Problem solving

PROBLEM SOLVING - DEMAND

HOME

AM GOV

ECONOMICS

AP GOV

AP ECON

SPECIAL EVENTS

ASSIGNMENT BOARD

Having studied the concepts behind demand, it’s time to apply them to real decisions. With a small group, work through each of the following problem solving activities - but, in all you do, make sure that the concepts are used appropriately to arrive at your answers.

A. Demand at the Grocery Store

Use the Weis flyer (or other provided store flyer) to locate examples of each of the following demand concepts. Cut out the portion of the flyer (picture or actual ad) that provides the appropriate example, post it on the given paper, and write a brief explanation on the paper of how it exemplifies the concept. When each have been completed, staple the the posted examples together as a packet. (15 points total - accurate examples & complete explanations)

1. Change in price of a substitute2.2. Change in price of a complement

3. Diminishing marginal utility4. Change in income (inferior vs. normal goods)5. Greater concern for health

2.B. Sell More Tickets - For the Same Price

Your group has become the officers for the Econ Club. You are selling tickets ($10) for an upcoming fundraising dance and sales are not what you hoped. How can you sell more tickets without dropping the price of the ticket? Create a plan that can be used to increase sales considering the concepts of demand. Explain your plan in a paragraph as a group and be able to share the idea.

We propose that problem-solving demand (PSD) is an important job attribute for employees' creative performance. Applying job design theory, we examined the relationship between PSD and employee creativity. The theorised model was tested with data obtained from a sample of 270 employees and their supervisors from three Chinese organisations. Regression results revealed that PSD was positively related to creativity, and this relationship was mediated by creative self-efficacy. Additionally, intrinsic motivation moderated the relationship between PSD and creative self-efficacy such that the relationship was stronger for individuals with high rather than low intrinsic motivation. We discuss our findings, implications for practice, and future research.

We propose that problem-solving demand (PSD) is an important job attribute

for employees’ creative performance. Applying job design theory, we examined

the relationship between PSD and employee creativity. The theorised model

was tested with data obtained from a sample of 270 employees and their

supervisors from three Chinese organisations. Regression results revealed that

PSD was positively related to creativity, and this relationship was mediated by

creative self-efficacy. Additionally, intrinsic motivation moderated the relationship between PSD and creative self-efficacy such that the relationship was

stronger for individuals with high rather than low intrinsic motivation. We

discuss our findings, implications for practice, and future research.

INTRODUCTION

Increasing global competition, financial crises, job restructuring, and the

flattening of organisational hierarchies have dramatically increased the creative problem-solving requirements of employees’ jobs (Shalley, Gilson, &

Blum, 2009). These trends have increased the velocity and frequency of

change (Eisenhardt & Tabrizi, 1995), reducing structure, predictability, and

supervision within the work environment (Moreland & Argote, 2003). As a

consequence, there has been a greater need for creativity at all levels and

across different types of jobs (Shalley et al., 2009). Reflecting the significance

of employee creativity, a growing body of literature has investigated individual and contextual influences on employee creativity (Oldham & Cummings, 1996; Shalley, Zhou, & Oldham, 2004). Yet despite these advances,

* Address for correspondence: Qin Zhou, ISCTE Business School, ISCTE-Instituto Universitário de Lisboa, 1649-026 Lisboa, Portugal. Email: [email protected]

APPLIED PSYCHOLOGY: AN INTERNATIONAL REVIEW, 2012, 61 (1), 56–80

doi: 10.1111/j.1464-0597.2011.00455.x

© 2011 The Authors. Applied Psychology: An International Review © 2011 International

Association of Applied Psychology. Published by Blackwell Publishing Ltd, 9600 Garsington

Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA.few studies have examined how basic elements of an employee’s job design

influence his or her creativity. There has, however, been an increasing recognition that creative ideas may be stimulated by employees’ work

experiences—and in particular aspects of job design (e.g. Mumford, Whetzel,

& Reiter-Palmon, 1997; Ohly, Sonnentag, & Pluntke, 2006; Oldham & Cummings, 1996; Tierney & Farmer, 2002). Further, a small but promising line of

research has found that a job’s design may be a potent source of creativity

(Oldham & Cummings, 1996; Unsworth, Wall, & Carter, 2005). These

studies illustrate that the cognitive requirements of a job, in combination

with individual characteristics, predict creative performance.

Extending previous research, the present study seeks to understand how

and when a job’s problem-solving demands influence employee creativity.

