Top Banner
http://jom.sagepub.com Journal of Management DOI: 10.1177/014920630102700502 2001; 27; 515 Journal of Management Ellen M. Whitener A cross-level analysis using hierarchical linear modeling Do "high commitment" human resource practices affect employee commitment?: http://jom.sagepub.com/cgi/content/abstract/27/5/515 The online version of this article can be found at: Published by: http://www.sagepublications.com On behalf of: Southern Management Association can be found at: Journal of Management Additional services and information for http://jom.sagepub.com/cgi/alerts Email Alerts: http://jom.sagepub.com/subscriptions Subscriptions: http://www.sagepub.com/journalsReprints.nav Reprints: http://www.sagepub.com/journalsPermissions.nav Permissions: http://jom.sagepub.com/cgi/content/refs/27/5/515 SAGE Journals Online and HighWire Press platforms): (this article cites 38 articles hosted on the Citations © 2001 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution. at Middlesex University on December 7, 2007 http://jom.sagepub.com Downloaded from
22
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Organizational Behavior

http://jom.sagepub.com

Journal of Management

DOI: 10.1177/014920630102700502 2001; 27; 515 Journal of Management

Ellen M. Whitener A cross-level analysis using hierarchical linear modeling

Do "high commitment" human resource practices affect employee commitment?:

http://jom.sagepub.com/cgi/content/abstract/27/5/515 The online version of this article can be found at:

Published by:

http://www.sagepublications.com

On behalf of:

Southern Management Association

can be found at:Journal of Management Additional services and information for

http://jom.sagepub.com/cgi/alerts Email Alerts:

http://jom.sagepub.com/subscriptions Subscriptions:

http://www.sagepub.com/journalsReprints.navReprints:

http://www.sagepub.com/journalsPermissions.navPermissions:

http://jom.sagepub.com/cgi/content/refs/27/5/515SAGE Journals Online and HighWire Press platforms):

(this article cites 38 articles hosted on the Citations

© 2001 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution. at Middlesex University on December 7, 2007 http://jom.sagepub.comDownloaded from

Page 2: Organizational Behavior

Do “high commitment” human resource practices affectemployee commitment?

A cross-level analysis using hierarchical linear modeling

Ellen M. Whitener*

Department of Management, McIntire School of Commerce, University of Virginia, Charlottesville,VA 22903, USA

Received 20 September 1999; received in revised form 7 September 2000; accepted 27 November 2000

Abstract

Relying on a cross-level paradigm and on social exchange theory (i.e., perceived organizationalsupport) I explore the relationships among human resource practices, trust-in-management, andorganizational commitment. Individual-level analyses from a sample of 1689 employees from 180credit unions indicate that trust-in-management partially mediates the relationship between perceivedorganizational support and organizational commitment. Cross-level analyses using hierarchical linearmodeling indicate that human resource practices affect the relationship between perceived organiza-tional support and organizational commitment or trust-in-management. © 2001 Elsevier Science Inc.All rights reserved.

1. Introduction

In a survey of over 7500 U.S. workers, Watson Wyatt International found that companieswith highly committed employees experienced greater 3-year total returns to shareholders(112%) than companies with low employee commitment (76%; Watson Wyatt, 1999). Theyalso found that human resource practices and trust in management had the strongest impacton building commitment.

As Watson Wyatt’s survey implies, human resource practices and trust may provide twoavenues that corporate executives can use to increase the commitment of their workforce.

* Tel.: �1-434-924-7091; fax: �1-434-924-7074.E-mail address: [email protected] (E.M. Whitener).

Pergamon

Journal of Management 27 (2001) 515–535

0149-2063/01/$ – see front matter © 2001 Elsevier Science Inc. All rights reserved.PII: S0149-2063(01)00106-4

© 2001 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution. at Middlesex University on December 7, 2007 http://jom.sagepub.comDownloaded from

Page 3: Organizational Behavior

Academic research conducted at the organizational level suggests that human resourcepractices affect organizational outcomes by shaping employee behaviors and attitudes(Arthur, 1994; Huselid, 1995; Wood & de Menezes, 1998). More specifically, systemsof “high commitment” human resource practices increase organizational effectiveness bycreating conditions where employees become highly involved in the organization andwork hard to accomplish the organization’s goals (Arthur, 1994; Wood & de Menezes,1998)—in other words, by increasing their employees’ commitment to the organization.However, ironically, these studies have provided estimates of the strength of therelationship between high commitment human resource practices and measures oforganizational effectiveness without investigating the relationship between human re-source practices and employee commitment.

Similarly, researchers have not systematically developed or tested theories linking trustand commitment; however, incidental to tests of other variables or theories, they have foundsignificant correlations between commitment and employees’ trust in a specific individual(e.g., their supervisor or leader) (Folger & Konovsky, 1989; Pillai, Schriesheim & Williams,1999; Podsakoff, MacKenzie & Bommer, 1996) or generalized trust-in-management (e.g.,Gopinath & Becker, 2000; Kim & Mauborgne, 1993; Pearce, 1993). These estimates suggestthat further study of the nature of the relationship between trust and commitment would befruitful, especially if theory-based.

The motivational processes of social exchange theory and the norm of reciprocity (e.g.,Blau, 1964; Homans, 1961) may explain the relationships among human resource practices,trust-in-management and employee commitment (Eisenberger, Fasolo & Davis–LaMastro,1990; Settoon, Bennett & Liden, 1996; Wayne, Shore & Liden, 1997). A well-establishedstream of research rooted in social exchange theory has shown that employees’ commitmentto the organization derives from their perceptions of the employers’ commitment to andsupport of them (Eisenberger et al., 1990; Hutchison & Garstka, 1996; Settoon et al., 1996,Shore & Tetrick, 1991; Shore & Wayne, 1993; Wayne et al., 1997). The research suggeststhat employees interpret organizational actions such as human resource practices (Settoon etal., 1996; Wayne et al., 1997) and the trustworthiness of management (Eisenberger et al.,1990; Settoon et al., 1996) as indicative of the personified organization’s commitment tothem. They reciprocate their perceptions accordingly in their own commitment to theorganization. Only a few studies have explored the role of human resource practices in thismodel (e.g., Allen, 1992; Guzzo, Noonan, & Elron, 1994; Miceli & Mulvey, 2000; Wayneet al., 1997) and none has explored the role of trust.

The purpose of this study is to investigate the relationships among human resourcepractices, trust-in-management, perceptions of organizational support, and organizationalcommitment. I rely on a “meso” paradigm recognizing that these variables exist at differentlevels of analysis (organizational and individual) but nonetheless interact and affect eachother (House, Rousseau & Thomas–Hunt, 1995; Rousseau, 1985). I use a social exchangelens relying specifically on the role of perceptions of organizational support to explore theprocesses that link these factors. I test the relationships with data from a sample of over 1600employees from 180 credit unions. I explore individual-level hypotheses using hierarchicallinear regression and cross-level hypotheses using hierarchical linear modeling.

516 E.M. Whitener / Journal of Management 27 (2001) 515–535

© 2001 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution. at Middlesex University on December 7, 2007 http://jom.sagepub.comDownloaded from

Page 4: Organizational Behavior

2. High commitment human resource practices

Human resource practices can be classified as “control” or “commitment” practices(Arthur, 1994; Walton, 1985; Wood & de Menezes, 1998). Control approaches aim toincrease efficiency and reduce direct labor costs and rely on strict work rules and proceduresand base rewards on outputs (Arthur, 1994). Rules, sanctions, rewards, and monitoringregulate employee behavior (Wood & de Menezes, 1998). In contrast, commitment ap-proaches aim to increase effectiveness and productivity and rely on conditions that encourageemployees to identify with the goals of the organization and work hard to accomplish thosegoals (Arthur, 1994; Wood & de Menezes, 1998).

