apllication methods
shauna101
nonlineareffectsofjobcomplexityandautonomyonthejobsatisfaction...pdf
The Nonlinear Effects of Job Complexity and Autonomy on Job Satisfaction, Turnover, and Psychological Well-Being
Greg A. Chung-Yan University of Windsor
This study examines the interactive relationship between job complexity and job autonomy on job satisfaction, turnover intentions, and psychological well-being. It was hypothesized that the positive or motivating effects of job complexity are only realized when workers are given enough autonomy to effectively meet the challenges of complex jobs. Results show that not only do job complexity and job autonomy interact, but that the relationships to the outcome variables are curvilinear in form. Job complexity is shown to be both a motivator and a stressor when job autonomy is low. However, the most beneficial effects of job complexity occur when it is matched by a high level of job autonomy. Theoretical and practical implications are discussed.
Keywords: work stress, job complexity, autonomy, job characteristics, work design
Characterizing the nature of work is an ongoing challenge for work researchers and practitioners; an enterprise made more difficult by the accelerated pace of change in the workplace. These changes are diverse: new technologies, new work processes, new sociotechnical demands, and limited resources are just some of the challenges affecting many industries (Howard, 1995). The psychological and physical ef- fects of the cognitive and social demands in the workplace have long been a topic of organizational research and theorizing: work redesign, job enrich- ment (e.g., Hackman & Lawler, 1971; Hackman & Oldham, 1980; Pearce & Dunham, 1976), and work stress (e.g., Karasek, 1979) are just some of the research areas focused on the motivational aspects of work and on matching the work environment to the needs of workers. Despite this research attention, more research is required concentrating on the inter- actions between elements of work characteristics and their effects on workers. This would, in part, help to reconcile inconsistent research findings and lead to a fuller understanding of the impact of job elements on workers (e.g., When is a job element a motivating
feature of work and when it is an aversive feature of work?). It might also help to explain why some workplace redesign initiatives fail despite the best practices suggested by the accumulated research ev- idence.
Job complexity is a particularly important job characteristic to investigate because of the growing complexity of jobs in “knowledge” societies and the attendant de-emphasis of manufacturing jobs. The United States, as an example, has seen a shift toward service-centered jobs, which emphasize cognitive and interpersonal skills, and away from manual labor jobs (Howard, 1995). Nevertheless, other job charac- teristics and resources, such as job autonomy, must also be available to workers for them to effectively deal with complex work. Drawing upon both work design and work stress theories, this study investi- gates the synergistic relationship between job com- plexity and job autonomy on worker attitudes, behav- iors, and well-being.
Job Complexity
Central to many definitions of job complexity is that complex jobs are mentally challenging and re- quire the worker to use a number of complex skills (e.g., Campbell, 1988; Campbell & Gingrich, 1986; Morgeson & Humphrey, 2006; Wood, 1986). They are also characterized by ambiguity, difficulty, and a lack of structure. Such jobs require novel approaches to problems or the determination of optimal solutions among many possible approaches (Campbell, 1988; Latham & Yukl, 1975; Man & Lam, 2003; March & Simon, 1958; Terborg & Miller, 1978). Simple jobs,
Greg A. Chung-Yan, Department of Psychology, Univer- sity of Windsor.
This research was supported by a University of Windsor Humanities and Social Sciences Research Grant. I thank Aaron C. H. Schat, Lori Francis, and Catherine T. Kwantes for their comments on an earlier version of this article. I also thank Andrea M. Butler for assisting with data collection.
Correspondence concerning this article should be ad- dressed to Greg A. Chung-Yan, Department of Psychology, University of Windsor, 401 Sunset Avenue, Windsor, On- tario, Canada N9B3P4. E-mail: [email protected]
Journal of Occupational Health Psychology 2010, Vol. 15, No. 3, 237–251
© 2010 American Psychological Association 1076-8998/10/$12.00 DOI: 10.1037/a0019823
237
T hi
s do
cu m
en t i
s co
py ri
gh te
d by
th e
A m
er ic
an P
sy ch
ol og
ic al
A ss
oc ia
tio n
or o
ne o
f i ts
a lli
ed p
ub lis
he rs
. T
hi s
ar tic
le is
in te
nd ed
s ol
el y
fo r t
he p
er so
na l u
se o
f t he
in di
vi du
al u
se r a
nd is
n ot
to b
e di
ss em
in at
ed b
ro ad
ly .
by contrast, can be completed by adhering to standard operating procedures and do not require the alteration of existing work methods.