Drawing upon job design theory (Hackman & Oldham, 1975) and research

(Dean & Snell, 1991; Wall, Corbett, Clegg, Jackson, & Martin, 1990), this

study introduces and applies the concept of problem-solving demand (PSD)

to examine its influence on employee creativity. PSD refers to the extent to

which a job requires employees to actively utilise their knowledge and skills

to “diagnose and solve problems” at work (Wall et al., 1990, p. 208),

thereby challenging employees to develop new solutions to problems

stretching their knowledge and skill bases. PSD is a specific aspect of job

complexity (Campbell, 1988; Dean & Snell, 1991). Job complexity refers to

the level of stimulating and challenging demands associated with a particular job (Valcour, 2007) and encompasses many different facets. Complex

jobs may require individuals to juggle different tasks, learn a great deal of

procedural knowledge, as well as engage in challenging problem-solving to

provide solutions to applied problems. It is this latter aspect, involving fluid

cognitive functioning (Horn & Noll, 1997), that we are most interested in,

since it captures the extent to which the job requires the individual to

develop new and useful solutions to problems. In our view, PSD differs

from the extent to which employees are motivated to engage in creative

processes. Such “creative engagement” refers to an employee’s motivation

to develop creative problem solutions, while PSD pertains to the extent to

which the job design “stretches” the individual to develop skills and new

solutions to problems.

PSD provides employees with opportunities to apply their skills and

stretch their capabilities. We propose that PSD may promote creative selfefficacy in several ways. First we propose, consistent with social learning

perspectives (e.g. Bandura, 1986; Davis & Luthans, 1980; Sims, 1983;

Wood & Bandura, 1989), that employees develop and acquire new skills

and insights about their tasks through this experience. In turn, as a consequence of skill acquisition and potentially mastering new tasks, employees

develop confidence and a greater belief that they are able to solve problems

creatively (i.e. creative self-efficacy; Gong, Huang, & Farh, 2009; Tierney &

PROBLEM-SOLVING DEMAND AND CREATIVITY 57


Association of Applied Psychology.Farmer, 2002). Thus, we propose and test whether PSD is related to creativity through creative self-efficacy.

When seeking to understand the effects of PSD on employee beliefs and

behavior, it is important to understand the individual’s motivation, as not all

employees will respond similarly to challenging job demands. Theory

(Amabile, 1996) and research (Tierney, Farmer, & Graen, 1999) highlight

that the extent to which an individual is intrinsically motivated to perform a

particular task plays an important role in understanding how the individual

reacts to PSD (as evidenced by the extensiveness of problem-focused strategies adopted to resolve these challenges). Intrinsic motivation is one’s interest

in activities because of the inherent learning, stimulation, and enjoyment of

these activities (Deci & Ryan, 1985). Individuals who are intrinsically motivated will be more predisposed to invest effort and persist when they face

difficulties, thereby leveraging the knowledge-related benefits of challenging

work conditions. Furthermore, they are more likely than those low in intrinsic motivation to capitalise on the opportunities provided by PSD and

develop higher levels of creative self-efficacy. Thus a further objective of this

research is to examine the moderating effects of intrinsic motivation on the

relationship between PSD and creative self-efficacy.

This research contributes to the creativity literature in at least three ways.

First, the study extends a promising and growing body of literature that

applies learning derived from job design theory to the creativity literature.

Second, by testing the mediating influence of creative self-efficacy, we specify

the processes by which a job’s design stimulates creativity. Further, we shed

light on the inconsistent association between intrinsic motivation and creativity, by testing Shalley and colleagues’ prediction that motivation may be

a necessary but not sufficient condition to promote creativity (Shalley et al.,

2004). Consistent with our focus that creativity is increasingly important in a

wide array of jobs, we obtained data from a multi-organisational sample

comprising a variety of work functions and job types in the People’s Republic

of China. This sampling strategy also allowed us to examine the predictive

utility of European/American theory in a different cultural context—a key

direction for future research (Drazin & Shoonhoven, 1996; Farmer, Tierney,

& Kung-McIntyre, 2003; Shalley et al., 2004). Figure 1 depicts the relationships examined in this study.

LITERATURE REVIEW AND HYPOTHESES

PSD and Creativity

We define creativity as employees’ generation of novel and useful ideas

concerning products, procedures, and processes at work (Amabile, 1988;

Oldham & Cummings, 1996; Shalley et al., 2004). Reflecting calls by scholars

58 ZHOU ET AL.


Association of Applied Psychology.that creativity research should focus not just on contexts where creativity is

anticipated (e.g. R&D teams) but also on contexts where creativity is not

necessarily expected as a matter of course (Ford & Gioia, 2000; Mumford &

Gustafson, 1988; Tierney & Farmer, 2002), we examine creative performance

that involves incremental developments or adjustments, which are common

in a context where creativity is not an expected outcome (Mumford &

Gustafson, 1988).