The practices that represent a high commitment strategy include sets of organization-widehuman resource policies and procedures that affect employee commitment and motivation.They include selective staffing, developmental appraisal, competitive and equitable com-pensation, and comprehensive training and development activities (Ichniowski, Shaw &Prennushi, 1997; MacDuffie, 1995; Snell & Dean, 1992; Youndt, Snell, Dean & Lepak,1996).

A review by Delery (1998) shows that early studies of human resource practices attemptedto find the universally best conduct of each independent practice. However, recently resultshave also shown that high commitment practices can work well synergistically, reflective ofa general commitment strategy. Across a variety of industries (e.g., automotive assemblyplants, steel companies and minimills, not-for profit organizations), organizations with highcommitment systems experience greater productivity, financial performance, and effective-ness than organizations with low commitment or control systems (e.g., Arthur, 1994;Delaney & Huselid, 1996; Huselid, 1995; Ichniowski et al. 1997; MacDuffie, 1995; Wood& de Menezes, 1998; Youndt et al., 1996). The organizational context (e.g., fit) and goals(e.g., outcomes) may impact whether particular human resource practices have synergistic orindependent effects on firm outcomes (Delery, 1998).

3. Social Exchange Theory

Social exchange theory (Blau, 1964; Homans, 1961) originally explained the motivationbehind the attitudes and behaviors exchanged between individuals. Eisenberger, Huntington,Hutchison, and Sowa (1986) expanded this work by proposing and establishing that thetheory of social exchange and the norm of reciprocity also explain aspects of the relationshipbetween the organization and its employees. They noted that employees form generalperceptions about the intentions and attitudes of the organization toward them from thepolicies and procedures enacted by individuals and agents of the organization, attributinghuman-like attributes to their employer on the basis of the treatment they receive (Levinson,1965). In this way, employees see themselves as having a relationship with their employerthat is parallel to the relationships individuals build with each other.

Recognizing this tendency to personify the organization, they applied social exchangetheory to the relationship between the personified organization and its employees. Inparticular, they predicted that “. . . positive, beneficial actions directed at employees by the

517E.M. Whitener / Journal of Management 27 (2001) 515–535

© 2001 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution. at Middlesex University on December 7, 2007 http://jom.sagepub.comDownloaded from

Page 5: Organizational Behavior

organization and/or its representatives contribute to the establishment of high-quality ex-change relationships. . . that create obligations for employees to reciprocate in positive,beneficial ways” (Settoon et al., 1996, p. 219).

3.1. Perceived organizational support, trust, and commitment

Eisenberger et al. (1986) developed the construct, perceived organizational support, toreflect employees’ beliefs about the organization’s support, commitment, and care for them.They proposed that perceived organizational support would be significantly related to avariety of employee attitudes and behaviors including organizational commitment and trust(Eisenberger et al., 1990).

Organizational commitment refers to identification with organizational goals, willingnessto exert effort on behalf of the organization, and interest in remaining with the organization(Mowday, Steers & Porter, 1979). Employees’ commitment to the organization would besignificantly related to their perceptions of the employer’s commitment to them (perceivedorganizational support) as they reciprocate their perceptions of the organization’s actions intheir own attitudes and behavior (Shore & Tetrick, 1991). Perceived organizational supporthas a high and significant correlation (0.38–0.71) with organizational commitment; yet,construct validity studies have verified that they are distinct though closely linked variables(Shore & Tetrick, 1991). In this study, I expect to replicate these results:

Hypothesis 1: Employees’ perceptions of organizational support will be positively andsignificantly correlated with their commitment to the organization.

Trust is a complex phenomenon that has long eluded precise definition because itencompasses many facets and levels that need to be carefully specified (Rousseau, Sitkin,Burt & Camerer, 1998; Whitener, 1997). Rousseau et al. (1998) recognized these difficultiesnoting that trust can be different depending on the focal object and level (such as interor-ganizational vs. interpersonal trust). However, they asserted that the primary components oftrust—risk, uncertainty, and interdependence—remain constant across contexts. Relying onthese elements, Rousseau et al. defined trust as a “psychological state comprising theintention to accept vulnerability based upon positive expectations of the intentions orbehavior of another” (Rousseau et al., 1998, p. 395). This definition suggests that employees’trust in management reflects employee faith in corporate goal attainment and organizationalleaders, and to the belief that ultimately, organizational action will prove beneficial foremployees (Kim & Mauborgne, 1993).

Like perceived organizational support, trust develops through a social exchange processin which employees interpret the actions of management and reciprocate in kind. “. . . Thegradual expansion of the exchange permits the partners to prove their trustworthiness to eachother. Processes of social exchange, consequently, generate trust” (Blau, 1964, p. 315).According to Eisenberger et al. (1986), perceived organizational support embodies the socialexchange process, reflecting employees’ interpretations and perceptions of the organization’sactions. Trust in management, who tends to be the primary purveyor of the personifiedorganization’s actions, is a likely response to their perceptions. On this basis, I predict asignificant relationship between perceived organizational support and trust such that:

518 E.M. Whitener / Journal of Management 27 (2001) 515–535

© 2001 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution. at Middlesex University on December 7, 2007 http://jom.sagepub.comDownloaded from

Page 6: Organizational Behavior

Hypothesis 2: Employees’ perceptions of organizational support will be positively andsignificantly correlated with their trust in management.

Similarly, researchers (e.g., Gopinath & Becker, 2000; Kim & Mauborgne, 1993; Pearce,1993) have estimated the strength of the relationship between organizational commitmentand employees’ trust in management. They have found significant correlations ranging from0.42 to 0.61. I expect to replicate these past results such that:

Hypothesis 3: Employees’ commitment to the organization will be positively andsignificantly correlated with their trust in management.

Hypotheses 1 through 3 suggest that trust, perceived organizational support, and commit-ment are inter-related. Indeed, Eisenberger et al. recognized links among perceived organi-zational support, trust and commitment when they noted “perceived organizational supportwould also enhance calculative involvement [an aspect of commitment] by creating trust thatthe organization will take care to fulfill its exchange obligations. . . ” (Eisenberger et al.,1990, p. 52). Although they did not elaborate, they seem to be suggesting that trust inmanagement mediates the relationship between employees’ perceptions of the organization’ssupport and commitment and their own commitment response.

Several studies looking at other kinds of perceptions (e.g., procedural fairness or individ-uals’ support) have generated results consistent with this notion. Individuals’ trust builds asa response to their perceptions of the actions of another and leads to an increase in theircommitment to that actor (Folger & Konovsky, 1989; Gopinath & Becker, 2000; Pillai et al.,1999; Podsakoff et al., 1996). These results suggest a perfectly mediated relationship inwhich perceptions are associated with commitment only indirectly through trust; however, inthis context, the relationship between perceived organizational support and commitment islikely to include a direct link as well. Because the correlations between perceived organi-zational support and commitment found in previous research (0.38–0.71) are so similar tothe correlations between trust and commitment (0.42–0.61), the relationship between per-ceived organizational support and trust would have to be perfect (1.0) for trust to perfectlymediate the relationship between perceived organizational support and commitment. Theseprevious results suggest a partially mediated model instead such that:

Hypothesis 4: Employees’ perceptions of organizational support will be related toemployee commitment directly as well as indirectly through their trust in management.