By and large, job complexity is thought to be a positive aspect of work— often equated with job en- richment (Pearce & Dunham, 1976)—and research is supportive of this contention. Job complexity is pos- itively related to well-being and job-related attitudes (for reviews, see Ilgen & Hollenbeck, 1991; Morge- son & Campion, 2003). Specifically, job complexity positively predicts job satisfaction and affective com- mitment and negatively predicts intentions to quit (Grebner et al., 2003). Simple jobs are also related to poorer psychological health and workers are more likely to cognitively disengage from work (Clegg & Wall, 1990).
In terms of the psychological reasons why job complexity is related to positive work outcomes, one of the most influential theories to guide motivational approaches to job characteristics research is the Job Characteristics Model (JCM) of work motivation (Hackman & Oldham, 1975, 1980). According to the JCM, jobs that have specific core dimensions, or characteristics, lead to higher levels of work motiva- tion by creating critical psychological states in work- ers. The five job characteristics are skill variety (the extent to which a job requires the use of different skills); task identity (the extent to which a job in- volves completing an entire, identifiable piece of work); task significance (the extent to which a job substantially impacts the lives or work of other peo- ple); autonomy (the extent to which a job allows employees the discretion to schedule their work and determine the procedures used to carry out the work); and job feedback (the extent to which the job itself results in employees obtaining information about the effectiveness of their performance; Hackman & Old- ham, 1980). These core dimensions in turn create the three critical psychological states: experienced mean- ingfulness of the work (the extent to which workers think their job is valuable and worthwhile); experi- enced responsibility for the work outcomes (the ex- tent to which workers feel personally accountable for work results); and knowledge of the results of the work activities (the extent to which workers under- stand how effectively they are performing their job). Collectively, the five core dimensions comprise an index of job complexity (also referred to as job scope by some researchers) and meta-analytic findings show associations with a number of attitudinal and behavioral outcomes, such as internal work motiva- tion, organizational commitment, job satisfaction, job
involvement, and job performance (Humphrey, Nahr- gang, & Morgeson, 2007).
Despite the generally positive findings between job complexity and various outcomes, there is also evi- dence that too much job complexity as well as too little job complexity can have a negative effect on workers. In keeping with activation theory (Hancock & Ganey, 2003; Scott, 1966), curvilinear relation- ships between job characteristics and performance have been theorized and investigated. Activation the- ory posits an inverted U-shaped relationship between arousal and performance, such that there is an opti- mal level of activation or stimulation that aids per- formance. Too little arousal results in boredom and too much leads to mental overload (Gaillard, 1993; Gardner, 1986). Thus, although complex jobs should provide greater stimulation to the worker, staving off boredom, too much complexity could result in too much activation, negatively impacting performance.
Warr (e.g., 1990, 1994)—with his Vitamin Mod- el—also posited that the relationships between job characteristics and well-being are curvilinear, similar to the relationship between vitamins and their effect on an individual’s health and well-being. Specifi- cally, although increasing one’s vitamin intake might be beneficial to the health of individuals, these ben- efits taper off and additional vitamin intake either does not result in further benefit to the individual (Constant Effect; CE) or causes the individual harm (Additional Decrement; AD).
Champoux (1980), seeking to extend the JCM to include curvilinear relationships between job com- plexity and psychological responses to the job, found an inverted-U relationship between job complexity (a summation of the five core dimensions) and general satisfaction, internal work motivation, and growth satisfaction. Work stress studies have similarly found that job complexity is curvilinearly related to impor- tant work-related outcomes. Job complexity has been found to be related to the strain variables of emo- tional exhaustion and job-related anxiety (de Jonge & Schaufeli, 1998; Xie & Johns, 1995), such that work- ers in jobs with moderate amounts of complexity have lower levels of strain as compared to those in jobs with high and low amounts of complexity.