According to Amabile (1996), when tasks are complex and intellectually

demanding, employees are likely to experience “interest, involvement, curiosity, satisfaction, or positive challenge” (p. 115). This, in turn, leads to

creativity. The positive relationship between complex job demands, such as

job complexity, and creativity has received some empirical support in extant

literature (see Shalley et al., 2004, for a review). PSD (Jackson, Wall, Martin,

& Davids, 1993; Wall et al., 1990; Wall, Jackson, & Mullarkey, 1995) is

defined as the extent to which individuals perceive their work to be challenging, exposing them to novel and unexpected events. PSD also relates to

whether the job requires the individual to apply job-specific accumulated

knowledge as well as adopting new approaches to develop solutions to problems. More importantly, however, we propose that PSD stands out as a

particularly important job attribute for creativity within a work context where

creativity is not explicitly required. It is arguable that creativity is not a

natural option in such a context. Employees may prefer familiar and routine

options and forsake novel creative options (Ford, 1996). While we acknowledge employees’ motivation as important in influencing their response to

challenging situations, we expect that on average PSD will “jolt” employees

out of their routines and point them in new directions (Csikszentmihalyi,

1993). When PSD is high, employees have to deal with problems that they

have not encountered before. In order to diagnose and solve these new

problems, employees are prompted to seek new information, knowledge, and

skills. Thus, PSD provides opportunities for employees to be open to possibilities and to let go of their usual ways of doing things. At the same time, the

Creativity

Problemsolving demand

Creative

self-efficacy

Intrinsic

motivation

FIGURE 1. Hypothesised model for the relationship between PSD and

creativity.



Association of Applied Psychology.extended knowledge base of jobs where PSD exists is likely to lead to creative

performance (Amabile, 1996). In contrast, when PSD is low, employees have

easy access to solutions. They will resort to routine approaches which consequently inhibit creative performance. Therefore, we hypothesised that:

Hypothesis 1: PSD is positively related to creativity.

The Mediating Influence of Creative Self-Efficacy

Reacting to the notion that one’s judgment of capability is domain specific

(Bandura, 1997), Tierney and Farmer (2002) developed the construct of

creative self-efficacy for applicability in a creativity context. Creative selfefficacy refers to “the belief that one has the ability to produce creative

outcomes” (Tierney & Farmer, 2002, p. 1138). It is a relatively new theoretical construct; even so, research evidence suggests that creative self-efficacy is

an important source of creativity (Gong et al., 2009; Tierney & Farmer, 2002,

2004). We hypothesise that creative self-efficacy is also a mediator of the

association between PSD and employee creativity.

PSD seems conducive to the formation and maintenance of employee

creative self-efficacy for several reasons. First, the experience of grappling

with complex problems will provide employees with greater confidence in

their capacity to deal with obstacles (Wood & Bandura, 1989), promoting

resilience and a sense of belief in one’s ability. Compared with low PSD tasks,

high PSD tasks stimulate employees to try new approaches to reduce

demands rather than follow established methods. Since these processes

promote flexibility and creativity in analyzing and identifying solutions,

employees working in high PSD situations will be more likely to experience

increased creative self-efficacy. Furthermore, challenging work offers opportunities to acquire new skills and task-related knowledge, enhancing the

arsenal of skills an individual possesses (cf. McCauley, Ruderman, Ohlott, &

Morrow, 1994). This, in turn, promotes self-belief, as well as the capacity to

develop creative solutions to problems. Further, researchers have concluded

that employees are unlikely to learn new skills and knowledge in repetitive

and restricted jobs (e.g. Frese, 1982) . When PSD is high, employees do not

have easy access to solutions. Rather, employees need to undergo extended

searches to obtain relevant information. For example, they may need to get

to know the operations of other work areas in order to diagnose and generate

solutions to problems that occur in their work area. PSD, therefore, engenders useful learning experiences which eventually lead to individuals’ belief in

their work abilities.

Finally, the definition and measurement of PSD in terms of the cognitive

problem-solving requirements of the task relates directly to an employee’s

perception of the extent to which the job encourages skill acquisition and

60 ZHOU ET AL.


Association of Applied Psychology.ultimately mastery of particular activities (as opposed to an evaluation of

their difficulty or non-job-related obstacles such as organisational politics).