3.2. Human resource practices, perceived organizational support, trust-in-management,and commitment

Ostroff and Bowen (2000) relied on social exchange and the norm of reciprocity indeveloping hypotheses about the relationships among human resource practices, attitudes,and performance. They proposed that human resource practices shape work force attitudes bymolding employees’ perceptions of what the organization is like and influencing theirexpectations of the nature and depth of their relationship with the organization. Employeeattitudes and behaviors (including performance) reflect their perceptions and expectations,reciprocating the treatment they receive from the organization. In their multilevel model

519E.M. Whitener / Journal of Management 27 (2001) 515–535

© 2001 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution. at Middlesex University on December 7, 2007 http://jom.sagepub.comDownloaded from

Page 7: Organizational Behavior

linking human resource practices and employee reactions, they depicted relationships sug-gesting that human resource practices are significantly associated with employee perceptionsand employee attitudes.

In the only study testing these relationships that explicitly recognizes the cross-levelnature of the research questions, Tsui and her colleagues (Tsui, Pearce, Porter & Tripoli,1997) found that employee attitudes (specifically employee commitment) were associatedwith the interaction of human resource practices and perceptions. They analyzed data froma sample of 10 organizations and over 900 employees in 85 jobs. They conducted theiranalyses at the individual-level but controlled for job-level differences in organizationalsupport. They found that employee commitment is associated with the interaction of humanresource practices (e.g., performance appraisal and rewards) and support.

These results are consistent with research on perceived organizational support. Sev-eral studies have indicated that perceived organizational support interacts with organi-zational actions (including human resource practices) in affecting employee commitment(Allen, 1992; Guzzo et al., 1994; Miceli & Mulvey, 2000; Wayne et al., 1997).Therefore, rather than just looking at direct relationships, I investigate the impact of theinteraction of human resource practices and perceptions on employee attitudes in thefollowing hypotheses:

Hypothesis 5: Human resource practices will moderate the relationship between per-ceived organizational support and organizational commitment such that the relationshipwill be stronger under commitment human resource practices than control humanresource practices.

Hypothesis 6: Human resource practices will moderate the relationship between per-ceived organizational support and trust-in-management such that the relationship willbe stronger under commitment human resource practices than control human resourcepractices.

These hypotheses are consistent with the general notion that human resource practicesinteract with perceptions of organizational support to affect employee commitment. How-ever, some human resource practices are more likely than others to have significant rela-tionships (Delery, 1998). Huselid (1995) suggested that human resource practices group intotwo categories—those practices that improve employee skills and those that enhance em-ployee motivation. In a study of over 900 organizations in the United States, he validatedthese two categories and their effects. He found that skill-enhancing human resourceactivities included selection and training activities and were associated with turnover andfinancial performance and that motivation-enhancing activities included performance ap-praisal and compensation activities and were associated with measures of productivity.

Because they can provide direct and substantial harm or benefit to employees (Mayer &Davis, 1999), motivation-oriented human resource activities are more likely to be associatedwith perceived organizational support and commitment than skill-oriented activities. Indeed,selection and training activities may not even be very salient to employees. Although jobcandidates pay close attention to selection procedures, current employees are more likely tobe focused on the job itself. Similarly training activities are infrequent occurrences. Even at

520 E.M. Whitener / Journal of Management 27 (2001) 515–535

© 2001 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution. at Middlesex University on December 7, 2007 http://jom.sagepub.comDownloaded from

Page 8: Organizational Behavior

the “100 best companies to work for” employees spend only 2% of their work-year (anaverage of 47 hr a year) in training (Levering & Moskowitz, 2000).

In contrast, performance appraisal and compensation processes have regular andpowerful effects on employees. Poorly designed or conducted appraisal systems can failto accurately evaluate the quality or quantity of performance. Compensation systems,especially those linked to performance assessments, can miss providing salient rewardsto the right people in a timely fashion. These flaws have the potential to under-rewarddeserving individuals and over-reward undeserving individuals. Aware of their vulner-ability to the vagaries of badly designed appraisal and reward systems, employees arelikely to perceive well-designed, developmental performance appraisals and internallyequitable and externally competitive compensation systems (Snell & Dean, 1992) asindicative of the organization’s support and commitment to them. As discussed above,employees would reciprocate their perceptions of the organization’s support and com-mitment conveyed by appraisal and compensation practices with their own commitmentto the organization.

In the same way, employees are likely to reciprocate their perceptions of support conveyedby appraisal and compensation activities in their trust in management. Whitener (1997) alsorelied on social exchange theory to predict that trust-in-management is associated withhuman resource practices through their effects on employees’ perceptions of support. In alongitudinal, quasi-experimental study conducted at the individual level, Mayer and Davis(1999) found that the implementation of a “more acceptable” performance appraisal systemincreased employees’ trust in management. They also found that employees’ perceptions ofmanagement’s ability, benevolence, and integrity mediated the relationship between theirperceptions of the performance appraisal system and trust-in-management. Mayer and Davis’purpose was to test somewhat different constructs and relationships; yet, their results reflectthe same pattern suggested by social exchange: human resource activities are associated withemployee attitudes through their perceptions.

Hypothesis 7: The patterns of relationships among human resource practices, perceivedorganizational support, trust-in-management, and organizational commitment will bestronger for performance appraisal and compensation human resource practices than forselection and training.

4. Summary

Taken together, these hypotheses imply a framework (depicted in Fig. 1) in whichtrust-in-management mediates the relationship between perceptions of organizationalsupport and employee commitment and human resource practices moderate the relation-ships between perceptions of organizational support and both trust and commitment. Asdescribed below, the individual-level relationships in the framework are tested usinghierarchical multiple regression and the cross-level relationships are tested using hier-archical linear modeling.

521E.M. Whitener / Journal of Management 27 (2001) 515–535

© 2001 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution. at Middlesex University on December 7, 2007 http://jom.sagepub.comDownloaded from

Page 9: Organizational Behavior

5. Materials and Methods

5.1. Sample and procedures

I worked with research associates from credit union associations to identify and contactthe sample. We drew the sample from the population of credit unions in the United States.In 1996, the year this study was conducted, there were approximately 12,000 credit unionsin the U.S. Most of them (75%) were small, with assets of less than $25 million and fewerthan 10 employees. One percentage was large, with assets of more than $500 million and anaverage of 350 employees.

The research associates generated a stratified random sample of 500 credit unions fromtheir database of credit unions. We excluded the small credit unions, those with assets below$25 million, because they have too few employees. We were concerned that it would bedifficult to preserve their confidentiality and that they would not have a formal human

Fig. 1. Cross-level framework depicting hypotheses linking human resource practices, perceived organizationalsupport, trust-in-management, and employee commitment.

522 E.M. Whitener / Journal of Management 27 (2001) 515–535

© 2001 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution. at Middlesex University on December 7, 2007 http://jom.sagepub.comDownloaded from

Page 10: Organizational Behavior

resource function. We slightly oversampled large credit unions, with assets over $1 billion,because there are so few.