Job Complexity and Autonomy: Complementary Job Characteristics
Given the unstructured nature of complex jobs, they require workers to exercise judgment, decision- making, creativity, and other discretionary behaviors.
238 CHUNG-YAN
T hi
s do
cu m
en t i
s co
py ri
gh te
d by
th e
A m
er ic
an P
sy ch
ol og
ic al
A ss
oc ia
tio n
or o
ne o
f i ts
a lli
ed p
ub lis
he rs
. T
hi s
ar tic
le is
in te
nd ed
s ol
el y
fo r t
he p
er so
na l u
se o
f t he
in di
vi du
al u
se r a
nd is
n ot
to b
e di
ss em
in at
ed b
ro ad
ly .
Frese and Zapf (1994) argue that those with discre- tion and control can more effectively resolve prob- lems because they have the freedom to choose strat- egies to deal with the situation. Conversely, simple jobs with well-defined, basic tasks do not require the exercise of discretion.
Job autonomy is a widely studied feature of work. Within the JCM, it refers to the “degree to which the job provides substantial freedom, independence, and discretion to the employee in scheduling the work and in determining the procedures to be used in carrying it out.” (Hackman & Oldham, 1975, p.162). This definition is similar to that used by other re- searchers who alternatively refer to autonomy as con- trol and decision latitude (see Ganster & Fusilier, 1989, for a review).
Like job complexity, job autonomy is positively associated with important job attitudes, organiza- tional behaviors, and health. Spector’s (1986) meta- analytic study found that autonomy was related to higher job satisfaction, organizational commitment, job involvement, job performance, and motivation; and lower physical and somatic symptoms, emotional distress, role conflict and ambiguity, absenteeism, turnover intentions, and turnover.
As an individual variable, autonomy is frequently studied under the job demands-control (DC) model (Karasek, 1979), a theoretical framework that de- scribes the effects of job characteristics on worker health and well-being. Although both demands and control have been inconsistently operationalized (see de Jonge & Kompier, 1997; Perrewé & Ganster, 1989; Wall, Jackson, Mullarkey, & Parker, 1996 for critiques), job control usually refers to how much discretion workers have over work processes and scheduling and is synonymous with job autonomy; job demands usually refer to the amount of work workers have to perform and the pace at which they must perform it. According to the DC model, the amount of job demands and job autonomy people have in their work predicts job-related strain. Specif- ically, job strain increases with higher job demands and lower job control.
Similar to matching job complexity with auton- omy, implied by the DC model is that workers must have high control in order to deal with high work demands otherwise they will not have the resources necessary to resolve problems. Following this premise, investigations of the DC model often test for an interaction between job demands and control. Al- though studies have shown robust main effect find- ings, supporting the conclusion that high job de- mands and low job control are associated with
negative health effects and negative work-related outcomes (see de Lange, Taris, Kompier, Houtman, & Bongers, 2003; van der Doef & Maes, 1999, for reviews), the research findings investigating the de- mand-control interaction have been mixed. In a re- analysis of the studies reviewed by van der Doef and Maes (1999); Taris (2006) determined that of the 90 D � C tests, only 10% resulted in unqualified support and another 24% provided only partial support (i.e., they depended on third variables such as personality factors).
One possible reason to explain why interactions are inconsistently found is the aforementioned incon- sistency in how demands and control are conceptu- alized and operationalized. For example, Karasek’s (1979) original demands-decision latitude scale largely measured demands with workload items (i.e., amount and pace of work). Decision latitude was measured with items assessing not just autonomy, but also items that could be equated with other constructs like job complexity (e.g., “to what extent is high skill level required?”) and task variety (e.g., “to what extent is your work nonrepetitious?”). However, when measurement problems are minimized or avoided, support for the interaction between demands and control is stronger (Wall et al., 1996).
