large, multinational, manufacturing organization whose

headquarters is located in Michigan. This organization engages in the development,

manufacture, distribution, and sale of various automotive products, primarily passenger

cars, light trucks, and commercial vehicles worldwide.

The organization agreed to participate in the study after being contacted by the

author. The organization agreed to participate in the study because they were interested in

learning about their employees’ perceptions of usefulness of the professional

organizational networks that the organization had recently created. The newly created

organizational-based networking groups were reportedly developed to minimize

unwanted turnover within the organization. Therefore, the organizational-based network

groups were developed by the organization as a way to reach out to specific groups

within the organization. For example, organizational-based networking groups were

created for individuals with physical disabilities, working mothers, individuals from

underrepresented minority groups (e.g. African-American and Latinos), and other groups

targeted by the organization’s diversity program.

Specifically, the organization was interested in understanding if the members of

their organization communicated with members outside of their network group. The

network groups within this organization were created based on various demographic

categories (e.g. African-American Professional Group). As a result, the organization

130

provided the researcher with the names of 1000 employees who worked throughout the

organization. The names of these 1000 employees were randomly selected from a larger

list of employees that were active in one of the organization-based networking groups.

That is, this list of names, while it included individuals from across the organization, was

restricted to individuals that were in some way active in one of the organizational-based

networking groups.

Although the organization created the employee networking groups for groups

within the organization targeted by their diversity program, the organization was

interested in learning to what extent individuals within the organization networked with

members outside of their interest group. Specifically the organization was interested in

learning if employees had cross-racial networking ties. Thus, in addition to the data that

was of interest for this dissertation, additional data was collected on the surveys related to

race and ethnic groups. Although additional questions were added, this did not increase

the amount of time the respondents had to spend completing the survey, for information

was gathered related to the race/ethnicity of their ties was not gathered on all ties the

respondent listed in their network.

It should be noted that individuals that did not fall into one of the organization’s

diversity groups (which was were not exclusive to racial and gender) did not have the

opportunity to participate in the study. That is, the organization only agreed to provide

the names of individuals whom both represented some diversity group within the

organization for whom an organizational-based network group had been created. This

restriction in sample size, may have resulted in the small representation of male

respondents. Also, it should be noted, that this sampling restriction may have produced

131

another bias among the respondents that agreed to participate in the study. Specifically,

because all of the respondents that participated in the study were in someway connected

to an organizational-based networking group, these individuals may see the advantage of

participating in organizational networks for the purpose of career enhancement and for

social support. Although there were no specific questions which asked the respondents to

provide feedback on relationship between organizational networks and career outcomes,

the individuals that participated in the networks may have felt some social desirability to

respond favorably to the survey questions related to either their careers or their network

characteristics. Also, during the time this data was collected, the organization made four

layoff announcements. The layoff announcements may have created feelings of

uncertainty among the survey respondents, which again may have introduced a biased in

their responses. Evidence of any kind of response bias would have produced limited

variability (i.e. a small standard deviation) in the surveys responses.

Per the requirements set forth by the study’s researcher, the participants in this

sample had to meet specific criteria to be included in this sample. The criteria for

inclusion in this study was the following: (a) on the job no less than 30 working hours a

week for a minimum of 9 months of the year (full-time employee), and (b) if parent, child

less than the age of six years old. As it turns out, there were many individuals who had

children that were not under the age of six. It was later decided that their responses (i.e.

survey respondents with children over the age of six) would remain in the study. This

decision was made based on the following criteria. Although previous research suggested

parents may experience the most significant interruptions in specific network

characteristics (e.g. network size) when their children are under the age of six,

132

there was no indication in the research that parents with children over the age of

six would not also experience significant interruptions in their network characteristics.

That is, parents with children over the age of six, also experience interruptions in their

networks. However, previous research does suggest that the most significant interruption

is most likely to occur for parents with children under the age of six (Smith-Lovin &

McPherson, 1993). As result, the decision was made to keep all working adults with

parental responsibility in the sample as long as they met the requirements of a full-time

employee.

The full-time employment measure (where individuals had to be on the job no

less than 30 working hours/per week, and work a minimum of 9 months per/year) used

in this study is consistent with the full-time employment measure used in previous studies

(e.g. Cron, 2001). Employees who are self-employed or working for a family business

were excluded from the study; individuals working in these conditions are “likely to

exhibit different patterns of career-related behaviors” (Forret & Dougherty, 2004, pp

424). This exclusion is consistent with previous studies (e.g. Forret & Dougherty, 2004).

Procedure

Consistent with the literature, surveys were used to collect data (See Appendices

D and E). Surveys and questionnaires soliciting self-reports are the predominant research

used when networking patterns are investigated (Marsden, 1990). The data collected was

self-reported data. Self-report data have two key advantages. First, self-report data is the

key way to obtain information related to an individual’s inner state (e.g. mood), beliefs

(e.g. beliefs related to careers), interpretations (e.g. is my behavior perceived as friendly

or unfriendly), cognitive processes (e.g. such as how was a decision reached), and

133

behaviors that can not be directly observed (Whitley, 2002). The second advantage of

self-report data is it is easy to collect, that is, questionnaires can be administered to

multiple individuals at one time (Whitley, 2002). Self-reports are also useful for

gathering information related to a psychological state of a respondent, (e.g. job attitudes

or motivation) (Podsakoff & Organ, 1986).

A common issues that arises with self-reported data is common method variance,

that is, variance that is attributable to the measurement method (e.g. survey) rather than

the constructs being measured (e.g. networking behaviors) (Podsakoff, Mackenzine, Lee

& Podsakoff, 2003). Common method bias is an issue in research because it is one of the

main sources of measurement error. Measurement error suggests that an individual may

make a mistake about the conclusions they draw about the relationships between

measures, that is, a researcher can draw misleading conclusions. To study the effect of

common method variance, some researchers have looked at the strength of relationships

between variables and control for common method variance (CMV). CMV can inflate or

deflate observed relationships between constructs (Podsakoff, et al., 2003). CMV is a

problem because it can lead to the artificial co-variance between self-reported measures.

Sources of CMV include: social desirability, item social desirability, item

complexity (e.g. double-barreled questions). It appears that one of the key ways to control

the possibly of CMV is to design the study well. This includes separating predictor and

criterion questions, making sure the wording of questions is clear (and there is no

presence of double-barreled questions), using existing measures- as these measures are

probably written clearly, including negatively worded or reverse-coded items, and using a

common rater. Ways to avoid CMV include: (1) Obtaining measures of the predictor and

134

criterion variables from different sources, (2) separating the measurement of the predictor

and criterion variables –this was completed in the dissertation, hence the 2-wave survey

(you can do this by introducing a time lag,), and using different response formats (e.g.

Likert scales, open-ended questions).

One recommendation for reducing common method variance is to

methodologically separate the measure of the predictor and criterion variables

(Podsakoff, et al., 2003). Benefits of separating predictor and criterion variables include

reducing the respondent’s ability and/or motivation to use previous answers to fill in

gaps, and prior responses are less salient available or relevant (Podsakoff, et al., 2003).

However, there are also challenges in trying to separating the predictor and criterion

variables. For example, if the time lag between measuring the predictor and criterion

variables is too long, this could mask a relationship that actually exists (Podsakoff, et al.,

2003). Also, if the time lag is too long, you may have an attrition problem. Therefore the

time lag between the measurement of the predictor and criterion has to be carefully

accounted for when designing the project. Other ways of avoiding CMV include

protecting the anonymity of the respondents. Also, you can ensure there are no right and

wrong answers as this is a reflection of question wording. Other suggestions for

improving CMV include being mindful of the manner in which questions are ordered and

to carefully construct scale items.

135

Consistent with the suggestion on how to reduce CMV, the predictor and criterion

variables were collected separately. That is, this dissertation used a 2-wave survey. In the

fist wave of the survey, all of the predictor variables were collected, and in the second

wave of the survey all of the criteria variables were collected. The first and second

waves of the surveys were collected approximately two months apart from each other.

The surveys used in this dissertation were web-based. The decision was made for

use web-based surveys for several reasons. The advantages of web-based surveys

include: (1) control over physical appearance, (2) web-base surveys can include radio

buttons and drop down lists that permit only one choice for the response, (3) check boxes

allow multiple answers, (4) data from web-based surveys can be easily imported into data

analysis programs, (5) data collection can be monitored (e.g. response rate), (6) it

eliminates costs associated with paper, postage, mail out, and data entry costs, (7)

reminder and follow-up for non-respondents is relatively easy, and (8) they can be

designed to allow for more dynamic interaction with respondents (Archer, 2005). Other

advantages of web-based surveys include cost efficiency (i.e. web-based surveys costs

between 5 and 20% of e-mail surveys), survey return speed (e-mail and web-based

surveys are usually completed within 8 days, while mail surveys are usually completed

within 12 days), tracking (i.e. researchers can do sample adjustments and timely follow-

up), better and more accurate responses), and web-based surveys can be easily down-

loaded (Kalaian & Kasim, 2005).

The disadvantages of web-based surveys include: (1) not everyone has access to

the internet, so this method will not work with all populations, (2) even if everyone was

connected, not all individuals are equally computer literate, (3) screen configurations may

136

appear significantly different from one respondent to another dependent on settings of

individual computers, (4) decision not to respond may be made more quickly, often

people are flooded with multiple e-mails, and (7) invitations to web-based surveys may

be detected as junk mail (Archer, 2005).

As a result, specific recommendations have been suggested when researchers are

designing web-based surveys. These recommendations include: (1) utilize a multiple

contact strategy much like the one that is used for regular mail surveys, (2) personalize all

e-mail contacts, (3) keeping the invitation brief, (4) beginning with an interesting, but

simple to answer questions, (4) using skip logic when possible, (5) making it possible for

each question to be visible on the screen at one time, (6) shortening the timing between

notice and reminders – e,g, reminder should be sent within 10-14 days of initial e-mail

inviting individuals to participate, (7) keeping the questionnaires short, (8) keep the

questionnaires short, (9) making limited used of open-ended questions, and (10) pilot

testing the survey before launching. Finally, research suggests that sending 2 reminders

(during an eleven day period) produces hat maximum number of responses.

Kalaian and Kasim (2005) suggest the best way to deal with non-response rates

on web-based surveys is to avoid them. In order to avoid them, they suggest that

researchers should be mindful of survey length, respondent contacts, compensation,

ensuring confidentiality, and questionnaire design (Kalaian & Kasim, 2005). All efforts

were made during the pilot testing phase to ensure that the survey length was as short as

possible and the survey responses were kept confidential (the responses were stored on a

secured server that was only accessible to researchers.

137

In inviting 1000 random selected employees to participate, an initial e-mail was

sent by the researcher which described the purpose of the project and the eligibility

requirements individuals had to meet in order to participate in the project. The researcher

was responsible for screening and performing applicant selection when the individuals

respond to the e-mail. The e-mail to the respondents stated "please respond if you meet

the following qualifications (1) working at least 30 hrs per week and (2) no parental

responsibility or (3) parental responsibility of children under the age of 6. In addition,

these same questions were asked on the survey. Although it was later decided that an

individual with a child would be allowed to participate in the project, that is people with

children over the age of six were not excluded from the study, individuals were still

disqualified if they did not work at least 30 hours per week.

The respondents were able to gain access to each survey via the link or URL

address that was included in the e-mails inviting the respondents to participate during

both the 1st and 2nd waves of the survey. Specifically, during the first wave of the survey,

the respondents were initially contacted via e-mail and they were told about the purpose

of the study, and the criteria that had to be met in order to be eligible to participate in the

study. They respondents were told to participate in the survey by selecting the web-link

or URL addresses that was embedded in the e-mail, if they were interested in voluntarily

participating in the survey and if they met the eligibility requirements to participate in the

study (See Appendix B, Wave 1 E-mail: Invitation to Participate in The Study).

For the second wave of the survey, the respondents were sent an e-mail with the

link/URL address also embedded within the survey. The participants were reminded in

the e-mail that they were only allowed to complete the second wave of the survey if

138

they’d completed they’d completed the first wave of the survey. The respondents were

told that their unique password they would need to complete the second wave of the

survey had been provided to them when they completed the first wave of the survey.

Specifically at the end of the first wave of the survey the respondents were instructed to

write down their unique password, which was provided on the last screen of the survey.

In addition, the respondents received an e-mail which made them aware that the answers

to the first wave of the survey had been recorded in the database and their unique

password was also provide in the e-mail. However, in the e-mail that was sent inviting

the respondents to participate in the 2nd wave of the survey, they were reminded that they

could contact the researcher for their unique password if they’d misplaced the e-mail

they’d received after completing the first wave of the survey, or they did not write down

their unique password at the close of the first wave of the survey. (See Appendix C,

Wave 2 E-mail: Invitation to Participate in The Study).

The first phase of the survey was used to collect the demographic variables, the

control variables (e.g. organizational tenure, job tenure, number of hours worked per

week, employment status, total number of years worked, and educational level), and all

independent variables of interest. The independent variables of interest included all

networking measures (e.g. network size, network ties, and network topics of

conversations), a family involvement measure, a job involvement measure, and a role

segmentation measure. The networking measures were being collected to potentially

identify some of the differences that may exist between working adults without children

and working adults with children specific to their network characteristics. Those

differences in parental status were expected in the following areas: family involvement,

139

job involvement, and role segmentation.

The 2nd phase of the field survey was distributed via e-mail approximately 6-8

weeks after the first phase of the survey was completed. The measures included in the

2nd phase of the survey are directly related to the key outcome measures of the study.

Those outcome measures include subjective/objective measures of career success and

career management perceptions. Career success can be defined as “the accomplishment

of desirable work-related outcomes at any one point in a person’s work experiences over

time”. A common distinction made in the literature is between objective and subjective

career success. Objective career success is success that is directly observable,

measurable, and verifiable by a third party. Examples of career success may include pay

increases or promotion to a higher position (e.g. from bank teller to loan officer within a

commercial banking setting). Subjective career success alludes more to an internal

satisfaction an individual experiences. Specifically, subjective career success is defined

by an individuals reactions (across any dimensions that are important to that individual)

to their career experiences and it is usually operationalized by a career satisfaction or a

job satisfaction measure. For the purposes of this survey both subjective and objective

measures of career success were collected. The perceptions of career self-management

measure were used to assess to extent to which individuals perceived they were able to

manage their own careers.

Survey Response Rate

The section describes the response patterns of the subjects that participated in this

study. The specific response rates that will be included in this discussion are the attrition

rate between Wave 1 and Wave 2 of the field surveys, and any reported response bias

140

between Wave 1 and Wave 2 of the survey. To determine if any response biases existed

between Wave 1 and Wave 2 response rates, a series of t-tests were conducted on the

demographic characteristics of the respondents whom completed only Wave 1 of the

survey in comparison to those respondents who completed both waves of the survey.

As mentioned previously, this survey was conducted in two separate waves. 1000

employees were contacted over e-mail and invited to participate in the study. The

participants were sent an initial e-mail and two reminder e-mails during the study. The

use of three total e-mails sent to those asked to participate is consistent with existing

research which suggests that the highest response rate on for survey-based data collection

efforts occurs when one initial e-mail is sent and two reminder e-mails are sent (Archer,

2005). Of the total 1000 participants invited to participate in the study, 441 respondents

participated in the first wave of the survey, which indicates a participation rate of 44.1%

for the first wave of the survey. Consistent with the methodologies explained previously,

only those individuals that participated in the first wave of the survey were invited to

participate in the second and final wave of the survey. After completing the first wave of

the survey, the respondents were asked to voluntarily submit their e-mail addresses,

which would allow the researcher to contact the participants and invite then to participate

in the 2nd wave of the survey. Of the 441 respondents who participated in the first wave

of the survey, 405 individuals submitted an e-mail address where they could be contacted

to complete the final wave of the survey. 36 participants whom completed the first wave

of the survey could not be contacted to participate in the second wave of the survey. A

total of 405 participants were contacted via e-mail and asked to participate in the second

and final wave of the survey. Of the 405 individuals contacted, 6 were no longer

141

employed with the company. Thus the total number of individuals invited to participate in

the second wave of the survey were 399. The participants were contacted a total of 3

times, where one initial e-mail was sent, and two reminders e-mails were sent within two

weeks of the initial e-mail. Of the 399 individuals contacted to participate in the survey,

295 individuals completed the second wave of the survey, which led to an attrition rate of

27.1% between Wave 1 and Wave 2 of the surveys (i.e. 295/405 equals 27.15).

Next, a series of T-tests were conducted to determine if there was any response

bias between those individuals that completed only the first wave of the survey (n=441)

and those individuals that completed both waves of the survey (n=295). A series of t-tests

were run on various demographic characteristics of the sample. The F-values and

significant levels are provided below in the table. In general, there very few differences

between those individuals that participated in the first wave of the survey, and those that

completed both waves of the survey. The two variables in which significant differences

were found between those participants that completed Wave 1 and 2 and those that

completed only Wave 1, included number of hours worked per week, average number of

ties, and child care responsibility of the ego. The average number of hours worked per

week was slightly lower for those individuals that completed both waves of the survey

(M=44.87, SD =6.85), in comparison to those individuals that completed only the first

wave of the survey (M=46.47, SD =8.74). Given that the group that did not complete the

second wave of the survey worked slightly more hours, they may have had less time to

complete the second wave of the survey. The mean number of ties per ego was slightly

higher for those individuals that completed both waves of the survey (M=8.61, SD

=5.11), in comparison to the mean number of ties for those individuals that completed

142

only the first wave of the survey (M=7.35, SD =5.84). Finally, for those individuals that

completed both waves of the survey, they reported higher frequencies of childcare

responsibility. That is, they were more likely to take care of their children themselves,

with the help of their spouses, or they were more likely to get help with childcare from a

third party (e.g. daycare). Overall, the results from the analysis of any response bias

between wave 1 and wave 2 of the survey suggested that because there were very few

differences between the respondents, the characteristics of the sample were consistent

across both waves of the survey, despite a slight decrease in response rate between Wave

1 and Wave 2 of the surveys. Table 4.1 shows the results of the response bias analysis.

143

F-Value Significance

Marital Status 9.048 .060

Gender .216 .643

Parental Status .760 .859

Child Living at Home 3.603 .059

Child Born in Last Five Years

2.022 .156

Number of Children .984 .323

Child Care Responsibility -Ego

9.876 .007**

Number of Hours Worked Per Week

4.169 .042*

Ego’s Age .638 .425

Employment Tenure Years

.055 .814

Job Tenure Years

.643 .423

Total Employment .821 .365

Total Number of Ties 4.766 .030*

Education 7.308 .063

* p= >.050 ** This is the Pearson Chi-Square value, not the F Value. Table 4.1: Non-Response Bias Analysis – Wave 1,2 and Wave 1 (Only) Participants

144

Measures

Control Measures

There were multiple control variables included in this study. The purpose in

including the control variables was to statistically account for factors other than the

independent variables of interest, that are known to influence the dependent variables

included in this study. In order to identify the control variables included in this study, a

literature review was conducted to identify those variables that had been statistically

controlled in previous studies where career outcomes have been investigated (e.g. Noe,

1996; Kossek et al., 1998; Eby et al., 2005; Gunz &Heslin, 2005; Ng et al., 2005). The

criteria for including the specific control variables used in this study included: (1) control

variables that were common across multiple studies related to network and career

research were included (e.g. age), (2) variables that were indicators of an individual’s

human capital, that is, an individual’s personal experiences (e.g. education, hours

worked) that are likely to enhance their careers (Ng et.al, 2005) were included as control

variables, and (3) socio-demographic variables (e.g. age) that have been found to

influence career outcomes in previous studies were included.

Of note, it is important to isolate an individual’s human capital factors, because

human capital variables may influence career success in addition to any networking

characteristics (e.g. size, ties). Based on the criteria described previously, the following

control variables were included in this study: age, organizational tenure (i.e. the total

amount of time an individual has spent in their current organization), job tenure (i.e. the

total amount of time an individual has spent in their current job), total employment (i.e.

the total amount of time an individual has been employed over their lifetime), number of

145

hours worked per week, number of work interruptions, organizational size, and

education level. In the following paragraph(s) a description will be provided as to how

each of the control variables included in this study were measured.

The first control variable included in this study was age. Age was measured with

the following open-ended question: “What is your age_________?”. In a recent meta-

analysis age was found to have a large statistically significant effect on multiple career

outcomes (e.g. age, and promotions) (Ng et al., 2005). Therefore, the decision to control

for age statistically is substantiated by results from pervious studies. The next control

variable included in this study was organizational tenure, which measured the amount of

time an individual has been with the current organization. Organizational tenure was

measured with the following open-ended question: “How many years and months have

you been with your current employer”? The next control variable included in this study

was job tenure. Job tenure was measured with the following open-ended question: “How

many months have your been employed in your current job_______?”. Next, total

employment experience, that is, the number of years an individual has been employed

over the career was used as a control variable. Total employment experience was

measured with the following open-ended question: “How many total years and months

have your worked (type your best estimate) _________”? Similar to findings related to

the variable age, recent findings in the Ng et al (2005) meta-analysis found organizational

tenure, job tenure, and total employment experience to all have a large statistically

significant effect on multiple career outcomes (e.g. promotions, salary). Thus

organizational and job tenure, and total employment experience were included as control

variables in this study

146

Level of education was another variable controlled in this dissertation. Similar to

many of the other human capital variables (e.g. job tenure, organizational tenure), level of

education was found to have a large, statistically significant effect on multiple career

outcomes (e.g. promotions and salary) (Ng, et. al, 2005). In this study, level of education

was measured using the following prompt: “Please indicate the highest level of education

achieved: (1) high school diploma, (2) associates degree, (3) bachelor degree, and (4)

graduate degree”. Organizational size is another control variable commonly included on

career-related studies. In this dissertation, the question used to measure organizational

size was: “Please indicate the approximate size of the organization in which you were

employed; where Small (2.500 employees or less), Midsize (2.501 employees to 10,000

employees), and Large (10,001 employees or more)”. Of note, although organizational

size was included as a control variable, it was not entered in the final analyses, as all data

was collected from a single organization.

Employment Status was also measured as a control variable. In measuring this

variable, employment status was coded (1) for exempt employees and (2) for non-exempt

employees. Number of hours worked per week, a control often included in many career-

related studies was measured by using the following open-ended question: “Please

indicate the number of hours you work per week (on average)____”. Finally, the last

control variable included in this study was work interruptions. This variable was

measured on a 4 point scale, where 0= I have not had any work interruptions to 4= I have

taken more than 3 months off from work. The specific question used was: “How many

work interruptions have you had over your career?”

147

A work interruption includes taking a leave from work for at least 30 days. This excludes

vacation time or time allocated for training and development. For example, if you had

two work interruptions which resulted in missing work for more than a total of 40 days,

you would select answer choice ‘2’ ”.

Family Status

There were many additional variables collected in this study that were related to

family status. These variables were not included as the control variables in the study, but

they were collected to gather a better perspective of the respondents completing this

survey. Also, some of the questions related to family status were included to verify the

parental status of the respondents. In addition, this study included a measure related to

elder care, previous studies have indicated that elder care is very similar to childcare in

that it takes away an individual’s time and may also impose a similar role constraint as

childcare.

Marital status was measured by asking the survey respondents to indicate their

marital status ( single, partnered, married, divorced, or widowed). As mentioned, there

were several questions used to verify the parental status of the respondents. This step was

taken, as the primary interest of this dissertation was to learn if network characteristics

and career outcomes differed by parental status. Therefore, it was clearly important to be

able to identify the respondent’s parental status. In addition, these questions were also

included to learn more about the ages of the children of those respondents whom were

identified as parents. Thus, if the respondent indicated they were a parent, they were

asked to answer the following questions: (1) “If you have at least one child, are these

children living at home - Yes (1) or No (2)”, (2) “If you have at least one child, was this

148

child born within the last five years”, (3) If you have at least one child, please indicate the

number of children that are in your home UNDER the age of six”, (4) “If you are a

parent, do you share parental responsibility with another adult”, and (5) If you share

parental responsibility with another person, please indicate their employment status”.

Many of the questions related to family status have been used in previous research (e.g.

Lyness and Thompson, 1997).

Network Constraint Measures (Gender, Family Involvement, Role Segmentation, and Job

Involvement)

As mentioned previously, the network constraint measures identified in this study,

including gender, job involvement, family involvement, and role segmentation are the

variables that have been hypothesized to create differences in the network characteristics

size, ties, and content across parental status. This next section provides a description of

how the network constraint variables were operationalized in this study.

Gender

The first network constraint variable, gender, was a dummy-coded variable in this study.

Therefore, Gender was be coded, (1) for females and (2) for males.

Job Involvement

The job involvement scale measure included in this study was drawn from the

Kanungo (1982) JIQ, or Job Involvement Questionnaire. The Kanungo JIQ scale was

developed to construct distinct measures of specific job involvement in comparison to

distinct measures of general work involvement. Further, the JIQ measure was developed

to address a key issue that Kanungo (1982) identifies about the widely used Lodahl &

Kejner job importance measure. In short, Kanungo (1982) argues that the items used in

149

Lodahl & Kejner scale (1965) actually tap a person’s “psychological identification”

(Kanungo, 1982, pp341) with the job (e.g. “I live, eat, and breathe my job”) and an

individual’s “intrinsic motivation at work for fulfilling self-esteem needs” (Kanungo,

1982, pp341). Taken together, Kanungo argues that previous measures of job

involvement have confused job involvement with intrinsic motivation on the job (e.g.

Lodahl & Kejner, 1965) , or the measures have confused identifying the antecedents of

job involvement rather than identifying job involvement as a state and identifying it’s

subsequent affects (e.g. Lodahl & Kejner, 1965). As a result, Kanungo developed a new

measure of job and work involvement. In developing the “distinct measures of specific

job and work involvement, three different measurement formats, including questionnaires

(e.g. Lodahl & Kejner, 1965) semantic differential and graphic techniques,” were used

(Kanungo, 1982, pp342).

The final JIQ scale contained 10 items including: “The most important things that

happen to me involve my present job role”; “To me, my job is only a small part of who I

am”; “I am very much involved personally on my job”; “I live, eat, and breathe my job”;

“Most of my interests are centered around my job”; “I have very strong ties with my

present job which would be difficult to break”; “Usually I feel detached from my job”;

“Most of my personal life goals-are job oriented”; “I consider my job to be central to my

existence”; and “I like to be absorbed in my job most of the time”. These items were

measured on a 6-point agreeability scale. The alpha reliability for these 10 items was

alpha =.87. The JIQ scale was found to be correlated with overall job satisfaction (r=.43).

150

Family Involvement

The measures for the family involvement scale were similar to the measures used

in the job involvement scale; except, the words “job” were replaced by “parent or

family”. This measure was taken from Kanungo (1982). The Kanungo (1982) scale

included 10 items, measuring job do you mean “family” involvement. The alpha

reliability for the 10-item Kanungo scale was alpha=.87. Of the 10 items included in the

scale, 4 items were included in this dissertation. The four items that were included in the

measure were: “The most important things that happen to me involve my present parental

role”; “Most of my interests are centered around my family”; “Most of my personal life

goals are family-oriented”; “I consider my family to be very central to my existence”.

The items were measured on a six-point agreeableness scale, where (1=Strongly disagree,

3= Neutral, and 6=Strongly Agree). The decision to made to use four items for the

family involvement scale is consistent with the items used to measure family involvement

in previous studies (e.g. Greenhaus et al., 1989).

Role Segmentation

Two measures were used to assess role segmentation, a perceptual measure of

role segmentation and an actual measure of role segmentation. The perceptual measure of

role segmentation was used account for an individual’s preferences to segment their work

and family roles/lives. The items used to measure role segmentation were drawn from

two scales inclduing the PEBS (Person-employment Boundary Scale) scale, and Edwards

and Rothbard’s (1999) four-item role segmentation scale. This section will begin with a

discussion of the PEBS scale.

151

The first set of items used to measure role segmentation were drawn from Person-

employment boundary scale (PEBS) developed by Hecht & Allen (2002). The PEBS

scale was developed to provide a measure of boundary strength between work and

nonwork domains. Specifically, segmentation between work and home boundaries occurs

when what belongs to work and what belongs to home are clearly distinct. That is, work

and home activities are clearly differentiated and these activities are not carried out in the

same time or space (Hecht & Allen, 2002; Nippert-Eng, 1996). People who prefer to

segment their work and family lives, see these two aspects of their lives as mutually

exclusive and they allocate their time and space to one activity or the other (Hecht &

Allen, 2002; Nippert-Eng, 1996). Because there are no other measures of segmentation

available in published literature, the purposes of this scale were to develop a measure of

boundary strength, and to examine its psychometric properties (Hecht &Allen, 2002).

The second goal of this research was to explore the relationship between boundary

strength and an individual’s well-being.

Within the scale, items were worded such that high scores reflected weak

boundaries. In addition to measuring for an individual’s preference for work and home

life to be integrated or segmented, this measure also assessed to what extent people prefer

strong or weak boundaries. A strong boundary was defined when the two domains home

and work, do not overlap (Ashforth et al., 2000). In comparison a weak boundary existed

when the two domains work and home overlap (Ashforth et al., 2000).

The Work and Home preferences scale is a newly developed measure. The PEBS

measure includes 19 items in total. The scale is divided into two factors, WTOH (work-

to-home) and HTOW (home-to work). An individual that prefers integrating their work

152

and home lives, should have high scores on the WTOH items. Nine items tapped the

strength of home-to-work boundary (HTOW) and the remaining 10 items tapped the

strength of the work-to-home (WTOH) boundary.

The WTOH items included: “I often do work at home”, “I do not work on

personal time (Reverse Coded)”, “I get work-related correspondence at home”, “I work

after hours”, “Work spills over to personal life”, “I never take work out of the office

(Reverse Coded)”, “My personal time is my own (Reverse Coded)”, “It is difficult to

keep work issues from coming home”, “It is not unusual to work over breakfast/dinner”,

and “I have no problem leaving work in the workplace (Reverse Coded)”. The reliability

for the WTOH scale was alpha=.94. The nine items measuring HTOW included: “I

rarely deal with personal matters when working (Reverse Coded)”, “I communicate with

friends & family during business hours”, “I often do personal errands on work time”,

“The office is reserved for doing work (Reverse Coded)”, “I often think of personal life

when working”, “When working, I am completely focused on work (Reverse Coded)”,

“My personal issues spill over to work”, “I schedule personal activities during business

hours”, and “I have no problem leaving personal life outside of the workplace”. The

reliability for the HTOW scale was alpha =.91. All items were measured on a seven-point

agreeableness scale, where 1= Strongly disagree and 7 =Strongly agree. Of note, there

were no other measures of boundary strength available at the time that Hecht & Allen

(2002) published this scale. Therefore, the authors did not measure convergent validity

with other boundary measures. In other words, the authors had no opportunity to

determine of there measure of boundary strength was similar to other measures of

boundary strength reported in published literature.

153

Of the 19 total items included in the PEBS scale, 15 items were included in this

study. Those 15 items included: “I often do work at home”, “I work after hours”, “I

schedule personal activities during business hours”, “I communicate with family and

friends during business hours”, “I think of personal or family-related issues while I am

working”, “I do not work on personal time”, “I take work out of the office”, “My

personal time is my own time”, “When working I am completely focused on work”, I

leave my personal life outside of the workplace”, “I rarely deal with personal matters

when working”, and “The office is reserved for doing work”.

The four items not included in this scale were the following items: “I get work-

related correspondence at home”, “It is not unusual to work over breakfast/dinner”,

“Work spills over to personal life”, and “I often do personal errands on work time”. The

items “Work spills over to personal life” although included in the work and home

preferences scale, really seems to be items that measure work-family conflict. Because

the focus of this study was to understand the preferences an individual makes to integrate

or segment their work and family lives, a construct that is separate and distinct from the

work-family construct, the decision was made to eliminate this item from the measure

used in this study. The items “I get work-related correspondence at home”, “It is not

unusual to work over breakfast/dinner”, and “I often do personal errands on work time”,

do seem to suggest some preferences for creating boundaries between work and home.

However, these items seem to be a better measure of an individual that prefers to

integrate their work and homes lives. That is, an individual that prefers to integrate their

work and home lives has few, if any, boundaries between work and home and would be

likely to do work while eating breakfast, complete personal errands while working, or

154

receive work-related correspondence at home. The purpose of this study was to measure

an individual’s preference for segmenting their work and home lives. Thus, these later 3

items were not included in the segmentation scale, because they appear to be items that

measure the extent to which an individual chooses to integrate their work and home lives.