We mailed letters describing the study and requesting participation to the CEO of eachcredit union. We also included the human resource and employee surveys. We instructedthe CEO to ask the senior human resource officer to complete the human resourcepractices questionnaire and to organize the distribution, collection, and return of em-ployee surveys in a way to preserve the anonymity of participants. We includedidentification information on the human resource survey; however, the employee surveyincluded no identifying information. In addition, the employee survey was contained ina booklet with an opaque cover sheet on the front and back. Employees stapled theirsurvey closed inside the cover sheet so that their responses were completely hidden. Thehuman resource contact returned all surveys unopened to the research team. The researchteam was instructed to identify any surveys that had been opened and restapled. Noneappeared to have been tampered with.

Of the 500 credit unions contacted, 185 returned all or part of the survey for a responserate of 37%. Respondents varied by asset size ($27 million to $8,700 million with a mean of$326 million), number of members (2000 to 2,000,000 with a mean of 65,912), and numberof full time equivalent employees (12 to 3035 with a mean of 133). A comparison ofrespondents to nonrespondents by the credit union research affiliate provided no evidence ofresponse bias.

The human resource practices survey asked human resource managers to describe theircredit union’s staffing, training, compensation, and performance appraisal practices. Of the185 organizations, 182 provided complete data on their human resource practices.

The employee survey assessed employees’ commitment to the organization, perceptionsof organizational support, and trust-in-management. Human resource managers were askedto pick 10 employees to complete the surveys and most indeed returned 10. The number ofemployees responding from each credit union averaged a mean of 9.37 and mode and medianof 10 employees per credit union. One credit union only returned one employee survey;another made copies and returned 18. The remaining credit unions provided at least 5 and nomore than 10 employee usable responses. The two outliers were removed from the database,yielding a total sample of 180 credit unions and 1689 employees.

The average employee respondent was 36 years old and had worked for 6.7 years at hisor her credit union. Most were full time (92.5%) and female (83%). Three percentage heldexecutive positions, 25% held management or supervisory positions, 43% held professionalstaff positions and did not supervise subordinates, and 29% were administrative assistants ortellers.

5.2. Measures

5.2.1. Human resource practicesThe human resource practices survey contained scales developed by Snell and Dean

(1992) to measure high commitment human resource practices: selective staffing measuresthe extensiveness of the firm’s selection process; comprehensive training measures theextensiveness of the firm’s training and development process; developmental appraisal

523E.M. Whitener / Journal of Management 27 (2001) 515–535

© 2001 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution. at Middlesex University on December 7, 2007 http://jom.sagepub.comDownloaded from

Page 11: Organizational Behavior

measures whether performance appraisal is used for developing employees; externallyequitable reward systems measured the extent to which the organization’s pay levels werecompetitive with similar organizations; and internally equitable reward systems measured theextent to which the organization’s pay structure was equitably construed. Scales ranged from1 to 5 but the anchors varied depending on the question. For example, the response for thestaffing item, “How extensive is the employee selection process for a job?” ranged from “notextensive (1) to ”very extensive (5). The response for the comprehensive training item, “Howmuch priority is placed on training employees?” ranged from “very little” (1) to “a greatdeal” (5).

Snell and Dean (1992) provided little evidence of the construct validity of their measures,conducting only an exploratory factor analysis, so I replicated their procedure and tookseveral additional steps to explore their psychometric properties (Hinkin, 1995; 1998). First,an exploratory factor analysis (principle components and varimax rotation) yielded fivefactors consistent with those proposed by Snell and Dean. A confirmatory factor analysis,however, did not fit the factors well yielding a �2 of 1276.78 (df. � 395), with NFI � 0.60and CFI � 0.68. To explore whether greater parsimony would increase the fit, I evaluated thetop four items for each scale using confirmatory factor analysis. The five factor model(developmental appraisal, selective staffing, comprehensive training, internally equitablerewards, and externally competitive rewards) fit the data significantly better yielding a �2 of393.72 (df. � 160), with NFI � 0.79 and CFI � 0.86. In addition, this five-factor model fitthe data better than a one factor model representing the whole system of high commitmenthuman resource practices (�2 � 1006.60, df. � 170, NFI � 0.46 and CFI � 0.50). On thisbasis the four-item scales were retained to measure human resource practices. Coefficientalpha for these scales fell within acceptable levels ranging from 0.70 (staffing) to 0.86(training).

5.2.2. Employee attitudesWell-established scales were used for each attitude. Perceived organizational support was

measured with Eisenberger’s scale (Eisenberger et al., 1986). The positively worded itemsfrom the Organizational Commitment Questionnaire (Mowday et al., 1979) measured orga-nizational commitment. Their items reflect the extent to which the employee is willing to putin a great deal of effort beyond that normally expected and the extent to which the employee“talks up” the organization as a great place to work. Robinson and Rousseau (1994)developed a scale to measure employees’ trust in their employer. Items include “I am not sureI fully trust my employer (R)” and “In general, I believe my employer’s motives andintentions are good.”

These items were also subjected to an exploratory factor analysis yielding three scales.Scales comprised of the top four items were evaluated using confirmatory factor analysis.The results indicated that a three-factor model (�2 � 521.72, df. � 51, NFI � 0.97 andCFI � 0.97) fit the data better than a one-factor model (�2 � 3741.47, df. � 54, NFI � 0.80and CFI � 0.80). Coefficient alpha for each scale fell in acceptable and usual ranges from0.85 (trust-in-management) to 0.90 (employee commitment).

524 E.M. Whitener / Journal of Management 27 (2001) 515–535

© 2001 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution. at Middlesex University on December 7, 2007 http://jom.sagepub.comDownloaded from

Page 12: Organizational Behavior

5.3. Data analysis

5.3.1. Individual-level analysesAs indicated above, I analyzed the individual-level data using multiple regression. With

Hypothesis 4, which predicts that trust-in-management mediates the relationship betweenperceived organizational support and commitment, I used hierarchical multiple regression toassess the three conditions to demonstrate mediation: (1) a significant relationship betweenthe mediator and the dependent variable, (2) a significant relationship between the indepen-dent variable and the dependent variable, and (3) the relationship between the independentvariable and the dependent variable decreasing or becoming nonsignificant when the medi-ator is added to the step (Baron & Kenny, 1986).

5.3.2. Cross-level analysesAs indicated above, hierarchical linear modeling allows for the iterative investigation of

multiple levels of relationships with individual-level dependent variables (Hofmann, 1997;Hoffman, Griffin & Gavin, 2000). A “level 1” analysis estimates parameters describing therelationship(s) between independent and dependent variables within each group, that is, atthe individual level. The parameters depicting the relationships (the intercept and slopeestimates) become the dependent variables for the “level 2” analysis that assesses the role ofthe higher order (e.g., group or organizational) variables. Significant coefficients on predic-tors of the intercepts and slopes provide evidence of the cross-level relationships.

Several conditions must be established before testing specific hypotheses using hierar-chical linear modeling. These conditions are investigated through a series of models. First,the purpose of this study is to investigate whether employee commitment is associated withindividual-level (trust-in-management and perceived organizational support) and organiza-tional-level (human resource practices) variables. Therefore, the first condition to be estab-lished is the existence of within- (individual) and between- (organizational) variance inemployee commitment. This condition is evaluated by estimating the “null model,” whichpartitions the variance in commitment into within and between-group components. A �2 teston the residual variance indicates whether the level-2 (between-group) variance is signifi-cantly different from zero.