Another possible reason for the inconsistent inter- action findings is that demands and control are not well matched. Specifically, control is not always a relevant resource brought to bear on demands. A large amount of work that must be completed in a short time-frame—the most common conceptualiza- tion of demands—might require a lot of effort on the part of workers, but workers do not necessarily need control or discretion over the procedures used to complete the work. Routine work such as an assem- bly line job is one example. Control should, however, be more relevant for complex jobs, which lack struc- ture and do not have set procedures or a single way to accomplish the work. Also, although mentally chal- lenging, complex jobs do not necessarily have a high workload in terms of quantity, frequency, or time investment. As Perrow (1970) argues, routine jobs could benefit from established procedures, rules and regulations, whereas jobs composed of nonroutine, novel tasks benefit from a more flexible, nonbureau- cratic structure. Therefore, job complexity is concep- tually a better match to control/autonomy than work- load.
In terms of past research looking at the interaction between complexity and autonomy, Schaubroeck and colleagues (Schaubroeck & Merritt, 1997; Schau- broeck, Jones, & Xie, 2001) investigated the interac-
239JOB COMPLEXITY AND AUTONOMY
T hi
s do
cu m
en t i
s co
py ri
gh te
d by
th e
A m
er ic
an P
sy ch
ol og
ic al
A ss
oc ia
tio n
or o
ne o
f i ts
a lli
ed p
ub lis
he rs
. T
hi s
ar tic
le is
in te
nd ed
s ol
el y
fo r t
he p
er so
na l u
se o
f t he
in di
vi du
al u
se r a
nd is
n ot
to b
e di
ss em
in at
ed b
ro ad
ly .
tion between job complexity, control, and individual differences (i.e., explanatory style and job self- efficacy) in predicting physical health (i.e., blood pressure, upper respiratory illness, and immune func- tion). Significant three-way interactions were found between complexity, control, and individual differ- ences, such that high control, high self-efficacy, and low feelings of responsibility for negative job out- comes buffered the negative impact of high job com- plexity on physical health. Although not specifically hypothesized, their results also showed significant two-way interactions between job complexity and control when predicting blood pressure.
The Current Study
The current study is an investigation of the inter- action between job complexity and autonomy in pre- dicting both organizational and stress outcomes. Al- though drawing on the basic rationale behind the theorized interaction between demands and control from the DC model, it is argued that job complexity is a better match to autonomy than workload, which is how demand is commonly assessed.
Work design research often measures job com- plexity by using a composite of the five core dimen- sions of the JCM—including autonomy. This con- founds job autonomy with complexity. Collectively, the JCM dimensions of skill variety, task identity, task significance, and job feedback also do not map well onto the definition of job complexity— characterized by ambiguous, difficult, and unstruc- tured work. The current study conceptualizes and measures job complexity more narrowly than the JCM in keeping with commonly accepted definitions of job complexity (e.g., Morgeson & Humphrey, 2006).
Work design research that stems from the motiva- tional investigations of the structure of work often concentrate on job attitudes and work behaviors, whereas DC model research—a work stress model— concentrates mostly on worker well-being (Ganster & Fusilier, 1989). The current study draws upon both perspectives and assesses the impact of work design on all three types of outcomes (i.e., attitudes, behav- ior, and well-being). The specific outcomes chosen were job satisfaction, turnover intentions, and psy- chological well-being. These outcomes were chosen because they are commonly assessed in the work design and work stress research (e.g., Baltes, Bauer, Bajdo, & Parker, 2002; Campion, 1988; Fletcher & Jones, 1993; Fried & Ferris, 1987; Spector & Jex, 1991), allowing for comparisons to this research.
Actual turnover data were not collected, but turnover intentions have been shown to be strongly linked to actual turnover behavior (Hom & Griffeth, 1995).