Of note, there were some items included in the measure used in the study that did

assess an individual’s preference for integrating work and home (e.g. “I often do work at

home”). Thus, this measure did include items that reflected both an individual’s

preference to integrate and segment their work and home lives. The measure used in this

dissertation had to include items that measured both preferences for integration and

segmentation, because these items are on the same continuum. However, because a

majority of the items were included in the Hecht & Allen (2002) scale, the decision was

made to eliminate some of the items (as described previously) that measured an

individual’s preference for integrating their work and home lives, as the purpose of this

study was really to understand an individual’s preference to segment their work and

home lives.

The second measure used to assess an individual’s desire to for segmentation

between their work and home roles was drawn from Edwards and Rothbard (1999). This

scale is an established scale in the literature, and it measures an individual’s desire for

segmentation by asking individuals to identify acceptable, rather than ideal, amounts of

segmentation between work and family roles (Rothbard et al., 2005; Edwards &

Rothbard, 1999). This scale was presented similar to the way in which it has been

presented in other published studies (i.e. Rothbard et al., 2005). Thus, the items were

assessed using a six-point scale ranging from strongly undesirable to strongly desirable.

155

The instructions which appeared prior to the actual items read: “When responding to

these questions consider how much of that characteristic you personally feel is

acceptable. Some people prefer more or less of some job characteristics than others”. The

items that followed the directions included: “I do not want to be required to work while at

home”, “I want to be able to forget work while I am at home”, “I do not want to think

about the work once I leave the workplace”, and “I do not want to be expected to take

work home”. The internal consistency reliability estimate for this scale was 0.77.

In addition to the work and home preferences scale, a perceptual measure of an

individual’s preference to segment their work and home lives, A separate network

measure was used to assess the actual measure of role segmentation among the

respondents. In order to assess this, the following steps were taken. First, within the

survey, the respondents were asked to identify both who they talk to (e.g. spouse) and

what topics they discuss (e.g. children/household). An actual measure of role

segmentation was calculated then to understand the proportion of the number of ties

within their network that the ego discusses both work and home conversation topics. The

proxy used to calculate this actual measure of role segmentation was N(work & family) /

n(Work OR family), where N is number of individuals with whom the respondents talk

about both work and family topics and n is the total number of people sampled in each of

the respondent’s network. This measure was developed specifically for this study, as

there are no other none measures of netrole segmentation measures of work and family

topics, available in published literature.

156

The outcome for this role segmentation variable ranged from 0 to 1. A score of

zero indicates that there is no overlap in topics of conversation that the ego discusses with

their ties. In other words, the ego has a completely segmented network, where they

discuss work topics and family topics with different people within their network. In

comparison, a score of one indicates that there is complete overlap in the types of topics

that the ego discusses within their network. That is, the ego discusses both family and

work topics with all members within their network. This suggests that the ego wants to

integrate both their work and family lives. Next, an illustrative example of the role

segmentation (actual) measure is provided.

User ID Either* Both** Overlap***

15 6 6 1.00

57 6 4 0.66

24 5 2 0.40

97 6 0 0.00 *Either-Number of people in respondent’s sample (Also, it’s possible for ego to discuss either work or family with each tie within their sample) **Both- Intersection of work and family topics. That is, these are individuals within the ego’s network with whom they discuss both work and family. ***Overlap (Work and Family)/Work or Family or Both/Either (in Table 4.2)

Table 4.2: Calculation of Role segmentation ‘Actual’ Measure

Organizational Network Size

A single-item question was used for individuals to list the individuals with whom

s/he has a direct tie. The single item measure was drawn from the General Social Survey

157

(GSS), and has been used in previous empirical research (e.g. Polodony & Baron, 1997).

The GSS is an ongoing survey that measures the social and political attitudes and various

behaviors of adults living in the United States; the GSS is administered by the National

Opinion Research Center (NORC) and is given every other year to a random sample of

households (Marsden, 2003). Data for the GSS survey is collected in person (interviews)

from a household representative at least eighteen years of age (Marsden, 2003).

For the purposes of this dissertation, the respondents were asked the following

(name generator) question (See Appendix for Full Survey):

“Please type BOTH the initials and first name (e.g. KLS – Kyra) or (e.g. KS – Kyra) of the most important people in your professional and personal life (up to the 20). This includes both people inside and outside of your organization, family members, friends, neighbors, members of professional organizations, supervisors, colleagues, and anyone else with whom you discuss important matters including your career plans and various aspects of your professional life”

The single item measure is called the name generating technique and it used to help the

respondent define their network, and it is a commonly used measure across social

network research (e.g. Polodony & Baron, 1997; Granovetter, 1972).

The name generator is the best known and one of the most widely used measures

for the collection of egocentric, that is, personal network data (Bailey & Marsden, 1999).

This question is consistent with the “important matters” name generator questions that

has been used in previous GSS surveys (e.g. Marsden, 1987; Marsden, 2003). Name

generator questions are usually followed by name interpreter questions that are used to

yield additional information on the individuals identified by the name generator

questions, and they are used to measure specific network characteristics (e.g.

composition, density) (Marsden, 2003).

158

For the purposes of this study, there are six name interpreter questions that were

included. The purpose of these name generator questions is to measure specific network

characteristics, including, network size, network ties, and network content of

conversation. First, network size will be measured by counting the number of direct ties

connected to the ego. That is, network size will be measured as the number of distinct

names given in response to the name generator question (Marsden, 2003). b Network

size has a reasonably high stability over short periods of time (Mardsen, 1990; Mouton, et

al, 1995). Both the gender and the total number of organizational members will be

counted. Networks can include work groups, associations, neighborhoods, and family

members.

When estimating network size, research indicates that often the number of

reported interactions will be an imperfect indicator of the number of interactions that an

individual has with the ties within their network (Feld & Carter, 2002). However,

research also suggests that there is a strong correlation between reported and actual

number of interactions that an individual has with their specific ties, thus the number of

ties an individual reports is not very far from the actual number of reported interactions

with their ties (Feld & Carter, 2002).

Calculation of Network Size:

Network size was simply calculated by summing the number of ties each

respondent identified within their network. This calculation of network size is consistent

with how network size is calculated in the social networking literature (Wasserman &

Faust, 1994). The range of network ties was from 1 to 20 with a M = 8.60 and SD =

5.123.

159

Organizational Members (Ties)

The average North American is said to have 1500 ties (Wellman, 1992), of which

4-8 are typically identified as close and supportive ties. Using the GSS survey instrument,

the respondent was asked to indicate up to the 20 people closest to them. They named the

individual(s) using initials only, and also identified the type of tie (e.g. kinship tie, friend

tie, work tie). The specific question used to measure the ego’s network ties was:

“Please indicate the type of relationship that you have with this person (e.g. KLS-Mentor)

as their name appears on the screen”. The answer choices included Supervisor/Boss

(Former or Current), Colleague/Coworker (Former or Current), Employee/Subordinate

(Former or Current), Mentor, Work Friend, Non-work Friend, Spouse/Partner,

Sibling/Parent, Other relative (e.g. daughter/son, in-laws), Neighbor and Other. The

answer choices were based on pilot data that was collected prior to the field survey being

initiated. In the field study, respondents were asked to identify the specific individuals

with whom they had conversations about important matters in their life. They were asked

to provide the individual’s name and their relationship to the ego. Based on this pilot

data, 10 categories/labels of relationship types were derived (e.g. Work Friend, Non-work

Friend) and used as the answer choices for this name interpreter question. This data was

interpreted as a dichotomous variables where (1) was used to code work ties and (2) was

used to code non-work/family ties. In addition to the relationship type, the survey

respondent was asked to provide the gender, age, parental status, and race of the

individuals with whom they have a relationship.

160

As mentioned previously, network matrix data was collected. That is, the survey

respondent described the extent to which each member within their network knows each

other. This measure is consistent with measures used in previous studies (e.g. Burt, 1992)

where individuals have been asked to respond to specific questions about the alters in

their network. The specific question stem used to generate the network matrix data was

discussed in the previous section, in which the measure for the network constraint,

segmentation, was described.

Operationalization – Network Ties

In this study, network ties were operationalized as the percent of network ties that

were kin. It was expected that the number of kin ties would be higher for working parents

than it would be for working adults without parental responsibility.

To calculate network ties, the following procedure was completed. This

procedure is consistent with other measures used in the social networking literature, when

the researcher is trying to determine what proportion of an individual’s network, a certain

type of tie (e.g. friendship tie or kin tie) comprises within the ego’s overall network (e.g.

Moody, 2001). First, by user, all network ties were coded into one of two broad

categories, work ties and non-work ties. The ties that were included in the work ties were:

supervisor/boss (former/current), colleague/coworker (former/current),

employee/subordinate (former or current), mentor, and work friend. The remaining ties

including non-work friend, spouse/partner, sibling/parent, other relative, and neighbor

were classified as non-work ties. Next all sampled ties were counted per respondent (this

was completed in SPSS). The amount of sample ties per respondent ranged from 0-6.

Next all kin ties were counted per respondent (this was also computed using SPSS). The

161

ratio of kin ties/all possible ties (sampled) was calculated. This ratio ranged from zero to

one, and the mean and standard deviation for this ratio were M = 0.566 and SD=.273.

Network Content

The conversation topics used to measure ego network content were first identified

during the focus group study described in Appendix A. As mentioned previously, during

the focus group study, individuals were provided with an open-ended question which

asked them to name 1-3 topics of conversation they discussed with each of the members

within their network. The purpose of including this measure in the focus group study was

to generate a pre-determined list of conversation topics, organized around multiple

themes, which would later be used in the field study. Specifically, all conversation

topics, by respondent, identified during the focus group study were recorded. After a full

set of conversation topics was generated per respondent, a conversation topic frequency

table was developed. The frequency table demonstrated by conversation topic, the

number of times it had been named across all participants in the focus group study. The

conversation topics with the highest frequencies were then categorized into one of two

broad themes, work and non-work. This pre-determined list of work and non-work

topics, was the final list of topics used during the field study. The list of conversation

topics used in the field study included, work-general (e.g. work expectations,

assignments), career/career progress (e.g. job hunting, career planning), continuous

education/training, work-related projects (e.g. new projects), networking, children/family

household, spouse/partner, marriage/relationship, health, and other. That is, the

participants in the field study selected conversation topics from a pre-determined list of

topics, where each topic fell into one of the two broader categories.

162

Specifically, on the field survey, the two specific question stems were used to

measure topics of conversations (i.e. network content). The two questions read as

follows: (1) “Please consider the topics of conversation discussed with this individual.

Thinking back to the most recent discussions you had about an important matter,

inclduing your career plans, please indicate all of the conversation topics you discussed

with this individual (you may select more than one conversation topic)”, and (2) “Of the

conversation topics you selected in the previous question, select the TWO topics that you

discuss most frequently when you discussed an important matter, including your career

plans with this individual”. In actuality, the variable of interest for this study was the later

questions, that is, this study was interested in gaining an understanding of what each

respondent talks about most frequently within their network. However, in order to avoid a

double-barreled question, that is, what do you talk about and what do you talk about most

frequently, these two questions were separated on the field survey.

This content of conversations measure has been used in previous studies to

investigate the conversation topics within the social networks of individuals (e.g.

Bearman &Parigi, 2004). Also, the content of conversations is directly related to the

name generator technique (described earlier). That is, when interpreted literally, the name

generator technique, asks respondents to indicate the important matters they had in mind

when they identified the list of alters within their network (Bailey & Marsden, 1999).

Thus when the name generator technique is used (i.e. identify up to twenty people with

whom you discuss important life matters), the respondents considers both the individuals

and the topics of conversations with these individuals, when answering this question.

163

Operationalization – Network Content (i.e. Network Topics of Conversation)

In this study, network content was operationalized as the percent of network

content that was family-oriented. It was expected that the number of family conversation

topics would be higher for working parents than it would be for working adults without

parental responsibility.

To calculate network content (i.e. topics of conversation), the following procedure

was completed. First, by user, all network content were coded into one of two categories,

work topics and non-work topics. The topics that were included in the work topics

category included work general (e.g. work expectations, assignments), Career/Career

Progress (e.g. job hunting, career planning), Continuous Education/Training, Work-

related Projects (e.g. new projects), and Networking. The remaining topics inclduing

children/family household, spouse/partner, marriage/relationship, health, and other were

included in the non-work topics. Next all ties sampled network content topics were

counted per respondent (computed in SPSS). Next all non-work topics were counted per

respondent (computed in SPSS). The total amount of topics discussed per respondent

ranged from 0-9. This procedure was followed by the number of non-work topics

counted per respondent. The total amount of non-work conversation topics discussed

ranged from 0-4. Finally, the ratio of non-work topics/all possible topics (sampled) was

calculated. This ratio ranged from zero to one, and the mean and standard deviation for

this ratio were M = 0.473 and SD=.198.

Of note, the procedure used to calculate proportion of non-work content is very

similar to the procedure used to calculate the percentage of kin ties. In this vain, the

procedure followed to calculate proportion of non-work content is very similar to

164

procedures followed in previous studies (e.g. Moody, 2001) where the researcher is

interested in calculating the proportion of friendship ties that exist within a network (i.e.

proportion of friendship ties/total network ties).

Objective Career Success

The perceptions of career success, both subjective and objective career success,

were used to assess the extent to which individuals are happy with their current career

progress. As mentioned previously, current research in the careers literature (e.g. Huselid,

2004) calls for more career-orientated research to be focused on both the subjective (e.g.

criteria to reflect an individual’s values and preferences for things such as level of pay,

challenging jobs, job security) and objective (e.g. attainments in various areas including

work performance, pay, and promotions) indicators of career success.

In total, five items were used to measure objective career success in this study.

Those five items included two items used to measure salary/salary progression, and the

other three items were used to measure promotion/likelihood of promotions. The two

salary items were drawn from the Forret & Dougherty (2004) study on networking

behaviors and career outcomes. The specific items used in the study to measure total

compensation and number of salary increases were: “ Please indicate your total

INDIVIDUAL pretax income (e.g. salary, bonus, stock, profit sharing) in 2005 and in US

Dollars”; and the answer choices included (a) less than 30,000 (b) 30,000- 45,000 (c)

45,000-60,000 (d) 60,000 – 75,000 (e) 75,000-90,000 (f) 90,000 – 105,000 (g) 105,000 -

120,000 (h) 120,000 – 135,000, (i) greater than 135, 000. The last item used to asses the

individual’s salary included: “Throughout your career, please indicate how many salary

increases you have received. Of note a salary increase includes both (a) changes in annual

165

salary; (b) qualifying for a performance-based company bonus, incentive, or stock plans”

Answer choices include 0, 1, 2, 3, 4, 5, and greater 5.

Three items were used on the survey to measure promotions, the second indicator

of objective career success. The first item based on Forret & Dougherty (2004) and

James (2000) included: “Since you’ve joined your current organization/company, please

indicate how many promotions you have received. Note: Promotions includes (a)

significant changes in salary; (b) lateral or horizontal promotions; (c) changes in offices

and/or type of furniture/décor in office; (d) significant changes in job scope or

responsibilities; and (e) changes in company level. The answer choices in this question

include 0,1,2,3, and 4, where 0 indicated zero promotions, 1 indicated one promotion, 2

indicated two promotions, 3 indicated three promotions, and 4 indicated four promotions.

The next question used to measure promotions included: “Please indicate how many

promotions you have received in your entire career. Note: Promotions include (a)

significant changes in salary; (b) lateral or horizontal promotions; (c) changes in office

and/or type of furniture/décor in office; (d) significant changes in job scope or

responsibilities; and (e) changes in company level. The answer choices for this question

included 0,1,2,3,4,5, and Greater Than 5, where the answer choices indicated, by number,

(e.g. 3 is 3 promotions) the number of promotions that individual received in their entire

career. The last question related to a promotions was drawn from the Stout, et al. (1988)

study , the question read: “How likely is that you will receive a promotion within the next

five years?” A five point Liker scale was used, where 0= No chance and 5= Very good

chance.

166

Subjective Career Success

As mentioned previously, the second set of career success items, included those

used to assess perceptions of subjective career success. There were two sets of items used

to measure subjective career success, which included the individual’s perspective of their

career success, and their career success when compared to other-referent criteria.

The first set of items used to measure intrinsic career success, focused on the

individual’s perspective of their career success. These items were drawn from the

Greenhaus et al. (1990) career satisfaction scale. The Greenhaus et al (1990) measure was

developed to measure both satisfaction with career success and the extent to which an

employee has made satisfactory progress towards goals for income level, advancement

and development of skills. The Greenhaus et al (1990) career satisfaction scale has been

shown to correlate positively with having a job in general management, salary level,

number of promotions received, perceptions of upward mobility, sponsorship within an

organization, acceptance, job discretion, supervisory support, career strategies, perceived

person-organizational value congruence, presence of an internal labor market, and job

performance (Fields, 2002). It has been shown to correlate negatively with having

reached a career plateau (Fields, 2002). Finally, a confirmatory factor analysis

demonstrated that general perceptions of career satisfaction are empirically distinct from

financial success and hierarchical success in an organization (Field, 2002).

In total there are seven items in the Greenhaus et al (1990) scale. The alpha

reliability for these items was 0.89 in the Greenhaus et al (1990) study. This study used

six of the seven items from the Greenhaus et al (1990) scale. The items, measured on an

agreeableness scale where 1 = strongly disagree to 6 = strongly agree, included: “Relative

167

to my career aspirations, I am satisfied with the progress I have made towards meeting

my goals for advancement”, “Relative to my career aspirations, I am satisfied with the

overall success I have achieved in my career”, “Relative to my career aspirations, I am

satisfied with the progress I have made toward meeting my goals for income”, “Relative

to my career aspirations, I am satisfied with the skill development I have attained”,

“Relative to my career aspirations, I am satisfied with the autonomy I have attained”, and

“Relative to my career aspirations I am satisfied with the intellectual stimulation I have

attained”. A decision was made after the pilot test was conducted to exclude one of the

items from the scale. The item, Relative to my career aspirations, I am satisfied with the

progress I have made toward meeting my overall career goals”, appeared to tap the same

dimension as the item “Relative to my career aspirations, I am satisfied with the overall

success I have achieved in my career”.

The second set of items measured the individual’s career success in comparison to

the career success of members within their peer group. These items were from Heslin

(2003) study. Peer-related career success really draws from Festinger’s (1954) social

comparison and Adam’s (1965) equity theories. These theories suggests that individuals

are motivated to evaluate outcomes they achieve and they attempt to do so by comparing

their outcomes to those of other people. Drawing from social comparison theory, Heslin

(2003) argued that the Greenhaus et al (1990) career success scale, although widely used,

is not the most accurate measure of assessing each respondent’s career success for at least

two reasons. First, Heslin (2003) argues that the Greenhaus et al (1990) scale is not

applicable to individuals who work on a contract basis, who run their own small business,

or individuals that value other features of their career besides hierarchical advancement.

168

Thus, Heslin (2003) argues that researchers need to do a better job of assessing what

really matters to each person in terms of their career satisfaction. The second criticism of

the Greenhaus et al (1990) draws from social comparison theory. Heslin (2003) argues

that individuals make social comparison when they evaluate career outcomes, including

perceptions of their career success. The Heslin (2003) peer-related career satisfaction

scale was adapted from Greenhaus et al (1990) career satisfaction scale.

In total there are seven items drawn included in the Heslin (2003) measure. The

coefficient alpha level for these seven items was 0.95. In addition, success relative to

other-referent criteria was found to be positively correlated with career success relevant

to self-referent criteria. That is, an individual’s perception of how successful they were in

their own career when compared to someone within their peer groups, was highly related

to their own perceptions of the success they achieved in their careers.

Six of the seven items from the Heslin (2003) study were included in this

measure. Those six items measured on a six-point agreeableness scale where 1=strongly

disagree to 6=strongly agree, included: “Relative to the people who I perceive as peers in

my career/profession, I am satisfied with the progress I have made towards meeting my

goals for advancement”, “Relative to people who I perceive as peers in my

career/profession, I am satisfied with the overall success I have achieved in my career, “

Relative to people who I perceive as peers in my career/profession, I am satisfied with the

progress I have made toward meeting my goals for income”, “Relative to the people who

I perceive as peers in my career/profession, I am satisfied with the skill development I

have attained”, “Relative to people who I perceive as peers in my career/profession, I am

satisfied with the autonomy I have attained”, and “Relative to people who I perceive as

169

peers in my career/profession, I am satisfied with the intellectual stimulation I have

received”. The seventh item, “Relative to people who I perceive as peers in my

career/profession, I am satisfied with the progress I have made towards meeting my

overall career goals”, was excluded as pilot test participants did not believe this item was

distinct from the item “relative to the people who I perceive as peers in my

career/profession, I am satisfied with the overall career success I have achieved in my

career)”.

Career Self-Management Perceptions

As described in the literature review, individuals bear the responsibility of

managing their own careers. The perceptions of career self-management measure were

measured by two specific set of items related to perceptions of career management and

job mobility preparedness (Kossek et al., 1998). Job mobility preparedness is a measure

assessing the extent to which an individual is proactive in gathering information about

new career opportunities. The later measure is a valid measure of career self-

management, as this study argues that individuals will have to manage their own careers

within the new protean career, which implies that individuals will proactively have to

gather information about jobs and be prepared to move between jobs frequently.

The first set of items, those related to perceptions of career management, was

drawn from Sturges et al. (2000) and Hall (1990). In total, there was a combined total of

19 items used from the two scales to measure the individual’s perception of career self-

management. Those 19 items were measured on a six point agreeableness scale, where 1=

strongly disagree to 6= strongly agree. Next a discussion is provided of the specific scales

from which the 19 items were drawn.

170

Eleven of the nineteen items used to assess career management perceptions were

drawn from the Hall (1990) scale. This scale was designed to assess career management

practices related to career planning and tactics. The eleven items used from the Hall

(1990) study were drawn from two subscales. The first five items drawn from the Hall

(1990) career subscale, had an alpha of 0.70 and they were measured on a 5-point likert

type scale (where 1= very untrue of me to 5=very true of me). Those items measured

career planning and they included: “I have definite goals for my career over my lifetime”;

“When I think of changing my job, I always consider whether the new job leads to

another one I want”, “I give a lot of thought to plans and schemes for achieving my

career goals”, “I know what my strengths and weaknesses are in relation to my career”,

and “Achieving my career goals is very important to me”. The later six items were

drawn from the Hall (1990) career tactics subscale where the coefficient alpha was 0.68.

These six items were developed to assess the career management tactics that seemed to

be the most widely generalizable across organizations. Those six items measured on a 5-

point likert-type scale (where 1 =very untrue of me to 5 = very true of me) included: “ I

am always very careful to avoid dead-end career paths”, “I try to have as much visibility

and exposure to my bosses as I can”, “I go out of my way to find a mentor or sponsor to

help in my career in the firm”, “I cultivate friendships with influential people for my

career outside of work”, “I actively seek opportunities, rather than wait to be chosen”.

Of note, Hall’s (1990) eleven item measure of career management was used in at

least one other study (Orpen, 1994). Orpen (1994) found that there was a positive

relationship between career self-management and career effectiveness. Orpen (1994)

171

also found (individual) career self-management to be positively correlated with salary

growth, promotions, career performance and career satisfaction.

The next set of items used to create the career management scale for this study

(where there were a total of 19 items) were drawn from the Sturges et al. (2000) study.

The Sturges et al (2000) were seeking to understand the relationship between an

individual’s experiences with multiple career management practices as it relates to their

organizational commitment. The Sturges et al. (2000) study was an exploratory study,

where the authors presented the respondents with a list of multiple career management

practices where they were asked about the frequency of their participation in these career

practices. The list of career management practices that appeared on the Sturges et al.

(2000) survey, included both existing career management practices found in the career

literature (e.g. Gould & Penley, 1984; Noe, 1996), and additional items developed by the

authors for the study. Sturges et al (2000) collected the data at two points in time (using

the same measure each time), and the items loaded on four factors including networking,

mobility-oriented behavior, practical things, and drawing attention. The coefficient alphas

for each scale were reported, where networking was 0.72, mobility-oriented behavior

0.80, practical things was 0.62, and drawing attention was 0.82.

The specific items used in this study were drawn from the items that loaded on

networking factor in the Sturges et al. (2000) study. The decision was made to include

the items from networking, because the researcher wanted to include a measure that was

directly related to networking. Specifically, a scale directly tapping networking behavior

was warranted in this study, as this attempted to better understand the relationship

between specific network characteristics, and how they relate to an individual’s to

172

manage their career, including their ability to engage in networking behaviors. The items

loading on the remaining three factors, that is, mobility-oriented behavior, practical

things, and drawing attention in the Sturges et al (2000) study were either not relevant to

this study or they appeared to tap dimensions of career management behaviors that were

better represented by other scales (e.g. Kossek et al., 1999 and Hall, 1990).

During the field survey, the respondents were asked to indicate to what extent

theyhad engaged in each of the networking behaviors, and these items were measured on

a 5-point agreeableness scale (where 1=strongly disagree to 5= strongly agree). The items

the respondents responded to on the survey included: “I have gotten myself introduced to

people who can influence my career”, “I have talked to senior management at the

company’s social gatherings”, “I have built contacts with people in areas where I would

like to work”, “I have taken the initiative to be involved in high profile projects”, “I have

asked for career advice from people even when it has not been offered”, “I have asked for

feedback on my performance even when it was not given”, “I have refused to accept a

new role because it would not help me to develop new skills”, and “I have monitored job

advertisements to see what is available outside of the organization”.

The last set of items that appeared on the career management scale in this study

were drawn from the Kossek et al (1998) career mobility preparedness scale. The scale

was developed from information gathered during interviews with supervisors, employees,

and human resource managers at a major large auto company with operations in the U.S.

and Canada. The purpose of this measure was to understand the extent to which an

individual gathers information about new career opportunities, and the degree to which

an individual prepares him/her-self to act on internal and external career opportunities

173

(Kossek et al., 1999). Examples of behaviors include proactiveness in obtaining job

information, keeping a resume current, developing internal and external networks of

contacts who provide job information, and keeping a current resume (Kossek et al.,

1999). The specific items used in this study were drawn from a larger scale used to assess

career management in a quasi-experimental setting. The items were measured on a five-

point likert-type scale (where 1=not at all to 5=a great deal). As reported in the Kossek et

al (1999) study, the coefficient alpha level for these items was 0.84. Further Kossek et al.

(1999) found career mobility preparedness to be positively correlated with career self-

efficacy, adaptability, and feedback seeking behaviors.

The nine items, drawn from the Kossek et al (1999) measure, used in this study

included: “Over the past 6 months to what extent have you reviewed internal postings”,

“Over the past 6 months to what extent have you discussed future job openings within

your network (where internal means your network within your current organization (e.g.

co-workers, supervisors)” , “Over the past six months to what extent have you discussed

future job openings within your external network (where external means members in your

network outside of your current organization)”, “Over the past six months to what extent

have you thought about what position you would like to have next”, “To what extent do

you actively seek out information about job opportunities outside your current

organization”, “To what extent have you sought out any new personal connections AT

WORK in the past 6 months for the purpose of furthering your career?”, and “To what

extent have you sought out any new personal connections outside of work for the purpose

of furthering your career”. These items were measured on a five point likert-type scale

(where 1=not at all to 5 = a great deal). The ninth item was measured on a separate five-

174

point scale (where 1=not at all current and 5=very current), and the item read “How

current is your resume”.

Plan for Data Analysis

The data analysis for this dissertation began by first transferring all data that had

been collected over the internet into a usable file developed in SPSS. During this process,

the raw data was manually scanned for completeness, and any large amount of data

missing for an individual respondent was noted. The raw data was organized such that

the each respondent was given a unique row in the SPSS program, where all data that

corresponded with that respondent was entered into the same row.

Next, an analysis was conducted to analyze both the overall response rate to both

waves of the survey, and the attrition response rate between Wave 1 and Wave 2 of the

survey. Also, a series of steps were conducted to clean the data, including running a

series of one-wave ANOVAS and Chi-square tests to determine if there was non-

response bias-present between those that completed wave 1 of the survey and those that

completed both waves of the survey. Specifically, several of the key control variables

including, marital status, gender, parental status, number of hours worked per week,

ego’s age, job tenure and others, were selected to be included in the non-response bias

analysis. The variables that were selected to be included in the non-response bias analysis

met one of two characteristics: (1) they were the variables used as control variables in the

field study, or (2) they were variables measuring specific network characteristics. That

is, these variables were likely to influence the career outcomes measures. After the

variables were selected, a series of Chi-Square tests (for the categorical variables, e.g.

education) and One-Way Anova (for the continuous variables) were conducted. After the

175

non-response bias analysis was completed, a series of descriptive and frequency analysis

were conducted which were used to generate the sample descriptive (e.g. average age of

sample, number of respondents by gender).

Next, validity and reliability analysis was conducted on each of the scales used in

the dissertation. This included reviewing the measures used in the study, and determining

if any of those measures had items that needed to be reverse-coded. Next factor analysis

and reliability checks were conducted on each of the scales. During this process it was

determined the extent to which items from the scales loaded on the appropriate factors,

and the reliabilities were double-checked to determine if the reliabilities were fairly

consistent with the reliabilities reported in the published literature. Next, a series of

network-related measures (e.g. network content and network ties) were generated from

the raw data. Correlations, means and standard deviations were generated among the

study variables. Finally, the hypotheses proposed in the conceptual model were tested

using multiple regression. Specifically, multiple regression analysis was used to tests the

hypotheses regarding the main and interaction effects of the independent variables on the

dependent variables. This procedure was conducted to test all hypotheses, proposed in the

conceptual model.

176

CHAPTER 5

RESULTS

The purpose of this chapter is to report the results of the tested hypotheses

presented in Chapter 3. In completing these tasks, several preliminary analyses were

conducted prior to testing the hypotheses. This chapter will begin with a description and

results found during the preliminary analysis and will conclude with the formal testing of

the hypotheses.

Preliminary Analysis

Descriptive Statistics

Of the 295 participants that completed both waves of the survey, 245 respondents

were female and 50 respondents were male. It may be argued that the small sample of

male respondents, n= 50, does not warrant including them in the analyses of the study.

However, given that at least 3 of the proposed hypotheses are directly related to gender, it

makes sense to keep the male respondents in this sample, despite the small

representation. The mean age of the respondents was 39.04 (9.561 SD).

The marital status of the respondents in the sample included those who were

Single (n=58), Married (N=194), Partnered (N=18), Divorced (N=20), and Widowed

(N=3). Two of the 295 respondents did not report their marital status. Of the 295

respondents, 171 reported having children, this represents 58% of the sample.

177

Of those that have children, 61 (36%) had at least one child, 82 (48%) had at least two

children, and 28 (16%) had at least 3 children. The average number of children per

household was 1.40 (SD .492). There were 147 (86%) respondents who reported their

child lived at their home, and 100 respondents (58%) reported at least one child being

born within the last five years. Most of the respondents reported that they shared

childcare responsibility with another person (n=144) or 84%. Usually childcare

responsibility was shared between the ego and their spouse/partner n=78(54%). A smaller

amount of the respondents shared primary childcare responsibilities with a third person

provider (e.g. daycare) n=21 (15%), or a family member, besides their spouse or partner

n=7 (5%). Of those parents that shared childcare responsibility, 119 (83%) of the

individuals with whom they shared childcare responsibility were also employed at least

30 hours per week. In addition to childcare, a small amount of the respondents reported

they had elder care responsibility (n=11).

All but four of the respondents (n=291) reported working at least 30 hours per

week. The average amount of time individuals reported they had been with their current

employer (i.e. employment tenure) was 10 years and 5 months. The average amount of

time respondents reported they had been employed in their current job (i.e. job tenure)

was 4 years and 5 months. Total employment, that is the amount of time an individual has

been employed over their career was 19.73 years (10.272 SD). Almost half of the

respondents reported a salary between 45,000-60,000 (48%). Most of the respondents

have either a bachelor degree (n=108) 36.6% or a graduate degree (n=164) 55.5%.