In turn, development promotes a greater sense of capability. Empirical

research supports this hypothesised association, illustrating that more

broadly defined challenging job attributes such as task complexity have been

shown to be conducive to creative self-efficacy, in turn enhancing employee

creativity (Tierney & Farmer, 2002). Thus we hypothesised that:

Hypothesis 2a: PSD is positively related to creative self-efficacy.

Theorists have suggested that self-efficacy constitutes an indispensable

dimension of the motivational process important for individual creativity

(Bandura, 1997; Ford, 1996). As Bandura stated, “Effective personal functioning is not simply a matter of knowing what to do and be motivated to

do”. Rather, one needs to have efficacy beliefs which “activate cognitive,

motivational and affective processes that govern the translation of knowledge and abilities into proficient action” (1997, pp. 36–37). Only when individuals are confident about their ability to be creative will they engage in the

activities leading to creative performance (Bandura, 1997; Ford, 1996) .

Tierney and Farmer (2002) provide two reasons why creative self-efficacy

may be related to employee creativity. First, they argue that creative selfefficacy constitutes a motivational mechanism important for creativity. Creative performance involves challenges, risks, and potential failures. It is

important for one to be persistent in the face of difficulties (Amabile, 1983).

When individuals have high levels of creative self-efficacy, they hold a

strong belief in their ability to be successful in spite of difficulties. This

belief will help them to set creative goals, to be persistent, and to put in

more effort in their creative endeavors (Bandura, 1997). Second, creative

self-efficacy also serves as a cognitive mechanism important for creativity.

Creativity requires creativity relevant processes as well as domain-relevant

knowledge (Amabile, 1983, 1996). When individuals have high levels of

creative self-efficacy, they will sustain effort at seeking work-related information (Bandura, 1997), leading to a better understanding of work-related

knowledge. Furthermore, individuals who hold a strong belief in their creativity abilities will not be satisfied with ordinary and routine ideas or solutions (Ford, 1996). Rather, they will put in more effort to use cognitive

resources (e.g. knowledge, memory, analytic skills) to come up with unique

ideas. This is consistent with the notions of “divergent thinking” and

breaking “mental set”, which are inherent in creativity relevant processes

(Amabile, 1983). Such notions are consistent with research evidence that

creative self-efficacy is positively related to creativity (Tierney & Farmer,

2002, 2004).

As discussed, PSD prompts employees to develop and apply problemsolving knowledge and skills by directly dealing with work-related problems.



Association of Applied Psychology.The resulting attainments or mastery experiences will lead to elevated creative self-efficacy. Since creative self-efficacy is related to creativity (Tierney

& Farmer, 2002, 2004), it constitutes an underlying motivational mechanism

through which PSD is related to creativity. Accordingly, we hypothesised

that:

Hypothesis 2b: Creative self-efficacy mediates the relationship between PSD and

creativity.

The Moderating Influence of Intrinsic Motivation

Intrinsic motivation relates to whether individuals are internally driven to

complete a task due to a personal interest in the task itself. Thus, intrinsic

motivation refers to the motivational state in which employees are attracted

to and energised by the task itself, rather than the external outcomes that

doing the task might yield (Deci & Ryan, 1985). Intrinsic motivation is

related to other stable personality traits, such as learning goal orientation

(Dweck & Leggett, 1988), mastery orientation (Kagan, 1972), and growth

need strength (GNS; Hackman & Oldham, 1975). However, intrinsic motivation is conceptually distinct from these individual differences. For some

tasks, individuals may have an orientation to learn, grow, and achieve (i.e.

high learning goal orientation, high mastery orientation, or high GNS), and

yet they may not enjoy working on the task (i.e. low intrinsic motivation)

(Shalley et al., 2009).

To date, to our knowledge no study has examined the moderating

influence of intrinsic motivation on the relationship between demanding job attributes (e.g. PSD) and individual outcomes. However, there is

strong evidence that when individuals are intrinsically as opposed to

extrinsically motivated they are more willing to tackle difficult tasks or

goals (cf. Meyer, Becker, & Vandenberghe, 2004), and persist at these (Vallerand, Pelletier, Blais, Briere, Senecal, & Vallieres, 1992; Ryan & Connell,

1989). Therefore, they are less likely to be affected by the failure and dif-

ficulties entailed by challenging tasks, e.g. PSD. We argue that the relationship between PSD and creative self-efficacy may be a function of an

individual’s intrinsic motivation: those with higher levels of intrinsic motivation will respond more favorably to PSD situations. Individuals high in

intrinsic motivation are more likely to accept difficult problems and be

persistent when they encounter difficulties and challenges in looking for

solutions.