The second model investigates the nature of the between-group variance, after controllingfor within-group variance. For example, this model explores the amount of variance ingroups’ intercepts and slopes of the regression equations representing the relationshipbetween the individual-level variables, organizational commitment, and perceived organi-zational support. A t test of the parameters in the level 1 equation provides evidence of thestrength of the relationship between organizational commitment and perceived organiza-tional support. A �2 test for the residual variances in the level 2 equations in which groupintercepts and slopes are regressed on unit vectors (no predictors) indicates whether thevariances in group intercepts and slopes are significantly different from zero.

Finally, the last set of models is only investigated if the second model indicates that thereis significant variance in the intercepts and/or slopes. If there is significant variance in theintercepts, then the “intercepts-as-outcomes” model estimates whether this variance isassociated with the organizational-level variable (human resource practices). For example,

525E.M. Whitener / Journal of Management 27 (2001) 515–535

© 2001 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution. at Middlesex University on December 7, 2007 http://jom.sagepub.comDownloaded from

Page 13: Organizational Behavior

the level 1 equation regresses organizational commitment on perceived organizationalsupport as in the second model above, but the level 2 analysis of intercepts-as-outcomes addshuman resource practices as predictors to the intercept equation. The t tests on the coeffi-cients associated with human resource practices test whether they are significantly related tothe variance in intercepts or in other words, whether human resource practices are related toorganizational commitment after controlling for perceived organizational support. If the �2

test on the residual variance is significant, it indicates that additional group-level predictorsmay be associated with the group-level variance in intercepts

Similarly, if the results of the second model indicate that there is significant variance inthe slopes, then the “slopes-as-outcomes” model estimates whether this variance is associ-ated with the organizational-level variable (tests of Hypotheses 5 and 6). This model issimilar to the intercepts-as-outcome model but adds human resource practices as a predictorto the slopes equation in the level-2 analysis. The t test on human resource practices indicateswhether they are significantly related to the variance in slopes. Such a result is consistent, forexample, with the notion that human resource practices affect the relationship with organi-zational commitment and perceived organizational support thus representing a cross-levelinteraction effect. The �2 test indicates whether additional level-2 predictors might beassociated with the variance in slopes.

Hierarchical linear modeling provides several “centering” options to assist in theinterpretation of results concerning the intercept term in the level-2 analyses (Bryk &Raudenbush, 1992; Hofmann, 1997; Hofmann & Gavin, 1998). “Grand-mean” centeringindicates that the intercept represents expected commitment for a person with an averagelevel of the predictor, perceived organizational support. “Group-mean” centering repre-sents expected commitment for a person with his or her group’s average perceivedorganizational support. The appropriate choice of centering depends on the model.Grand-mean centering provides better estimates and interpretability with most models;however, group-mean centering plus the addition of an aggregate measure of the meanof the individual scores in perceived organizational support in each organization,facilitates estimation and interpretability of cross-level moderation effects (as in theslopes-as-outcomes model).

The data used for the hierarchical linear modeling analyses were gathered from over 180credit unions with a median of 10 employees from each. Recent studies have providedevidence that this data set should have sufficient power to detect differences. For example,results indicate that a sample of 150 groups requires only five persons per group to obtain apower of 0.90 (Hofmann, 1997). This data set surpasses this estimate.

Even though ten subjects per group may be adequate for power analyses, they may notadequately represent a group. To establish whether recognizing them as a group isappropriate, I calculated intraclass correlations and inter-rater agreement indices toexamine within-group agreement and between-group variation (James, Demaree & Wolf,1984; Klein & Kozlowski, 2000). ICC(1) ranged from 0.18 to 0.19, ICC(2) ranged from0.69 to 70, and rwg ranged from 0.91– 0.93 indicating aggregating the measures ofemployee commitment, perceptions of organizational support, and trust to the organi-zational level was appropriate.

526 E.M. Whitener / Journal of Management 27 (2001) 515–535

© 2001 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution. at Middlesex University on December 7, 2007 http://jom.sagepub.comDownloaded from

Page 14: Organizational Behavior

6. Results

Table 1 contains descriptive statistics across all levels. The individual level data allow forthe assessment of Hypotheses 1 through 3, which predict significant relationships among thethree attitudes: perceived organizational support, trust, and organizational commitment. Thecorrelations among these variables, presented in the top third of Table 1, indicate that the dataare consistent with these hypotheses. The relationships are strong, for example, the corre-lation between perceived organizational support and organizational commitment is 0.70,consistent with results found in prior studies.

In Hypothesis 4, I predicted that trust mediates the relationship between perceivedorganizational support and organizational commitment such that perceived organizationalsupport is both directly and indirectly related (through trust) to organizational commitment.Table 2 presents the results of the hierarchical regression testing whether trust acts as amediator. The result is consistent with this prediction. The significant relationship betweenperceived organizational support and organizational commitment declines substantially whentrust is added to the equation. However, the relationship between perceived organizationalsupport and organizational commitment remains significant when controlling for trust,

Table 1Descriptive statisticsa

Mean SD 1 2 3 4 5 6 7 8

1. Perceived organizationalsupport

3.60 .83 (.88)

2. Trust 3.86 .87 .66** (.85)3. Organizational commitment 3.96 .85 .70** .61** (.90)4. Appraisal 3.01 .80 .20** .13 .10 (.78)5. External rewards 3.65 .73 .12 .08 .08 .16* (.74)6. Internal rewards 3.41 .68 .19** .21** .11 .37** .37** (.70)7. Staffing 3.72 .54 .05 .06 .07 .35** .17* .34** (.70)8. Training 3.34 .90 .02 .01 .09 .42** .25** .23** .47* (.86)

a n � 180 for inter-correlations among HR practices and cross-level correlations between HR practices andaggregated employee attitudes; n � 1689 for inter-correlations among individual-level employee attitudes.Coefficient alphas are on diagonal.

* p � .05.** p � .01.

Table 2Results of regressing employee commitment on perceived organizational support and trust (n � 1689)

Step Independentvariable

� Adj. R2 �R2 F p

1 Perceived organizationalsupport

.70 .49 1658.69 .000

2 Perceived organizationalsupport

.53 .53 963.77 .000

Trust .26 .04 137.67 .000

527E.M. Whitener / Journal of Management 27 (2001) 515–535

© 2001 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution. at Middlesex University on December 7, 2007 http://jom.sagepub.comDownloaded from

Page 15: Organizational Behavior

consistent with the hypothesis, that trust only partially mediates the relationship betweenperceived organizational support and organizational commitment.

I investigated the cross-level hypotheses (5–7) using hierarchical linear modeling. How-ever, before testing the hypotheses I needed to explore the conditions associated withhierarchical linear modeling. The first condition, whether there is systematic within- andbetween-group variance, is explored by looking at the null model. The �2 test on the nullmodel (�2 � 377.85, df. � 179, p � .001) indicated that employee commitment variedsignificantly by organizations, satisfying the first condition. In addition, the intraclasscorrelation (Hofmann, 1997) indicated that 10.5% of the variance in organizational com-mitment lies between organizations.

Similarly, the null model for trust indicated that there is systematic within and between-group variance. The �2 test (�2 � 331.96, df. � 179, p � .001) showed that trust variedsignificantly by organizations. The intraclass correlation (Hofmann, 1997) indicated that8.2% of the variance in trust lies between organizations.

The second model estimated the between-group variance in the intercepts and slopes andthe amount of variance in organizational commitment or trust explained by perceivedorganizational support (R2). The results for organizational commitment replicate the findingsof previous studies indicating that organizational commitment and perceived organizationalsupport are strongly related (� � 0.70, t � 33.58, p � .001; R2 � 0.49). They also indicatethat, after controlling for perceived organizational support, sufficient variance remains in theintercepts and slopes to investigate their relationship with human resource practices (�2 �273.45, df. � 179, p � .001 for intercepts; �2 � 230.69, df. � 179, p � .001 for slopes),satisfying the second condition.