The average number of network size across respondents was 8.61 (5.109 SD). The

minimum number of reported network members was 1 and the maximum number of

178

reported network members was 20, as participants were not allowed to name more than

20 network members. In addition to network size, respondents were asked to provide

additional information describing the relationship they hold with the network members

identified (e.g. work-friend), the gender of those members, and the age of the network

members and the content of conversation. As mentioned previously, respondents could

identify up to 20 members within their network. Of those members, respondents were

asked to answers a series of additional questions on up to 6 members within their

network. Of those 6 network members, the first 3 members were consistent with the first

three members that the respondents identified as part of network. The remaining three

members were randomly selected from the larger set of network ties identified. Across all

respondents, information (e.g. tie type, gender, age, conversation topic) was shared for

approximately 1,340 network members. This total of 1,340 network members reported

was lower than the total number of network ties expected to be generated by this study.

If all 295 members had shared information about 6 members (i.e. the maximum number

of network members which the respondents had to answer follow-up questions), there

would have been information reported about 2,360 network members. There are several

reasons why a higher total number of ties was not reported. First, during the data

collection process there was a glitch in the computer program. As a result, the individual

network member data for 65 individuals was lost. All of the 65 individuals were

contacted and asked to retake the section of the survey that was lost, that is Section B of

the network relationship, portion of the survey, and 31 individuals did retake that portion

of the survey. In addition to the computer glitch, some individuals may have experienced

respondent fatigue and may not have fully completed all questions related to their

179

network members (up to six members). However, all precaution was taken to minimize

respondent fatigue. This was accomplished by asking respondents to only answer

questions for up to 6 members within their network.

More than half of the network members (56.19%) identified by survey

respondents were non-work ties (n=753). The types of family ties most often identified as

a network member included Sibling/Parent ties (n=287), followed by Spouse/Partner Ties

(n=191) and Non-Work Friend Ties (n=151). Very few ties were identified as neighbor

ties (n=4). Work ties (n= 487) represented about 44% of the total ties identified by the

respondents. The two types of work tie most often identified were the supervisor (former

or current) n= 228 ties and the colleague (former or current) ties n=214. Surprisingly,

very few respondents identified mentors (n= 39) as members of their network with whom

they discuss important matters including their career plans. In identifying the gender of

ties, the distribution across gender was about even, where males ties made up 49% of all

ties identified in this data set and female ties made up the remaining 51% of all ties

identified in this sub-sample. This was an interesting finding given that the distribution

of female and male egos was not even represented in this dissertation. As mentioned

earlier of the 295 respondents, 245 of those individuals were female. That is, the gender

across network ties was fairly equal across gender. However, the gender across ego was

not even, where there were more female egos or respondents represented in this sample.

Therefore, if the distribution of gender across network ties had been consistent with the

distribution of gender across ego, there would have been a larger number of female

network ties (as over 80% of the respondents in this sample were female). Thus it appears

that females are likely to be networking with cross-gender ties (i.e. males) while males

180

may be more likely to network with same-sex gender ties. The average age of the ties

identified by the respondents was 44 years old (13.40 SD), just slightly higher than the

average age of the egos (i.e. 39 years old). When the egos were asked to share the

conversation topics they talk about most frequently amongst the ties in their network,

nearly 61% of the time, individuals are discussing work-related topics with the members

of their networks. The most common work-related topics discussed included general

work topics (e.g. work expectations and assignments) and topics related to career/career

progress (e.g. job hunting and career planning). One of the factors that may have

contributed to these discussion topics was the members of this organization were

experiencing a very turbulent work environment, as while data was being collected this

organization announced two rounds of layoffs. In addition, the work-ties most often

identified were supervisors and colleagues. Therefore it is likely that individuals would

discuss general work topics including work assignments and expectations with

supervisors and other colleagues. The remaining 39% of the topics discussed were

family-related topics, especially topics related to children and general household

concerns. See Table 5.1 for a full detail of the sample characteristics described in this

section.

181

Demographic Characteristic(s)

Ego Gender Female (n=245) Male (n=50)

Ego Age Mean= 39 (SD=9.561)

Ego Marital Status Single (n=58) Married (n=194) Partnered (n=18) Divorced (n=20) Widowed (n=3)

Ego Parental Status Child Living at Home Child Born Within Last Five Years Average Number of Children Shared Parental Responsibility Shared Primary Childcare (i.e. “If you have someone with whom you share parental responsibility with, who has the primary childcare responsibility”) Job Status (person with whom ego shares childcare responsibility)

Zero Children (n=124) One child (n=61) Two children (n=82) Three Children (n=28) Yes (n=146) No (n=21) Yes (n=100) No (n=68) Mean= 1.24 (SD =.429) Yes (n=144) No (n=17) Myself (n= 26) Both Myself and Spouse/Partner (n=78) My Spouse (n=16) Third Party Care Provider (n=21) Family Member (non spouse/partner) (n=7) Employed (n=119) Self-Employed (n=11) Unemployed (n=15)

Continued

Table 5.1: Descriptive Statistics for Demographic Variables –Wave 1 and 2 Participants

182

Table 5.1 continued Demographic Characteristic(s)

Elder Care Responsibility Yes (N=11) ; No (N=281)

Job Characteristics Average Number of Work Interruptions Working Greater than 30 Hrs Per Week Average Number of Hours Per Week Employment Status Employment Tenure (years) Employment Tenure (months) Job Tenure (years) Job Tenure (months) Total Employment (years)

Mean=1.09 (SD=1.234)

Yes (n=291); No (n=4)

Mean=44.87 (SD =6.854)

Exempt (N=275); Nonexempt (N=18)

Mean= 13.53 (SD =9.546)

Mean= 5.51 (SD =3.289)

Mean= 4.32 (SD =5.398)

Mean= 5.32 (SD =3.206)

Mean= 19.73 (SD =10.272)

Education High School Diploma (n=6) Associates Degree (n=17) Bachelor Degree (n=108) Graduate Degree (n=164)

Continued

183

Table 5.1 Continued

Members within Networks By Type (Work Ties) Colleague Mentor Work Friend Supervisor Employee Member 1 15 9 10 42 Member 2 33 8 17 53 Member 3 34 8 17 40 2 Member 4 48 6 16 30 5 Member 5 39 2 13 43 5 Member 6 45 6 16 20 5 Column Totals

214 39 89 228 17

Sum of Column Totals

587

Members within Networks By Type (Family Ties) Neighbor Spouse/

Partner Non-Work Friend

Other Relative

Sibling

Member 1 136 14 11 30 Member 2 18 22 29 82 Member 3 21 29 28 67 Member 4 1 9 26 20 55 Member 5 2 5 29 22 28 Member 6 1 2 31 14 25 Column Totals

4 191 151 124 287

Sum of Column Totals

753*

*Respondents were allowed to select an “other” when identifying their relationship with a tie. There were a total of 23 ‘other’ ties identified. The above totals do not reflect the ties identified as ‘other’.

Continued

184

Table 5.1 continued Frequency of Conversation By Topics (Work) Career Networking Work-related

(e.g. New Projects)

Work-general (e.g. Expectations Assignments)

Continuous Education

Sum

629 80 167 705 68 1649

Frequency of Conversation By Topics (Family) Children/ Household

Marriage Health

Spouse/ Partner

Other

Sum 588 257 125 63 11 1033

185

Data Checks and Cleaning

Prior to testing the hypotheses proposed in the conceptual model, several analyses

were conducted to insure the integrity of the data. First, as mentioned in Chapter 5, all

data was collected over the internet. Therefore all data was entered electronically through

syntax on the Web-based questionnaire. Participants had to respond to the survey in the

order that the questions appeared and they were provided reminders at the bottom of the

web-page which made them aware of when they had completed a given section. Per the

consent form, the respondents were given the option of skipping any questions that they

did not feel comfortable answering. As the respondents responded to each of the

questions, the data was stored in a series of Excel files. The data was stored in multiple

Excel files because there were multiple variables collected per respondent, especially

given the network measures that were collected.

The first step of preliminary analysis that was conducted was synthesizing the

data collected across Wave 1 and Wave 2 of the survey, into a single Excel spreadsheet,

where each user was given a unique row where all variables collected from that user were

stored. This single Excel spreadsheet was then uploaded into SPSS, where again, all

variables collected in each unique respondent were entered into a single row which was

identified by the first cell in that row where the respondent’s unique user ID appeared.

After the data had been uploaded to SPSS, a series of analyses were conducted to

ensure the accuracy of the data. First, the entire data set was scanned for completeness

and coherence. In conducting this analysis, the total amount of missing data was assessed

on all of the primary variables collected across both waves of the study. Overall, it was

discovered that just less than 2% of the data was missing across Wave 1 and Wave 2 of

186

the surveys. Specifically, in Wave 1, there was 306 (or 1.98%) missing data points out of

15,383 data points (this 15,383 data points excludes the specific network questions such

as content of conversation with network member, as there were great variance in the

number of responses addressing specific follow-up questions per network member). In

the second wave of the survey, there were 167 (or 1.27%) missing data points out of a

possible 13,275 data points. These outcomes suggest a reasonably high level of

completeness and accuracy of data in this dataset. Further, with each wave of the survey

reporting less than 2% of missing data this was especially impressive given that the

organization from which the data was collected experienced three rounds of layoffs

during the time the data was being collected. In addition, the missing data was evenly

distributed across all primary study variables (e.g. 2-3 responses were missing on average

per variable), therefore, there was no reason to exclude the data from one specific

primary variable included in the study. Finally, although this will be discussed later in

this section, research guidelines suggests that the pairwise deletion method is appropriate

to use when there is less than 2% of missing data (Roth, 1994). Consistent with this

notion, the pairwise deletion method was used during the hypotheses testing.

Every precaution was taken to ensure that all data was properly stored in the

Excel files as the respondents completed the survey. However, approximately 65

respondents were impacted by a glitch in the computer system. This resulted in a loss of

data on one specific question in the first wave of the study. Specifically, the data being

collected over the internet was stored in a series of Excel files. In the MS Excel program,

there is a default of the number of rows into which data can be entered. In the specific file

where the data related to network ties was being stored, the limit was not initially

187

removed. Therefore, although the respondents completed that section of the survey, their

answers were not recorded. This occurred until the error was recognized at which time

the maximum number of rows row default in that specific Excel file was removed.

Because the data was collected across a series of Excel files, other data collected for

those individuals was not impacted by the this error. All 65 individuals were contacted

and asked to retake the portion of the survey where there was missing data. Of the 65

individuals contacted, 31 individuals retook the missing portion of their survey. Although

the remaining 34 individuals had missing data, it was determined that those cases would

remain in the data set, as the missing data was applicable to only one questions on the

survey. The missing data was coded appropriately.

Next, all scale ranges were analyzed to ensure that the responses fell within the

correct parameters that were consistent with the scales and anchors used to assessed the

variables of interest. All responses were found to fall within the correct parameters.

Finally, there were two control variables that were included in the survey that were used

to ensure that the respondents met the eligibility criteria that was presented on the set of

directions that accompanied the survey. Those specific set of criteria included that

individual must be working at least 30 hours per week and no less than 9 months per

year. These criteria were used to assess whether or not that individual was working full-

time. In total 4 of the respondents worked less than 30 hours per year. Specifically two of

the respondents worked 28 hours per week and 2 worked only 24 hours per week. Of

those 4 network members that worked less than 30 hours, 2 network members also

worked less than 9 months per year (that is, the two that worked 24 hours per week). It

was determined that the individuals that worked 28 hours per week would be kept in the

188

data set. In comparison the individuals that worked 24 hours per week also worked less

than 9 months per year. Therefore it was determined those two respondents would be

excluded from the data set. This resulted in the total sample size for this study of 293.

Next, a series of analyses were run on the data including an analysis of response

bias between waves was conducted (i.e. a series of one-way ANOVAs and chi-square

tests were run), sample descriptives and frequencies were generated, and an analyses of

the scale reliabilities were performed on the study data.

189

16

.44**

-.01

-.01

-/06

-.19**

-.12*

-.1

1

-.13*

-.09

15

-.08

-.01 .07

.13*

.19*

.01 -

17**

.03

.06

.06

14

(.87)

.0

9 -.1

9**

-.26**

.15**

.22**

.00

.14*

-.05

.00

.16**

.16**

13

-.05

.04

.16**

.17**

-.06

-.09

-.01

-.04

-.08

-.07

-.06

-.05

12

(..86

) .0

2 -.3

1**

-.02

.21**

.18**

-.06

-.09

-.01

-.04

-.06

-.07

-.06

-.06

11

(..93

) .0

5 .1

9**

-.17**

.05

.08

.09

-.01

-.09

-.08

.00

.02

.09

.06

.05

10

-.11

-.10

-.02

.09

-.08

.04

-.08

-.04

-.03

.09

.13*

.08

-.09

-.11

-.10

9 .02

.20**

.00

.08

-.10

-.06

-.06

.09

-.07

-.17**

-.15*

.17**

.32*

* .1

3*

.03

.01

8 .4

3***

-.18**

.06

.02

.01

-.16**

-.02

-.04

.08

-.06

-.10

-.03

.09

.27*

* .1

7**

.00

.00

7 -.1

8** -.1

5*

.09

-.06

-.02

-.07

.28**

.06

-.07

-.14* .05

.12*

.13*

.33**

.00

.16**

.09

.07

6 -.0

8 -.0

1 -.0

5 .1

0 -.0

9 .0

3 -.0

4 -.0

7 -.1

0 .0

9 .0

2 .04

-.01

-.03

-.11

-.03

-.11

-00

.00

5 -.1

8**

.14

-.02

-.05

.10

-.09

.03

-.03

.03

.03

-.08

-.06

.03

.04

.21**

.07

.27**

.04

-.07

4 -.2

0**

.09

.06

.21**

.26**

.11

-.04

-.06

-.14*

.01

-.15*

-.18**

.01

-.10

-.17**

-.19**

.39**

.54**

.30**

-.04

-.04

3 .3

8**

-.2188

.01

-.05

.11

.15*

.0

8 -.0

6 -.1

1*

-.13*

.00

-.10

-00

.12*

-.12*

-.18*

-.32**

.13*

-.02

.10

.03

.06

2 .4

1**

.79**

-.10

-.04

.05

.19**

.23**

.08

-.06

-.06

-.06

.02

-.11

-.16**

.02

-.13*

-.22**

-.23**

.46**

.47**

.41**

.03

.02

1 .7

8**

.41**

.93**

-.10

.03

.02

.22**

.29**

.19**

-.13*

-.09

-.14*

.05

-.10

-.24**

-.03

-.13*

-.19**

-.22**

.44**

.54**

.27**

-.09

-.10

SD

9.59

11

6.08

62

.02

10.2

9 .6

99

.241

6.

65

1.23

5 1.

04

.377

1.

37

1.13

.2

9 .9

4 5.

12

.28

.21 .93

1.01

.9

5 1.

81

1.47

1.

91

1.06

1.06

Mea

n 39

.07

163.

69

48.8

4 19

.78

2.45

1.

06

45.0

1 1.

09

1.

04

1.17

4.

73

4.86

.7

6 3.

07

8.60

.5

6 .4

4 4.

61

3.86

3.

00

4.48

5.

24

2.59

4.11

4.21

Var

iabl

esC

ON

TRO

L1.

Age

2.O

rgTe

nure

3.Jo

bTe

nure

(Mth

s)4.

Empl

oym

ent

5.Ed

ucat

ion

6.Em

ploy

men

t7H

ours

Wor

ked

8..W

ork

IVs

9..P

aren

talS

tatu

s10

.Gen

der

11.F

amily

12.S

egm

enta

tion-

13.S

egm

enta

tion-

14.J

obIn

volv

emen

t15

.Eg

oN

etw

ork

16.

Ego

Net

wor

k17

.Eg

oN

etw

ork

DV

s18

..C

M-C

aree

r19

..C

M-C

aree

r20

.C

M-M

obili

ty21

.CS-

Sala

ry22

.CS-

Sala

ryG

rwth

23. C

S- P

rom

otio

ns

24.

Sat-

Indi

vidu

al

25.

Sat

-. -P

eers

Con

tinue

d

T

able

5.2

Des

crip

tive

Stat

istic

s and

Cor

rela

tions

on

Car

eer S

ucce

ss a

nd C

aree

r Man

agem

ent

Not

e: N

=29

3

Alp

ha re

liabi

litie

s in

diag

onal

.

*

p<.0

5;**

p<.0

1

190

Tab

le 5

.2 D

escr

iptiv

e St

atis

tics a

nd C

orre

latio

ns o

n C

aree

r Suc

cess

and

Car

eer M

anag

emen

t N

ote:

N =

293

A

lpha

relia

bilit

ies i

n di

agon

al.

*p<

.05;

**p<

.01

Tabl

e 5.

2 C

ontin

ued

25

(.89)

24

(.86)

.8

7**

23

.1988

.2288

22

..35**

.02

.01

21

.39*

*.3

3**

.07

.08

20

(.86)

.0

0 - .08

.02

-.07

-.05

19

(.79)

.3

7**

-.05

-.10

-.03

.19**

.24**

18

(.82)

.6

8**

.38**

00

-.05

.03

.13*

.15**

17

-.10

-.17**

-.10

.02

.04

.01

-.09

-.07

SD

.21

.93

1.01

.9

5 1.

81

1.47

1.91

1.

06

1.06

Mea

n

.44

4.61

3.

86

3.00

4.

48

5.24

2.59

4.

11

4.21

Var

iabl

esC

ON

TRO

L1.

Age

2.O

rgTe

nure

3.Jo

bTe

nure

(Mth

s)4.

Empl

oym

ent

5.Ed

ucat

ion

6.Em

ploy

men

t7H

ours

Wor

ked

8..W

ork

IVs

9..P

aren

talS

tatu

s10

.Gen

der

11.F

amily

12.S

egm

enta

tion-

13.S

egm

enta

tion-

14.J

obIn

volv

emen

t15

.Eg

oN

etw

ork

16.

Ego

Net

wor

k17

.Eg

oN

etw

ork

DV

s18

..C

M-C

aree

r19

..C

M-C

aree

r20

.C

M-M

obili

ty21

.CS-

Sala

ry22

.CS-

Sala

ry23

.CS-

Prom

otio

ns24

.Sa

t-In

divi

dual

25.S

at-P

eers

191

Scale Reliability Analysis

The coefficient alphas for the scales used ranged from .79 to .93 (see diagonals on

Table 5.2). In addition, factor analysis was used to ensure that the items loaded on the

same factors that they did in the published literature. Overall, all items loaded on the

appropriate factors for each of the scales of interest. To this end, the following

observations were made about each of the scales.

The first scale used in the study was the family involvement scale. As mentioned

previously, the original scale consisted of ten items. Of those ten, four items were used to

create the family involvement scale for this study. The family involvement items were

taken from the Kanungo (1982) scale. The original scale consisted of ten items, of which

four were used in this study. Of note, none of the 4 items had to be reverse coded.

Consistent with existing research, results from the factor analysis suggested that the four

items loaded one factor. This four item scale had an internal consistency reliability of .93.

The alpha reliability estimated in this study was slightly higher than the alpha reported in

the study from which this measure was drawn (alpha =0.87). The mean of this scale was

M=4.74. This suggests that most of the sample moderately agrees that family is important

to them. The factor loadings from the family involvement scale are presented in Table

5.3.

192

RESULTS OF FACTOR ANALYSIS – SCALE ITEMS

Item

Factor Loadings

1. The most important things that happen to me involve my present parental role

2. Most of my interests are centered around my family

3. Most of my personal life goals are family-oriented

4. I consider my family to be very central to my existence

Eigenvalue (3.332) Percent of variance explained (83.289%)

.857

.886

.882

.902

Table 5.3: Family Involvement Scale, Items and Loadings

193

The next two scales used in this dissertation were scales measured work and home

preferences, that is, this measure was used to understand the extent to which individuals

wanted their home and work lives segmented. Two scales were used to tap the work and

home preferences measure. Those two scales included Hecht and Allen (2002) PEBS

scale and the Edwards and Rothbard (1999) role segmentation scale. Prior to conducting

both the factor analysis and reliability analysis on the Hecht and Allen (2002) scale,

several items had to be reverse coded as was done in the original scale (e.g. I do not

work on personal time, When working, I am completely focused on my work. Consistent

with the original Hecht and Allen (2002) scale the twelve items loaded on two separate

scales. The five items that loaded on the WHOT scale (i.e. those items that suggest that

work is spilling over to home) had an internal consistency reliability of 0.85. This alpha

is slightly lower than the alpha reported in the study from which these items were drawn

(where alpha =0.94 for the WTOH scale). In comparison the seven items that loaded on

the HTOW scale (i.e. those items that suggest home is spilling over to work) had an

internal consistency of 0.82. This reliability estimate was also slightly lower than the

reliability reported in the study from which these items were drawn (HTWO alpha =.91).

The mean for the WHOT scale was M= 3.49 (SD= 1.31) , and the mean for the HTOW

scale was M= 3.12 (SD= 0.92). In interpreting the scale, the items were worded such that

high scores reflect weak boundaries between work and home, and low scores reflect

strong boundaries between work and home. Given that the mean of the scales was M=

3.5 and M=3.1 on a 6-point scale for the WHOT and HTOW scales respectively, this

would suggest moderately weak boundaries between work and home.

194

In anticipation of not knowing exactly how well the PEBS scales would

accurately measure an individual’s desire for segmenting their work and home

preference, that is, the construct of interest, another scale was included to measure an

individual’s desire for segmenting their home and work lives. As mentioned previously,

this third scale measured preferences for segmenting work and home lives (see Edwards

& Rothbard 1999). This measure included four items (e.g. I do not desire to be expected

to take work home) and loaded on one factor, which is consistent with the original scale.

The reported internal consistency for the scale in this study was 0.86, which was slightly

higher than the estimate reported in the original study (alpha =0.77). In addition, the

mean on this scale was M =4.9 (SD= 1.13) on a six point scale. In this case, the items in

the original scale were written such that high scores suggest that individuals desire to

have a clear segmentation between their work and home lives.

Of the three scales used to measure an individual’s preferences to segment, the

decision was made to use the Edwards & Rothbard (1999) preferences scale in the final

analyses. The rationale for this decision was based on two key ideas. First, of the three

scale used to measure work and home preferences, the Edwards & Rothbard scale had the

highest estimated internal consistency (alpha =.86) in comparison to the other two scales

drawn from the Hecht & Allen (2002) scales (WHOT =0.85 and HTOW =0.82). Further,

once the data was collected, a further examination of the items used in the three scales,

revealed that the items drawn from the Edwards & Rothbard (1999) scale (e.g. “I do not

desire to be required to do work while I am at home) were really a better measure of an

individual’s desire to segment their work and home roles.

195

The Hecht & Allen (2002) HTOW and WHOT scales are really a better measure of an

individual’s perception of which role, that is work or home, seems to have the most

spillover into the opposite role. The factor loadings from all three segmentation scales are

presented in Tables 5.4, 5.5 and 5.6.

196

RESULTS OF FACTOR ANALYSIS – SCALE ITEMS CONT.

Item

Factor Loadings

1. I often do work at home 2. I work after hours 3. I do not work on personal time (R) 4. My personal time is my own (R) 5. I take work out of the office

Eigenvalue (3.138) Percent of variance explained (62.752%)

.812

.739

.700

.599

.793

Table 5.4: Role segmentation Scale A- “Work to Home Boundary”, Items and Loadings.

197


Item Factor Loadings

1 2

1. I schedule personal activities during business hours

.189 .363

2. I communicate with family and friends during work hours

.161 .973

I 3. I think of personal or family-related issues

while I am working. .336 .521

4. When working I am completely focus on work (R)

.565 .185

5. I leave my personal life outside of the workplace (R)

.744 .213

6. I rarely deal with personal matters while working (R)

.785 . 362

7. The office is reserved for doing work(R) .695 .239

Eigenvalue (3.427) Percent of variance explained (48.962%) Table 5.5: Role segmentation Scale B- “Home to Work Boundary”, Items and Loadings

198



1. I do not desire to be required to work while at home

.641

2. I desire to be able to forget work while I am at home

.739

3. I do not desire to have to think about work once I leave the workplace.

.921

4. I do not desire to be expected to take work from home

.852

Eigenvalue (2.852) Percent of variance explained (71.310%) Table 5.6: Role segmentation Scale C- “Work and Home Desirability Scale”, Items and Loadings.

199

This items that comprise the job importance scale, as mentioned previously, were

drawn from the Kanungo (1982) scale. Prior to completing the factor analysis, two of the

ten items had to first be reverse coded (i.e. “To me, my job is only a small part of who I

am”; “Usually I feel detached from my job”).In this study, the job involvement items

loaded on two factors. However, a reliability analysis was conducted and the results

revealed this scale had the highest internal consistency when all items were loaded

together. That is the internal consistency for this scale was higher when all items were

loaded together rather than when the items were spilt between two scales. The internal

consistency for this scale was 0.87, which is consistent with the reliability reported in the

original study (Kanungo, 1982). In addition, the mean score on this scale was M= 3.07

(SD =0.94). This suggests that the respondents on this survey slightly disagree that their

jobs are very important in their lives. The factor loadings from the job involvement scale

are reported in Table 5.7.

200

RESULTS OF FACTOR ANALYSIS – SCALE ITEMS


1 2

1. The most important things that happen to me involve my present job

.436 .471

2. To me, my job is only a small part of who I am (R)

.371 .335

3. I am very much involved with

personally with my job .285 .652

4. I live, eat, and breathe my job. .536 .534 5. Most of my interests are centered

around my job .678 .395

6. I have very strong ties with my present job which would be difficult for me to break.

.310 .531

7. Usually I feel detached from my job (R).

.064 .607

8. Most of my personal life goals are job-oriented

.769 .080

9. I consider my job to be very central to my existence

.646 .253

10. I like to be absorbed in my job most of the time.

.609 .303

Eigenvalue 1 Factor – (4.554) Variance Explained 1 Factor – (45.542%)

Table 5.7: Job Involvement Scale: Items and Loadings

201

The next set of scales used in this dissertation included those scales used to

measure career management. In total there were twenty seven items used to assess

perceptions of career management. A factor analysis was conducted on all twenty seven

items. There were multiple factor analysis conducted on the career management scales.

Specifically, an exploratory factor analysis was conducted, and a one factor, three factor,

and four factor solutions were each tested separately. The exploratory factor analysis was

conducted using the maximum likelihood extraction method, and the rotated factor matrix

was conducted using the orthogonal rotation method. In addition, a separate reliability

analysis was conducted on the one-factor, three factor, and four factor tests. The initial

factor analysis resulted in a 3 factor solution, as evidenced by a scree plot analyses which

indicated three factors. After this initial factor analysis was conducted, a one-factor and

four-factor solution were forced. All items were found to load during the one-factor

solution, and the internal consistency reliability estimate for the one-factor solution was

0.90. However, the decision was made not to use the one-factor solution, as this solution

was not consistent with the manner in which the original scales were used in previous

research. That is, the career management scale was created by combining four separate

career management scales (i.e. career planning scale (Hall, 1990); career tactics scale

(Hall, 1990); career networking scale (Sturges et al., 2000), and career mobility

preparedness scale (Kossek et al., 1999). Next, the four factor solution was analyzed.

The conclusion drawn from this analysis was, although the career management measure

written for the survey used in this dissertation was done by combining four separate

scales, there was no evidence to suggest that these scales loaded on four separate factors.

202

That is, in investigating the scales, there was no evidence to support that each of the 27

items was loading uniquely on one of the four factors. That is, several of the items loaded

on multiple factors, and there was no evidence to suggest that the entire set of items (i.e.

B1,B2, B3, B15, B16, B17, B18, and B19) that loaded on multiple factors loaded on any

one factor better than the other (i.e. several of the items would load the same on more

than one factor). As a result, the decision was made to use a three-factor solution. The

three factor solution was used for this data set for several reasons. First, this solution

allowed consistency between the manner in which the original scales were written, that

is, this solution allowed for multiple career management scales to be tested in the final

analyses. In addition, the outcome of the three factor solution allowed the items drawn

from Hall (1990) and Kossek et al (1999) study, that is, career planning, career tactics,

and career mobility preparedness, to load separately on one of the three factors. There

was one exception to this rule. The items drawn from the Sturges et al (2000) study did

not load uniquely on a single. Rather, the items from the Sturges et al (2000) study loaded

on multiple factors. Further, the reliability of the items from the Sturges et al (2000)

study, that is, the career networking scale had the lowest reliability of the four scales used

to create the career management measure used in this study (internal consistency

reliability estimate was 0.76). Finally, the career networking scale was highly correlated

with the remaining three scales (e.g. career mobility preparedness). As a result, the

decision was made not to use the career networking scale in the final analyses. Thus, the

scales used to measure career management in the final analyses included the Career

Planning (Hall, 1990), Career Tactics (Hall, 1990), and Career Mobility Preparedness

(Kossek et al., 1999).

203

The first career management scale, career planning, included five items from

(Hall 1990). This scale measured the extent to which an individual knows their

individual strength and weaknesses and makes specific plans that will allow them to

achieve their career goals. The internal consistency reliability estimate for the career

planning scale was 0.82. This alpha is slightly higher than the one reported in the study

from which these items were drawn (alpha =0.70). The mean for this scale was M=4.618

(SD=0.92) on a six-point scale which suggests that individuals in the sample moderately

agree that they are able to plan their own careers. The factor loadings from the career

planning scale reported in this study appear in Table 5.8. Of note, the factor loadings

reported are the factor loadings from the 3-factor solution that was used in the final

analyses.

The second scale, career tactics, was also drawn from Hall (1990), and it

comprised of six items. The scale really assesses the extent to which the individual goes

out of their way to find mentoring and to take positions that will help enhance their

career. The internal consistency reliability estimate for the career tactics scale was 0.79.

This alpha was slightly higher than the original study where the reported alpha was 0.68.

The mean for this scale was M=3.976 (SD=0.93) on a six-point scale. The factor

loadings from the career tactics scale reported in this study appear in Table 5.8 As

mentioned previously, the factor loadings reported are the factor loadings from the 3-

factor solution that was used in the final analyses.

The final scale used to assess perceptions of career management, that is career

mobility preparedness, included eight items from Kossek et al. (1999). This scale was

used to assess an individual’s perceptions that they were engage in mobility preparedness

204

behaviors (e.g. to what extent do you actively seek out information about job

opportunities outside of your organization). The internal reliability estimate for this scale

was 0.86. The reliability estimate found is identical to the reliability estimate reported in

the study from which the items were drawn (where alpha =0.84). The mean for this scale

was M=3.041 (SD=0.95) on a five-point scale. This suggests to moderate extent,

individuals are engaging in mobility preparedness behaviors in order to manage their

careers. The factor loadings from the career mobility preparedness scale reported in this

study appear in Table 5.8.

205

FACTOR

1 2 3 “Career Tactics” “Career Mobility

Preparedness” “Career Planning”

Items B1 .413 .107 .152 B2 .578 .050 .108 B3 .458 .152 .329

.488 B4 .302 .120

.622 B5 .345 .255

.889 B6 .292 .203

.306 B7 .246 .080

.581 B8 .345 .128 .445 B9 .088 .582 .652 B10 .016 .283 .583 B11 .148 .366 .493 B12 .205 .309 .507 B13 .230 .252 .431 B14 -.053 .150

B15 .729 .129 .149 B16 .642 .264 .288 B17 .577 .130 .176 B18 .181 .204 .187 B19 -.005 .520 .129

206

C1 .100 .524 .162 C2 .048 .615 -.019 C3 .186 .610 .091 C4 .045 .770 .084 C5 .154 .595 .269 C6 -.007 .774 .069 C7 .335 .601 .106 C8 .170 .624 .099 *Scale Reliability 0.79 0.86 0.82

Eigenvalue 4.360 4.045 3.053 Variance Explained (%) 16.149% 14.981% 11.306% *Scale Reliabilities (Loadings from Rotated Factor Matrix) Table 5.8: 3-Factor Solution, Career Management Items

The last set of scales that were analyzed included two scales used to measure

career satisfaction, which is subjective career success. The scales described previously all

measured career management, a construct treated separately from subjective career

success in the hypothesized model in this study. As mentioned in Chapter 4, there were

two separate scales used to assess career success. The first scale used to assess career

success measured an individual’s perceptions of their own career success. This measure

drawn from Greenhaus et al. loaded on a single factor and the internal reliability estimate

was 0.86. The reliability estimate found in this study is identical to the reliability

estimate reported in the Greenhaus et al (1990) study. The mean for this scale was M =

4.12 (SD = 1.054) on a 6-point scale.