Furthermore, individuals high in intrinsic motivation are likely to see

PSD in terms of the opportunities it presents for them to fully apply capabilities and develop better problem-solving skills. Such positive attitudes

towards PSD will help them achieve better learning outcomes and higher

62 ZHOU ET AL.


Association of Applied Psychology.creative self-efficacy. In contrast, low intrinsic motivation individuals may

feel threatened by PSD situations. Instead of trying to overcome problem

situations and develop problem-solving skills, individuals low in intrinsic

motivation are likely to avoid problems or not take advantage of opportunities to apply and develop their knowledge and skills. Consequently,

individuals low in intrinsic motivation are less likely to experience increased

creative self-efficacy. On the basis of these arguments, we hypothesised

that:

Hypothesis 3a: Intrinsic motivation will moderate the relationship between PSD

and creative self-efficacy such that the relationship between PSD and creative

self-efficacy will be stronger for individuals with high rather than low intrinsic

motivation.

If creative self-efficacy mediates the relationship between PSD and creativity, as predicted in Hypothesis 2b, and the impact of PSD on creative selfefficacy is dependent on intrinsic motivation, as predicted in Hypothesis 3a,

it is likely that intrinsic motivation may moderate the strength of the mediated relationship between PSD and creativity via creative self-efficacy, such

that the mediated relationship will be stronger for individuals high rather

than low in intrinsic motivation. This effect pattern is referred to as moderated mediation (Muller, Yzerbyt, & Judd, 2005; Preacher, Rucker, &

Hayes, 2007).

Hypothesis 3b: Intrinsic motivation will moderate the mediated effect of PSD on

creativity via creative self-efficacy such that the mediated effect will be stronger for

individuals with high rather than low intrinsic motivation.

METHOD

Sample and Procedure

Three issues were considered in the selection of participating organisations.

First, this study aims to examine creativity in an environment where the

requirements for creativity are not salient. This is different from creativity

studies that focus on R&D teams, where creativity is the expected outcome

(e.g. Amabile, Conti, Coon, Lazenby, & Herron, 1996; Scott & Bruce, 1994;

Tierney et al., 1999). Second, it is important to have participants from different functional groups. By so doing, a representative sample (of a general

work environment) can be achieved. Lastly, like many other creativity

studies, supervisor ratings were used to measure employee creativity. Therefore, it is important to identify supervisor–subordinate dyads, whereby the

supervisor is in an appropriate position to observe subordinates’ creative

performance. Three organisations located in the city of Wuxi, the People’s



Association of Applied Psychology.Republic of China satisfied the preceding requirements and were invited to

participate in this study. In meetings with Human Resources (HR) managers

of each of the companies, the first and second authors explained the aims of

the study and asked them to identify individual work units within the

company for possible participation in the research. We made it clear that we

were not focusing specifically on work units with high creative performance,

but instead examining job-related characteristics across all or most functional areas of the business. Units were selected in order to be representative

of the organisation as a whole.

Employees in the identified units were informed of this survey through

the HR department of each company before the questionnaires were distributed. A survey coordinator was assigned by each HR department to

help the first author distribute questionnaire packages to respondents.

Separate questionnaires were administered to subordinates and supervisors.

Subordinate questionnaires were distributed to 320 employees while

supervisor questionnaires were distributed to 60 immediate supervisors of

the subordinates. Employees completed a questionnaire that included

measures of PSD, creative self-efficacy, intrinsic motivation, and demographics variables. Separately, each supervisor was asked to rate the creativity of an average of five subordinates. A cover letter attached to each of

the questionnaires informed respondents of the purpose of the survey.

Respondents were assured of the confidentiality of their responses and their

personal ID (provided at the top right hand corner of the questionnaire)

would only be used to match their responses to the ratings provided by

their supervisors.