Past research, already previously discussed, indi- cates that job complexity and autonomy should be positively related to job satisfaction and psychologi- cal well-being and negatively related to turnover in- tentions. Therefore, the following preliminary hy- potheses were made:
Hypothesis 1: Job complexity is positively re- lated to job satisfaction (1a), negatively related to turnover intentions (1b), and positively re- lated to psychological well-being (1c).
Hypothesis 2: Job autonomy is positively related to job satisfaction (2a), negatively related to turnover intentions (2b), and positively related to psychological well-being (2c).
Although past research supports the expectation of positive main effects for job complexity and job autonomy, the level of job complexity could be both a motivator and a stressor to workers, depending on the amount of discretion workers have over their work procedures. Job complexity can be engaging for workers assuming they are provided the resources to successfully complete their work. However, without sufficient resources job complexity becomes a road- block, because the work can no longer be completed.
Hypothesis 3: The relationships between job complexity and job satisfaction (3a), turnover intentions (3b), and psychological well-being (3c) are moderated by autonomy. The moder- ated relationships also include curvilinear rela- tionships.
Specifically, when there is sufficient job autonomy (i.e., moderate to high), job complexity should be positively related to satisfaction and psychological well-being and negatively related to turnover inten- tions. When there is not sufficient job autonomy (i.e., low to moderate), job complexity should be related to job satisfaction and emotional well-being in an in- verted-U pattern and to turnover intentions in a U pattern. In other words, up to a certain point, job com- plexity is expected to be motivating for people with even low levels of job autonomy, because some degree of stimulation is needed by workers. But once complex- ity begins to exceed the resources available to workers (i.e., low to moderate autonomy), it will change from a motivator to a stressor (i.e., an aversive stimulus).
240 CHUNG-YAN
T hi
s do
cu m
en t i
s co
py ri
gh te
d by
th e
A m
er ic
an P
sy ch
ol og
ic al
A ss
oc ia
tio n
or o
ne o
f i ts
a lli
ed p
ub lis
he rs
. T
hi s
ar tic
le is
in te
nd ed
s ol
el y
fo r t
he p
er so
na l u
se o
f t he
in di
vi du
al u
se r a
nd is
n ot
to b
e di
ss em
in at
ed b
ro ad
ly .
By contrast, workers with a high degree of autonomy will still be able to meet the challenges of a very complex job. And because of this increased match between job complexity and autonomy, not only should workers with a lot of job autonomy be buff- ered by the negative impact of highly complex jobs, the match should actually have a positive impact on job satisfaction, turnover, and psychological well- being.
Method
Participants and Procedures
Participants were solicited through the StudyRe- sponse Project— hosted by the School of Information Studies at Syracuse University (Syracuse, New York)—which facilitates online questionnaire re- search by distributing participation requests (see Stanton et al., n.d., for details). Piccolo and Colquitt (2006) and Inness, LeBlanc, and Barling (2008) are examples of studies that have used StudyResponse for sampling.
Participants were given a chance to win one of 12 $51 gift certificates to the online retailer, Amazon. Three hundred and forty-two participants completed an online questionnaire. Eighty-three participants in- dicated they did not have a full-time job (e.g., held part-time jobs); their data were removed from the study. The final sample consisted of 259 people (125 men, 132 women, 2 did not identify their sex). The majority of participants were either Caucasian (n � 148) or Asians/Pacific Islanders (n � 55). Ages ranged from 19 to 65 years (M � 38.3, SD � 10.9) and job tenure ranged from 0 to 38 years (M � 7.19, SD � 7.46). Due to an error in the online survey, job tenure was not recorded for 52 respondents. All par- ticipants had a full-time job and were asked to self- classify their job according to the occupational area it belonged to and its training and educational require- ments. The jobs represented in the sample were spread across all occupational and educational/ training categories outlined in the National Occupa- tional Classification System (Human Resources De- velopment Canada, 1993).
Measures
Subscales from the Work Design Questionnaire (WDQ; Morgeson & Humphrey, 2006) were used to measure job complexity and autonomy. The WDQ is a more contemporary measure of job characteristics than the Job Diagnostic Survey (JDS; Hackman &
Oldham, 1980) and distinguishes between complex- ity and autonomy in keeping with the purpose of this study.