The second sets of items used to measure perceptions of career success (i.e.

subjective career success) were drawn from the Heslin (2003) study. These items

measured an individual’s perception of career success in comparison to others in their

peer group. Consistent with the scale originally published in the Heslin (2003) study, all

six items loaded on one factor and the internal consistency reliability estimate for this

scale was 0.89. This estimate is slightly lower than the alpha reported in the Heslin

(2003) study where the reported estimate was 0.95. The mean for this scale was M =

4.203 (SD = 1.065) on a six-point scale. Taken together, both scales suggest that the

respondents in this sample slight agree that they are satisfied with the career success.

The factor loadings from the Individual Career Satisfaction and Peer-Related Satisfaction

appear in Tables 5.9 and 5.10, respectfully.

207



1. Relative to my career aspirations, I am satisfied with the progress I have made towards meeting my goals for advancement.

.767

2. Relative to my career aspirations, I am satisfied with

the overall success I have achieved in my career. .860


the progress I have made towards meeting my goals for income.

.690


the skill development I have attained. .681


the autonomy I have attained. .659


the intellectual stimulation I have attained. .658

Eigenvalue (3.620) Percent of Variance Explained (60.335%) Table 5.9: Career Success-Individual Career Satisfaction Scale, Items and Loadings

208



1. Relative to the people I perceive as peers, I am satisfied with the progress I have made towards meeting my goals for advancement.

.886

2. Relative to the people I perceive as peers, I am satisfied with the overall success I have achieved in my career.

.924

3. Relative to the people I perceive as peers, I am satisfied with the progress I have made towards meeting my goals for income.

.698

4. Relative to the people I perceive as peers, I am satisfied with the skill development I have attained.

.665

5. Relative to the people I perceive as peers, I am satisfied with the autonomy I have attained.

.671

6. Relative to the people I perceive as peers, I am satisfied with the intellectual stimulation I have attained.

.660

Eigenvalue (3.983) Percent of Variance Explained (66.376 %) Table 5.10: Career Success-Peer/Related Career Satisfaction Scale, Items and Loadings

209

Tests of Hypotheses

Main Effects of Network Constraints on Ego’s Network

Hypotheses 1-4 predicted that there are differences in the networks of working

adults with and without children. Specifically, theses hypotheses suggested that the

network characteristics that differ between working adults and working adults with

children include network size, network composition (i.e. network ties), and conversation

topics discussed within networks. The variables that were thought to contribute to theses

differences in network characteristics across parental status included gender, family

involvement, role segmentation, and job involvement would moderate the relationship

between parental status and ego’s network.

Prior to testing the hypotheses, a set of one-way ANOVAs were run to denote if

there were differences in network characteristics across parental status. Of note, parental

status was a dummy-coded variable, where 0 was used to identify working adults without

parental responsibilities and 1 was used to denote working adults with parental

responsibility. The results from these ANOVAs appear in Table 5.11. Of note, the

sample sizes across network size, network ties, and network content are not the same, as

some of the survey respondents did not questions related to all three network

characteristics. Similar to the missing data within the entire dissertation, any missing data

within the ANOVA analysis was coded appropriately. As noted previously, there was

less than 2% missing data, and the missing data was evenly distributed across all primary

study variables (e.g. 2-3 responses were missing on average per variable). Therefore,

there was no reason to exclude the data from one specific primary variable included in

the study.

210

Differences across parental status were not found among network size. Instead, it

was found that regardless of parental status, the average network size for the respondents

included in this sample was 8.61 ties. This finding suggests, that despite the prevailing

notion (e.g. Smith-Lovin & McPhearson, 1993) that the network size for parents will be

smaller than the networking size for adults without parental status, the results of this

dissertation suggests that network size of parents is not significantly different from the

networking size of adults without parental responsibility. This may suggest that parental

status alone will not impact network size, and working parents should not expect to have

a smaller network than individuals without parental responsibility.

In addition, differences across parental status were not found among network ties.

However, it should be noted that the average proportion of kin ties among the individuals

sampled in this dissertation was 0.56. This finding suggests that more than 50% of the

ties within each respondent’s network were kin ties. As a result, one can conclude that

parental status has little to do with the ties within an individual’s network. In fact,

research suggests that most North American networks are usually kin-centered networks

(Bearman & Parigi, 2004). Prevailing notion suggests that the nature of kin-centered

networks exists in the US because individuals make personal choices to talk to kin ties

about matters that are of great importance to them (Bearman & Parigi, 2004).

Differences across parental status were among the remaining network

characteristics, that is, network content. Specifically, it was found working adults with

parental responsibility were more likely to discuss family-oriented topics among

members of their network. Differences in network content also occurred within parental

status groups. Specifically, parents with only one child reported that 50 percent of the

211

conversations among the members of network were related to family topics (e.g.

children/household, marriage, health). In comparison, working adults with two or at least

three children reported 45% and 44%, respectively, as family-oriented conversation

topics, which is less than that reported by the working adults with one child.

After the initial set of one-way ANOVAs were conducted, a series of moderated

regressions were used to test Hypotheses 1-4. Moderated regression analyses were used

to tests the hypotheses; as this type of analysis, in comparison to ANOVA, allows the

researcher to include control variables in the analyses. Moderated regressions were also

used to test for both the main effects and interactions of the network constraints (i.e.

gender, family involvement, role segmentation, and job involvement) and their

relationship with parental status, on the network characteristics of interest (i.e. network

size, network content, and network ties). The specific interaction variables (e.g. gender x

parental status) were created in SPSS. Specifically, interaction terms were created by

multiplying parental status by each of the by each independent variable (e.g. gender,

family involvement). In total five interactions variables were created (parental status x

gender, parental status x family involvement, parental status x role segmentation (actual),

parental status x role segmentation (perceptual), and parental status x job involvement

from the existing data, prior to beginning the hypotheses testing. Of note, these variables

were not centered prior to the interaction terms being formed. The variables were not

centered prior to the interaction action terms being created because centering only needs

to be done for interactions between variables that do not have meaningful zero values. In

this case, parental status can have a meaningful zero category (either working adults

without kids or working adults with kids can be recoded as 0 and the other category 1).

212

Therefore, the main effect and any interactions with parental status (i.e. parental status x

gender, parental status x family involvement, parental status x role segmentation (actual),

parental status x role segmentation (perceptual), and parental status x job involvement)

should not be centered.

The control variables included in these analyses were respondent age, number of

hours worked per week, number of work interruptions, and highest level of education

completed. The purpose in including the control variables was to statistically account for

factors other than the independent variables of interest, that are known to influence the

dependent variables included in this study. Also, the decision was made not include

organizational tenure, job tenure, and employment tenure (see Table 5.2). Each of these

variables was highly correlated with age (i.e. correlations were greater than r=.70). As a

result the decision was made to keep age as one of the control variables, and eliminate

organizational tenure, job tenure, and employment tenure. Age is control variable that is

included is most studies related to careers and networks, thus is it made sense to keep this

variable. Finally, the decision was made to exclude employment status as one of the

control variables. Employment status was not correlated with any of the variables

included in the study thus it was unnecessary to keep this as one of the control variables

in the study.

213

Variable

214

N = Working Adults Without Children

N= Working Adults With Children

Total MeanN

STD Df F Sig

Network Size

118 152 270 8.60 5.123 269 1.098 .350

Network Ties

116 150 266 0.57 0.28 265 1.97 .120

Network Content

116 151 267 0.44 0.21 266 3.05 .029**

Table 5.11: One-Way ANOVAs between Parental Status and Ego Network Characteristics **p > 0.05

215

Parental Status and Gender on Network Size, Ties and Content

Hypotheses 1a predicted that working adults with parental responsibility will have

a smaller network (i.e. network size) than working adults without parental status; and

among working parents, working mothers will have a small network than working fathers

(i.e. the interaction between gender and parental status will result in a negative

relationship with network size). Tables 5.12 – 5.14 illustrates the results of the regression

analyses used to investigate this hypothesis.

Hypothesis 1a was not supported. That is, there was no support found for the main

effect of parental status on network size. Also, there was no support found for the

interaction of gender x parental status on network size. Results from this are shown in

Table 5.12.

Hypothesis 1b suggested that working adults without parental responsibility will

have more kin ties within their network than working adults without parental

responsibility; and among working parents, working mothers will have more kin ties

within their network than working fathers (i.e. the interaction between gender and

parental status will result in a negative relationship with network ties). Hypothesis 1b

was not supported. That is, there was no support found for the main effect of parental

status on network size. In addition, there was no support found for the interaction of

gender x parental status on network ties. Results from this are shown in Table 5.13.

Lastly, Hypotheses 1c suggested that working adults with parental responsibility

would have more non-work network content than working adults without parental

responsibility; and among working parents, working mothers would have a higher

proportion of non-work content than working fathers (i.e. the interaction between gender

and parental status would result in a negative relationship with network ties). Hypothesis

1c was not supported. That is, there was no support found for the main effect of parental

status on network content. Also, there was no support found for the interaction of gender

x parental status on network content.

Overall, there was no support found that the interaction of gender with parental

status had any impact on network size, ties, or content. Further, there was no support

found for the main effect of gender on any of the network characteristics (i.e. network

size, network ties, and network content). While, Hypotheses 1a,1b, and 1c were not

supported, this outcome is interesting and meaningful. Specifically, several empirical

studies within the networking literature have found that demographic variables,

especially gender, do effect network size and network ties. Specifically, previous research

has suggested that men tend to have larger networks than women (e.g. Ragins and

Sundstrom, 1989), men tend to have a higher proportion of co-worker ties, while women

tend to have a higher proportion of kin ties (Mardsen, 1990), and women have been

found to avoid talking about their families at work (e.g. Singh et al. 2002). While,

previous research has found that gender predicts network size, network ties, and network

content, this study suggests that gender does not produce significant differences in

network characteristics. This finding suggests that while initial research found that gender

was helpful in explaining variance in networks, other factors may now be more important

in explaining the differences found across network characteristics. Consistent with this

notion, more recent research has begun to examine the role of personality traits in

predicting differences in networks (e.g. Bozionelos, 2003).

216

The regression analyses results for Hypotheses 1a, 1b, and 1c are shown in Tables

5.12, 5.13, and 5.14. All tables show each of the steps included in the regression

analysis. The first step includes the control variables, the second step tested for the main

effects, and the third step tested the hypothesized interactions (e.g parental status x

gender on network ties).

217

218

Table 5.12: Regression Results of the Relationship between Parental Status and Network Constraint (Gender) on Network Size, Hypothesis 1a

DV(s) Beta R2 Adjusted R

Df Change R

F2 2

change

F Overall

Step 1: .015 .000 4, 263 .015 1.033 1.033 Control Variables

Education .024 Hours

Worked Per Week

.059

Work Interruptions

.026

Ego’s Age -.091 Step2: Independent Variables

.021 Parental Status

-.197 -.002 6,261 .005 .697 .919

Gender -.118 Step3: Interaction Term Parental Status x Gender

.023 .181 -.003 7, 260 .003 .695 .886

2DV(s) Beta R Adjusted

RDf Change

RF

219

Table 5.13 Regression Results of the Relationship between Parental Status and Network Constraint (Gender) on Network Ties, Hypothesis 1b (* = p< 0.05)

2 2 change F Overall

Step 1: .075 .061 4, 259 .075 5.237 5.237 Control Variables

Education -.117 Hours

Worked Per Week

-.066

Work Interruptions

.014

Ego’s Age -.274 Step2: .088 .067 6,257 .013 1.884 4.143 Independent Variables

Parental Status

.188

Gender .178 Step3: .092 .067 7,256 .004 .994 3.693 Interaction Term Parental Status x Gender

-.210

220

Table 5.14 Regression Results of the Relationship between Parental Status and Network Constraint (Gender) on Network Content, Hypothesis 1c (* = p< 0.05)


Df Change R

F2 2

change

F Overall

Step 1: .029 .014 4,260 .029 1.920 1.920 Control Variables


Worked Per Week

-.116

Work Interruptions

.014

Ego’s Age -.043 Step2: .035 .013 6, 258 .006 .854 1.563 Independent Variables

Parental Status

.207

Gender -.020 Step3: .037 .011 7, 257 .002 .448 1.401 Interaction Term Parental Status x Gender

-.145

Parental Status and Family Involvement on Network Size, Ties and Content

Hypothesis 2a suggested that the parental status-network size relationship will be

moderated by family involvement. That is, family involvement will interact with parental

status, such that parents that are highly involved with their families will have smaller

networks compared to parents that are less involved with their families (i.e. when family

involvement and parental status interact, there will be a negative relationship with

network size). Hypothesis 2a was not supported. That is, there was no support found for

the interaction of family involvement and parental status on network size.

Hypothesis 2b suggested that the parental status-network ties relationship will be

moderated by family involvement. That is, family involvement will interact with parental

status, such that parents that are highly involved with their families will have a higher

proportion of kin ties in their network compared to parents that are less involved with

their families. Hypothesis 2b was not supported. That is, there was no support found for

the interaction of family involvement x parental status on network ties.

Finally, Hypothesis 2c suggested that the parental status-network content

relationship will be moderated by family involvement. That is, family involvement will

interact with parental status such that parents that are highly involved with their families

will have a higher proportion of kin/non-work network content compared to parents that

are less involved with their family. Hypothesis 2c was not supported. That is, there was

no support found for the interaction of family involvement x parental status on network

content.

Overall, there was no support found that interaction of parental status with family

involvement had any impact on network size, ties, or content. Further, there no support

221

found for the main effect of family involvement on any of the network characteristics. As

a result, it appears that family involvement, alone, is not a useful variable in explaining

the variance in networks. As mentioned in the literature review, family involvement is

measure used to assess the importance of a specific role in one’s life. Further, role

involvement is thought to lead to conflict among individuals because (1) high levels of

involvement that role may lead to increased amount of time spent in that role, therefore

allowing less time to be allocated to the second role (Greenhaus & Beutell, 1985).

Although family involvement is often measured in work-family studies, the impact of

family involvement on the criterion variable(s) of interest is not studied independently.

Rather, usually work-family research studies the interaction between role involvement

and work-family conflict, and the impact of that interaction on various outcomes (e.g. job

satisfaction). Thus, future research that is interesting in examining if family involvement

impacts networks, should look at the interaction of family involvement with work-family

conflict, and determine if that interaction explains variance across network

characteristics.


5.15, 5.16, and 5.17. The results shown in Tables 5.15 -5.17 present each step of the

regression analyses, where the first step of the analyses included the control variables, the

second step tested for the main effects, and the third step tested the hypothesized

interactions (e.g. parental status x family involvement on network size).

222

223

Table 5.15: Regression Results of the Relationship between Parental Status and Network Constraint (Family Involvement) on Network Size, Hypothesis 2a (* = p< 0.05)


Df Change R

F2 2

change

F Overall



Worked Per Week

.061

Work Interruptions

.026

Ego’s Age -.089 Step2: .019 -.004 6,261 .004 .470 .843 Independent Variables

Parental Status

-.052

Family Involvement Composite

.053

Step3: Interaction Term

.019 Parental Status x Family Involvement Composite

.006 -.007 7,260 .000 .001 .720

224

Table 5.16 Regression Results of the Relationship between Parental Status and Network Constraint (Family Involvement) on Network Ties, Hypothesis 2b (* = p< 0.05)


Df Change R

F2 2

change

F Overall

Step 1: .075 .061* 4, 259 .075 5.237 5.237* Control Variables


Worked Per Week

-.065

Work Interruptions

.004

Ego’s Age -.246 Step2: .076 .054 6,257 .001 .158 3.521 Independent Variables

Parental Status

.049


.052

Step3: .076 .051 7,256 .000 .083 3.019 Interaction Term Parental Status x Family Involvement Composite

-.067

225

Table 5.17 Regression Results of the Relationship between Parental Status and Network Constraint (Family Involvement) on Network Content, Hypothesis 2c (* = p< 0.05)


Df Change R

F2 2

change

F Overall



Worked Per Week

-.121

Work Interruptions

.044


Parental Status

.241


.105

Step3: .037 .011 7,257 .003 .793 1.430 Interaction Term Parental Status x Family Involvement Composite

-.212

Parental Status and Role Segmentation (Actual and Perceptual) on Network Size, Ties

and Content

As mentioned previously, two measures of role segmentation were used in this

study. The perceptual measure of role segmentation was used to determine the

respondent’s desirability for clear segmentation between their work and home lives (e.g.

“I do not desire to be required to do work while at home”). In comparison, the actual

measure of role segmentation was a network measure created by looking at the overlap

between the types of conversation topics an ego discusses among the ties within their

network. A case of segmentation occurred when the ego did not have any overlapping

work and family conversation topics among the members in their network. That is, the

ego talked with some members in their network about work topics and they talked with

other members in their network about family topics, and there was no overlap between

the two. If an ego spoke with members within their network about both work and family

topics, this was an indication that the ego prefers to integrate the work and family lives.

As such, both actual and perceptual measures of role segmentation were examined to

understand the influence these variables have with parental status on the three network

characteristics (size, content, and size).

Hypothesis 3a suggested the parental status-network size relationship will be

moderated by role segmentation. That is, role segmentation will interact with parental

status, such that parents that clearly segment their work and family roles will have

smaller networks compared to parents that do not clearly segment their work and family

roles. Hypothesis 3a was not supported. That is, there was no support found for the

226

interaction of role segmentation (i.e. perceptual role segmentation or actual role

segmentation) x parental status on network size.


moderated by role segmentation. That is, role segmentation will interact with parental

status, such that parents that clearly segment their work and family roles will have a

higher proportion of kin/network ties compared to parents that do not clearly segment

their work and family roles. Hypothesis 3b was not supported. That is, there was no

support found for the interaction of role segmentation (i.e. perceptual role segmentation

or actual role segmentation) x parental status on network size.

However, there was support found for the main effect of role segmentation

(perceptual) on network ties. That is, the reported beta between role segmentation

(perceptual) was significant (B= .245), and it indicated a positive relationship between

role segmentation (perceptual) and network ties. In other words, as role segmentation

increases, the proportion of kin ties within an individual’s network also increases.

Therefore, individuals that clearly segment their work and home lives, such that there is

no overlap between these two domains, tend to have a higher proportion of kin ties within

their network. Results from this analysis are displayed in Table 5.19. Of note, there was

no significant relationship found for the main effect of actual role segmentation on

network ties (See Table 5.22).

Finally, Hypothesis 3c suggested the parental status-network content relationship

will be moderated by role segmentation. That is, role segmentation will interact with

parental status, such that parents that clearly segment their work and family roles will

have a higher proportion of kin/non-work related network content compared to parents

227

that do not clearly segment their work and family roles. Hypothesis 3c was not

supported. That is, there was no support found for the interaction of role segmentation

(i.e. perceptual role segmentation or actual role segmentation) x parental status on

network content.

However, there was support found for the main effect of role segmentation

(perceptual) on network content. That is, the reported beta between role segmentation

(perceptual) was significant (B= .219), and it indicated a positive relationship between

role segmentation (perceptual) and network ties. Results from this are displayed in Table

5.20. In other words, as role segmentation increases, the proportion of non-work content

also increases. Therefore, individuals that clearly segment their work and home lives,

such that there is no overlap between these two domains, tend to have a tendency to

discuss a higher proportion of non-work content that they discuss with members of their

network. Thus, these individuals have very little overlap in their work and non-work

conversations. Therefore, they talk to certain members in their network about work topics

(e.g. career, current projects), and the talk to a separate and distinct group of members

within their network about non-work topics (e.g. children, household issues).

A similar finding was discovered for the main effect of actual role segmentation

on network content. That is, there was a positive and significant relationship found for the

main effect of actual role segmentation found on network content. Specifically, the

reported beta between role segmentation (actual) was significant (B= .166). This result

also suggests that individuals that clearly segment their work and family lives, speak with

certain members within their network about work topics, and they speak to a separate and

distinct group of individuals about non-work content. Results from this analysis are

228

displayed in Table 5.23. Of note, both actual and perceptual measures of role

segmentation were included in this study. Perceptual measure was used to assess an

individual’s attitude toward separating their work and home lives, while the actual role

segmentation measure was a true measure of the extent to which there is overlap between

work and family topics a member discusses with each member within their network.

The regression analysis results for Hypotheses 3a, 3b, and 3c are shown in Tables

5.18 thru Tables 5.23. The results shown in Tables 5.18 thru Tables 5.23 are

demonstrative of each step of the regression analyses, where the first step of the analyses

included the control variables, the second step tested for the main effects, and the third

step tested for the proposed interactions (e.g. actual segmentation x network content).

229

230

Table 5.18 Regression Results of the Relationship between Parental Status and Role Segmentation (Perceptual) on Network Size, Hypothesis 3a (* = p< 0.05)


Df Change R

F2 2

change

F Overall



Worked Per Week

.060

Work Interruptions

.026

Ego’s Age -.103 Step2: .017 -.005 6,260 .002 .254 .767 Independent Variables

Parental Status

-.166

Perceptual Role Segmentation

-.065

Step3: .018 -.008 7,259 .001 .232 .688 Interaction Term Parental Status x Perceptual Role Segmentation

.140

231

Table 5.19 Regression Results of the Relationship between Parental Status and Role Segmentation (Perceptual) on Network Ties, Hypothesis 3b (* = p< 0.05)


Df Change R

F2 2

change

F Overall



Worked Per Week

-.063

Work Interruptions

-.002

Ego’s Age -.235 Step2: .112 .092* 6,256 .037* 5.404* 5.398* Independent Variables

Parental Status

.215


.245*

Step3: .115 .090 7, 255 .002 .670 4.717 Interaction Term Parental Status x Perceptual Role Segmentation

-.228

232

Table 5.20 Regression Results of the Relationship between Parental Status and Role Segmentation (Perceptual) on Network Content, Hypothesis 3c (* = p< 0.05)


Df Change R

F2 2

change

F Overall



Worked Per Week

-.120

Work Interruptions

.039


Parental Status

.269


.219*

Step3: .063 .038 7,256 .002 .559 2.465 Interaction Term Parental Status x Perceptual Role Segmentation

-.214

233

Table 5.21 Regression Results of the Relationship between Parental Status and Role Segmentation (Actual) on Network Size, Hypothesis 3a (* = p< 0.05)


Df Change R

F2 2

change

F Overall



Worked Per Week

.056

Work Interruptions

.019


.018 Parental Status

-.262 -.005 6,257 .002 .291 .771

Actual Role Segmentation

-.035

Step3: .024 -.003 7, 256 .006 1.526 .881 Interaction Term Parental Status x Actual Role Segmentation

.255

234

Table 5.22 Regression Results of the Relationship between Parental Status and Role Segmentation (Actual) on Network Ties, Hypothesis 3b (* = p< 0.05)


Df Change R

F2 2

change

F Overall



Worked Per Week

-.060

Work Interruptions

.005


Parental Status

-.039


.122

Step3: .091 .066 7,256 .000 .014 3.650 Interaction Term Parental Status x Actual Role Segmentation

.024

235

Table 5.23 Regression Results of the Relationship between Parental Status and Role Segmentation (Actual) on Network Content, Hypothesis 3c (* = p< 0.05)


Df Change R

F2 2

change

F Overall



Worked Per Week

-.113

Work Interruptions

.046


Parental Status

.067*


.166*

Step3: .056 .031 7,256 .000 .013 2.185 Interaction Term Parental Status x Actual Role Segmentation

-.023

Parental Status and Job Involvement on Network Size, Ties and Content

Hypothesis 4a proposed that the parental status-network size relationship will be

moderated by job involvement. That is, job involvement will interact with parental status,

such that parents that are not highly involved with their jobs will have smaller networks

compared to parents that are not highly involved with their jobs. Hypothesis 4a was not

supported. That is, there was no support found for the interaction of job involvement x

parental status on network size.


moderated by job involvement. That is, job involvement will interact with parental status,

such that parents that are not highly involved with their jobs will have a higher proportion

of kin ties within their network compared to parents that are not highly involved with

their jobs. Hypothesis 4b was not supported. That is, there was no support found for the

interaction of job involvement x parental status on network ties. However, there was

support found for the main effect of role job involvement on network ties. Specifically,

there was a significant, negative relationship found for the main effect of job involvement

on network ties, (B= -.260). Results for this can be found in Table 5.25. This outcome

suggests that there is a negative relationship between job involvement and network ties,

or when job involvement increases, proportion of kin ties within an individual’s network

decreases. That is, it suggests that when individuals are highly involved with their jobs

they will have less kin ties in their network, and vice versa . In sum, job involvement and

proportion of kin ties are negatively related.

236

Finally, hypothesis 4c suggested that the parental status-network content relationship will

be moderated by job involvement. That is, job involvement will interact with parental

status, such that parents that are not highly involved with their jobs will have a higher

proportion of kin/non-work network content compared to parents that are highly involved

in with their jobs. Hypothesis 4c was not supported. That is, there was no support found

for the interaction of job involvement x parental status on network ties. However, there

was a significant, negative relationship found for the main effect of job involvement on

network content. The beta for job involvement on network contact was (-.235). Results

from this can be seen in Table 5.26. This suggests that as job involvement are negatively

related. That is, as job involvement increases, the proportion of non-work content

decreases and vice-versa.


5.24, 5.25, and 5.26, respectfully. The results shown in Tables 5.24 -5.26 present each

step of the regression analyses, where the first step of the analyses included the control

variables, the second step tested for the main effects, and the third step tested for the

proposed interactions.

237

238

Table 5.24 Regression Results of the Relationship between Parental Status and Network Constraint (Actual) on Network Size, Hypothesis 4a (* = p< 0.05)


Df Change R

F2 2

change

F Overall



Worked Per Week

.037

Work Interruptions

.034


Parental Status

.046

Job Involvement

.111

Step3: .024 -.002 7,260 .000 .133 .914 Interaction Term Parental Status x Job Involvement

-.078

239

Table 5.25 Regression Results of the Relationship between Parental Status and Network Constraint (Actual) on Network Ties, Hypothesis 4b (* = p< 0.05)


Df Change R

F2 2

change

F Overall



Worked Per Week

-.020

Work Interruptions

-.010

Ego’s Age -.244 Step2: .102 .082 6,257 .028* 3.961* 4.891* Independent Variables

Parental Status

-.307

-.260* Job Involvement

Step3: Interaction Term Parental Status x Job Involvement

.111 .314 .086 7,256 .008 2.312 4.544

240

Table 5.26 Regression Results of the Relationship between Parental Status and Network Constraint (Actual) on Network Content, Hypothesis 4c (* = p< 0.05)


Df Change R

F2 2

change

F Overall



Worked Per Week

-.062

Work Interruptions

.019

Ego’s Age -.044 Step2: .082 .061 6,258 .053* 7.474* 3.835* Independent Variables

Parental Status

.010

-.248* Job Involvement

Step3: .082 .057 7,257 .000 .053 3.283 Interaction Term Parental Status x Job Involvement

.048

Network Characteristics and Career Success

Hypotheses 5a, 5b, and 5c proposed there would be a relationship between each

of the network characteristics and each of the career success indicators. This section

begins with a discussion of Hypothesis 5a, that is, the relationships between network size

and salary, network size and salary growth, network size and promotions, network size

and individual career success, and network size and peer-related career success.

Specifically Hypothesis 5a proposed that network size will negatively influence

career success, such that as network size decreases, objective indicators of career success

(i.e. salary, salary growth, and promotions) will also decrease. Further, there will also be

a negative relationship between network size and career satisfaction; where if network

size decreases, career satisfaction will also decrease. Hypothesis 5a was partially

supported. That is, there was one significant relationship found between network size and

objective/subjective indicators of career success. Network size was related to salary

growth (see Table 5.29, Beta = -.122; change R2 = .015, p<.05). Although this

relationship was significant, the direction of this relationship was different from that

proposed in the hypothesized model. That is, this finding suggests that there is a negative

relationship between network size and salary growth, such that smaller network size

results in larger salary growth. This finding is counterintuitive to several findings in

previous studies (e.g. Burt, 1992), as most literature suggests that the relationship

between network size and salary are positive. That is, as network size increases, salary

also increases.

241

The regression analysis results for Hypothesis 5a are shown in Tables 5.27 (i.e.

network size on salary), 5.28 (i.e. network size on salary growth), 5.29 (i.e. network size

on promotions), 5.30 (i.e. network size on individual career satisfaction), and 5.31 (i.e.

network size on peer-related satisfaction). All control variables were entered into Step 1

of the regression and the independent variable of interest (e.g. network content) was

entered during Step 2 of the regression analysis.

242

243

Table 5.27 Regression Results of the Relationship of Network Size on Salary (Hypothesis 5a)


Df Change R

F2 2

change

F Overall

Step 1: .393 .383 4, 261

.393 42.170 42.170 Control Variables


Worked Per Week

.311

Work Interruptions

.045

Ego’s Age .465 Step2: Independent Variables

.393 Network Size .030 .382 5, 260

.001 .376 33.731

244

Table 5.28 Regression Results of the Relationship of Network Size on Salary Growth (Hypothesis 5a) * p= <.05


Df Change R

F2 2

change

F Overall

Step 1: .342 .332 4, 261

.342 33.950 33.950 Control Variables


Worked Per Week

.026

Work Interruptions

.164


.357 Network Size -.122* .345 5,260 .015* 5.916* 28.855*

245

Table 5.29 Regression Results of the Relationship of Network Size on Promotions

(Hypothesis 5a) * p= <.05


Df Change R

F2 2

change

F Overall



Worked Per Week

.179

Work Interruptions

.151


.126 Network Size .050 .111 5,260 .002 .737 7.639

246

Table 5.30 Regression Results of the Relationship of Network Size on Individual Career Satisfaction (Hypothesis 5a) * p= <.05


Df Change R

F2 2

change

F Overall



Worked Per Week

.103

Work Interruptions

.045


Network Size .043

247

Table 5.31 Regression Results of the Relationship of Network Size on Peer-Related Career Satisfaction (Hypothesis 5a) * p= <.05


Df Change R

F2 2

change

F Overall



Worked Per Week

.088

Work Interruptions

.043


Network Size .048

Hypothesis 5b proposed a negative relationship between network ties and the

objective indicators of career success. That is, as the proportion of kin ties increases,

objective indicators of career success (i.e. salary, salary growth, and promotions) will

decrease. Also, Hypothesized 5b proposed a negative relationship between network ties

and subjective indicators of career success. That is, as the proportion of kin ties increases,

subjective indicators of career success (i.e. individual career satisfaction and peer-related

career satisfaction) will also decrease.

There was partial support found for Hypothesis 5b. Specifically, two significant

relationships were found between network ties and the objective/subjective indicators of

career success. First, a significant relationship was found between network ties and

individual career satisfaction (Beta = -.174; change R2 = .028, p<.05). The results for the

regression analysis of network ties on individual career satisfaction are shown in Table

5.35. The results from this analysis suggest that individual career satisfaction declines as

the proportion of kin ties increases.

Next, a significant relationship was found between network ties and peer-related

career satisfaction (Beta= -.138; change R2 = .018, p<.05). The regression results are

presented in Table 5.36. The results from this analysis suggest that peer-related career

satisfaction declines, as the proportion of kin ties increases within a network.

As mentioned previously, the network ties measure was operationalized as the

percentage of non-work/kin ties within an individual’s network. The results of these

findings are consistent with Hypothesis 5b which suggested a negative relationship

between network ties and career satisfaction.

248

That is, as the proportion of kin or non-work ties increase within an individual’s network,

the less satisfied they will be with the careers; this includes the satisfaction an individual

has for their career when they consider their own career goals, and the satisfaction they

have with their career when they compare their career relative to individuals within their

peer group. The regression analysis results are shown in Tables 5.32 (network ties on

salary), 5.33 (network ties on salary growth), 5.34 (network ties on promotion), 5.35

(network ties on individual career satisfaction), and 5.36 (network ties on peer-related

career satisfaction). All control variables were entered into Step 1 of the regression and

the independent variable of interest (e.g. network content) was entered during Step 2 of

the regression analysis.