Completed and usable questionnaires from 270 supervisor–subordinate

dyads were received. This represented a response rate of 84 per cent for

subordinates and 90 per cent for supervisors. Of the 270 respondents, 66

per cent were male. Respondents reported an average age of 28.35 years

(SD = 5.25) and average job tenure of 2.93 years (SD = 3.03). In terms of

highest level of education achieved, 23 per cent (62) respondents had completed high school (12 years of education), 50 per cent (135) college degree

(15 years of education), 25.2 per cent (68) Bachelor’s degree (16 years of

education), and 1.9 per cent (5) Master’s degree (19 years of education). It

should be noted that in China a college degree, “Da Zhuan”, is a qualifi-

cation lower than a bachelor degree, “Ben Ke”. The duration of a college

degree is normally three years whereas that of a bachelor degree is normally

four years. Participants were from different functions of the companies:

administration and HR (88 respondents, 33.1%), production (78 respondents, 29.3%), finance/accounting department and quality control (57

respondents, 21.4%), logistics (20 respondents, 7.5%), and sales and marketing and others accounted for 23 respondents (8.7%). Four respondents

did not indicate their job function.

64 ZHOU ET AL.


Association of Applied Psychology.Measures

Following procedures suggested by Brislin (1980), the questionnaire was

developed originally in English and translated into Chinese. The Chinese

version of the questionnaire was back-translated into English. A third

person, an English native speaker, compared the original version with the

back-translation.

PSD. A five-item scale originally developed by Jackson et al. (1993) and

later improved and validated by Wall et al. (1995) was used to measure PSD.

Items include “To what extent are you required to deal with problems which

are difficult to solve?” “To what extent do you have to solve problems which

have no obvious correct answer?” “To what extent do you need to use your

knowledge of work processes to help prevent problems arising in your job?”

“To what extent do the problems you deal with require a thorough knowledge of the work process in your area?” and “To what extent do you come

across problems in your job you have not met before?” Response options

ranged from (1) “not at all” to (5) “a great deal”. The scale’s alpha reliability

is .76.

Creative Self-Efficacy. A three-item scale developed by Tierney and

Farmer (2002) was used to measure creative self-efficacy. A sample item is “I

have confidence in my ability to solve problems creatively”. Response

options ranged from (1) “strongly disagree” to (7) “strongly agree”. The

scale’s alpha reliability is .87.

Intrinsic Motivation. A 12-item scale originally developed by Vallerand

and his colleagues in an academic context (see Vallerand, 1997), and later

adapted by Van Yperen and Hagedoorn (2003) to a work context, was used

to measure intrinsic motivation. It represents three types of intrinsic motivation: intrinsic motivation to know (items 1–4), to accomplish things (items

5–8), and to experience stimulation (items 9–12). Respondents were asked,

“Why do you do this job?” Sample items are: “for the pleasure it gives me to

know more about my job” (intrinsic motivation to know); “because I feel a

lot of personal satisfaction while mastering certain difficult job skills” (intrinsic motivation to accomplish things); “for the excitement I feel when I am

really involved in my job” (intrinsic motivation to experience stimulation).

Response options ranged from (1) “strongly disagree” to (7) “strongly

agree”. We conducted both exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) to test: (1) whether a three-factor structure holds

for intrinsic motivation; and (2) whether the three factors attributed to one

factor. EFA illustrated that two items displayed cross-loading. They were

therefore deleted. The remaining 10 items loaded onto thee factors, respecPROBLEM-SOLVING DEMAND AND CREATIVITY 65


Association of Applied Psychology.tively, i.e. intrinsic motivation to know (three items), to accomplish things

(three items), and to experience stimulation (four items). The CFA results

confirmed that three first-order latent variables loaded onto one secondorder latent factor (c

2

= 97.39, df = 32, p < .001, CFI = .97, TLI = .96, RMSEA

= .08). Consequently, we combined the three subscales to create a composite

index for intrinsic motivation. The scale’s alpha reliability is .94.

Creativity. A 13-item scale developed by Zhou and George (2001) was

used to measure creativity. Supervisors rated the creative performance for

each of their subordinates who participated in the survey. A sample item is:

“This employee is a good source of creative ideas”. Response options ranged

from (1) “not at all” to (5) “to a great extent”. The scale’s alpha reliability is

.96. Since supervisors rated more than one employee, there was a risk that the

creativity rating scores received by individual employees were dependent on

rater identity. Following Dansereau and Yammarino (2000), we conducted

within and between analysis (WABA) (Dansereau, Alutto, & Yammarino,

1984) to test the assumption of independence of creativity ratings that each

supervisor provided for multiple subordinates. The E ratio (tests of practical

significance) for creativity was .96, less than 1.0, and therefore indicated that

the variation within groups was significantly greater than the variation

between groups. Furthermore, the corrected F-test (tests of statistical signifi-

cance) was statistically nonsignificant (1/F = .28). Together, these WABA

results supported the assumption of independence for creativity and the

appropriateness of conducting the analysis at the individual rather than the

group level.