Job complexity. Job complexity was assessed with the 4-item, job complexity subscale of the WDQ. The items were scored on a 5-point strongly agree to strongly disagree scale. Example item: “The job requires that I only do one task or activity at a time” (reverse scored).
Autonomy. Job autonomy was assessed with a combination of three autonomy subscales of the WDQ: work scheduling autonomy (3 items), deci- sion-making autonomy (3 items), and work methods autonomy (3 items). The items were scored on a 5-point strongly agree to strongly disagree scale. Example items from each subscale respectively: “The job allows me to plan how I do my work; the job allows me to make a lot of decisions on my own; the job gives me considerable opportunity for indepen- dence and freedom in how I do the work.”
Job satisfaction. Job satisfaction was assessed with the 3-item Overall Job Satisfaction scale (Cam- mann, Fichman, Jenkins, & Klesh, 1983) and recent meta-analytic findings (Bowling & Hammond, 2008) support its reliability and validity. Each item was rated on a 7-point strongly disagree to strongly agree scale. Example item: “In general, I like work- ing here.”
Turnover intentions. Turnover intentions were measured with the 5-item Turnover Cognition scale (Bozeman & Perrewé, 2001). Each item was rated on a 5-point strongly disagree to strongly agree scale. Example item: “I will probably look for a new job in the near future.”
Psychological well-being. The General Health Questionnaire (GHQ-12) is a widely used measure of psychological well-being and mental health in work- related research, demonstrating utility in work set- tings (e.g., Jackson, Stafford, Banks, & Warr, 1983; Wall, Kemp, Jackson, & Clegg, 1986). Questions relate to disturbances such as worry and depression. Each item is measured on a 7-point not at all to all of the time scale. Higher mean scores indicate greater psychological well-being. Example item: “Over the past year, have you lost much sleep from worry?” (reverse coded).
Results
Descriptive statistics and correlations among the study variables are presented in Table 1.
Contrary to expectations, job complexity was not a significant predictor of any of the outcome variables.
241JOB COMPLEXITY AND AUTONOMY
T hi
s do
cu m
en t i
s co
py ri
gh te
d by
th e
A m
er ic
an P
sy ch
ol og
ic al
A ss
oc ia
tio n
or o
ne o
f i ts
a lli
ed p
ub lis
he rs
. T
hi s
ar tic
le is
in te
nd ed
s ol
el y
fo r t
he p
er so
na l u
se o
f t he
in di
vi du
al u
se r a
nd is
n ot
to b
e di
ss em
in at
ed b
ro ad
ly .
Thus, hypotheses 1a-c were not supported. Job au- tonomy, however, was significantly related to the outcomes in the anticipated directions: it was posi- tively related to job satisfaction (r � .39, p � .001) and psychological well-being (r � .33, p � .001), and negatively related to turnover intentions (r � �.21, p � .01). Hypotheses 2a-c were thus sup- ported.
Regression Analyses
Three moderated hierarchical multiple regressions (MHMRs) were used to test for curvilinear by curvi- linear interactions between job complexity and job autonomy on job satisfaction, turnover intentions, and psychological well-being. Mean substitution was used to account for missing data. Consistent with Cohen, Cohen, West, and Aiken (2003), the predictor variables were first mean centered. Product and squared terms were created from the centered vari- ables. The linear effects of job complexity and job autonomy were entered in the first step. The squared terms were entered in the second step (squared terms were entered in the step before the linear interaction term because of the possibility that curvilinear main effects may be confused for linear interactions; Cor- tina, 1993; Ganzach, 1997). The linear interaction terms were entered in the third step. The linear by curvilinear interaction terms were entered in the fourth step. The curvilinear by curvilinear interaction terms were entered in the final step.