The results also indicate that perceived organizational support and trust are significantlyrelated (� � 0.70, t � 31.86 p � .001; R2 � 0.44). They reveal that, after controlling forperceived organizational support, sufficient variance remains in the intercepts and slopes toinvestigate their relationship with human resource practices (�2 � 263.54, df. � 179, p �.001) for intercepts; �2 � 247.75, df. � 179, p � .001 for slopes).

The intercepts-as-outcomes model explores whether human resource practices were as-sociated with the variance in intercepts in organizational commitment or trust-in-manage-ment after controlling for perceived organizational support. t tests indicated that none of theregression coefficients representing the relationship between human resource practices andorganizational commitment were significant. Human resource practices did not explainadditional variance in organizational commitment after controlling for perceived organiza-tional support. However, sufficient between-group variance remained in organizationalcommitment after controlling for perceived organizational support that a search for alterna-tive between-group predictors would be warranted.

I also used the intercepts-as-outcomes model to explore whether human resource practiceswere associated with the variance in intercepts in trust after controlling for perceivedorganizational support. t tests indicated the coefficient representing the relationship betweeninternal equity of rewards and trust was significant (� � 0.06, t � 2.05 p � .001; R2 � 0.08).Fairness of rewards predicts variance in trust beyond that explained by perceived organiza-tional support. Sufficient between-group variance remained to warrant a search for additionalbetween-group predictors.

528 E.M. Whitener / Journal of Management 27 (2001) 515–535

© 2001 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution. at Middlesex University on December 7, 2007 http://jom.sagepub.comDownloaded from

Page 16: Organizational Behavior

After exploring these conditions, I used the slopes-as-outcomes model to test Hy-potheses 5 and 6. First I tested Hypothesis 5, whether human resource practices wereassociated with the variance in slopes in organizational commitment after controlling forperceived organizational support. Perceived organizational support was added in thelevel-1 equation with group-mean centering and organizational means of perceivedorganizational support were added in the intercept equation in the level-2 analysis(Hofmann & Gavin, 1998). As shown in Table 3, t tests indicated that internal equitysignificantly affects the relationship between perceived organizational support and or-ganizational commitment such that it is stronger when organizations have high internalequity of rewards. It accounted for 16% of the variance in the organizational commit-ment—perceived organizational support slopes. The �2 indicated that a search foradditional between-group predictors would be warranted.

Second, I explored Hypothesis 6, whether human resource practices were associatedwith the variance in slopes in trust after controlling for perceived organizational support.Similar to the previous analysis, perceived organizational support was added in thelevel-1 equation with group-mean centering and organizational means of perceivedorganizational support were added in the intercept equation in the level-2 analysis(Hofmann & Gavin, 1998). As shown in Table 3, t tests indicated that the interactionsof developmental appraisal and comprehensive training with perceived organizationalsupport were related to trust. The relationship between perceived organizational supportand trust is stronger in organizations with highly developmental appraisal processes butweaker in organizations with highly comprehensive training opportunities. These humanresource practices accounted for 23.5% of the variance in the trust—perceived organi-zational support slopes. The �2 indicated that a search for additional between-grouppredictors would be warranted.

Finally, the hierarchical linear modeling analyses summarized in Table 3 provide evidenceof Hypothesis 7, that appraisal and rewards processes are more likely to be associated withcommitment and trust than selection and training. The results are only partially supportive ofthis hypothesis. Appraisal and internal reward human resource practices moderated therelationships of perceived organizational support with organizational commitment or trust.However, comprehensive training also moderated the relationship between perceived orga-nizational support and trust (albeit in an unpredicted, negative direction).

Table 3Results of the level-2 analysis for slopes-as-outcomes (controlling for perceived organizational support)

Dependentvariable

Appraisal Externalrewards

Internalrewards

Staffing Training

Organizational � .04 �.06 .08 �.04 �.03Commitment S.E. .03 .03 .04 .05 .03

p .209 .054 .025 .421 .364Trust � .07 �.05 .05 .00 �.06

S.E. .03 .03 .04 .05 .03p .045 .157 .216 .949 .047

529E.M. Whitener / Journal of Management 27 (2001) 515–535

© 2001 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution. at Middlesex University on December 7, 2007 http://jom.sagepub.comDownloaded from

Page 17: Organizational Behavior

7. Discussion

The purpose of the study was to explore the relationships among human resourcepractices, trust-in-management, and organizational commitment using social exchange the-ory. Research on social exchange theory has shown that employees’ commitment to theorganization derives from their perceptions of the employers’ commitment to and support ofthem. It also implies that employees interpret human resource practices and the trustwor-thiness of management as indicative of the personified organization’s commitment to them.They reciprocate their perceptions accordingly in their own commitment to the organization.

Human resource practices and trustworthiness of management exist at different levels ofanalysis, organizational and individual respectively, and thus require cross-level analyses.Hierarchical linear modeling complemented analyses conducted at the individual level usinghierarchical multiple regression to investigate these relationships.

The first set of Hypotheses (1–4) summarized relationships among the individual-levelvariables culminating in the integrating prediction (reflected in Hypothesis 4) that trustpartially mediates the relationship between perceived organizational support and organiza-tional commitment (e.g., Eisenberger et al., 1990; Settoon et al., 1996). The results of thehierarchical linear regression were consistent with a case of partial mediation: perceivedorganizational support has direct and indirect relationships, through trust-in-management,with organizational commitment.

These results indicate that employees’ trust and commitment are stronger when theyperceive that the organization is committed to and supportive of them. Researchers exploringemployees’ perceptions of organizational support have proposed such relationships but rarelyexplored them empirically. This study contributes to that literature by demonstrating thestrong relationships. However, as a cross-sectional study, it cannot more deeply contribute tothe understanding of the processes of social exchange. As Blau predicted (Blau, 1964), trust(also commitment) grows as partners’ exchanges gradually expand. To more fully investigatethese relationships, we need to conduct longitudinal studies with multiple measurements ofperceptions and attitudes.

The second set of Hypotheses (5–7) explored the cross-level relationships between humanresource practices (an organizational-level variable) and employee perceptions and attitudes(individual-level variables). Using hierarchical linear modeling, I explored whether employ-ees’ attitudes (trust and commitment) were related to the organization’s human resourcepractices. The analysis first established that 8% of the variance in employee trust and 10%of the variance in commitment were associated with organizational differences. Second, itindicated that a search for organizational-level predictors was warranted for both trust andcommitment but that only internal equity of rewards directly accounted for differences acrossorganizations in trust after controlling for perceived organizational support.

Finally, I used hierarchical linear modeling to explore the cross-level hypotheses. Thehypotheses represented the social exchange-based notion that employees’ attitudes reflecttheir interpretation of the actions of the personified organization, including the organization’shuman resource policies. In the most specific and integrated Hypothesis (7), I predicted thatappraisal and compensation activities interact with perceived organizational support to affectboth organizational commitment and trust. The results of the analyses are partially consistent

530 E.M. Whitener / Journal of Management 27 (2001) 515–535

© 2001 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution. at Middlesex University on December 7, 2007 http://jom.sagepub.comDownloaded from

Page 18: Organizational Behavior

with these predictions. Two of six predicted relationships were significant: the relationshipbetween perceived organizational support and organizational commitment was strongerwhen organizations have high internal equity of rewards and the relationship betweenperceived organizational support and trust in management was stronger when organizationsconduct developmental appraisals. I also did not find evidence to reject three of the four nullpredictions (for selection and training); however, contrary to expectations, training did havea significant, and negative, interaction with perceived organizational support. Specifically,the relationship between perceived organizational support and trust was stronger whenorganizations offer less comprehensive training opportunities.