Control Variables. We controlled for education level and job tenure,

because both reflect individual knowledge level (Tierney & Farmer, 2002),

which has been related to individual creativity (Amabile, 1983; Woodman,

Sawyer, & Griffin, 1993). Education level was measured on a 4-point scale

(1 = “high school and below”, 2 = “college degree”, 3 = “Bachelor’s degree”, 4

= “Master’s degree”). Job tenure was measured in years. Age has been associated with creativity (Amabile, 1983) and self-efficacy (Bandura, 1997). There

are gender differences in perceptions of capabilities to do a certain type of job.

For instance, compared with men, women generally perceived themselves to

be less efficacious in doing some types of jobs, such as scientific jobs (Matsui &

Tsukamoto, 1991). Consequently, we also controlled for age and gender. Two

dummy variables (org 1 and org 2) were created to control for the difference

in creativity receptivity or perceived expectation for creativity (Ford, 1996)

that might exist among the three organisations. Finally, we controlled for job

type. Following Tierney and Farmer (2002), we distinguished two job types:

production versus operations (including administration and HR, finance/

66 ZHOU ET AL.


Association of Applied Psychology.accounting, logistics, sales and marketing, and others). A dummy variable was

created for job type: production = 0 and operations = 1.

Data Analysis Procedures

We conducted CFA to test the distinctiveness of the variables. The overall

model chi-square measure (c

2

), the Tucker-Lewis index (TLI; Tucker &

Lewis, 1973), the comparative fit index (CFI; Bentler, 1990), and the root

mean square error of approximation (RMSEA; Browne & Cudeck, 1993)

were used as key indicators of overall model fit (Hoyle & Panter, 1995). For

TLI and CFI, a value of .90 is seen as a reasonable minimum for model

acceptance (Bentler & Bonett, 1980), whereas for RMSEA, a value of .08 or

less is indicative of a good model fit (Browne & Cudeck, 1993).

To test for mediation (Hypothesis 2b), we followed procedures suggested

by Baron and Kenny (1986). According to Baron and Kenny (1986), the

following conditions must be met to demonstrate a mediating effect: (1)

independent variable (PSD) must be related to mediator (creative selfefficacy); (2) independent variable (PSD) must be related to dependent variable (creativity); (3) mediator (creative self-efficacy) must be related to

dependent variable (creativity); and (4) independent variable (PSD) must

have no effect on dependent variable (creativity) when mediator (creative

self-efficacy) is held constant (full mediation) or should become significantly

smaller (partial mediation).

Hypothesis 3b proposes moderated mediation. To test for this possible

relation, we followed the procedures described by Muller et al. (2005),

Edwards and Lambert (2007), and Preacher et al. (2007). Specifically, we

examined four conditions: (1) significant effect of PSD on creativity; (2)

significant interaction effect between PSD and intrinsic motivation in predicting creative self-efficacy and creativity; (3) significant effect of creative

self-efficacy on creativity; and (4) indirect effect of PSD on creativity signifi-

cantly differs at high and low levels of intrinsic motivation.

RESULTS

We compared the fit of our hypothesised four-factor model to a number of

nested alternative models: (1) a three-factor model where two motivational

constructs, intrinsic motivation and creative self-efficacy were loaded on one

factor; (2) a two-factor model where all self-report variables, PSD, intrinsic

motivation, and creative self-efficacy were loaded on one factor; and (3) a

one-factor model where all variables loaded on one factor. The fit indices

indicate that our hypothesised four-factor model fit the data best (c

2

= 537.72,

df = 246, p < .001, TLI = .91, CFI = .93, RMSEA=.07), suggesting support for

the distinctiveness of the variables in this study.



Association of Applied Psychology.Descriptive statistics and zero-order correlations are presented in Table 1.

PSD was significantly and positively correlated with creativity (r = .21,

p < .01) and creative self-efficacy (r = .32, p < .01), while creative self-efficacy

was significantly and positively correlated with creativity (r = .20, p < .01),

indicating preliminary support for the relationships suggested in Hypotheses

1, 2a, and 2b.

To test the hypotheses, we conducted hierarchical regression analysis.