Negative affect (NA) was not controlled for in the analyses. Some researchers argue that NA might bias self-reports of job-related variables and, in turn, in- flate correlations between stressors and strains (e.g., Brief, Burke, George, Robinson, & Webster, 1988;
Watson, Pennebaker, & Folger, 1986). However, Spector and colleagues have demonstrated that NA does not bias reports of job characteristics (Spector, Fox, & van Katwyk, 1999) and that partialing out negative affect might, in fact, remove important vari- ance associated with variables of interest (Spector, Zapf, Chen, & Frese, 2000). Although the study follows this latter view, for the interested reader, the analyses were also performed controlling for NA and the pattern of results were similar. NA was assessed by the international Positive and Negative Affect Schedule Short Form (Thompson, 2007).
The results of the MHMR analyses are presented in Tables 2, 3, and 4. Job autonomy— but not job complexity—was linearly related to job satisfaction (� � .39, p � .001), turnover (� � �.21, p � .01), and psychological well-being (� � .33, p � .001). In step 2 of all the MHMR analyses, neither of the squared terms for job complexity or job autonomy was significantly related to the outcome variables, indicating no curvilinear main effects. In step 3 of the MHMR analyses, none of the linear interaction terms were significant.
As predicted, in the fifth step, the interaction be- tween the squared components of job complexity and job autonomy (C2 � A2) was significantly related to job satisfaction (� � �.38, p � .05) explaining 2% of additional variance (p � .05). Although the C2 � A2 interaction was not significantly related to turn- over intentions or psychological well-being, in the fourth step of these analyses, the C2 � A interaction was significant for turnover intentions (� � �.33, p � .01) and the C � A2 interaction was significant for psychological well-being (� � �.31, p � .01). It should be noted, however, that the C2 � A interac- tion for psychological well-being approached statis-
Table 1 Descriptive Statistics and Correlations Between Variables
Variable M SD 1 2 3 4 5 6 7 8
1. Negative affect 2.01 0.67 .78 2. Job complexity 3.40 1.00 �.14� .88 3. Job autonomy 3.85 0.79 �.24��� .03 .94 4. Job complexity2a 1.00 1.16 .00 �.28��� .14� — 5. Job autonomy2a 0.63 1.08 .05 �.02 �.43��� .09 — 6. Job satisfaction 5.26 1.36 �.29��� .08 .39��� .04 �.20�� .85 7. Turnover intentions 2.61 1.04 .20�� �.02 �.22�� �.05 .07 �.69��� .86 8. Psychological well-being 5.07 1.02 �.55��� .05 .33��� .03 �.08 .52��� �.39��� .90
Note. N � 257, listwise deletion. Internal consistency coefficients (Cronbach Alpha) are on the diagonal. a Calculated from the centered variable. � p � .05. �� p � .01. ��� p � .001.
242 CHUNG-YAN
T hi
s do
cu m
en t i
s co
py ri
gh te
d by
th e
A m
er ic
an P
sy ch
ol og
ic al
A ss
oc ia
tio n
or o
ne o
f i ts
a lli
ed p
ub lis
he rs
. T
hi s
ar tic
le is
in te
nd ed
s ol
el y
fo r t
he p
er so
na l u
se o
f t he
in di
vi du
al u
se r a
nd is
n ot
to b
e di
ss em
in at
ed b
ro ad
ly .
tical significance (� � .20, p � .06). Step 4 of the turnover intentions and psychological well-being MHMR analyses explained 3% (p � .05) and 4% (p � .01) of additional variance, respectively.
Figures 1a, 2a, and 3a are three-dimensional sur- face plots of the full regression equations represented in Tables 2, 3, and 4 for job satisfaction, turnover intentions, and psychological well-being. To make
the interaction effects easier to interpret, they are also accompanied by cross-sectional plots at both high (�1 SD), the mean, and low (�1 SD) levels of job complexity and job autonomy.