The results of the cross-level analyses are not inconsistent with Ostroff and Bowen’s(2000) framework suggesting that human resource practices are significantly associated withemployee perceptions and employee attitudes. Organizations’ employee groups do vary intheir attitudes—differences in organizational membership were significantly associated withdifferences in trust and commitment. In addition, motivation-focused human resource prac-tices such as developmental appraisals and equitable rewards seem to have a stronger andmore meaningful relationship with those organizational differences than selection and train-ing. Though not representing strong effects, the results indicate that managers can beencouraged that their commitment-oriented actions are associated with positive perceptionsand attitudes.

However, the study also yielded an unexpected result—that commitment is related to adisordinal interaction between training and perceived organizational support. The graph ofthis result indicates employees with low perceptions of organizational support expressedhigher commitment when they worked for organizations with more comprehensive trainingbut employees with high perceptions of organizational support expressed high commitmentwhen they worked for organizations with less comprehensive training. The complexities ofthis result suggest that other, unmeasured variables, perhaps related to employees’ perceptionof special treatment might also be interacting with perceptions of support and trainingcomprehensiveness in affecting commitment. For example, employees in organizations withnumerous opportunities for training may not see those opportunities as conveying theorganization’s support and commitment to them personally because the opportunity andbenefit of training are widely available. And, employees in organizations with fewer oppor-tunities for training may see these relatively rare opportunities as special treatment thatconveys the organization’s personal support and commitment.

This result needs to be explored further to see if it is an idiosyncratic result. In similarfashion to Mayer and Davis (1999), future researchers exploring these relationships shouldmeasure employees’ perceptions of the characteristics of human resource practices as anintervening variable between managers’ descriptions of human resource practices and em-ployees’ perceptions of support.

In addition, this study addresses but does not resolve the dilemma over whether humanresource practices contribute to organizations’ effectiveness as individual practices or assystems of high commitment versus high control practices (Delery, 1998). In this study,specific practices (e.g., internal equity of rewards and developmental performance appraisal)were associated with a specific employee-centered outcome—employee commitment. Inother studies, systems of practices have been associated with more global measures such as

531E.M. Whitener / Journal of Management 27 (2001) 515–535

© 2001 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution. at Middlesex University on December 7, 2007 http://jom.sagepub.comDownloaded from

Page 19: Organizational Behavior

financial performance. These results provide some support for Delery’s proposal that therelationship between human resource practices and outcomes depends on the fit between thepractices and outcomes—a proposal, however, that needs to be tested more directly anddeliberately.

Finally, this study has several strengths, for example, sufficient power and multiplesources of data; however, its limitations suggest that the findings should be interpreted withcaution. First, common method variance might inflate the estimates of the strength of therelationships among the individual-level variables (Harris & Schaubroeck, 1990). However,because the confirmatory factor analysis indicated that the variables were sufficiently distinct(Hinkin, 1995) and the intercorrelations were consistent with previous results, this threat isprobably minimal. Second, the data on human resource practices were provided by only onesource. The reliability and validity of this person’s perceptions of the credit union’s humanresource practices could not be verified. Third, I could not ensure managers randomlyselected employee respondents. Sampling bias could have artificially influenced the withingroup agreement, reducing the variation among employees within each organization. Finally,the domain of high commitment human resource practices includes a wide variety ofmanagement and human resource activities. The survey measures do not include all possibleactivities, policies, and procedures.

Future research should remedy these deficiencies. In addition, it should be designed tocapitalize on the analytic capabilities offered by hierarchical linear modeling to investigatethe relationship between organizational-level practices and employee attitudes and behavior.This study explored only a few relationships but demonstrated the usefulness of hierarchicallinear modeling in testing cross-level relationships. For example, none of the correlationsbetween human resource practices and aggregated organizational commitment was signifi-cant; but hierarchical linear modeling analyses indicated that employees’ commitment variedby organization and that human resource practices were associated with some of thatbetween-organization variation. Finally, future research should continue to explore whetherhuman resource practices affect organizational variables synergistically (e.g., in a bundle) orindependently.

In conclusion, employees interpret human resource practices and the trustworthiness ofmanagement (Eisenberger et al., 1990; Settoon et al., 1996) as indicative of the personifiedorganization’s commitment to them. The results of this study, consistent with social ex-change theory, implies that they reciprocate their perceptions accordingly in their owncommitment to the organization.

Acknowledgments

An earlier version of this study was presented at the 1998 Meeting of the SouthernManagement Association. Thank you to David Hofmann and Jeff Vancouver for theirassistance with hierarchical linear modeling and to John Delery, Adelaide Wilcox King, andthree anonymous reviewers for their comments on earlier drafts. Thanks also to the FileneResearch Institute, the Center for Credit Union Research at the University of Wisconsin—

532 E.M. Whitener / Journal of Management 27 (2001) 515–535

© 2001 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution. at Middlesex University on December 7, 2007 http://jom.sagepub.comDownloaded from

Page 20: Organizational Behavior

Madison, and the Center for Financial Services Studies at the University of Virginia for theirsupport of this project.

Ellen M. Whitener is a professor of management at the University of Virginia’s McIntireSchool of Commerce. She earned her Ph.D. in management at Michigan State University.Her current research interest focuses on the impact of human resource practices on employeecommitment, trust, and perceptions of organizational support.

References

Allen, M. W. (1992). Communication and organizational commitment: perceived organizational support as amediating factor. Communication Quarterly, 40, 357–367.

Arthur, J. B. (1994). Effects of human resource systems on manufacturing performance and turnover. Academyof Management Journal, 37, 670–687.

Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychologicalresearch: conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology,51, 1173–1182.

Blau, P. M. (1964). Exchange and power in social life. New York: Wiley.Bryk, A. S., & Raudenbush, S. W. (1992). Hierarchical linear models. Newbury Park, CA: Sage.Delaney, J. T., & Huselid, M. A. (1996). The impact of human resource management practices on perceptions of

organizational performance. Academy of Management Journal, 39, 949–969.Delery, J. E. (1998). Issues of fit in strategic human resource management: implications for research. Human

Resource Management Review, 8, 289–309.Eisenberger, R., Fasolo, P., & Davis–LaMastro, V. (1990). Perceived organizational support and employee

diligence, commitment, and innovation. Journal of Applied Psychology, 75, 51–59.Eisenberger, R., Huntington, R., Hutchison, S., & Sowa, D. (1986). Perceived organizational support. Journal of

Applied Psychology, 71, 500–507.Folger, R., & Konovsky, M. (1989). Effects of procedural and distributive justice on reactions to pay raise

decisions. Academy of Management Journal, 32, 115–130.Gopinath, C., & Becker, T. E. (2000). Communication, procedural justice, and employee attitudes: relationships

under conditions of divestiture. Journal of Management, 26, 63–83.Guzzo, R. A., Noonan, K. A., & Elron, E. (1994). Expatriate managers and the psychological contract. Journal

of Applied Psychology, 79, 617–626.Harris, M. M., & Schaubroeck, J. (1990). Confirmatory modeling in organizational behavior/human resource

management: issues and applications. Journal of Management, 16, 337–360.Hinkin, T. R. (1995). A review of scale development practices in the study of organizations. Journal of

Management, 21, 967–988.Hinkin, T. R. (1998). A brief tutorial on the development of measures for use in survey questionnaires.