249

250

Table 5.32 Regression Results of the Relationship of Network Ties on Salary (Hypothesis 5b) * p= <.05


Df Change R

F2 2

change

F Overall

Step 1: .393 .383 4,257 .393 .000 41.524 Control Variables


Worked Per Week

.311

Work Interruptions

.454

Ego’s Age .046 Step2: .393 .382 5,256 .001 .371 33.212 Independent Variables

Network Ties -.031

251

Table 5.33 Regression Results of the Relationship of Network Ties on Salary Growth (Hypothesis 5b) * p= <.05


Df Change R

F2 2

change

F Overall



Worked Per Week

.020

Work Interruptions

.533


Network Ties .023

252

Table 5.34 Regression Results of the Relationship of Network Ties on Promotions (Hypothesis 5b) * p= <.05


Df Change R

F2 2

change

F Overall



Worked Per Week

.180

Work Interruptions

.236


Network Ties -.037

253

Table 5.35 Regression Results of the Relationship of Network Ties on Individual Career Satisfaction (Hypothesis 5b) * p= <.05


Df Change R

F2 2

change

F Overall



Worked Per Week

.094

Work Interruptions

-.165

Ego’s Age .046 Step2: .057 .039 5,258 .028* 7.650* 3.113* Independent Variables

Network Ties -.174*

254

Table 5.36 Regression Results of the Relationship of Network Ties on Peer-Related Career Satisfaction (Hypothesis 5b) * p= <.05


Df Change R

F2 2

change

F Overall



Worked Per Week

.082

Work Interruptions

.044


-.138* Network Ties

Hypothesis 5c proposed there will be a negative relationship between network

content and objective career success. That is, as the proportion of non-work content (i.e.

types of conversations) increases, objective indicators of career success (i.e. salary, salary

growth, and promotions) will decrease, that is, there is a negative relationship expected

between network content and the objective indicators of career success. In addition,

Hypothesis 5c proposed a negative relationship between network content and subjective

indicators of career success. That is, as the proportion of non-work content (i.e. types of

conversations) increases, subjective indicators of career success (i.e. individual career

satisfaction and peer-related career satisfaction) will decrease.

There was partial support found for Hypothesis 5c. Specifically, one significant

relationship was found between network content and the objective/subjective indicators

of career success. First, a significant relationship was found between network content and

salary (Beta = .104; change R2 = .010, p<.05). Results are presented in Table 5.37. The

results of the finding between network content and salary are not consistent with the

relationship proposed in the hypothesized model. Rather, it was expected that as the

proportion of non-work topics increased, salary would decrease. However, this finding

does demonstrate that network content effects salary. Research suggests that is important

for individuals to have access to the vocabulary and topics of conversation that are

relevant to the group (Tonsing & Alant, 2004). That is, an employee needs to be able to

talk about topics that are of interest to their coworkers in order to build and maintain

relationships at work (Tonsing & Alant, 2004). Therefore, one can conclude that

individuals that are interested in building their careers do not have to limit their network

content to work topics. Perhaps, speaking about topics other than work content helps the

255

individual identify topics that are of interest to their coworkers, inclduing topics related to

family-issues. Interestingly, a study found that when individuals were asked to identify

the topics of conversation they shared with coworkers, family, was not a topic that was

discussed with great frequency among coworkers (Tonsing & Alanta, 2004). Instead, the

topics that were discussed with most frequency among coworkers included food,

interpersonal relations, and work (e.g. work processes, work activities, and work

equipment) (Tonsing & Alanta, 2004). Thus, it appears that selecting or being mindful of

conversation topics among the members of your network is important. It appears that

individuals find success in building and maintaining relationships (i.e. which leads to the

gathering of job and career-related information) when they select topics that are familiar

to all members within a given group (Tonsing & Alanta, 2004).

The regression analysis results are shown in Table(s) 5.37 (network content on

salary), 5.38 (network content on salary growth), 5.39 (network content on promotions),

5.40 (network content on individual career satisfaction), and 5.41 (network content on

peer-related career satisfaction). All control variables were entered in the first step of the

regression, the predictor variable of interest (e.g. network size, network content, or

network ties) was entered during the second step of the regression.

256

257

Table 5.37 Regression Results of the Relationship of Network Content on Salary (Hypothesis 5c) * p= <.05


Df Change R

F2 2

change

F Overall



Worked Per Week

.326

Work Interruptions

.467


.403 .391* 5, 257 .010 4.516* 34.706* Network Content

.104*

258

Table 5.38 Regression Results of the Relationship of Network Content on Salary Growth (Hypothesis 5c) * p= <.05


Df Change R

F2 2

change

F Overall



Worked Per Week

.025

Work Interruptions

.158

Ego’s Age .529 Step2: .345 .332 5,257 .003 1.055 27.065 Independent Variables

Network Content

.053

259

Table 5.39 Regression Results of the Relationship of Network Content on Promotions (Hypothesis 5c) * p= <.05


Df Change R

F2 2

change

F Overall



Worked Per Week

.188

Work Interruptions

.148

Ego’s Age .248 Step2: .128 .111 5,

257 .002 .566 7.514

Independent Variables

Network Content

.044

260

Table 5.40 Regression Results of the Relationship of Network Content on Individual Career Satisfaction (Hypothesis 5c)


Df Change R

F2 2

change

F Overall



Worked Per Week

.094

Work Interruptions

.051


Network Content

-.091

261

Table 5.41 Regression Results of the Relationship of Network Content on Peer-Related Career Satisfaction (Hypothesis 5c) * p= <.05


Df Change R

F2 2

change

F Overall



Worked Per Week

.082

Work Interruptions

.049


Network Content

-.076

Network Characteristics and Career Management

The network characteristics size, ties, and content were expected to have a

relationship with each of the career management perceptions (i.e. career planning, career

tactics, and career mobility-preparedness). As mentioned previously, career management

is a process where employees gather information to help them make key decisions about

their careers. In order to assess this process, career management perceptions must be

measured by a set of items that reflect the multiple dimensions of this construct. As a

result, there were three different dimensions of career management (career planning,

career tactics, and career mobility preparedness) measured in this study.

Hypothesis 6a, 6b, and 6c suggested that network size, network ties, and network

content, will each have a relationship with career management perceptions. Specifically,

Hypothesis 6a proposed there will be a negative relationship between network size and

career management perceptions. That is, as network size decreases, an individual’s

perceptions of their ability to manage their career will also decrease. Also, if network size

increases, an individual’s perception of their ability to manage their career will also

increase.

Partial support was found for Hypothesis 6a. Specifically, one significant

relationship was found between network size and career management perceptions. There

was a significant relationship found between network size and career mobility

preparedness (Beta = .152; change R2 = .023, p<.05). Results from this finding are

presented in Table 5.44. This finding suggested that as network size increases an

individual’s perception of their ability to prepare to mobilize their career also increases

(e.g. actively discuss internal career opportunities with members within their network).

262

The regression analysis results are shown in Tables 5.42 (network size on career

planning), 5.43 (network size on career tactics), and 5.44 (network size on career

mobility preparedness). The results in Table 5.42 – 5.44 are demonstrative of all steps

included in the regression analysis. All control variables were entered into Step 1 of the

regression and the independent variable of interest (e.g. network size) was entered during

Step 2 of the regression analysis.

263

264

Table 5.42 Regression Results of the Relationship of Network Size on Career Planning (Hypothesis 6a) * p= <.05


Df Change R

F2 2

change

F Overall



Worked Per Week

.047

Work Interruptions

-.019


Network Size .056

265

Table 5.43 Regression Results of the Relationship of Network Size on Career Tactics (Hypothesis 6a) * p= <.05


Df Change R

F2 2

change

F Overall

Step 1: Control Variables

.054 Education -.034 .039 4,263 .054 3.725 3.725 Hours

Worked Per Week

.111

Work Interruptions

-.041


.065 Network Size .108 .047 5,262 .012 3.247 3.655

266

Table 5.44 Regression Results of the Relationship of Network Size on Career Mobility Preparedness (Hypothesis 6a) * p= <.05


Df Change R

F2 2

change

F Overall



Worked Per Week

.118

Work Interruptions

.036

Ego’s Age -.196 Step2: .023 .106* 5,262 .023 6.783* 7.300* Independent Variables

.152* Network Size

Hypothesis 6b proposed a negative relationship between network ties and career

management perceptions. That is, as the proportion of non-work/kin ties (in comparison

to non-work ties) increases, perceptions of career management will decrease. Hypothesis

6b was not supported. That is, there was no support found for the main effect of network

content on career planning, career tactics, or career mobility preparedness.

The regression analyses results are shown in Tables 5.45 (network ties on career

planning), 5.46 (network ties on career tactics), and 5.47 (network ties on career mobility

preparedness). All control variables were entered into Step 1 of the regression and the

independent variable of interest (e.g. network ties) was entered during Step 2 of the

regression analysis.

267

268

Table 5.45: Regression Results of the Relationship of Network Ties on Career Planning (Hypothesis 6b) *p= <.05


Df Change R

F2 2

change

F Overall



Worked Per Week

.047

Work Interruptions

-.018


Network Ties -.050

269

Table 5.46: Regression Results of The Relationship of Network Ties on Career Tactics (Hypothesis 6b) *p=<.05


Df Change R

F2 2

change

F Overall



Worked Per Week

.115

Work Interruptions

-.039


Network Ties -.050

270

Table 5.47 Regression Results of the Relationship of Network Ties on Career Mobility Preparedness (Hypothesis 6b)


Df Change R

F2 2

change

F Overall



Worked Per Week

.121

Work Interruptions

.039


Network Ties -.091

Finally, Hypothesis 6c proposed a negative relationship between network content

and career management perceptions. That is, as the proportion of non-work content (i.e.

topics of conversation) increases, perceptions of career management will decrease. There

was partial support found for Hypothesis 6c. Specifically, one significant relationship

was found between network content and career management perceptions. A significant,

negative relationship was found between network content and career tactics (Beta= -.165;

change R2 = .026, p<.05). Results are presented in Table 5.49. The results of this analysis

are consistent with those made in the hypothesized relationships. The negative beta

between (-.165) and career mobility tactics suggested that as the proportion of non-work

topics (in comparison to work topics) increases, an individual’s perception of their ability

to engage in career tactic behaviors (e.g. networking, seeking developmental feedback), a

facet of career management, will decrease.

The regression analyses results for the relationships proposed in Hypothesis 6c

are shown in Tables 5.48 (network content on career planning), 5.49 (network content on

career tactics), and 5.50 (network content on career mobility preparedness). All control

variables were entered into Step 1 of the regression and the independent variable of

interest (e.g. network content) was entered during Step 2 of the regression analysis.

Finally, Table 5.51 presents an overview of the hypotheses and results presented

in Chapter 5. A discussion of the findings will be presented in Chapter 6.

271

272

Table 5.48 Regression Results of the Relationship of Network Content on Career Planning (Hypothesis 6c) * p= <.0 5


Df Change R

F2 2

change

F Overall



Worked Per Week

.038

Work Interruptions

-.011

Ego’s Age -.135 Step2: .032 .013 5, 529 .010 2.671 1.685 Independent Variables

Network Content

-.101

273

Table 5.49 Regression Results of the Relationship of Network Content on Career Tactics (Hypothesis 6c) * p= <.0 5


Df Change R

F2 2

change

F Overall



Worked Per Week

.097

Work Interruptions

-.028


.080 Network Content

-.165* .062* 5,259 .026 7.451* 4.509*

274

Table 5.50 Regression Results of the Relationship of Network Content on Career Mobility Preparedness (Hypothesis 6a) * p= <.0 5


Df Change R

F2 2

change

F Overall



Worked Per Week

.116

Work Interruptions

.044


Network Content

-.086

Number Hypothesis Result 1A Not Supported Hypothesis 1a: Working adults with parental

responsibility will have a smaller network (i.e. network size) than working adults without parental status; also among working parents, working mothers will have a smaller network than working fathers (i.e. the interaction between gender and parental status will result in a negative relationship with network size).

1B Not Supported Hypothesis 1b: Working adults with parental responsibility will have a higher proportion of kin ties within their network than working adults without parental responsibility; also among working parents, working mothers will have a higher proportion of kin ties within their network than working fathers (i.e. the interaction between gender and parental status will result in a negative relationship with network ties).

1C Not Supported Hypothesis 1c: Working adults with parental responsibility will have a higher proportion of non-work network content, than working adults without parental responsibility; also among working parents, working mothers will have a higher proportion of non-work content than working fathers (i.e. the interaction between gender and parental status will result in a negative relationship with network ties).

Continued

Table 5.51: Summary of Study Hypotheses and Findings

275

Table 5.51 continued

2A Not Supported Hypothesis 2a: The parental status-network size relationship will be moderated by family involvement. That is, family involvement will interact with parental status, such that parents that are highly involved with their families will have smaller networks compared to parents that are less involved with their families. (i.e. When family involvement and parental status interact, there will be a negative relationship with network size).

2B Not Supported Hypothesis 2b: The parental status-network ties relationship will be moderated by family involvement. That is, family involvement will interact with parental status, such that parents that are highly involved with their families will have a higher proportion of kin ties in their network compared to parents that are less involved with their families (i.e. when family involvement and parental status interact, there will be a positive relationship with network ties).

2C Not Supported Hypothesis 2c: The parental status-network content relationship will be moderated by family involvement. That is, family involvement will interact with parental status, such that parents that are highly involved in their families will have a higher proportion of kin/non-work network content compared to parents that are less involved with their families (i.e. when family involvement and parental status interact, there will be a positive relationship with network content).

Continued

276


3A Not Supported Hypothesis 3a: The parental status-network size relationship will be moderated by role segmentation. That is, role segmentation will interact with parental status, such that parents that clearly segment their work and family roles will have smaller networks compared to parents that do not clearly segment their work and family roles. (i.e. when role segmentation and parental status interact, there will be a negative relationship with network size).

3B Not Supported Hypothesis 3b: The parental status-network ties relationship will be moderated by role segmentation. That is, role segmentation will interact with parental status, such that parents that clearly segment their work and family roles will have a higher proportion of kin network ties compared to parents that do not clearly segment their work and family roles. (i.e. when role segmentation and parental status interact, there will be a positive relationship with network ties).

3C Not Supported Hypothesis 3c: The parental status-network content relationship will be moderated by role segmentation. That is, role segmentation will interact with parental status, such that parents that clearly segment their work and family roles will have a higher proportion of kin/non-work related network content compared to parents that do not clearly segment their work and family roles. (i.e. when role segmentation and parental status interact, there will be a positive relationship with network content).

Continued

277


4A Not Supported Hypothesis 4a: The parental status-network size relationship will be moderated by job involvement. That is, job involvement will interact with parental status, such that parents that are not highly involved with their jobs will have smaller networks compared to parents that are highly involved with their jobs. (i.e. when job involvement and parental status interact, there will be a negative relationship with network size).

4B Not Supported Hypothesis 4b: The parental status-network ties relationship will be moderated by job involvement. That is, job involvement will interact with parental status, such that parents that are not highly involved with their jobs will have a higher proportion of kin ties within their network compared to parents that are highly involved with their jobs. (i.e. when job involvement and parental status interact, there will be a negative relationship with network ties).

4C Not Supported Hypothesis 4c: The parental status-network content relationship will be moderated by job involvement. That is, job involvement will interact with parental status, such that parents that are not highly involved with their jobs will have a higher proportion of kin/non-work network content compared to parents that are highly involved with their jobs. (i.e. when job involvement and parental status interact, there will be a negative relationship with network content).

Continued

278


5A Partial Support Network size will negatively influence career success, such that as network size decreases, objective indicators (i.e. salary, salary growth, and promotions) will also decrease. Further, there will also be a negative relationship between network size and career satisfaction; where if network size decreases, career satisfaction will also decrease.

5B Partial Support There will be a negative relationship between network ties and objective career success. That is, as the proportion of kin ties increases, objective indicators of career success (i.e. salary, salary growth, and promotions) will decrease. Also, there will be a negative relationship between network ties and subjective indicators of career success. That is, as the proportion of kin ties increases, subjective indicators of career success (i.e. individual career satisfaction and peer-related career satisfaction) will decrease.

5C Partial Support There will be a negative relationship between network content and objective career success. That is, as the proportion of non-work content (i.e. types of conversations) increases, objective indicators of career success (i.e. salary, salary growth, and promotions) will decrease. Also, there will be a negative relationship between network content and subjective indicators of career success. That is, as the proportion of non-work content (i.e. types of conversations) increases, subjective indicators of career success (i.e. individual career satisfaction and peer-related career satisfaction) will decrease.

Continued

279


6A Partial Support There will be a negative relationship between network size and career management perceptions. That is, as network size decreases, an individual’s perceptions of their ability to manage their career will also decrease.

6B Not Supported There will be a negative relationship between network content and career management perceptions. That is, as the proportion of non-work/kin ties increases, perceptions of career management will decrease.

6C Partial Support Hypothesis 6c: There will be a negative relationship between network content and career management perceptions. That is, as the proportion of non-work content (i.e. topics of conversation) increases, perceptions of career management will decrease.

280

CHAPTER 6

DISCUSSION

This chapter provides a discussion of the findings presented in Chapter 5. The

chapter begins with a brief summary of the results, which includes a discussion of the

theoretical implications. Next the practical implications of this study are discussed. This

section closes with a discussion of future research questions, study limitations, and a brief

discussion of post-hoc statistical analysis.

Overview of Findings

As discussed in Chapter 1, the purpose of this dissertation was to determine if

there were differences in the three network characteristics, (network size, ties, and

content) across parental status. In addition, this study sought to understand the

relationship that each of the three network characteristics had with one’s ability to

achieve career success and their relationship with career management perceptions.

The reason is it is important to understand networks, is because research suggests

that employees are now working in the age of the protean career, where the responsibility

to manage an individual’s career, has now shifted from the employer to the employee

(Hall, 2002). Therefore, one way an individual can manage their career is through their

networks.

281

If it is the case that networks are important to careers, it is important to understand (1) are

their any constraints that may lead to differences in network characteristics across

employees (i.e. parental status), and (2) it’s important to understand the relationship that

network characteristics has with various career outcomes. As a result this dissertation

used a social networking perspective to understand how individuals exchange

information for the purpose of advancing their careers. That is, a social networking

framework was used to study the influence of relationships (and how information is

exchanged within these relationships) on various career outcomes.

In an attempt to address these issues, this study began by seeking answering for

the following three questions: First, how do networks differ after the birth of a child for

males vs females? Second, how do networks differ between working adults with and

without children? Third, what constraints produce those differences? In order to address

these questions, a series of One-Way ANOVAs were run and series of hypotheses were

tested, per the results presented in Chapter 5.

First, a series of One-Way ANOVAs were run to determine if there were any

differences in the network characteristics, network size, network ties, and network

content, across parental status. The results from the One-Way ANOVAs presented in

Chapter 5 (see Table 5.14), and the conclusion drawn from this analysis is, that working

adults with children and working adults without children, only differ on one of the three

network characteristics studied, that is, network content. As discussed in Chapter 5,

working adults with children, reported a higher percentage of their network content (i.e.

topics of conversation) to be related to non-work topics (e.g. children/household, health).

An interesting finding from this study was that the percentage of non-work content

282

individuals discuss with the members of their network, decreases as the number of

children increases. Specifically, parents with only one child reported that 50 percent of

the conversations among the members of network were related to family topics (e.g.

children/household, marriage, health). In comparison, working adults with two or at least

three children reported 45% and 44% , respectively, as family-oriented conversation

topics, which is less than that reported by the working adults with one child. Thus it

appears that working parents reported a higher proportion of non-work content if they

had just one child. As the number of children for which the working parent is responsible

for increases, the proportion of non-work content (e.g. family-related conversation topics)

also decreases.

The results showed no significant differences found across parental status for

network size and network ties. Overall, the average network size for the respondents

included in this sample was 8.61 ties. This is interesting to because it suggests that

despite the prevailing notion (e.g. Smith-Lovin & McPhearson, 1993) that the network

size for parents will be smaller than the networking size for adults without parental status,

the results of this study suggest that the network size of parents is not significantly

different from the networking size of adults without parental responsibility. This suggests

that parental status alone will not impact network size, and working parents should not

expect to have a smaller network than individuals without parental responsibility. In

addition, differences across parental status were not found among network ties. However,

it should be noted that the average proportion of kin ties among the study participants

was 0.56. This suggests that more than 50% of the ties within each respondent’s network

were kin ties. As a result, one can conclude that parental status has little to do with the

283

ties within an individual’s network. In fact, research suggests that most North American

networks are usually kin-centered networks (Bearman & Parigi, 2004). Prevailing notion

suggests that the nature of kin-centered networks exists in the US because individuals

make personal choices to talk to kin ties about matters that are of great importance to

them (Bearman & Parigi, 2004).

A series of regressions were run to test Hypotheses 1-4. Each of these hypotheses

proposed tested first for the main effect of parental status on each of the three network

characteristics (e.g. network size, ties, and content). Next, the hypotheses proposed tested

for the interaction effects of each of the 4 moderators with parental status (i.e. gender x

parental status, family involvement x parental status, job involvement x parental status,

and role segmentation x parental status) on the network characteristics included in the

study (i.e. network size, network ties, and network content).

As discussed in Chapter 5, when Hypotheses 1a, 1b, and 1c were tested, there was

no support found for the main effects of gender found for any of the three network

characteristics, size, ties, or content. Nor was there any support found for the interaction

between gender and parental status on any of the three network characteristics. That is,

there was no support found for Hypotheses 1a, 1b, and 1c. While, Hypotheses 1a,1b, and

1c were not supported, this outcome is interesting and meaningful. Specifically, several

empirical studies within the networking literature have found that demographic variables,

especially gender, relate to network size and network ties. Specifically, previous research

has suggested that men tend to have larger networks than women (e.g. Ragins and

Sundstrom, 1989), men tend to have a higher proportion of co-worker ties, while women

tend to have a higher proportion of kin ties (Mardsen, 1990), and women have been

284

found to avoid talking about their families at work (e.g. Singh et al. 2002). While

previous research has found that gender predicts network size, network ties, and network

content, this study suggests that gender does not produce significant differences in

network characteristics. This finding suggests that while initial research found that gender

was helpful in explaining variance in networks, other factors may be more important in

explaining the differences found across network characteristics. Consistent with this

notion, recent research has begun to examine the role of personality traits in predicting

differences in networks (e.g. Bozionelos, 2003).

In addition to gender, three other moderators, including family involvement, job

involvement, and role segmentation preferences, were hypothesized (Hypotheses 2-4) to

result in differences in network characteristics (i.e. network size, network ties, and

network content) across parental status. These hypotheses were not supported. That is,

the interaction between parental status and family involvement, job involvement, or role

segmentation, were not found to be related to network size, network ties, and network

content.

In examining the results from Hypotheses 2a-c, that is, the interaction between

parental status and family involvement on the three network characteristics, size, ties, and

content, the following observations can be made. As mentioned previously, no support

was found for the main effect of family involvement on any of the network

characteristics. As a result, it appears that family involvement alone is not a useful

variable for explaining the variance in networks. As mentioned in the literature review,

family involvement is used to assess the importance of a specific role in one’s life.

Further, role involvement is thought to lead to conflict among individuals. This occurs

285

among individuals because high levels of involvement in a role may lead to an increased

amount of time spent in that role, therefore allowing less time to be allocated to a second

role (Greenhaus & Beutell, 1985). Although family involvement is often measured in

work-family studies, the impact of family involvement on the criterion variable(s) of

interest is not studied independently. Rather, work-family research studies typically

investigate the interaction between family involvement and work-family conflict, and the

impact of that interaction on various outcomes (e.g. job satisfaction). Thus, future

research that is interested in examining if family involvement impacts networks, should

look at the interaction of family involvement with work-family conflict, and determine if

that interaction explains variance across network characteristics.

Related to the findings from Hypotheses 3a-c, which tested for the main effects of

role segmentation on the three network characteristics, and it tested for the interaction

between parental status and role segmentation on the three network characteristics, the

following observations can be made. There was support found for the main effect of two

of the moderating variables, that is, role segmentation and job involvement, network ties

and network content. First, there was a significant and positive main effect of perceptual

role segmentation on network ties. This finding suggests that there is a positive

relationship between role segmentation and network ties, such that as role segmentation

increases, the proportion of kin ties within a person’s network also increases. This

suggests that individuals that clearly want their work and family lives segmented, are

likely to have a higher number of kin ties within their network. In addition, there was also

a significant, main effect found for both actual and perceptual segmentation on network

content. The perceptual measure of role segmentation measures an individuals attitudes

286

or preferences for segmenting their work and home lives (e.g. “I desire to be able to

forget work while I am at home”). In comparison, the actual segmentation measure is

used to understand the extent to which an individual separates their work and home lives,

by considering their topics of conversations with members within their network. The

actual segmentation measure ranges from 0 to 1, where 0 indicates an individual that

talks to some people within their network about work only. Meanwhile, a person scoring

a one on the role segmentation measure, completely integrates their work and family

lives, and they talk about both work and family matters with all members within their

network. The positive significant relationship between role segmentation (both actual and

perceptual) and network content, suggests that the more an individual segments their

work and family lives, they will have a higher proportion of non-work/kin related topics

that they discuss among the members of their network. Thus, individuals that segment

their work and family lives are more likely to talk about non-work issues (e.g. parental

responsibility, health) with the people in their network. What is interesting about these

findings is there was no significant interaction found between role segmentation x

parental status on network content or network ties. Thus, it may be the case that an

individual’s parental status does not directly impact the types of ties that they have within

their network (kin vs non-kin ties), nor does it seem to impact what they discuss with the

people in their network (i.e. network content). Rather, the preferences for segmentation

between work and family lives, an individual difference receiving a lot of attention in the

work-family literature (e.g. Rothbard et al., 2004) may help increase understanding of

specific network characteristics, network ties and network content. As a result, future

studies may want to measure role segmentation preferences in future studies using

287

network analyses, if the researcher is interested in identifying other individual differences

besides gender, (something that is assumed to impact various network characteristics (e.g.

network size) ) to really develop a clearer picture of the type of individual differences

that cause variance in network characteristics.

In addition to the significant main effect found for role segmentation, a significant

main effect was also found for job involvement on both network ties and network

content. As expected, the relationship between job involvement and network ties was

found to be significant, but negative (as discussed in Chapter 5). The negative

relationship between job involvement and network ties, suggests that an individual that is

highly involved in their job, will have a smaller number of kin ties within their network.

This is not unexpected because an individual that is highly involved with their job is not

likely to spend a lot of time with kin ties outside of those in their immediate household

(e.g. wife, parent, in-law, child). In fact, previous research suggests that when individuals

are highly involved in one role, this will lead to an increased amount of time spent in that

role (e.g. Greenhaus & Beutell, 1985). Moreover, this finding does not suggest that

individuals highly involved with their jobs have zero kin ties within their network, they

are just likely to have fewer kin ties within their network than an individual that is not

highly involved with their job. One explanation for why individuals that are highly

involved in their jobs is directly related to one of the key factors thought to contribute to

high job involvement. Specifically, a factor that often contributes to high job involvement

is the supportive relationships an individual develops with their coworkers and

supervisors (Lodahl & Kejner, 1965). Therefore, if an individual is highly involved with

their job and is benefiting from the supportive relationships they have developed with

288

their coworkers and supervisors, than they are less likely to have a high proportion of kin

ties within their network; instead they are likely to have a higher proportion of work ties

within their network. Finally, there was also a significant main effect of job involvement

on network content (i.e. topics of conversation). The beta for job involvement on network

content was negative. This suggests that as job involvement increases the non-work

network content (i.e. non-work topics of conversation) also decline. The direction of this

relationship is understandable, as one would expect that an individual that is highly

involved with their job will also spend a lot of time discussing their jobs or other topics

related to their jobs (e.g. specific projects, management styles). In conclusion, individuals

that are highly involved with their jobs are likely to have smaller proportions of kin ties

within their network, and they will discuss fewer non-work/kin topics amongst members

of their network.

As mentioned previously, there were no main effects found for parental status any

of the three network characteristics included in this study, that is, network size, network

ties, or network content. However, prior to the hypotheses being tested, a series of One-

Way ANOVAs were conducted on the data. The ANOVA results shown in Table 5.14

suggests that network content differs across parental status. That is, of the three network

characteristics proposed to differ across parental status, network content was the only

characteristic found to have significant differences across parental status. In addition,

although network content differed across parental status, the difference across parental

status was not moderated by the four variables suggested in this study, that is, gender,

family involvement, role segmentation, and job involvement. Future research needs to

determine what antecedents contribute to the differences in network content across

289

parental status. One antecedent that may contribute to the differences found in network

content and in differences in other network characteristics of interest may be work-family

conflict.

Work-family conflict is the conflict individuals feel related to the pressure they

experience in balancing dual roles, that is their role at work and their role at home. That

is, work-family conflict is the role conflict individuals feel that makes compliance with

one role, difficult to comply with the other role (Kahn et al., 1964). Also, related to the

idea of work-family conflict is the theory of role strain, which suggests that individuals

experience a felt difficulty in fulfilling multiple role obligations (Goode, 1960). Finally,

Greenhaus & Beutell (1985) describe work-family conflict as three different types of

conflict an individual can face, which includes times-based conflict (where involvement

in one role is impeded by pressures in the other role), strained-based conflict (where

performance in role is impacted by performance in another role), and behavior-based

conflict (where performance in one role is made more difficult by the behavior required

in another role). In conclusion, the argument can be made that researchers interested in

using network analysis as a way of understanding the career challenges of working

parents should consider work-family conflict as one of the key antecedents in future

research. Work-family conflict is probably one of the key variables that is studied in the

work-family literature, and existing research demonstrates that work-family conflict is

empirically linked to other variables, for example job satisfaction (Kossek & Ozeki,

1998). Thus, in integrating the network, careers, and work-family literature, future

research should consider including antecedents from the work-family literature (i.e.

work-family conflict) in an attempt to explain the unique challenges working parents may

290

have in managing their careers, including maintaining their social capital (networks),

when they are compared to working adults without parental responsibility.

This study also sought to understand if a relationship exists between the network

characteristics size, ties, and content, and two specific career outcomes, career success

and career management perceptions (see Hypotheses 5-6). Hypotheses 5a – 5c predicted

the relationships between the three network characteristics and objective career success

indicators (e.g. salary, salary growth), while Hypotheses 6a-6c predicted the relationships

between the thee network characteristics and career management perceptions (e.g. career

planning, career tactics).

First, Hypotheses 5a, 5b, and 5 c proposed that network size (5a), network

ties(5b), and network content(5c) would have a relationship with each of the five career

success indicators (i.e. salary, salary growth, promotions, individual career success and

peer-related career success). Partial support was found for Hypotheses 5a, 5b, and 5c.

Specifically, network size, that is, the average number of ties in an individual’s network

was found to have a significant relationship with salary growth. The beta weight for

network size predicting salary growth was (-.122). This indicates for every -.122 unit

decrease in network size, is accompanied by a unit increase in salary growth. It was

expected that salary growth would increase as network size increases. A summary of the

regression results for Hypotheses 5a, 5b, and 5c can be seen in Table 6.1.

291


Salary Salary Growth

Promotions Individual Peer Career Sat Career Sat

Step 1

292

Education .294 .131 .057 -.098 -.128

Number of Hours/Week

.311 .026 .179 .103 .088

Ego’s Age .465 .514 .251 -.116 -.121

Work Interruptions

.045 .164 .151 .045 .043

Step 2

Network Size .030 -.122* .050 .043 .048

Change in R2 .001 .015* .002 .002 .002 (Step 2)

Overall Adjusted R

.382 .345* .111 .012 .017 2

Table 6.1: Summary Results of Network Size on Career Objective Career Success Indicators (H5a) *p >= 0.05

Hypotheses 5b proposed that network ties would be related to salary, salary

growth, promotions, and career satisfaction indicators. Network ties was found to have a

significant relationship on both individual career satisfaction and peer-related career

satisfaction. The beta weight for network ties predicting individual career satisfaction was

(-.174). This indicates for every -.174 unit decrease in network ties, is accompanied by an

unit increase in individual career satisfaction. The beta weight for network ties predicting

peer-related career satisfaction was (-.138). This indicates for every -.138 decrease in

network ties, is accompanied by an unit increase in peer-related career satisfaction.

Further this finding suggests that a high number of kin ties in an individual’s network

leads to negative relationship with career satisfaction. Finally, it is also important to

point out that this that this relationship between network ties and career satisfaction is

true regardless of parental status, as both working parents and working adults without

parental responsibility reported that greater than 50% of the network, on average, is

composed of kin ties.