Table 2 shows the results for the tests of mediation as suggested in Hypotheses 1, 2a, and 2b. The results supported Hypotheses 1, 2a, and 2b as follows:

(1) PSD was significantly related to creative self-efficacy in model 1. (2) PSD

was significantly related to creativity in model 2. (3) The regression coefficient

for creative self-efficacy was significantly related to creativity and the coeffi-

cient for PSD became non-significant (b = .13, p = .06) in Model 3, indicating

the mediating effect of creative self-efficacy on the relationships between PSD

and creativity (Baron & Kenny, 1986). We also conducted a Sobel test (Sobel,

1982) with “Calculation for the Sobel Test” provided by Preacher and

Leonardelli (2004). The results revealed that PSD had significant indirect

effects on creativity through creative self-efficacy (Sobel = 2.08, p < .05),

providing support for Hypothesis 2b.

Table 3 displays the regression results for testing the moderating effect of

intrinsic motivation on the relationship between PSD and creative selfefficacy (Hypothesis 3a) and the mediated effect of PSD on creativity via

creative self-efficacy (Hypothesis 3b). Following Aiken and West (1991), we

centered all the variables used in the interaction analysis. In Model 4 creative

self-efficacy was regressed on the control variables, PSD, intrinsic motivation, and the interaction term of PSD and intrinsic motivation. In Model 5,

creativity was regressed on the control variables, PSD, intrinsic motivation,

and the interaction term of PSD and intrinsic motivation. In Model 6,

creativity was regressed on the control variables, PSD, intrinsic motivation,

creative self-efficacy, the interaction term of PSD and intrinsic motivation,

and the interaction term of intrinsic motivation and creative self-efficacy

(Muller et al., 2005). In support of Hypothesis 3a, the change of the multiple

squared correlation coefficient (DR

2

) for the interaction term of PSD and

intrinsic motivation was statistically significant, explaining a significant

amount of variance in creative self-efficacy (DR

2

= .02, p < .05). To interpret

the nature of this interaction, we calculated regression equations for the

relationship between PSD and creative self-efficacy at the high and low levels

of intrinsic motivation. Following Cohen and Cohen (1983) we defined the

high and low values as plus and minus one standard deviation from the mean.

Figure 2 shows that the form of the interaction was as predicted in that the

relationship between PSD and creative self-efficacy was stronger for individuals with high intrinsic motivation. For those with low intrinsic motivation, the link between PSD and creative self-efficacy was not as significant. A

68 ZHOU ET AL.


Association of Applied Psychology.TABLE 1

Means, Standard Deviations, and Correlations

a Mean SD 1 2 3 4 5 6 7 8 9 10 11

1 Org1

a

– – – –

2 Org2

a

– – – -.81** –

3 Job type

b

– – – -.82** .65** –

4 Gender

c

– – – .24** -.27** -.29** –

5 Education – – – -.36** .17** .45** -.20** –

6 Age – 28.35 5.25 .02 -.04 .05 -.08 .01 –

7 Job tenure – 2.93 3.03 .06 -.03 -.04 -.13* -.04 .41** –

8 PSD .72 3.03 0.60 -.14* .04 .09 .20** .16* .02 .06 –

9 Creative self-efficacy .87 5.00 1.02 .16* -.15* -.08 .19** .08 .03 .01 .32** –

10 Intrinsic motivation .94 5.47 1.12 .11 -.11 -.09 .09 -.08 .01 -.15** .22** .28** –

11 Employee creativity .96 2.97 0.66 .07 -.13 -.02 .06 .04 .12 .19** .21** .20** .01 –

a

Dummy variable

b

Production = 0, Operations = 1

c

Female = 0, Male = 1

* p < .05; ** p < .01.



Association of Applied Psychology.further simple slope analysis revealed that when intrinsic motivation was

high (i.e. one standard deviation above the mean), the slope was significant

and the relationship between PSD and creative self-efficacy was positive

(simple slope = .64, p < .001). However, when intrinsic motivation was

low (i.e. one standard deviation below the mean), the slope was insignificant

(simple slope = .17, p = ns), indicating that the relationship between PSD

and creative self-efficacy became insignificant when intrinsic motivation

was low.

To test Hypothesis 3b, i.e. moderated mediation effects, we examined

four conditions (described in the data analysis section). PSD was positively

related to creativity in Model 5 (b = .20, p < .05), showing an overall effect

of PSD on creativity, demonstrating that condition one was present. As

described in the results previously, the interaction effect between PSD and

intrinsic motivation was positively related to creative self-efficacy (Model

Problem solving

Documents

scales alpha

item scale

intrinsic

low intrinsic

job design

psd andintrinsic

intrinsic

job requires