Organizational Research Methods, 1, 104–121.Hofmann, D. A. (1997). An overview of the logic and rationale of hierarchical linear models. Journal of

Management, 23, 723–744.Hofmann, D. A., & Gavin, M. B. (1998). Centering decisions in hierarchical linear models: implications for

research in organizations. Journal of Management, 24, 623–641.Hofmann, D. A., Griffin, M. A., & Gavin, M. B. (2000). The application of hierarchical linear modeling to

organizational research. In K. J. Klein & S. W. J. Kozlowski (Eds.), Multilevel theory, research, and methodsin organizations: foundations, extensions, and new directions (pp. 467–511). San Francisco: Jossey–Bass.

Homans, G. C. (1961). Social behavior. New York: Harcourt, Brace, and World.House, R., Rousseau, D. M., Thomas–Hunt, M. (1995). The meso paradigm: a framework for the integration of

533E.M. Whitener / Journal of Management 27 (2001) 515–535

© 2001 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution. at Middlesex University on December 7, 2007 http://jom.sagepub.comDownloaded from

Page 21: Organizational Behavior

micro and macro organizational behavior. In L. L. Cummings & B. M. Staw (Eds.), Research in organiza-tional behavior (pp. 71–114). Greenwich, CT: JAI Press.

Huselid, M. A. (1995). The impact of human resource management practices on turnover, productivity, andcorporate financial performance. Academy of Management Journal, 38, 635–672.

Hutchison, S., & Garstka, M. L. (1996). Sources of perceived organizational support: goal setting and feedback.Journal of Applied Social Psychology, 26, 1351–1366.

Ichniowski, C., Shaw, K., & Prennushi, G. (1997). The effects of human resource management practices onproductivity. American Economic Review, 87, 291–313.

James, L. R., Demaree, R. G., & Wolf, G. (1984). Estimating within-group interrater reliability with and withoutresponse bias. Journal of Applied Psychology, 69, 85–98.

Kim, W. C., & Mauborgne, R. A. (1993). Procedural justice, attitudes, and subsidiary top management compli-ance with multnationals’ corporate strategic decisions. Academy of Management Journal, 36, 502–526.

Klein, K. J., & Kozlowski, S. W. J. (2000). From micro to meso: critical steps in conceptualizing and conductingmultilevel research. Organizational Research Methods, 3, 211–236.

Levering, R., & Moskowitz, M. (2000). 100 best companies to work for in America. Fortune, January 10,82–110.

Levinson, H. (1965). Reciprocation: the relationship between man and organization. Administrative ScienceQuarterly, 9, 370–390.

Mayer, R. C., & Davis, J. H. (1999). The effect of the performance appraisal system on trust for management:a field quasi-experiment. Journal of Applied Psychology, 84, 123–136.

MacDuffie, J. P. (1995). Human resource bundles and manufacturing performance: organizational logic andflexible production systems in the world auto industry. Industrial and Labor Relations Review, 48, 197–221.

Miceli, M. P., & Mulvey. P. W. (2000). Consequences of satisfaction with pay systems: two field studies.Industrial Relations, 39, 62–87.

Mowday, R. T., Steers, R. M., & Porter, L. W. (1979). The measurement of organizational commitment. Journalof Vocational Behavior, 14, 224–247.

Ostroff, C., & Bowen, D. E. (2000). Moving HR to a higher level: HR practices and organizational effectiveness.In K. J. Klein & S. W. J. Kozlowski (Eds.), Multilevel theory, research, and methods in organizations:foundations, extensions, and new directions (pp. 211–266). San Francisco: Jossey–Bass.

Pearce, J. L. (1993). Toward an organizational behavior of contract laborers: their psychological involvement andeffects on employee co-workers. Academy of Management Journal, 36, 1082–1096.

Pillai, R., Schriesheim, C. A., & Williams, E. S. (1999). Fairness perceptions and trust as mediators fortransformational and transactional leadership: a two-sample study. Journal of Management, 25, 897–933.

Podsakoff, P. M., MacKenzie, S. B., & Bommer, W. H. (1996). Transformational leader behaviors and substitutesfor leadership as determinants of employee satisfaction, commitment, trust, and organizational citizenshipbehaviors. Journal of Management, 22, 259–298.

Robinson, S. L., & Rousseau, D. M. (1994). Violating the psychological contract: not the exception but the norm.Journal of Organizational Behavior, 15, 245–259.

Rousseau, D. M. (1985). Issues of level in organizational research: multi-level and cross-level perspectives. InL. L. Cummings & B. M. Staw (Eds.), Research in organizational behavior (pp. 1–37). Greenwich, CT: JAIPress.

Rousseau, D. M., Sitkin, S. B., Burt, R. S., & Camerer, C. (1998). Not so different after all: a cross-disciplineview of trust. Academy of Management Review, 23, 393–404.

Settoon, R. P., Bennett, N., & Liden, R. C. (1996). Social exchange in organizations: perceived organizationalsupport, leader-member exchange, and employee reciprocity. Journal of Applied Psychology, 81, 219–227.

Shore, L. M., & Tetrick, L. E. (1991). A construct validity study of the Survey of Perceived OrganizationalSupport. Journal of Applied Psychology, 76, 637–743.

Shore, L. M., & Wayne, S. J. (1993). Commitment and employee behavior: comparison of affective commitmentand continuance commitment with perceived organizational support. Journal of Applied Psychology, 78,774–780.

534 E.M. Whitener / Journal of Management 27 (2001) 515–535

© 2001 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution. at Middlesex University on December 7, 2007 http://jom.sagepub.comDownloaded from

Page 22: Organizational Behavior

Snell, S., & Dean, J. (1992). Integrated manufacturing and human resource management: a human capitalperspective. Academy of Management Journal, 35, 467–504.

Tsui, A. S., Pearce, J. L., Porter, L. W., & Tripoli, A. M. (1997). Alternative approaches to the employee-organization relationship: does investment in employees pay off? Academy of Management Journal, 40,1089–1121.

Walton, R. E. (1985). From control to commitment in the workplace. Harvard Business Review, 63 (2), 77–84.Watson Wyatt. (1999). WorkUSA 2000: employee commitment and the bottom line. Bethesda, MD: Watson

Wyatt.Wayne, S. J., Shore, M., & Liden, R. C. (1997). Perceived organizational support and leader-member exchange:

a social exchange perspective. Academy of Management Journal, 40, 82–111.Whitener, E. M. (1997). The impact of human resource activities on employee trust. Human Resource Manage-

ment Review, 7, 389–404.Wood, S., & de Menezes, L. (1998). High commitment management in the U.K.: evidence from the Workplace

Industrial Relations Survey and Employers’ Manpower and Skills Practices Survey. Human Relations, 51,485–515.

Youndt, M., Snell, S., Dean, J., & Lepak, D. (1996). Human resource management, manufacturing strategy, andfirm performance. Academy of Management Journal, 39, 836–866.

535E.M. Whitener / Journal of Management 27 (2001) 515–535

© 2001 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution. at Middlesex University on December 7, 2007 http://jom.sagepub.comDownloaded from