One factor that may contribute to the relationship found between network ties and

it’s negative effect on career satisfaction is a result of what some researchers describe as

network position (Smith-Lovin & McPherson, 1993). Specifically, an individual’s

network ties, that is the types of relationship one develops through their network,

determines the roles that that an individual enacts in their daily lives (Smith-Lovin &

McPherson, 1993). In other words, if an individual has a network comprised of a higher

proportion of kin ties and fewer coworker ties, that individual is likely to spend more

time occupying the identity(s) that are closely tied to their family-oriented roles. That is,

an individual that has a greater amount of kin ties in comparison to non-kin ties will have

293

a stronger identity and spend more time occupying their roles that are associated with

their kin ties. For example, a woman may spend more time and identify with their role of

mother or daughter-in-law, as that is the role that is reinforced by the relationships they

hold with the people that are dominant in their network. In this case, an individual with a

higher proportion of kin ties, will spend less time occupying their role as coworkers, and

therefore will have less career satisfaction, as they are less likely to spend a large amount

of time or set goals related to their role and identity as a member of an organization.

In addition to network position, another factor that can help explain the negative

effect that network ties had on career satisfaction draws from Granovetter’s Weak Ties

Theory. It was expected that individuals with a higher proportion of kin ties within their

network would experience a decline in both individual career satisfaction and career-

related career satisfaction. Specifically, previous research has found a relationship

between career satisfaction and weak ties. That is, individuals that experience high levels

of career satisfaction also have a larger number of weak ties, as described in Chapter 2

and 3. As a result, if an individual has a greater proportion of kin ties, these ties are likely

to be strong ties. Strong ties, are good for providing an individual social and emotional

support, but they are typically less beneficial in providing an individual with

nonredundant information, as that is the role of weak ties. Thus, an individual with a high

proportion of kin ties, will likely have strong ties to those individuals. Empirical research

shows that the relationship between ties and career satisfaction is positive (perceived as

helpful) when an individual has weak ties.

Further, the study results are consistent with ideas expressed in previous research

related to kin ties. Specifically, previous research indicates that when network are

294

comprised of a high percentage of kin ties, kin ties are less likely to acquire new,

nonredundant information as it related to jobs and careers (Wellman, 1992). Thus, the

study results suggests that as the percentage of kin ties increases, individual will be less

satisfied with their careers. This is likely because as the number of kin increase,

individuals will acquire very little new information (Wellman, 1992). Further kin ties

tend to be strong ties, and previous research suggests that individuals will gather the most

job and career –related information from weak ties (Granovetter, 1972). That is,

instrumental actions, such as job searching, require diverse resources and are more likely

to be accomplished through weak ties (Wellman, 1992), and extensive, weak ties to

nonfamilial sources are most useful for finding a job and achieving higher income. One is

likely to feel more satisfied with their jobs if they believe they have access to both

information and social capital. This becomes difficult to achieve as the percentage of kin

ties increase within an individual’s network. Thus, this study demonstrated that it is

important to consider an individual’s network ties, as the nature of type of network tie

affects the type of support that an individual receives from that member in their network

(i.e. strong vs weak ties, where kin ties tend to be strong ties).

Lastly, very little research on career success has included a measure for peer-

related career success (Heslin, 2005). Therefore, this study contributes to the careers

management literature by finding empirical support for the idea that individuals use

multiple referent points, including peers, to evaluate their career success (Heslin, 2005).

Specifically, this study demonstrated that individuals do evaluate their career success

relative their personal criteria and relative to the outcomes attained by other people (i.e.

the significant, main effect found of network ties on peer-related career satisfaction).

295

This finding suggests that future studies that include career success should measure both

individual and peer-related career success. In other words, the meaning of career success

to an individual is more likely to be explained if both their self-referent and other-referent

career success are measured (Heslin, 2005). A summary of the results can be seen in

Table 6.2.

296




Step 1

297

Education .291 .131 .055 -.116 -.141


.311 .020 .180 .094 .082

Ego’s Age .454 .533 .236 -.165 -.161

Work Interruptions

.046 .162 .151 .046 .044

Step 2

Network Ties -.031 .023 -.037 -.174* -.138*

Change in R2 .001 .001 .001 .028* .018* (Step 2)

Overall Adjusted R

.382 .330 .110 .039* .032* 2

Table 6.2: Summary Results of Network Ties on Career Objective Career Success Indicators (H5b) *p >= 0.05

Finally, network content (i.e. Hypothesis 5c) was predicted to have a relationship

with both the career success indicators and career satisfaction indicators, that is, the

objective indicators of career success. Network content influenced salary, where the beta

was (Beta = .104). It was surprising that network content had a positive relationship with

salary, as it was expected that network content would have a negative relationship with

salary. Network content was measure of the amount of non-work content that individual

discussed with members in their network. Thus it was expected that individuals that

experienced salary growth would discuss a high proportion of work-content with their

network members. A summary of the regression results for Hypotheses 5a, 5b, and 5c can

be seen in Table 6.3.

One rationale for the finding between network content and salary maybe related to

what members discuss within their network and the composition of their network

members. Specifically, if an individual discusses their family responsibilities with

members of their networks, and if that individual’s network is comprised of powerful

individuals, that is, individuals that can influence their salary, then an individual may be

able to leverage conversations with members in their networks to express a need for

greater income due to changes within their family obligations. For example, suppose an

employee expresses to a manager who is also a member of their network a concern they

have related to their ability to complete work at home. The employee may suggest they

would be a more productive employee if they were able to work in the office. However,

in order to work in the office they will need additional money to pay for their child to go

to daycare. In this case, if an individual is able to make a link between their ability to

298

improve their productivity, a manager may be willing to increase the salary of this

individual, which in turn will allow them to work at the office and use the additional

money to pay for childcare. In sum, network content appears to be related to salary

growth. This seems most likely to occur when an individual directly links an increase in

their salary to work performance improvements, especially when the increase in salary

will allow the worker to be more efficient and be able to spend more time in the office,

without neglecting their childcare responsibilities.

299




Step 1

300

Education .300 .132 .061 -.102 -.131


.326 .025 .188 .094 .082

Ego’s Age .467 .529 .248 -.125 -.130

Work Interruptions

.039 .158 .148 .051 .049

Step 2

Network Content

.104* .053 .044 -.091 -.076

Change in R2 .010* .003 .002 .008 .006 (Step 2)

Overall Adjusted R

.391* .332 .128 .018 .020 2

Table 6.3: Summary Results of Network Content on Career Objective Career Success Indicators (H5c) * p >= .005

Hypotheses 6a,b,and c proposed a relationship between the network

characteristics and each of the career management perceptions. Hypothesis 6a proposed

that network size would influence career planning, career tactics. Network size was

found to significantly predict career mobility preparedness where the beta for network

size on career mobility preparedness was (.152). This suggests that a .152 increase in

career mobility preparedness will lead to an increase in network size. This relationships is

consistent with that propose in Hypothesis 5c which suggested that network size will

positively influence career mobility preparedness. See Table 6.4 for a summary of the

regression analyses.

Network size was successful in predicting both salary (Beta= -.122) and career

mobility preparedness ( Beta= .152). This study contributes to the career literature by

examining the impact that network size has on both career success indicators and career

management perceptions. In most previous studies, both career management and career

success indicators have not been investigated in the same study.

301


Career Planning

Career Tactics

Career Mobility Preparedness

Step 1

302

Education .012 -.034 .175


.047 .111 .118

Ego’s Age -.124 -.175 -.196

Work Interruptions

-.019 -.041 .036

Step 2

Network Size .056 .108 .152*

Change in R2 .003 .012 .023* (Step 2)

Overall Adjusted R

.006 .047 .106* 2

Table 6.4: Summary Results of Network Size on Career Management Indicators (H6a) * p >= .005

Hypothesis 6b proposed that network ties would influence career planning, career

tactics. Network ties was not found to significantly predict any of the career management

indicators. This finding seems to suggests that while network ties, does seem to influence

the objective success career indicators (e.g. salary), network ties do not appear to

perceptions of career management. See Table 6.5 for a summary of the regression

analyses. One can conclude that while network ties does have a small, but significant

relationship with subjective career success indicators (i.e. individual career satisfaction

and peer-related satisfaction), network ties do not explain any variance in career

management perceptions.

303


Career Planning

Career Tactics


Step 1

304

Education .007 -.038 .169


.047 .115 .121

Ego’s Age -.143 -.199 -.235

Work Interruptions

-.018 -.039 .039

Step 2

Network Ties -.050 -.050 -.091

Change in R2 .002 .002 .008 (Step 2)

Overall Adjusted R

.005 .038 .090 2

Table 6.5: Summary Results of Network Ties on Career Management Indicators (H6a) * p >= .005

Finally, there was one significant relationship found when network content

predicts career management perceptions. Specifically, network content was found to

predict career tactics (Beta =-.165). Career tactics are specific behaviors (e.g. networking

or seeking developmental feedback) that an employee engages in, when they are trying to

collect job and career-related information. The summary results from the relationships

proposed by Hypothesis 6c are shown in Table 6.6.

As mentioned previously, one significant relationship was found between network

content and career management indicators. That is, a significant relationship was found

between network content and salary (Beta = .104). Also, network content, had a

relationship with career tactics (Beta= -.165). However, in comparing the beta weights

(and the variance explained) network content seems to have a stronger relationship with

career management perceptions.

The negative relationship found between network content and career tactics was

consistent with the notion that topics of conversation matter. Specifically, research

suggests that is important for individuals to have access to the vocabulary and topics of

conversation that are relevant to the group (Tonsing & Alant, 2004). That is, an

employee needs to be able to talk about topics that are of interest to their coworkers in

order to build and maintain relationships at work (Tonsing & Alant, 2004). Therefore,

individuals must be selective about the conversation topics they discuss among people

from who they want to gather career or job-related information, as it is important for

individuals to discuss topics that are of relevance and interest to the individuals with

whom they want to build or maintain relationships. Interestingly, Tonsing and Alanta

(2004) found that when individuals were asked to identify the topics of conversation they

305

shared with coworkers, family, was not a topic that was discussed with great frequency

among coworkers, Instead, the topics that were discussed with most frequency among

coworkers included food, interpersonal relations, and work (e.g. work processes, work

activities, and work equipment) (Tonsing & Alanta, 2004). Thus, it appears that selecting

or being mindful of conversation topics among the members of your network is

important. It is likely the case that individuals find success in building and maintaining

relationships (i.e. which leads to the gathering of job and career-related information)

when they select topics that are familiar to all members within a given group (Tonsing &

Alanta, 2004).

306


Career Planning

Career Tactics


Step 1

307

Education .007 -.041 .173


.038 .097 .116

Ego’s Age -.135 -.194 -.217

Work Interruptions

-.011 -.028 .044

Step 2

Network Content

-.101 -.165* -.086

Change in R2 .010 .026* .007 (Step 2)

Overall Adjusted R

.013 .062* .089 2

Table 6.6: Summary Results of Network Content on Career Management Indicators (H6a) * p >= .005

In sum, the statistically significant relationships found between the three network

characteristics size, ties, and content, and the career outcomes career success indicators

and career management perceptions were all relatively small (that is the change in R-

squared did not exceed (.028). However, it does appear that each of the network

characteristics, that is, size, ties, and content were useful in predicting either a career

success indicator and/or a career management indicator. Specifically network size

predicted both salary and career mobility preparedness. Network ties predicted career

satisfaction, and network content predicted salary and career tactics. Therefore, based on

the study results in the influence of all three network characteristics on career outcomes

are worthy of future investigation . Alternatively, of the five career success indicators

included in the model (i.e. salary, salary growth, promotions, individual career

satisfaction, and peer-related career satisfaction), the network characteristics were not

useful in explaining any variance in promotions. Further, of the career management

perceptual measures included in the model (i.e. career planning, career tactics, and career

mobility preparedness), the network characteristics were not useful in predicting career

planning. Thus, if these same three network characteristics were tested in future studies

related to careers, it does not appear that it would be necessary to test for a relationship

between the three network characteristics and promotions, or the three network

characteristics and career planning.

308

Theoretical Implications

The hypothesized model described in Chapter 3 was developed by drawing from

two key theoretical frameworks. Those two theoretical frameworks include the weak ties

theory and boundary theory. Granovetter’s (1973) weak tie theory argues that individuals

should maximize the number of nonredundant ties within their network, as this will lead

to the individual learning more information about future job and career opportunities. For

the purposes of this study, the weak tie framework was used to identify the network

characteristics that were important to career outcomes (i.e. network size and network

ties), and this framework was used to understand and identify the relationships that were

likely to exist between the network characteristics (network size, network ties) and the

various career indicators (e.g. career mobility preparedness) included in this study (i.e.

the relationships proposed in Hypotheses 5-6).

Specifically, the components of the weak tie theory that are important to this

study are the distinction Granovetter makes between weak and strong ties, that is, the

weak tie theory is useful in this study in distinguishing the types of relationships that may

exist in an individual’s network. Weak ties are defined as relationships with individuals

that lack intimacy, where contact may be infrequent, and they usually provide an

individual with nonredundant information. Typical examples of weak ties include

neighbors, co-workers, etc. In comparison, strong ties are usually defined by close,

intimate relationships and ones where individuals usually rely on their ties for social

and/or emotional support. Examples of strong ties include spouse, parents, siblings, ect.

The network kin measure used in this studied, was a measure of the types of relationships

309

that individuals have with members of their network. Specifically network kin, was a

measure of the proportion of network ties, that were kin ties. The findings from the study

suggest that network ties did predict career success, specifically individual and peer-

related career satisfaction, and specifically network kin had a significant negative effect

on the career success indicators. Thus, it appears that using the weak tie theory to

distinguish among the types of ties that individuals have within their network, does

appear to be a sound theoretical framework to use when making this distinction, as

empirical evidence support that types of ties matter. In addition, another distinction made

in the weak tie theory is related to network size. The weak ties theory argues that bigger

is better, and individuals should maximize their network size in order to maximize the

amount of nonredundant career and job-related information they receive from members

in their network. This notion of network size was supported empirically in this study.

Specifically, network size was found to have a significant positive relationship with

career mobility preparedness. That is, the finding between network size and career

mobility preparedness suggested that network size is able to predict career mobility

preparedness, that is, larger network results in individuals having positive perceptions of

their ability to move within their career. Thus, it appears that the weak tie theory is at

least a good starting place for future empirical work that is interested in investigating the

relationships between network characteristics and career outcomes; especially if network

size and network ties are included as network characteristics in the study.

310

Thus, it appears that it was rationale to use the weak tie theory to develop some of the

relationships at the back-end of the conceptual model, that is, the portion of the model

that investigated the relationships between the network characteristics and career

outcomes.

It should be noted that in addition to network size and network ties, network

content was also found to predict several of the career outcome measures. However,

network content, that is, identifying what people are discussing with various members of

their networks is not a variable that can be directly explained by using the weak tie

theory. Therefore, it makes sense to investigate other theories that may be useful in

understanding the relationship between network content and career outcomes. In

conducting a literature search in Business Source Premier, fewer than 20 hits were found

when the key words communication (or conversation) and careers were entered. The

articles that were retrieved were from this search were practioner articles and they mostly

described how individuals should have conversations about careers with mentors or other

people if influence within organizations. Thus, there appears to be a conceptual need for

understanding the relationship between conversation content and career outcomes. There

has been some initial research done in sociology related to conversation topics.

Specifically, Bearman and Parigi (2004) looked at what people talk about when they

discuss important matters. This article presents empirical support for the idea that people

talk about different topics amongst the people in their network. However, it does not

provide any theoretical support for why these differences in conversation occur.

In addition to the weak tie theory, this study also used boundary theory as a

framework to guide the development of some of the hypothesized relationships displayed

311

in the model. Boundary theory was used to understand how one of the antecedents, role

segmentation, may moderate the relationship between parental status and network

characteristics. Boundary theory suggests that meeting individual preferences to integrate

or segment their work and family roles is a key determinant of role conflict. Further,

according to boundary theory, individuals have preferences for the extent to which they

want their work and home life integrated or segmented. Support was found to suggest

that role segmentation was useful in predicting two network characteristics, network

content and network ties. Specifically, a main effect was found for role segmentation on

network content and network ties. The main effect of role segmentation on network

content suggested that individuals that separate their work and family lives, tend to talk

about more non-work topics. Also, the main effect of role segmentation on network ties,

suggested that individuals that segment their work and family lives, have a higher

proportion of kin ties in their network. Given these findings, it made sense to use role

segmentation as a theoretical framework to hypothesis the relationship parental status

may have on network characteristics. However, although role segmentation had a main

effect on network content and network ties, it did not interact with parental status. Thus,

it can be concluded that role segmentation, regardless of parental status, is important in

determining the network characteristics size and content. As mentioned earlier, to truly

understand the unique challenges that parents face in maintaining their social networks

for the purposes of managing their careers, it makes sense to understand the relationship

between parental status and network characteristics using role theory. As mentioned

previously, role theory and work-family conflict (which is derived from role theory)

appears to be a key theoretical framework that should be used in future research related to

312

parents, networks, and careers. Work-family conflict is probably one of the key variables

that is studied in the work-family literature, and existing research demonstrates that

work-family conflict is empirically linked to other variables (Kossek & Ozeki, 1998).

Thus, in integrating the network, careers, and work-family literature, it makes sense to

include antecedents from the work-family literature (i.e. work-family conflict) in an

attempt to explain the unique challenges working parents may have in managing their

careers, including maintaining their social capital (networks), when they are compared to

working adults without parental responsibility.

Lastly, this dissertation also speaks to a research need initially addressed by

Sullivan (1999) who suggested that individual characteristics, such as gender, age, and

race, need to be investigated in terms of the overall development of large non-redundant

networks. The findings from this study do seem to suggest that while demographic

variables such as age and marital status do not explain variance in networks, there is

some evidence to suggest that role involvement and role segmentation are two

preferences that help explain variance in network characteristics. Specifically, individuals

that are highly involved with their jobs do seem to have a smaller proportion of kin ties

and they are less likely to discuss kin/and non-work related content among the members

of their network. Also, those individuals that clearly prefer to segment their work and

family roles, do have a higher proportion of kin ties within their network, and they are

more likely to discuss non-work/kin-related topics among the members in their network.

Individuals that have a higher proportion of kin ties in their network, are also individuals

that have a larger percentage of strong ties within their network. Social identity theory

would suggest that as an individual’s social identity becomes stronger they are more

313

likely to pick individuals within their network who are more similar to themselves. Thus,

one could argue that individuals that have a strong identity and prefer to segment their

work and family roles, view their family role as the most prominent identity in their lives.

As a result, they will have a tendency to have a higher proportion of kin ties, as they will

likely interact with individuals that are family members, that is, they select individuals

that are also part of their families.

Thus, it appears that one area of fruitful research is to further understand why an

individual’s role or identity impact specific network properties. This also suggests that

future research can benefit from the inclusion of psychological variables (Kalish &

Robbins, 2006) including role involvement. As of now, very few empirical studies have

examined the impact of various individual differences and psychological variables on

network characteristics. Thus far, it appears that the individual differences that have been

studied to understand their impact on network characteristics includes the Big 5 traits (i.e.

neuroticism & extraversion), self-monitoring, locus of control, and social identity.

Practical Implications

There are several practical implications of the study results. First, it does appear

that network size matters. Specifically, network size was related to salary growth, where

salary growth was measured as the number of salary increases an individual has

experienced over their career. However, in this study, network size had a negative

relationship with salary growth. This finding was surprising, as one would expect that as

network size increases salary growth would also increase. Network size did have a

positive relationship with career mobility preparedness, which was a measure of career

management. This finding suggests that individual should increase their network size, as

314

they will be in a better position to engage in multiple activities which are related to

managing their careers (e.g. seeking external contacts to learn about new job

opportunities). In addition, network ties were found to have a negative relationship with

both individual career satisfaction and peer-related career satisfaction. The practical

implication from this finding was individuals have to be mindful of the proportion of kin

ties in their network, in comparison to the proportion of work ties. An individual with a

higher proportion of kin ties within their network may feel less satisfied with the career,

because they are unable to leverage the kin ties in their network for various kinds of kin

support. Specifically, an individual receives different types of support from different ties

within their network. Often kin ties offer an individual a lot of social support, while work

ties may offer an individual career/job-related support. Perhaps, when an individual

receives a lot of social support when compared to job/career support, they are less

satisfied with their overall careers. Finally, network content, that is, what topics are being

discussed amongst the members of the network was related to both career success and

career management perceptions. This finding provides evidence that individuals must be

aware not only of who they are talking to, but what they are talking about when they

interact with members of their network. This is especially true for working parents, where

they are likely to have a higher percentage of non-work topics they discuss with members

within their network. This finding is consistent with the view, that one of the key

purposes of networks, as it relates to careers, are these relationships allow individuals to

gather and receive job and career-related information. Therefore, if an individual is

interested in better managing their own careers and experiencing higher levels of career

315

success, they must be cognizant of making an effort to discuss career-related topics with

members within their network.

From the organization’s perspective, there should be an effort made to help

employees join and become active in organizational networks, as there do seem to be

several benefits related to networks, as it relates to career management. Currently, many

organizations do have employee-based network groups present. However, a lot of these

network groups were developed in an effort to help individuals from underrepresented

groups (e.g. women) meet each other and provide social support. An example of this kind

of organizational group is found at AT Kearney Consulting Firm, where there was an

African-American Networking Group that was started by several of the partners in the

firm. The mission of this network group was to provide social support for current

African-American employees, and to address issues of diversity within the firm, for

example increasing the number of African-American partners within the firm. Another

example of an organizational-based networking group is found at McKinsey & Company,

a strategic consulting form. Within this firm, a networking group was formed among the

female partners to foster support among the limited number of female partners within the

firm. Secondly, the group works to help other women interested in becoming partners

within the firm, meet the goals necessary to make partner.

As mentioned, several organizations have similar organizational-based

networking groups. While, this effort is something that organizations can continue,

organizations should also consider forming organizational-based networking groups

around other themes. Specifically, the results from this research are consistent with

previous findings where network size was found to be important. In several of the current

316

organizational-based network groups, while a need may be met of a specific population

within the firm, organizations should consider creating groups that meet the needs of a

larger set of employees. Thus, perhaps they may develop organizational-based

networking groups and attract all members of the organization that are golf and tennis

fans/players. The point is, organizations should strive to create groups that will attract

large numbers of individuals. In doing so, each employee that willingly joins these

groups has the potential to increase their network size, through meeting a number of

employees, and in increasing their network size they also will have an opportunity to

receive a larger amount of non-redundant information (another benefit of Granovetter’s

weak ties theory). By forming organizational-based networks around themes/interests that

will attract a larger group of employees, would be an example of organizations taking a

relational approach to the development of their employees. That is, “the employer will

provide opportunities and flexibility and resources, particularly people resources, to

enable the employee to develop identity and adaptability and thus be in charge of their

own career” (pp 40) (Hall, 2002).

Finally, individual employees are not the only ones that would benefit from

organizations assisting them in forming networks or social capital. Organizations also

benefit from their employee’s participation in networks. As mentioned previously,

organization benefit from their employee’s participating in network in at least four ways:

(1) social capital developed among an organization’s employees facilitates the flow of

information (by providing an individual with useful information about opportunities and

choices not otherwise available), (2) the social ties one employees develop, may be able

to influence certain agents (e.g. recruiters or supervisors) who play an important role in

317

making critical decisions within organization, (3) an employee’s social ties, that is their

acknowledged relationships with others in an organization, may help highlight an

individual’s social credentials, and (4) social capital should help reinforce an individual’s

identity within an organization (Lin, 2001), thereby reducing the likelihood of turnover.

In addition, in helping employees develop their social capital through

organizational network, organizations are also helping employees manage their own

careers. This notion of organizations helping employees manage their careers was also

discussed in Kossek et al. (1998). In this case, if organizations help employees join and

participate in networks, they can also encourage employees to manage their own careers

by training employees on how to seek job and career-related information, especially for

those individuals that are interested in hiring internally, and individuals that are seeking

to mobilize their careers within their current organizations. Organizations will benefit

from training their employees in how to seek career and job-related information, by

eliminating the costs that are associated with external recruiting. Also, training

employees to seek job and career-related information internally should also reduce the

employee’s likelihood of leaving the organization. This is especially true for individuals

that are highly committed to their jobs, but may be dissatisfied with their current jobs.

While the argument can be made that organizations should become more involved in

training employees to manage their own careers through their participation in networking

groups, implementing these training programs may be difficult. Some of the challenges

that organizations may face include determining if this training should be mandatory or

voluntarily, determining the timing of this training (when in an employee’s career is it

318

appropriate to teach them how to manage their own careers), and determining which

employees might benefit the most form this training (Kossek et al., 1999).

Finally, there is a prevailing notion that suggests if employees receive support

from their employers, then employees will in turn feel obligated to reciprocate, that is

provide help to their organization. In this case, if organizations provide employees with

help in managing their careers and establishing networks they can join within the

organization, the employees are likely to feel motivated to help the organization (Sturges

et al., 2005). That is, career management support and opportunity to join organizational

networks that will benefit the employees, is viewed by the employees as perceived

organizational support. Perceived organizational support has been positively related to

job performance and negatively related to absenteeism and turnover (Sturges et al.,

2005).

Future Research Questions

The key issues related to conducting research on parents, networks and careers

are discussed below. First, it in order to identify some of the key antecedents that may

cause differences in network characteristics across parental status, future research should

draw variables from existing work-family literature. This would begin by examining two

key factors in work-family literature, role theory, and work-family conflict. Role theory

can be used a theoretical framework to describe the difficulty individuals experience in

attempting to balance multiple roles, that is, their work and family role. Drawing from

role theory, future research can also use boundary theory to explain the preferences

individuals have for segmenting their work and family roles. By drawing on these two

theoretical frameworks, many antecedents discussed in the work-family literature could

319

be used to describe factors that may impact network characteristics. Some of those

antecedents discussed in the work-family literature include work-family conflict, role

conflict, life satisfaction, absenteeism, parental demands (e.g. hours spent on housework),

and stress.

In addition, to variables that have been previously been studied in the work-

family literature and/or are directly related to role and boundary theories, a variable that

is often important to understand when investigating working parents is the element of

time. Time is an important variable to understand, as anecdotal evidence suggests that

working parents shared that they have limited time to participate in activities outside of

the immediate job responsibilities. In order to assess how time might impact the

relationship between parental status and network characteristics, one could study how an

individual allocates their time on a daily basis, including any time they allocate to

maintaining their social capital. Alternatively a longitudinal design also allows

researchers to study the impact of time, and very few longitudinal studies have been

conducted within the work-family literature (Macdermid, 2005).

In addition to time, it also appears that individual differences may play a role in

explaining the relationship between parental status and network characteristics.

Specifically, extroversion may play a role in determining differences in network

characteristics, especially network size. Extroverts usually prefer to spend time with other

people and do not enjoy being alone for longs amounts of time (Lee & Tsang, 2001). For

example, Van de Ven et al. (1984) found that individuals that were highly extroverted

tended to maintain a broad and complex network of multiple, concurrent organizations

relationships with ties both internal and external to their organization. As a result,

320

individuals that are generally outgoing and enjoy interacting with people may tend to

have a larger network size, regardless of parental status.

Finally, this study demonstrated there is a negative relationship between network

ties and career indicators, as the percentage of kin ties increase within a network.

However, previous research suggests that this increase of kin ties within the working

parents network, does not remain high. Rather, results from a longitudinal study suggest

that the percentage and contact with kin ties within an adult’s network seems to increase

the most within the first 24 months after the birth of a child, and begins to decline after

24 months (Bost et al., 2002). Thus, similar to most studies on social networks, the

findings in this study suggest research should be conducted using a longitudinal design.

That is, one should examine the how the relationship between network size, network ties,

and network content, varies with the multiple career indicators across time. This would

be helpful in understanding, for example, if the percentage of non-work topics changes,

as the age of the child increases.

Study Limitations

This study has several strengths helping it make a contribution to both the work-

family literature and the literature on careers, using network analysis. First, this study was

able to demonstrate that role segmentation, a variable in the work-family that although

discussed has had little empirical support, was able to predict network content and

network size. This finding is significant, as role segmentation is often discussed in the

work-family literature, but it has not been widely tested. Further, this study makes a

contribution to the literature on careers, as it was able to provide evidence that the

network characteristics ties and content are important to career outcomes. This was an

321

important contribution, as most previous research on career has focus on network size as

one of the key predictors of career-related outcomes. Finally, this study made a

contribution by introducing a new variable, network content, and demonstrated that

topics of conversation are related to career outcomes.

While this study made a contribution to the field, there were several study

limitations that should be noted. First, this data was collected from an organization that

was experiencing a significant amount of layoffs and turnover. Thus, when the

individuals were answered questions related to career satisfaction and career success,

there may have been some response bias introduced into their answers. That is, individual

may have responded positively to the career outcome scales, as they were working in a

very uncertain, unstable environment, and they may have been concerned about the true

nature of the study, that is, they may have been concerned that this survey was being

constructed to learn about their specific career attitudes with that organization.

Secondly, a limitation of this study was related to sample restriction. Individuals

that did not fall into one of the organization’s diversity group were not included in the

study. That is, the organization only agreed to provide the names of individuals whom

both represented some diversity group within the organization for whom an

organizational-based network group had been created. This restriction in sample also

resulted in a small representation of male respondents (only 50 respondents (or 17%)

were men). As a result, this may have made it difficult to assess any of the gender-related

assumptions proposed in the model (see Hypotheses 1a, 1b, and 1c).

322

Evidence of any kind of response bias would have produced limited variability (i.e. a

small standard deviation) in the surveys responses. In closing, a larger sample size may

have also produced larger effects sizes and increased the power of the study. Another

limitation of the study was the used of a cross-sectional design. When a cross-sectional

design is used, no assumptions can be made about causality. Rather only inferences can

be made about the relationships found between the variables included in the study.

In addition, another limitation of this study is related to the generalizability of the

study’s findings. Specifically, this sample was restricted to exempt white collar workers.

It is not certain that these same findings would generalize to other populations. In

addition, this study relied solely on self-reported data. Thus, there may have been a bias

that was introduced when the respondents were completing the survey. Finally, this

sample of respondents was drawn from a single organization, in comparison to the

general working population. Thus, the results of this study may not generalize to all

organizations. However, it should be noted that while this data was collected from a

single organization, the organization does have multiple sites which should lend itself to

some variance in response rate, that is, all data was not collected from employees in a

single location within the organization.

Another study limitation is the use of the ego-network technique. The ego-

network technique is a sampling method generally used when a researcher is

investigating a large or less definable network (e.g. the researcher is interested in

studying the networks of people across several organizations). The ego-network

technique uses a name-generator survey in which the ego identifies a list of alters within

their network and they are able to answer questions related to various characteristics of

323

their alters (frequency of contact, physical proximity of alters, etc). One common

shortcoming in using the ego-network technique is it usually elicits strong (people with

whom the ego is close too) rather than weak ties, that is, the ego usually names

individuals with whom they have more intimate relationships with or people with whom

they have more frequent contact (Lin, 2001). In order to overcome this challenge, that

name generator used in this study was written to say: “Please type BOTH the initials and

first name (e.g. KLS-Kyra) or (e.g. KS-Kyra) of the most important people in your

professional and personal life (up to 20). This includes BOTH people inside and outside

of your organization, family members, friends, neighbors, members of professional

organizations, supervisors, colleagues, and anyone else with whom you discuss important

matters INCLUDING your career plans and various aspects of your professional life”.

That is, the name generator questions was written to help the respondent elicit a broad set

of ties, including both weak (e.g. members of professional organizations, colleagues) and

strong ties (e.g. family members, friends).

It is important to speculate how future studies in this area could be improved.

First, future studies should include the work-family conflict measure, as one of the

antecedents that would moderate the relationship between parental status and network

characteristics. As described previously, it makes sense to include a variable well

established in the work-family literature and also one found to be related to various career

outcomes (e.g. career satisfaction) if the researcher is interested in understanding the

unique challenges working parents face in terms of managing their careers. In addition, a

longitudinal study should be conducted. In doing so, it would allow one to observe the

specific changes individuals face in their network characteristics. The cross-sectional

324

model used in this dissertation does not permit observing changes in network

characteristics pre- and post childbirth. Future studies should increase the sample size in

at least two ways. First, there should be a larger representations of males included in the

sample. Given that there has been a growth in the number of dual career couples, it is

quite possible that working fathers are facing some of the same challenges in maintaining

contact with members within their network, especially since childcare responsibility is

often shared between both parents. The overall study sample size should be increased oby

soliciting the participation of a larger number of people. This could easily be done by

conducting the study across multiple organizations, which would also be helpful in

generalizing the findings from the studies across populations. Lastly, from those

respondents that have parental responsibility, the children’s ages should be collected as

an additional variable. This would be useful for understanding if in fact there is a

difference in the network characteristics of parents across age of the child. That is, further

tests could be conducted to learn if network characteristics vary, and if so how, as the age

of the children change.

325

Works Cited

Agassi, J. (1982). Comparing the Work Attitudes of Men and Women. Lexington Books. Almer, E., Chen, J., & Single, L. (2004). Is it the Kids or the Schedule? The Incremental Effect of Families and Flexible Scheduling on Perceived Career Success. Journal of Business Ethics, 54, 51-66. Archer, T.M. (2005). Web Based Surveys Basics. Paper presented at the meeting of the Mid-Western Educational Research Association Annual Conference, Measurement and Research Method Division, Columbus Ohio. Arthur, M., Khapova, S.,& Wilderom, C. (2005). Career success in a boundaryless career world. Journal of Organizational Behavior, 26,177-202. Arthur, M. & Rousseau, D. (1996). A Career Lesson for the 21st Century. Academy of Management Executives, 10, 28-39. Arthur, M., Claman, P.H.& Defillippi, R.J. (1995). Intelligent enterprise, intelligent Career. Academy of Management Executive, 9 (4), 7-20.

Ashforth, B.E., Kreiner, G.E., Fugate, M. (2000). All in a Day’s Work: Boundaries and Micro Role Transitions. Academy of Management Review, 25, 472-491.

Bailey, S. & Marsden, P. (1999). Interpretation and interview context: examining the General Social Survey name generator using cognitive methods. Social Networks, 21, 287-309. Barham, L., Gottlieb, B. & Kelloway, E. (1997). Variables Affecting Managers’ Willingness to Grant Alternative Work Arrangements. The Journal of Social Psychology, 138, 291-302. Barnett, R & Baruch, G. (1985). Women’s involvement in multiple roles and psychological distress. Journal of Personality and Social Psychology, 49, 135-145. Bearman, P. & Parigi, P. (2004). Cloning Headless Frogs and Other Important Matters: Conversation Topics and Network Structure. Social Forces, 83, 535-557. Becker, G. (1993). Human Capital. Chicago: The University of Chicago Press.

326

Belsky, J. & Rovine, M. (1984). Social-network Contact, Family Support, and the Transition to Parenthood. Journal of Marriage and Family, 46, 455-462. Blair-Loy, M. & Wharton, A. (2002). Employees’ Use of Work-Family Policies and the Workplace Social Context. Social Forces, 80, 813-845. Bost, K., Cox, M. & Payne, C. (2002). Structural and Supportive Changes in Couples’ Family and Friendship Networks Across the Transition to Parenthood. Journal of Marriage and Family, 64, 517-513. Bozionelos, N. (2003). Intra-Organizational Network Resources: Relation to Career Success and Personality. International Journal of Organizational Analysis, 11, 41-66. Brannen, J. (1989). Childbirth and Occupational Mobility: Evidence From a Longitudinal Study. Work, Employment, & Society, 3, 179-201. Brass, D. (1984). Being in the Right Place: A Structural Analysis of Individual Influence in an Organization. Administrative Science Quarterly, 29, 518-539. Brass, D. (1985). Men’s and Women’s Networks: A Study of Interaction Patterns and Influence in an Organization. Academy of Management Journal, 28, 327-343. Brass, D. (1995). A Social Network Perspective on Human Resources Management. In Research in Personnel and Human Resources Management, 13, 39-79. Brass, D., Galaskiewicz, J., Greve, H. & Tsai, W. (2004). Taking Stocks of Networks and Organizations: A Multilevel Perspective. Academy of Management Journal, 47, 795-817. Broadbridge, A. (2004). It’s not what you know, it’s who you know. Journal of Management Development, 23, 551-562. Broadbridge, A. (2003). “Retail management career progression: are the gender differences narrowing?”, paper presented at the Gender, Work and Organization Conference, University of Keele, Keele, 25-27 June. Burke, R. (2000). Effects of Sex, Parental Status, and Spouse Work Involvement in Dual-Career Couples. Psychological Reports, 87, 919-928.

327

Burt, R. (1992). Structural Holes. Cambridge: Harvard University Press. Burt, R. (1997). The Contingency Value of Social Capital. Administrative Science Quarterly, 42, 339-365. Callahan, V. (1993). Subordinate-manager communication in different sex dyads: Consequences for job satisfaction. Journal of Occupational and Organizational Psychology, 66, 13-27.

Campbell, K. (1988). Work and Occupations. Beverly Hills: Sage. Cannings, K. & Montmarquette, C. (1991). Managerial Momentum: A Simultaneous Model of the Career Progress of Male and Female Managers. Industrial and Labor Relations Review, 44, 212-228. Carroll, G. & Teo, A. (1996). On the Social Networks of Managers. Academy of Management Journal, 39, 421-440. Chernesky, R. (2003). Examining the Glass Ceiling: Gender Influences on Promotion Decisions. Administration in Social Work, 27, 13-18. Clark, S.C. (2000). Work/family border theory: A new theory of work/family balance. Human Relations, 53, 747-770.

Cron, E. (2001). Job Satisfaction in Dual-Career Women at Three Family Life Cycle Stages. Journal of Career Development, 28, 17-28. Cross, R., Parker, A.,& Cross, R.(2004). The Hidden Power of Social Networks: Understanding How Work Really Gets Done in Organizations. Boston: Harvard Business School Publishing Corporation. Davis-Netzley, S.A.(1998). Women above the glass ceiling: Perceptions on corporate mobility and strategies for success. Gender and Society, 12, 339-355. Davidson, M.J. and Cooper, C.L. (1992). Shattering the Glass Ceiling: The Women Manager. Paul Chapman, London. Eby, L. T., Casper, W. J., Lockwood, A., Bordeaux, C., & Brinley, A. (2005). A twenty-year retrospective on work and family research in IO/OB: A review of the literature. Journal of Vocational Behavior, 66, 124-197. Eddleston, K., Baldridge, D. & Veiga, J. (2004). Toward modeling the predictors of managerial career success: does gender matter? Journal of Managerial Psychology, 19, 360-385. Edwards, J.R. & Rothbard, N.P. (2000). Mechanisms linking work and family: Clarifying the relationship between work and family constructs. Academy of Management Review, 25, 178-199.

Edwards, J.R. & Rothbard, N.P. (1999). Work and family stress and well-being: An examination of person-environment fit in the work and family domains. Organizational Behavior & Human Decision Processes, 77, 85-129.

328

Elloy, D. & Smith, C. (2003). Patterns of Stress, work-family conflict, role conflict, role ambiguity, and overload among dual-career and single-career couples: an Australian study. Cross Cultural Management, 10, 55-67. Erickson, B. (2001). Good Networks and Good Jobs: The Value of Social Capital to Employers and Employees. in N. Lin, K. Cook, and R.S. Burt (eds) Social Capital: Theory and Research. New York: Aldine de Gruyter. Evans, P., and Bartolome, F. (1984). The Changing Pictures of the Relationship between Career and Family. Journal of Occupational Behavior, 5, 9–21. Fern, E. (1982). The Use of Focus Groups for Idea Generation: The Effects of Group Size, Acquaintanceship, and Moderator on Response Quality and Quality. Journal of Marketing Research, 19, 1-13. Fields, D. (2002). Taking The Measure of Work: A Guide to Validated Scales For Organizational Research and Diagnosis. Thousand Oaks: Sage Publications. Forret, M. & Dougherty, T. (2001). Correlates of Networking Behavior For Managerial and Professional Employees. Group and Organization Management, 26, 283-311. Forret, M. & Dougherty, T. (2004). Networking behaviors and career outcomes: differences for men and women? Journal of Organizational Behavior, 25, 419-437. Frone, M. & Rice, R. (1987). Work-family Conflict: The Effect of Job and Family Involvement. Journal of Occupational Behavior, 8, 45-53. Gallinsky, E., Hughes, D., & Shinn, M. (1986). The corporate work and family life study. New York: Bank Street College of Education. Gattiker, U. & Larwood, L. (1988). Predictors for managers’ career mobility, success and satisfaction. Human relations, 41, 569-591. Golembiewski, R. (1977). Testing some stereotypes about the sexes in organizations: Differential centrality of work? Human Resources Management, 16, 21-24. Goode, W. (1960). A Theory of Role Strain. American Sociological Review, 25, 483-496. Gould, S. & Penley, L. (1984). Career strategies and salary progression: A study of their relationship in a municipal bureaucracy. Organizational Behavior and Human Performance, 34, 244-265. Gould, S., & Werbel, J. D. (1983). Work involvement: a comparison of dual wage earner and single wage earner families. Journal of Applied Psychology, 68, 313-319.

329

Granovetter, M.S. (1973). The strength of weak ties. American Journal of Sociology, 6, 1360-1380. Greenhaus, J. (2003). Career Dynamics. In W.C. Burman, D.R. Ilgen, & R.J.Klimoski (Eds.). Comprehensive Handbook of Psychology. Industrial and Organizational Psychology, (Vol 12, pp 519-540). New York: Wiley. Greenhaus, J.H. & Powell, G.N. (2002). When work and family collide: Deciding between competing role demands. Organizational Behavior and Human Decision Processes, 90, 291-303.

Greenhaus, J., Parasuraman, S. & Collins, K. (2001). Career Involvement and Family Involvement as Moderators of Relationships Between Work-Family Conflict and Withdrawal from a Profession. Journal of Occupational Health Psychology,6, 91-100. Greenhaus, J., Callanan, G., & Godshalk, V. (2000). Career Management. Fort Worth: The Dryden Press- Harcourt College Publishers. Greenhaus, J. H., Parasuraman, S., & Wormley, W. M. (1990). Effects of race on organizational experiences, job performance evaluations, and career outcomes. Academy of Management Journal, 33, 64-86. Greenhaus, J. & Beutell, N. J. (1985). Sources of Conflict Between Work and Family Roles, Academy of Management Review, 10, 76-88. Hall, D.T. (2004). The protean career: A quarter-century journey. Journal of Vocational Bheavior,65, 1-13. Hall, D.T. (2002). Careers In and Out of Organizations. Thousand Oaks: Sage Publications. Hall, D.T. (1990). Careers in Organizations. Goodyear, Santa Monica, CA. Hall, D.T. (1976). Careers In Organizations. Glenview, IL: Scott, Foresman. Hall, D.T. (1996). Final reflections: long live the career! In D.T. Hall & Associates, The Career is Dead: Long Live the Career. (pp337-345). San Francisco: Jossey-Bass. Hall, D.T., Briscoe, J.P. & Kram, K.E. (1997). Identity, values and learning in the protean career. In C.L. Cooper & S.E. Jackson (Eds.), Creating tomorrow’s organizations: A handbook for future research in organizational behavior (pp 321-337). London: John Wiley.

330 Hall, D.T. & Hall, F.S. (1979). The Two Career Couple. Reading, MA: Addison-Wesley.

Hall, D. & Mirvis, P. (1996). The new protean career: Psychological success and the path with a heart. In D.T. Hall & Associates (Eds.), The career is dead-Long live the career: A relational approach to careers (pp 15-45). San Francisco: Jossey-Bass. Hall, D. & Nougaim, K. (1968). An Examination of Maslow’s Need Hierarchy In An Organizational Setting. Organizational Behavior and Human Performance, 3, 12-35. Haller, M., & Rosenmayr, L. (1971). The pluridimensionality of work commitment. Human Relations, 24, 501-518. Harris, H. (2004). Global Careers: Work-Life issues and the adjustment of women international managers. Journal of Management Development, 23, 818-833. Heslin, P. (2005). Conceptualizing and Evaluating Career Success, Journal of Organizational behavior, 26, 113-136. Heslin, P. (2003). Self and Other-Referent Criteria of Career Success. Journal of Career Assessment, 11, 262-286. Hecht, T.D. & Allen, N.J. (2002). Drawing the line: Developing a measure of work/non-work boundary strength. Paper presented at the 17th annual conference of the Society for Industrial and Organizational Psychology, Toronto, Canada (April, 2002). Higgins, M. C. (1998). Early career change: Toward a portfolio theory of relations (Working Paper No. 98-016). Boston: Harvard Business School, Division of Research. Higgins, M. & Thomas, D. (2001). Constellations and careers: toward understanding the effects of multiple developmental relationships. Journal of Organizational Behavior, 22, 223-247. Higgins, M. & Kram, K. (2001). Reconceptualizing Mentoring at Work: A Developmental Network Perspective, Academy of Management Review, 26, 264-298. Hill, E.J., Ferris, M., & Martinson, V. (2003). Does it matter where your work? A comparison of how three work venues (traditional office, virtual office, and home office) influence aspects of work and personal/family life? Journal of Vocational Behavior, 63, 220-241. Hopkins, K. (2002). Interactions of Gender and Race in Workers’ Help Seeking for Personal/Family Problems. Journal of Applied Behavioral Science, 38, 156-176. Hughes, D. & Galinsky, E. (1994). Gender, job, and family conditions, and psychological symptoms. Psychology of Women Quarterly, 18, 251-270.

331

Hughes, E.C. (1937). Institutional Office and the person. American Journal of Sociology, 43, 404-413. Hughes, E.C. (1958). Men and their work. Glencoe: Free Press. Hyden, L.C. & Bolow, P.H. (2003). Who’s talking: drawing conclusions from focus-groups-some methodological considerations. International Journal of Social Research Method, 6, 305-321. Ibarra, H & Smith-Lovin, L. (1997). New Directions in Social Network Research on Gender and Organizational Careers. In C.L. Cooper and S.E. Jackson (Eds), Creating Tomorrow’s Organizations. New York: John Wiley and Sons. Ibarra, H. (1993). Personal Networks of Women and Minorities in Management: A Conceptual Framework. Academy of Management Review, 18, 56-87. James, E. (2000). Race-related Differences in Promotion and Support: Underlying Effects of Human and Social Capital. Org Science, 11, 493-508 James, D. & Drakich, J. (1993). Understanding gender differences in amount of talk: a critical review of research. In Deborah Tannen (ed). Gender and Conversational Interaction. New York and Oxford: Oxford University Press. Kahn, R.L., Wolfe, D.M., Quinn, R.P., Snoek, J.D., & Rosenthal, R.A. (1964). Organizational Stress: Studies in Role Conflict and Ambiguity. Nee York: Wiley. Kanungo, R. (1982). Measurement of Job and Work Involvement. Journal of Applied Psychology, 67, 341-349. Katz, D. & Kahn, R. (1978). The Social Psychology of Organizations. New York: John Wiley & Sons. Kaufman, G. & Uhlenberg, P. (2000). The Influence of Parenthood on the Work Effort of Married Men and Women. Social Forces, 78, 931-949. Kearns, J. & Leonard, K. (2004). Social Networks, Structural Interdependence, and Marital Quality over the Transition to Marriage: A prospective Analysis. Journal of Family Psychology, 18, 383-395. Kidd, J., Jackson, C. & Hirsh, W. (2003). The outcomes of effective career discussions at work. Journal of Vocational Behavior, 62, 119-133. King, Z. (2003) Career Self-Management: Its nature, causes and consequences. Journal of Vocational Behavior,65, 112-133.

332

Kossek, E. E., Roberts, K., Fisher, S., & DeMarr, B. 1998. Career self-management: A quasi-experimental assessment of a training intervention. Personnel Psychology, 51: 935-962. Kram. K. (1988). Mentoring at Work: Developmental Relationships in Organizational Life. Lanham, Maryland: University Press of America. Kram, K. (1996). A Relational Approach to Career Development. in D.T. Hall (ed.) The Career Is Dead: Long Live The Career, (pp. 132-157) Jossey-Bass Publishers. Lambert, S. (1991). The combined effects of job and family characteristics on the job satisfaction, job involvement and intrinsic motivation of men and women workers. Journal of Organizational Behavior, 12, 341-363. Levinson, D. (1978). The season’s of a man’s life. New York: Knopf. Levinson, D. (1986). A conception of adult development, American Psychologist, 41, 3-13. Lewis, S. & Taylor, K. (2000?). Evaluating the Impact of Family-Friendly Employer Practices: A Case Study. In S Lewis and J. Lewis (Eds), The Work-Family Challenge. London: Sage Publications. Lin, N. (2001). Building a Network Theory of Social Capital” in N. Lin, K. Cook, and R.S. Burt (eds) Social Capital: Theory and Research. New York: Aldine de Gruyter. Lin, N. & Dumin, M. (1986). Access to Occupations Through Social Ties. Social Networks,8, 365-385. Linehan, M. (2001). Networking for female managers’ career development. Journal of Management Development, 20, 823-829. Lodahl, T. M., & Kejner, M. (1965). The definition and measurement of job involvement. Journal of Applied Psychology, 49, 24-33. London, M. (1993). Relationships between career motivation, empowerment and support for career development. Journal of Occupational & Organizational Psychology,66, 55-69. Lorence, J. (1987). A Test of “Gender” and “Job” Models of Sex Differences in Job Involvement, Social Forces, 66, 121-142. Lyness, K. & Thompson, D. (2000). Climbing the Corporate Ladder: Do Female and Male Executives Follow the Same Route? Journal of Applied Psychology, 85, 86-101. MacDermid, S. (2005). (Re)Considering Conflict Between Work and Family. In Kossek, E. E. & Lambert, S. (Eds.) Work And Life Integration: Organizational, Cultural and

333

Psychological Perspectives. Mahwah, N.J.:Lawrence ErlbaumAssociates. Marsden, P. (2003). Interviewer effects in measuring network size using a single name generator. Social Networks, 25, 1-16. Marsden, P. (1990). Network Data and Measurement. Annual Review of Sociology, 16, 435-463. Marsden, P. (1987). Core Discussion Networks of Americans. American Sociological Review, 52, 122-31. Martin, N. & Strauss, A. (1956). Patterns of Mobility within Industrial Organizations. Journal of Business, 29, 101-110. McGuire, G. (2000). Gender, Race, Ethnicity, and Networks. Work and Occupations, 27, 501-523. Mehra, A., Kilduff, M., & Brass, D. (1998). At the Margins: A Distinctiveness Approach to the Social Identity and Social Networks of Underrepresented Groups. Academy of Management Journal, 41, 441-452. Melamed, T. (1996). Validation of a Stage Model of Career Success. Applied Psychology, 45, 35-66. Metz, I. & Tharenou, P. (2001). Women’s Career Advancement. The Relative Contribution of Human and Social Capital. Group and Organization Management, 26, 312-342. Miree, C. E. & Frieze, I. H. (1999). Children and Careers: A Longitudinal Study of the Impact of Young Children on Critical Career Outcomes of MBAs. Sex Roles,41, 787-808. Moen, P. & Sweet, S. (2004). From ‘work-family’ to ‘flexible careers’. Community, Work & Family, 7, 209-227.

Race, School Integration, and Friendship Segregation in AmericaMoody, J. (2001). , American Journal of Sociology, 107, 679-716. Moore, G. (1988). Women in elite positions: Insiders or outsiders? Sociological Forum, 3, 566-585. Morgan, D. (1996). Focus Groups. Annual Review of Sociology, 22, 129-152. Munch, A., McPherson, J. & Smith-Lovin, L. (1997). Gender, Children, and Social Contact: The Effects of Childbearing for Men and Women. American Sociological Review, 62, 509-520.

334

http://www.sociology.ohio-state.edu/jwm/ajs_reprint.pdf

Murrell, A., Frieze, I., & Olson, J. (1996). Mobility Strategies and Career Outcomes: A Longitudinal Study of MBAs. Journal of Vocational Behavior, 49, 323-335. Ng, C. & Fosh, P. (2004). The Effect of Career Ambition and Satisfaction on Attitudes Towards Equal Opportunities and Family-Friendly Policies for Women- A case study in Hong Kong. Community, Work, & Family, 7, 43-71. Nippert-Eng, C. (1996). Calendars and keys: The classification of “home” and “work”. Sociological Forum, 11, 563-582. Noe, R. (1996). Is career management related to employee development and performance? Journal of Organizational Behavior, 17, 119-133. Open, C. (1994). The Effects of Organizational and Individual Career Management on Career Success. International Journal of Manpower, 15, 27-37. Palgi, M. & Moore, G. (2004). Social Capital: Mentors and Contacts. Current Sociology, 52, 459-480. Parasuraman, S., Purohit,Y., Godshalk, V.M., & Beutell, N.J. (1996). Work and Family Variables, Entrepreneurial Career Success, and Psychological Well-Being. Journal of Vocational Behavior, 48, 275-300. Parker, P., Arthur, M., & Inkson, K. (2004). Career communities: a preliminary exploration of member-defined career support structures. Journal of Organizational Behavior, 25, 489-514. Plake, B., Murphy-Berman, V., Derscheid, L., Gerber, R., Miller, S., Speth, C., & Tomes, R. (1987). Access decisions by personnel directors: Subtle forms of sex bias in hiring. Psychology of Women Quarterly, 11, 255-264. Podsakoff, P., Mackenzie, S., Lee, J-Y,& Podsakoff, N (2003). Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88, 879-903. Podsakoff, P. & Organ, D. (1986). Self-reports in organizational research: Problems and prospects. Journal of Management, 12(4), 31-544. Polodny, J. & Baron, J. (1997). Resources and Relationships: Social Networks and Mobility in the Workplace. American Sociological Review, 62, 673-693. Portes, A. (1998). Social Capital. Annual Review of Sociology, 24, 1-24. Ragins, B. & Sundstrom, E. (1989). Gender and Power in Organizations: A Longitudinal Perspective. Psychological Bulletin, 105, 51-88.

335

Reeve, C. & Smith, C. (2001). Redefining Lodahl and Kejner’s Job Involvement Scale With a Convergent Evidence Approach: Applying Multiple Methods to Multiple Samples. Organizational Research Methods, 4, 91-111. Rigg, C. & Sparrow, J. (1994). Gender, Diversity, and Working Styles. Women in Management Review, 9, 9-16. Ridgeway, C. & Smith-Lovin, L. (1999). The Gender System and Interaction. Annual Review of Sociology, 25, 191-216. Roth, P. (1994). Missing Data: A Conceptual Review for Applied Psychologists. Personnel Psychology, 47, 537-560. Rothbard, N., Phillips, K. &Dumas, T. (2005). Managing Multiple Roles: Work-Family Policies and Individuals’ Desires for Segmentation. Organization Science, 16, 243-258. Saltzstein, A.L., Ting, Y., & Saltzstein, G.H. (2001). Family-friendly Balance and Job Satisfaction: The Impact of Family-Friendly Policies on Attitudes of Federal Government Employees. Public Administration Review, 4, 452-467.

Saxeninan, A. (1996). Beyond Boundaries: Open labor markets and learning in Silicon Valley. In M.B. Arthur & D.M. Rousseau (Eds). The Boundaryless Career. New York: Oxford University Press. Schmidt, Frank L. & Hunter, John E. (1998). The Validity and Utility of Selection Methods in Personnel Psychology: Practical and Theoretical Implications of 85 Years of Research Findings. Psychological Bulletin, 124, 262-274. Schneer, J. & Reitman, F. (2002). Managerial Life Without a Wife: Family Structure and Managerial Career Success. Journal of Business Ethics, 37, 25-38. Schneer, J. & Reitman, F. (1993).Effects of Alternative Family Structures on Managerial Career Paths. Academy of Management Journal, 33, 391-406. Scott, D.B. (2001). The costs and benefits of women’s family ties in occupational context: women in corporate-government affairs management. Community, Work & Family, 4, 5-27. Seibert, S., Kraimer, M. & Liden, R. (2001). A Social Capital Theory of Career Success. Academy of Management Journal, 44, 219-237. Singh,P., Gupta,U.& Rastogi,M. (1981). Job-involvement in relation to marital status and sex. Psychological Studies, 26, 2, 82-83.

336

Singh, V., Kumra, S. & Vinnicombe, S. (2002). Gender and Impression Management: Playing the Promotion Game. Journal of Business Ethics, 37, 77-89.

Smith-Lovin, L. & McPhearson, M. (1993). You are who you know: A network approach to gender. In P. England (Ed.), Theory on gender/feminism theory. New York: Aldine. Straits, B. (2000). Ego’s important discussants or significant people: an experiment in varying the wording of personal network name generators. Social Networks, 22, 123-140. Stout, S., Slocum,J., & Cron,W.L. (1988). Dynamics of the career plateauing process. Journal of Vocational Behavior, 32, 74-91 Stroh, L., Brett, & Reily, A.(1992). All the Right Stuff: A Comparison of Female and Male Managers’ Career Progression. Journal of Applied Psychology, 77, 251-260. Sturges, J., Guest,D., & Davey, K. (2000). Who’s in charge? Graduates’ attitudes to and experiences of career management and their relationship with organizational commitment. European Journal of Work and Organizational Psychology, 9, 351-370. Sturges, J. & Guest, D. (2004). Working to live or living to work? Work/Life balance in the early career. Human Resources Management Journal, 14, 5-20. Sullivan, S. (1999). The Changing Nature of Careers: A Review and Research Agenda. Journal of Management, 25, 457-484. Sutton, K.L. & Noe, R.A. (2004). Family Friendly Programs and Work-Life Integration: More Myth Than Magic. In Kossek, E. E. & Lambert, S. (Eds.) Work And Life Integration: Organizational, Cultural and Psychological Perspectives. Mahwah, N.J.: Lawrence ErlbaumAssociates. Super, D. (1957). The Psychology of Careers. New York: Harper & Row. Super, D. (1990). A life-span, life-space approach to career development. In D. Brown & L. Brooks (Eds.), Career choice and development (pp 197-261). San Francisco: Jossey- Bass. Swiss, D. (1996). Women breaking through: Overcoming the final 10 obstacles at work. Princeton, NJ: Peterson’s /Pacesetter Books. Tannen, D. (1992). You Just Don’t Understand. Virago: London. Tharenou, P. (2001). Going Up? Do Traits and Informal Social Processes Predict Advancing in Management? Academy of Management Journal, 44, 1005-1017. Tharenou, P. (1999). Is there a link between family structures and women’s and men’s managerial career advancement? Journal of Organizational Behavior, 20, 837-863.

337

Tharenou, P. (1995). Correlates of women's chief executive status: Comparisons with men chief executives and women top managers. Journal of Career Development, 21, 201-212. Tharenou, P., Latimer, S., & Conroy, D. (1994). How do you make it to the top? An examination of influences on women’s and men’s managerial advancement? Academy of Management Journal, 37, 899-931. Turban, D. & Dougherty, T. (1994). The role of protege personality in receipt of mentoring and career success. Academy of Management Journal, 37,688-702. Valcour, P. & Tolbert, P. (2003). Gender, family and career in the era of boundarylessness: determinants and effects of intra- and inter-organizational mobility. International Journal of Human Resource Management, 14, 768-787. Vardi, Y. (1980). Organizational Career Mobility: An Integrative Model. Academy of Management Review, 5, 341-355. Waite, L., Haggstrom, G., & Kanouse, D. (1986). The Effects of Parenthood on the Career Orientation and Job Characteristics of Young Adults. Social Forces, 65, 43-73. Wallace, J.E. (2001). The benefits of mentoring for female lawyers. Journal of Vocational Behavior, 58, 366-391. Wasserman, S. & Faust, K. (1994). Social Network Analysis. Methods and Applications. Cambridge: Cambridge University Press. Wasserman, S. & Galaskiewicz, J. (1994). Advances in social network analysis: Research in the social and behavioral sciences. London: Sage. Wellman, B. (1992). Men in Networks: Private Communities, Domestic Friendships. in P. Nardi (ed) Men’s Friendships. Newbury Park, CA: Sage. Wellman, B. & Frank, K. (2001). Network Capital in a Multilevel World: Getting Support from Personal Communities. in N. Lin, K. Cook, and R.S. Burt (eds) Social Capital: Theory and Research. New York: Aldine de Gruyter. Whitley, B. (2002). Principles of Research in Behavioral Science. Boston: McGraw Hill.

338

Appendix A

Focus Group Interview Survey

339

Code Number____________

1. Please indicate your martial status

_______ Single ________ Married _________Divorced

__________Widowed

2. Please indicate your parental status

______ no children _______ 1 child _______ 2 children______ 3+ children_______

3. If you have at least one child, are these children living in your home? If

not applicable, please skip this question.

Yes___________ No___________

4. Please indicate the number of children in your home that are under the

age of four? If not applicable, please skip this question.

____________ (write exact number)

5. What is your job title?

________________________________________________________

6. Please write down the initials (e.g. KLS) of up to the 10 most important

people in your professional lives (this includes members inside/outside of

your organization (e.g. family, neighbors, members of professional

organizations, supervisors, colleagues).

1. _____________ 2.________________ 3._______________

340

4. _______________ 5. _______________6. _______________

7.________________8._______________ 9.

________________10._____________


7. Please record, in the spaces below, the initials of these individuals.

Beside each set of initials, please indicate the type of relationship (Ex. KLS

– Wife)

Initials Relationship

1. _________________ _______________________

2. _________________ _______________________

3. _________________ _______________________

4. _________________ _______________________

5. _________________ _______________________

6. _________________ _______________________

7. _________________ _______________________

8. _________________ _______________________

9. _________________ _______________________

10. _________________ _______________________

341


8. This question asks you to consider the frequency of interaction with the

individuals named previously (Person 1-10). For each individual, please

indicate how often you interact with that person either in-person or over

the phone. Please circle the number that indicates the number of times you

are in contact with that individual.

Person 1: 1= Daily 2=Weekly 3=Monthly 4=Less than Monthly 5= Never










342


9. This question asks you to consider the topics of conversation that are

discussed among the members of your network. Thinking back to the more

recent discussions you had about an important matter, including your

career plans, would you describe briefly, what was the general topic of

discussion (e.g. community issue, news/economy, kids and education,

politics, life and health, relationships, money & house), ideology/religion,

work, career)”. List up to 5 topics that you have discussed with each of the

members of your network.

Person Topics of Conversation Person 1: Person 2: Person 3: Person 4: Person 5: Person 6: Person 7: Person 8: Person 9:

343

Person 10:

Code Number____________ 10. Please list the name of professional associations/organizations/groups to which you belong or have been active in within the last 3 years. ____________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________

344

APPENDIX B

WAVE 1 E-MAIL: INVITATION TO PARTICIPATE IN THE STUDY

345

Department of Management and Human Resources 700 Fisher Hall

2100 Neil Avenue Columbus, OH 43210-1144

fisher.osu.edu

To: Organization(s) X From: Kyra Sutton, Co-Investigator Re: Organizational Networks My name is Kyra Sutton and I am writing this letter on behalf of myself and my investigator, Ray Noe of the Fisher College of Business at The Ohio State University. The purpose of our project is to learn about the experiences individuals have with their informal/formal networks, and understand if there is any relationship between their networking and career experiences. Specifically, we are interested in comparing the networking and career experiences of working adults with parental responsibility to the experiences of working adults without parental responsibility. We are writing today to ask that you take our survey on a voluntary basis. The survey should be a fun, interesting experience for you, as we will ask you a number of questions related to your informal/formal networks (including the members of your network, the frequency of contact you have with these individuals). The survey will be conducted in two separate waves, in order to minimize participant fatigue. Thus, the first wave of the survey is available at the website provided at the end of this note. The second wave of the survey will be available at the same website approximately 4 weeks after the closing date of the first survey. Of note, the dates of the survey will be provided on the website. In the second wave of the survey, you will be asked question related to your perceptions of you careers. In exchange for your time, the employees that complete both waves of the survey will have a chance to enter a drawing. The winners of the drawing will be randomly selected and they will have a chance to select from one of the following prizes each valued at $100.00. If selected, the choices in prizes include (1) 100 gift certificate to Toys R’Us, (2) 100 gift certificate to Best Buy, (3) IPOD, or (4) an opportunity to donate 100.00 to their favorite charity. The participants will enter the drawing on a voluntary basis and they will not be penalized if they choose not to enter the drawing. Briefly, I will describe the survey, inclduing your time commitment, should you decide to voluntarily participate in this survey. As mentioned, this survey is a web-based survey. The link for this survey is http://fisher.osu.edu/~sutton_162/survey/. At the beginning of the survey you will find a consent form which you will be required to read (and we also suggest you print the consent form) prior to beginning the survey. After reading the consent form, you will begin Wave 1 Survey which consists of five parts. At

346

most, you will commit 30 minutes of your time in completing this survey. We ask that you complete the entire survey and answer the questions on the survey to the best of your ability. After completing the first wave of the survey, you will receive an e-mail notification approximately 3-4 weeks later, stating that the 2nd wave of the survey is now available. The second wave of the survey is shorter than the first wave, and you expect to commit no more than 15 minutes of your time in completing the second survey. We have made this survey available on-line in order to facilitate the ease of completing the survey. You are free to complete this survey at your convenience and please note that your answers will remain confidential. Further, we do not ask you to provide any identifying information on either wave of the survey, and the only individuals will access to the survey are the principle and co-investigator, Ray Noe and Kyra Sutton, respectively. Once again, we appreciate your assistance in this study. Please proceed to the link below to begin the first wave of the survey if you choose to voluntarily participate in our study. If you have any questions, please feel free to contact Kyra Sutton at [email protected] or 614-538-8839 regarding the study. The survey is now available at the following link: Thank you again for your time. Warm regards, Kyra Sutton

347

mailto:[email protected]

APPENDIX C

WAVE 2 E-MAIL: INVITATION TO PARTICIPATE IN THE STUDY

348

Department of Management and Human Resources

700 Fisher Hall 2100 Neil Avenue

Columbus, OH 43210-1144 fisher.osu.edu

Good Afternoon! Thank you very much for time in and interest in participating in the first wave of the Organizational Networks and Careers survey. I am appreciative of your willingness to participate in our study and it is in this vain that I write you today! As mentioned in the instructions of the first wave, we are now ready to launch the second and final wave of the survey. This part of the survey is specifically related to your career experiences and you will find this wave of the survey to be much shorter!! In total you should be able to complete this wave of the survey in no more than 10 minutes. Similar to the first wave, your answers will remain confidential. Specifically, all answers are stored in a main database and no personal identification is stored with the answers. Further, to assist in maintaining your confidentiality, we have provided you with a password that you will have to enter (along with the e-mail address you provided upon completing the first wave of the survey) in order to begin the survey. This unique ID was provided to you after you completed the first wave of the survey. The link for the second and final wave of the survey can be found at: http://fisher.osu.edu/~sutton_162/survey2/ After selecting the link to the survey, you will read the survey instructions and the consent form. Following those first two screens you will be prompted to enter your e-mail address (the one you provided after completing the 1st wave of the survey) and unique ID/password, which is a 6 digit-code (as previously mentioned). You will then begin the survey. Since I am aware that this e-mail may be passed along to other colleagues, I have not included your password/unique ID within the text of this e-mail. However, if you need a copy of your unique ID I am happy to share that with you via e-mail or telephone! As you know, your participation in this survey is meant to:

349

• Assist in a research project conducted by Ohio State’s Fisher College of Business in which we want to learn about the experiences individuals have with their informal/formal networks, and understand if there is any relationship between their networking and career experiences.

As always, should you have any questions please feel free to reach out to Kyra (Sutton) at [email protected] or 313-600-7935. I am pleased to answer any questions that you may have. Thank you in advance for your time and willingness to complete this 2nd and final wave of the survey! Kind regards, Kyra

350

mailto:[email protected]

APPENDIX D

FIELD SURVEY WAVE 1

351

Welcome Page, Consent Form and Wave 1 Survey Questions




We thank you for your interest in our study, Organizational Networks and Careers. What is the purpose of the study?

1. Understand the experiences you've had with participating in informal/formal networks either within your organization or outside of your organization.

2. Understand to what extent you see a relationship between your participation in formal/informal organizational networks and any relationship that may have with your career (e.g. your ability to manage your career, your happiness with your career).

How do I participate in the study?

1. Complete two surveys 2. First survey is attached to this link. 3. Second survey will be available in 4-6 weeks. You’ll receive an e-mail

notification asking you to complete the survey. How long will it take me to complete each survey?

1. The first survey will take no longer than 20-30 minutes. 2. The second survey (available in 4-6 weeks) will take no longer than 10-15

minutes. How do I know when I have completed the survey?

1. You will complete the survey once you arrive at the Thank You page. This page follows the screen where you will be given a Unique ID Number.

2. Remember to close your browser after you reach and read the Thank You Page.

352

Is there any reward for my participation in the study?

1. After completing both surveys, you will be eligible to participate in a drawing (the directions will appear after you complete the second and final survey).

2. There will be five prizes awarded each valued at $100.00. 3. The winners of the drawing will be able to select from one of four prizes

including: (1) An Ipod, (2) A Best Buy Gift Certificate, (3) A Toys “R” Us Gift Certificate, or (4) we will donate $100.00 to your favorite charity on your behalf.

4. Also, in exchange for your participation we will send you a copy of the survey results. If you are interested in receiving a summary copy of the results, please e-mail Kyra Sutton at [email protected]. No individual survey results will be shared

How do I know if I am eligible to participate in the study?

1. 18 years or older 2. Currently employed at least 30 hours/per week. 3. Both working adults with and without children are eligible to participate. 4. However, if you are a working adult with a child, at least one child must be

under the age of 5. 5. Be able to read and understand this web page. 6. Participation in study is completely voluntary.

Before proceeding to the study, please read (and print) the Consent form on the next screen.

353




Consent for Participation in Social and Behavioral Research Protocol title: Organizational Networks and Career Mobility, A Relational View Protocol Number: 2005B0369 Principal Investigator: Raymond A Noe I consent to my participation in research being conducted by Raymond Noe and Kyra Sutton of The Ohio State University. The description provided on the previous screen explained the purpose of the study, the procedures that will be followed, and the amount of time it will take to complete this survey. I understand the possible benefits, if any, of my participation. I know that I can choose not to participate without penalty to me. If I agree to participate, I can skip any question I do not want to answer and I can withdraw from the study at any time, and there will be no penalty. I have had a chance to ask any questions and to obtain answers to my questions. I can contact investigator Kyra Sutton at (614) 538-8839 or [email protected] if I have any further questions about this research. If I have questions about my rights as a research participant, I can call the Office of Responsible Research Practices of The Ohio State University at (614) 688-4792. I have read this form or I have had it read to me. I voluntarily agree to participate in this study. If you disagree with any of the previous statements and do not wish to participate, simply close this browser to end this session. If you agree to all of the above statements, print a copy of this page by selecting the print button on your web browser or by pressing the "Ctrl" key and the "P" key at the same time. Once the page has been printed, click the "submit" button below to continue the survey.

354

Part A A1. What is your gender? Male � Female � A2. What is your martial status? Single � Partnered � Married � Divorced � Widowed � A3. What is your parental status? Zero Children � At least one child � Two children � Three children � If you are not a parent, please proceed to Question A10. If you are parent, please answer questions A4-A9. A4. If you have at least one child, are these children living at home? If not applicable, please skip this question. Yes � No � A5. If you have at least one child, have you had this child (children) within the last five years? Yes � No � A6. If you have at least one child, please indicate the number of children in your home that are UNDER the age of five? If not applicable, please skip this question. 1� 2� 3� 4� 5� 6� 7� 8� 9� A7. If you are a parent, do you share parental responsibility with another adult? Yes � No �

355

A8. If you answered yes to Question A7, please answer the following question. Please indicate the job status of the person with whom you share parental responsibility? Employed (Full-Time; at least 30 hrs/week) � Employed (Part-Time; less than 30 hours/week) � Self-Employed � Unemployed � A10. If you are a parent, please indicate who has primary care giving responsibility for your child/children? If not applicable, please skip this question. Myself � My Spouse/Partner � Both myself and my spouse/partner � Family Care Provider (e.g. mother, father, sibling) � Third Person Care Provider (e.g. Daycare, Nanny, Au Pair) � A11. Do you work at least 30 hours or more per week (on average) Yes � No � A12. Please indicate the number of hours you work per week (on average) ____________ (type/write your best estimate ) A13. Do you work at least 9 months or more out of the year (on average)? Yes � No � A14. Please indicate your employment status. Exempt � Non-exempt �

356

A15. Please indicate the approximate size of the organization in which you are employed. Small (2,500 employees or less) � Midsize (2,501 employees to 10,000 employees) � Large (10,001 employees or more) � A16. Which of the following best describes your Race/Ethnicity? White or Caucasian � African-American or African Descent � Native American � Native Hawaiian � Spanish � Mexican � Puerto Rican � Cuban � Other Hispanic � Chinese � Japanese � Korean � Other Asian � Other race � A17. What is your age? __________________________(years) A18. How many months have you been with your current employer? _________________ (type/write your best estimate )

357

A19. How many months have you been employed in your current job (i.e. company/organization)? _________________ (type/write your best estimate ) A20. How many total years have you worked? _____________________________________(type/write your best estimate ) A21. Please indicate the highest level of education achieved? High school Diploma � Associates Degree � Bachelor Degree � Graduate Degree � A22. Do you have a physical disability? Yes � No �

358

Part B: Relationships Overview/Directions: In this section we ask you to identify individuals that are important to your professional life. After identifying those individuals, we ask you to respond to a series of questions about each of the individuals. B1. Please type BOTH the initials and first name (e.g. KLS- Kyra) of the most important people in your professional and personal life (up to 20). This includes both people inside and outside of your organization, family members, friends, neighbors, members of professional organizations, supervisors, colleagues, and anyone else with whom you discuss important matters including your career plans and various aspects of your professional life. Initials First Name 1. _______ _______________ 2. _______ _______________ 3. _______ _______________ 4. _______ _______________ 5. _______ _______________ 6. _______ _______________ 7. _______ _______________ 8. _______ _______________ 9. _______ _______________ 10. _______ _______________ 11. _______ _______________ 12. _______ _______________ 13. _______ _______________ 14. _______ _______________ 15. _______ _______________ 16. _______ _______________ 17. _______ _______________ 18. _______ _______________ 19. _______ _______________ 20. _______ _______________

359

Part B: Relationships (continued) Overview/Directions B2A – B2F

• The next set of questions will relate to the group of people that you have named (on the previous screen) as important people in your life, including those people with whom you discuss your career plans.

• At the top of each screen you will see the first name and the initials of the most

important people in your professional life. The first name and initials appear to the left of the phrase Network Member.

• On this same screen you will answer a series of questions related to the individual

whose name and initials appear at the top of the screen.

• After you answer the questions related to that specific network member, please press NEXT at the bottom of the page. Each time you proceed to the next screen- you will answer a set of questions about a DIFFERENT person in your network.

• The previously mentioned activity will repeat itself until you arrive at the screen

for question B3.

• Please note, you will not answer these questions for all individuals you have identified as part of your network. Rather you will answer a series of questions of a randomly selected set of important people you identified in question B1.

360

Part B: Relationships (continued) Network Member: KLL (Katie) B2. Please answer the following questions about members you have identified in your network. The categories of questions include the type of relationship, gender, parental status, and the topics of conversation you discuss with these individuals. B2a. Please indicate the type of relationship you have with this person (e.g. KLL – Mentor) as their name appears on the screen. Supervisor/Boss (Former or Current) Colleague/Coworker (Former/Current) Employee/Subordinate (Former or Current) Mentor Work Friend Non-Work Friend Spouse/Partner Sibling/Parent Other Relative (e.g. daughter/son, in-laws) Neighbor Other B2b. Please indicate the gender of this individual. Male � Female � B2bb. Please indicate the approximate age of this individual. ____________ Years (type your best estimate) B2bbb. Please indicate the race of this individual. You are free to select more than one race. White or Caucasian � African-American or African Descent � Native American � Native Hawaiian � Spanish � Mexican � Puerto Rican � Cuban � Other Hispanic � Chinese � Japanese �

361

Korean � B2c. Please indicate the parental status of this individual. Zero children � At least one child � At least two children � At least three children � B2d. If this individual has at least one child, please indicate if their child/children are ________ (select one below) than your children. Younger � About the same age � Older � I don’t have children � B2e. This question asks you to consider the topics of conversation discussed with this individual. Thinking back to the most recent discussions you had about an important matter, including your career plans, please indicate all of the conversation topics you discussed with this individual (you may select more than one topic): Work - General (e.g. work expectations, assignments) � Career/Career Progress (e.g. job-hunting, career planning) � Continuous Education/Training � Work-related Projects (e.g. new projects) � Networking � Children/Family Household � Spouse/Partner � Marriage/Relationship � Health � Other � B2f. Of the conversation topics you selected in question B2e, select the TWO topics that you discuss most frequently when you discussed an IMPORTANT matter, including your career plans with this individual. Work - General (e.g. work expectations, assignments) � Career/Career Progress (e.g. job-hunting, career planning) � Continuous Education/Training � Work-related Projects (e.g. new projects) � Networking � Children/Family Household � Spouse/Partner � Marriage/Relationship � Health � Other �

362

Part B Relationships (continued) Overview/Directions: In this section we ask you to identify individuals that are important to your professional life. After identifying those individuals, we ask you to respond to a series of questions about each of these individuals. B3. Who knows who within your network? Directions (B3). This next set of questions asks you to consider how well the contacts within your network know one another. Type/write the number that best describes how well each pair of individuals in your network know each other. Please type the following number per the following guidelines:

Enter [ 0 ] for people within your network who are total strangers as far as you know.

Enter [ 1 ] for people within your network who are ‘distant’. A distant relationship is best described as one in which two people rarely spend time together.

Enter [ 2 ] for people within your network who are ‘close’. A close relationship is best described as one in which the two people know each other but are not in frequent contact, or their interaction may be specific to a certain setting (i.e. work, professional)

Enter [ 3 ] for people within your network who are ‘especially close’. An especially close relationship is best described as one in which people work very closely together, or have a high level of friendship, and they are in touch with each other on a regular basis.

Complete this table by beginning at the far left corner with Person 1. Indicate his or her relationship with the person in each column (from left to right) in one of four ways: ‘0-> no contact’, ‘1->distant’, ‘2->close’, ‘3->especially close’.

For example, first consider the relationship Person 1 in your network has with person 2. If Person 1 and Person 2 within your network are personal friends, you would identify this relationship as [3] or especially close. In comparison if Person 1 and Person 2 within your network work at different organizations, do not share the same profession, and have only met once (briefly) you would identify this relationship as [1] distant. Next, consider the relationship Person 1 had with Person 3 and the remaining persons in your network. Complete the table for each pair of relationships. For example, Your contacts WGG KLL (0)

363

Part C: Families We want you to share your experiences of your involvement with your family. Indicate the extent of your agreement with each of the following statements: 1 = Strongly Disagree, 2 Moderately Disagree, 3 Slightly Disagree, 4 Slightly Agree, 5 Moderately Agree, 6 = Strongly Agree.

Strongly Disagree

Moderately Disagree

Slightly Disagree

Slightly Agree

Moderately Agree

Strongly Agree

C1. 1=� 2=� 3=� 4=� 5=� 6=� The most important things that happen to me involve my present parental role C2. Most of my interests are centered around my family

1= � 2=� 3=� 4=� 5=� 6=�

C3. Most of my personal life goals are family-oriented

1= � 2=� 3=� 4=� 5=� 6=�

C4. I consider my family to be central to my existence

1= � 2=� 3=� 4=� 5=� 6=�

364

Part D: Work and Home Preferences

The following questions ask about your experiences with managing your workload in the office and at home. Indicate the extent of your agreement with each of the following statements: 1 = Strongly Disagree, 2 Moderately Disagree, 3 Slightly Disagree, 4 Slightly Agree, 5 Moderately Agree, 6 = Strongly Agree. Strongly

Disagree Moderately Disagree

Slightly Disagree

Slightly Agree

Moderately Agree

Strongly Agree

D1. I often do work at home.

1= � 2=� 3=� 4=� 5=� 6=�

D2. I work after hours.

1= � 2=� 3=� 4=� 5=� 6=�

D3. I schedule personal activities during business hours.

1= � 2=� 3=� 4=� 5=� 6=�

D4. I communicate with family and friends during business hours.

1= � 2=� 3=� 4=� 5=� 6=�

D5. I think of personal or family-related issues while I am working.

1= � 2=� 3=� 4=� 5=� 6=�

D6. I do not work on personal time.

1= � 2=� 3=� 4=� 5=� 6=�

D7. I take work out of the office.

1= � 2=� 3=� 4=� 5=� 6=�

365

D8. My personal time is my own.

1= � 2=� 3=� 4=� 5=� 6=�

D9. When working, I am completely focused on my work.

1= � 2=� 3=� 4=� 5=� 6=�

D10. I leave my personal life outside of the workplace.

1= � 2=� 3=� 4=� 5=� 6=�

D11. I rarely deal with personal matters when working.

1= � 2=� 3=� 4=� 5=� 6=�

D12. The office is reserved for doing work.

1= � 2=� 3=� 4=� 5=� 6=�

366

Part D: Work and Home Preferences (continued)

Below we ask you to consider additional questions related to managing your workload between the office and home. When responding to these questions consider how much of that characteristic you personally feel is acceptable. Some people prefer more or less of some job characteristics than others.

Indicate the extent to which you desire each of the following statements:

1= Strongly Undesirable, 2=Moderately Undesirable, 3=Slightly Undesirable, 4=Slightly Desirable, 5=Moderately Desirable, 6= Strongly Desirable.

Strongly Undesirable

Moderately Undesirable

Slightly Undesirable

Slightly Desirable

Moderately Desirable

Strongly Desirable

D13. I do not desire to be required to work while at home.

1=� 2=� 3=� 4=� 5=� 6=�

D14. I desire to be able to forget work while I am at home.

1=� 2=� 3=� 4=� 5=� 6=�

D15. I do not desire to have to think about work once I leave the workplace.

1=� 2=� 3=� 4=� 5=� 6=�

D16. I do not desire to be expected to take work home.

1=� 2=� 3=� 4=� 5=� 6=�

367

Part E Job Importance

The last set of questions asks you to consider how important your job is to your life.

Indicate the extent of your agreement with each of the following statements: 1 = Strongly Disagree, 2 Moderately Disagree, 3 Slightly Disagree, 4 Slightly Agree, 5 Moderately Agree, 6 = Strongly Agree.

Strongly Disagree

Moderately Disagree

Slightly Disagree

Slightly Agree

Moderately Agree

Strongly Agree

E1. The most important things that happen to me involve my present job.

1=� 2=� 3=� 4=� 5=� 6=�

E2. To me, my job is only a small part of who I am.

1=� 2=� 3=� 4=� 5=� 6=�

E3. I am very much involved personally with my job.

1=� 2=� 3=� 4=� 5=� 6=�

E4. I live, eat, and breathe my job.

1=� 2=� 3=� 4=� 5=� 6=�

368

E5. Most of my interests are centered on my job.

1=� 2=� 3=� 4=� 5=� 6=�

E6. I have very strong ties with my present job which would be difficult for me to break.

1=� 2=� 3=� 4=� 5=� 6=�

E7. Usually I feel detached from my job.

1=� 2=� 3=� 4=� 5=� 6=�

E8. Most of my personal life goals are job-oriented.

1=� 2=� 3=� 4=� 5=� 6=�

E9. I consider my job to be very central to my existence.

1=� 2=� 3=� 4=� 5=� 6=�

E10. I like to be absorbed in my job most of the time.

1=� 2=� 3=� 4=� 5=� 6=�

369

APPENDIX E

ORGANIZATIONAL NETWORKS AND CAREERS SURVEY- WAVE 2

370

Department of Management and Human Resources 700 Fisher Hall

2100 Neil Avenue Columbus, OH 43210-1144

fisher.osu.edu Welcome! We thank you for your involvement with the first wave of our survey and we thank your for returning to complete the second wave of our study, Organizational Networks and Careers. What is the purpose of the second wave of the survey? In the second wave of the survey we ask you to complete a series of questions related to your careers (e.g. career satisfaction, career management) How do I know if I can participate in the second wave of the survey? After completing the first wave of the survey you were given a unique password. This password was sent to you via e-mail immediately after you completed the first wave of the survey. You must enter your unique password in order to complete the second wave of the survey. If I lost my Unique ID Number, what should I do?

1. If you have misplaced your Unique ID Number, please e-mail Kyra Sutton at [email protected].

2. If you need to e-mail Kyra, please include the e-mail address (only) that you provided when your unique password was provided.

3. Of note, your survey responses were not stored in the same file as your password. Therefore, your answers from both the first and second survey remain confidential.

How long will it take me to complete the second wave of the survey? This wave of the survey will take no longer than 10-15 minutes. Is there any reward for my participation in the study?

1. If you complete both waves of the survey, you will be eligible to participate in a drawing.

2. There will be several prizes awarded. 3. The winners of the drawing will be able to select from one of four

prizes including: (1) An Ipod, (2) A Best Buy Gift Certificate, (3) A Toys “R” Us Gift Certificate, or (4) we will donate $100.00 to your favorite charity on your behalf.

4. Also, in exchange for your participation we will send you a copy of the survey results. If you are interested in receiving a summary copy of the results, please e-mail Kyra Sutton at [email protected]. No individual survey results will be shared.

371

How do I know if I am eligible to participate in the study? 1. Completed the first wave of the survey. 2. 18 years or older 3. Currently employed at least 30 hours/per week. 4. Both working adults with and without children are eligible to

participate. 5. However, if you are a working adult with a child, at least one child must

be under the age of 5. 6. Be able to read and understand this web page. 7. Participation in study is completely voluntary. How do I know when I have completed the survey?

1. You will complete the survey once you arrive at the Thank You page. 2. If you are interested in entering the drawing, you will be prompted to

provide your name and e-mail address. 3. Remember to close your browser after you reach and read the Thank You

Page! Before proceeding to the study, please read (and print) the Consent form on the next screen.

372




Consent for Participation in Social and Behavioral Research Protocol title: Organizational Networks and Career Mobility, A Relational View

Protocol Number: 2005B0369 Principal Investigator: Raymond A Noe I consent to my participation in research being conducted by Raymond Noe and Kyra Sutton of The Ohio State University. The description provided on the previous screen explained the purpose of the study, the procedures that will be followed, and the amount of time it will take to complete this survey. I understand the possible benefits, if any, of my participation. I know that I can choose not to participate without penalty to me. If I agree to participate, I can skip any question I do not want to answer and I can withdraw from the study at any time, and there will be no penalty. I have had a chance to ask any questions and to obtain answers to my questions. I can contact investigator Kyra Sutton at (614) 538-8839 or [email protected] if I have any further questions about this research. If I have questions about my rights as a research participant, I can call the Office of Responsible Research Practices of The Ohio State University at (614) 688-4792. I have read this form or I have had it read to me. I voluntarily agree to participate in this study. If you disagree with any of the previous statements and do not wish to participate, simply close this browser to end this session. If you agree to all of the above statements, print a copy of this page by selecting the print button on your web browser or by pressing the "Ctrl" key and the "P" key at the same time. Once the page has been printed, click the "submit" button below to continue the survey.

373

Part A: Work Interruptions Overview: The following asks you to share your work history. A1.

How many work interruptions have you had over your career? A work interruption includes taking a leave from work for at least 30 days. This excludes vacation time or time allotted for training and development. (Next Add a Space between F1 description and example). Example: For example, if you have two work interruptions (e.g. maternity leave, personal leave) of at least 30 days each, you would select number 3. � I have had no work interruptions of 30 days � I have had one work interruption for at least 30 days � I have had two work interruptions for at least 30 days each � I have had three work interruptions for at least 30 days each � I have had four work interruptions for at least 30 days each � I have had five work interruptions for at least 30 days each � I have had six work interruptions for at least 30 days each

374

Part B: Career Management Overview: The following questions ask you about how you manage your career. Indicate the extent of your agreement with each of the following statements: 1 = Strongly Disagree, 2 Moderately Disagree, 3 Slightly Disagree, 4 Slightly Agree, 5 Moderately Agree, 6 = Strongly Agree.

Strongly Disagree

Moderately Disagree

Slightly Disagree

Slightly Agree

Moderately Agree

Strongly Agree

B1. 1=� 2=� 3=� 4=� 5=� 6=� I have gotten myself introduced to people who can influence my career. B2. 1=� 2=� 3=� 4=� 5=� 6=� I have talked to senior management at the company's social gatherings. B3. 1=� 2=� 3=� 4=� 5=� 6=� I have built contacts with people in areas where I would like to work. B4. 1=� 2=� 3=� 4=� 5=� 6=� I have definite goals for my career over my lifetime.

375

1=� 2=� 3=� 4=� 5=� 6=� B5.

When I think of changing my job, I always consider whether the new job leads to another one I want. B6. 1=� 2=� 3=� 4=� 5=� 6=� I give a lot of thought to how the specific plans I make for my career, are going to be useful in achieving my career goals. B7. 1=� 2=� 3=� 4=� 5=� 6=� I know what my strengths and weaknesses are in relation to my career. B8. 1=� 2=� 3=� 4=� 5=� 6=� Achieving my career goals is very important to me. B9. 1=� 2=� 3=� 4=� 5=� 6=� I am always very careful to avoid dead-end career paths.

376

B10. 1=� 2=� 3=� 4=� 5=� 6=� I try to have as much visibility and exposure to my bosses as I can. B11. 1=� 2=� 3=� 4=� 5=� 6=� I go out of my way to find a mentor or sponsor to help in my career in the firm. B12. 1=� 2=� 3=� 4=� 5=� 6=� I cultivate friendships with influential people for my career outside of work.

377

B13. 1=� 2=� 3=� 4=� 5=� 6=� I actively seek opportunities, rather than wait to be chosen. B14. 1=� 2=� 3=� 4=� 5=� 6=� I try to help my superiors achieve things that are important to them, even if it is not what I want. B15. 1=� 2=� 3=� 4=� 5=� 6=� I have taken the initiative to be involved in high profile projects. B16. 1=� 2=� 3=� 4=� 5=� 6=� I have asked for career advice from people even when it has not been offered. B17. 1=� 2=� 3=� 4=� 5=� 6=� I have asked for feedback on my performance even when it was not given. 1=� 2=� 3=� 4=� 5=� 6=� B18. I have refused to accept a new role because it would not help me develop new skills. B19. 1=� 2=� 3=� 4=� 5=� 6=� I have monitored job advertisements to see what is available outside the organization.

378

Part C: Career Management- Part Two Overview: In this section, the following questions ask about how you feel about your career.

In this first section, we ask you to consider your resume.

Not at All

Current Very

Current 1* 2* 3* 4* 5* C1 How

current is your resume?

� � � � �

In this section reflect on whether you have done the following activities over the past 6 months?

Not at All Somewhat A Great Deal 1 2 3 4 5 C2 Over the past 6

months to what extent have you reviewed internal postings?

1=� 2=� 3=� 4=� 5=�

1=� 2=� 3=� 4=� 5=� C3 Over the past 6 months to what extent have you discussed future job openings within your INTERNAL network (where internal network members include people working at your current organization including co-workers, supervisors, etc.)?

379

1=� 2=� 3=� 4=� 5=� C4 Over the past 6

months to what extent have you discussed future job openings within your EXTERNAL network (where external includes members in your network outside of your current organization)

C5 Over the past 6 months to what extent have you thought about what position you would like to have next?

1=� 2=� 3=� 4=� 5=�

C6 To what extent do you actively seek out information about job opportunities outside your current organization?

1=� 2=� 3=� 4=� 5=�

C7 To what extent have you sought out any new personal connections AT WORK in the past 6 months for the purpose of furthering your career?

1=� 2=� 3=� 4=� 5=�

C8 1=� 2=� 3=� 4=� 5=� To what extent have you sought out any new personal connections outside of work for the purpose of furthering your career?

380

Part D: Career Satisfaction Overview: The following questions ask about YOUR INDIVIDUAL attitudes toward your career.


Strongly


Slightly Disagree

Slightly Agree

Moderately Agree

Strongly Agree

D1 Relative to my career aspirations, I am satisfied with the progress I have made towards meeting my goals for advancement.

1=� 2=� 3=� 4=� 5=� 6=�

D2 Relative to

my career aspirations, I am satisfied with the overall success I have achieved in my career.

1=� 2=� 3=� 4=� 5=� 6=�

1=� 2=� 3=� 4=� 5=� 6=� D3 Relative to

my career aspirations, I am satisfied with the progress I have made toward meeting my goals for income

381

D4 Relative to

my career aspirations, I am satisfied with the skill

1=� 2=� 3=� 4=� 5=� 6=�

development I have attained.

D5 Relative to my career aspirations, I am satisfied with the autonomy

1=� 2=� 3=� 4=� 5=� 6=�

I have attained.

D6 Relative to my career aspirations, I am satisfied with the intellectual

1=� 2=� 3=� 4=� 5=� 6=�

stimulation I have attained.

382

Part E : Career Satisfaction And My Peer Group Overview: The following questions ask how you feel about your career compared to your peers.


Strongly


Slightly Disagree

Slightly Agree

Moderately Agree

Strongly Agree

1=� 2=� 3=� 4=� 5=� 6=� E1 Relative to people who I perceive as peers in my career/profession, I am satisfied with the progress I have made towards meeting my goals for advancement.

E2 Relative to people who I perceive as peers in my career/profession, I am satisfied with the overall success I have achieved in my career.

1=� 2=� 3=� 4=� 5=� 6=�

1=� 2=� 3=� 4=� 5=� 6=� E3 Relative to

people who I perceive as peers in my career/profession, I am satisfied with the progress I have made toward meeting my goals for income.

383

E4 Relative to

people who I perceive as peers in my career/profession, I am satisfied with the skill

1=� 2=� 3=� 4=� 5=� 6=�

development I have attained.

E5 Relative to people who I perceive as peers in my career/profession, I am satisfied with the autonomy

1=� 2=� 3=� 4=� 5=� 6=�

I have attained.

E6 Relative to people who I perceive as peers in my career/profession, I am satisfied with intellectual

1=� 2=� 3=� 4=� 5=� 6=�

stimulation I have attained.

384

F1. Please indicate your total INDIVIDUAL pretax income (e.g. salary, bonus, stock, profit sharing) in 2005 and in US Dollars. Less than 30,000 30,000- 45,000 45,000-60,000 60,000- 75,000 75,000 - 90,000 90,000-105,000 105,000- 120,000 120,000-135,000 Greater than 135,000 F2. Throughout your career, please indicate how many salary increases you have received. Note: Salary increases include (a) changes in annual salary; and/or (b) qualifying for a performance-based company bonus, incentive or stock plan.

0 1 2 3 4 5 Greater than 5 F3. Since you've joined your current organization/company, please indicate how many promotions you have received. Note: Promotions includes (a) significant changes in salary; (b) lateral or horizontal promotions; (c) changes in offices and/or type of furniture/décor in office; (d) significant changes in job scope or responsibilities; and (e) changes in company level.

0 1 2 3 4 5 Greater than 5

385

F4. Please indicate how many promotions you have received in your entire career. Note: Promotions includes (a) significant changes in salary; (b) lateral or horizontal promotions; (c) changes in offices and/or type of furniture/décor in office; (d) significant changes in job scope or responsibilities; and (e) changes in company level. 0 1 2 3 4 5 Greater than 5 F5. How likely is that you will receive a promotion within the next five years? 0 - No Chance 1 2 3 4 5 - Very Good Chance

386




Thank you for your participation in our study, Parenthood and Organizational Networks: A Relational View of the Career Mobility of Working Parents- Part 2. In appreciation of completing both waves of the survey, we would like to give you the opportunity to enter a drawing. Five individuals will be randomly selected to receive a prize each worth 100 US Dollars. Please read the disclaimer below and enter this drawing should you be interested in your name being included in the drawing. Of note, all winners will be notified directly over e-mail by the Co-Investigator, Kyra Sutton. Disclaimer: I understand that I am entering this drawing on a voluntary basis. I have completed Wave 1 AND Wave 2 of the surveys from the Parenthood and Organizational Networks: A Relational View of the Career Mobility of Working Parents- Part 2 study. I understand that 5 winners will be randomly selected, and that each winner will be awarded a prize worth 100 US dollars. I understand there will be a choice in the prizes awarded. Specifically, the winners will have a choice of (1) an Ipod, (2) a Toys R’ Us Gift certificate, (3) a Best Buy Gift Certificate, or (4) an opportunity to donate 100.00 US Dollars to their favorite charity. I am aware that I am entering this drawing on a voluntary basis, and I do not have to enter this drawing if I do not desire to be considered for the drawing. I have given my name and e-mail addresses in the boxes below, and I am aware that all winners will be contacted directly by e-mail. Name (First, Last): E-mail Address (Preferred):

387

Sutton Kyra Leigh

Documents