The results are consistent with expectations of the interactive relationship between job complexity and job autonomy. As predicted, the association between job complexity and the outcomes is dependent on the
Table 2 Moderated Hierarchical Multiple Regression Analysis for Job Satisfaction
Variables entered
Steps
1 2 3 4 5
Step 1 Job complexity (C) .07 .07 .07 .12 .17�
Job autonomy (A) .39��� .37��� .37��� .20� .25��
Step 2 C2 .00 .00 �.07 .04 A2 �.04 �.04 �.04 .13
Step 3 C � A .01 .10 .11
Step 4 C � A2 �.15† �.37��
C2 � A .32�� .24�
Step 5 C2 � A2 �.38�
Overall R2 .15��� .16��� .16��� .19��� .21���
�R2 .00 .00 .03�� .02�
Note. The displayed coefficients are beta weights at each step. † p � .10. � p � .05. �� p � .01. ��� p � .001.
Table 3 Moderated Hierarchical Multiple Regression Analysis for Turnover Intentions
Variables entered
Steps
1 2 3 4 5
Step 1 Job complexity (C) �.02 �.02 �.02 �.07 �.07 Job autonomy (A) �.21�� �.22�� �.22�� �.04 �.05
Step 2 C2 �.02 �.02 .05 .04 A2 �.02 �.02 �.02 �.03
Step 3 C � A �.03 �.13 �.13
Step 4 C � A2 .15 .16 C2 � A �.33�� �.32��
Step 5 C2 � A2 .03
Overall R2 .05�� .05� .05� .08�� .08��
�R2 .00 .00 .03� .00
Note. The displayed coefficients are beta weights at each step. � p � .05. �� p � .01. ��� p � .001.
243JOB COMPLEXITY AND AUTONOMY
T hi
s do
cu m
en t i
s co
py ri
gh te
d by
th e
A m
er ic
an P
sy ch
ol og
ic al
A ss
oc ia
tio n
or o
ne o
f i ts
a lli
ed p
ub lis
he rs
. T
hi s
ar tic
le is
in te
nd ed
s ol
el y
fo r t
he p
er so
na l u
se o
f t he
in di
vi du
al u
se r a
nd is
n ot
to b
e di
ss em
in at
ed b
ro ad
ly .
level of job autonomy. The relationships are also nonlinear in nature. For low autonomy jobs, as job complexity rises, job satisfaction and psychological well-being increase while turnover intentions de- crease. This trend, however, reverses when job com- plexity becomes moderate to high: job satisfaction and psychological well-being decrease, while turn- over intentions increase. However, when job auton- omy is high, moderate to high job complexity is positively associated with job satisfaction and psy- chological well-being. These qualified relationships might partly explain why job complexity was not found to be linearly related to the outcomes (i.e., Hypotheses 1a-c). Unexpectedly, at a high level of autonomy, job complexity is related to turnover in- tentions in an inverted-U pattern, such that turnover intentions is higher at moderate levels of complexity than at low and high levels of complexity.
Finally, as can be seen in Figures 1c, 2c, and 3c, the worst outcomes are observed when job complex- ity is high and job autonomy is low. Hypotheses 3a-c are thus supported.
Discussion
Proposed in the current study was that job com- plexity and job autonomy are work characteristics that are synergistically related. The patterning of relationships with work attitudes, behaviors, and
worker well-being were also expected to be nonlin- early related. The expectations were borne out: job complexity and job autonomy significantly interacted to predict job satisfaction, turnover intentions, and psychological well-being. In addition, at some levels of job autonomy, job complexity was found to be curvilinearly related to the outcome variables. So too, at some levels of job complexity, job autonomy was curvilinearly related to the outcome variables.
Theoretical Implications
The results support and extend both motivational work design research and work stress research. From the work design perspective, job complexity is a positive feature of work that should be motivating for workers. From the work stress perspective, if job complexity is considered a stressor, it should be aversive (i.e., strain) to workers. The results of this study show that job complexity is not uniformly a motivator or a stressor, but shows features of both depending on the level of job autonomy. Even with low levels of job autonomy, job complexity is asso- ciated with increasing job satisfaction, decreasing turnover intentions, and increasing psychological well-being. However, when job complexity becomes high, the trend reverses and it becomes associated with decreasing job satisfaction, increasing turnover intentions, and decreasing psychological well-being.
Table 4 Moderated Hierarchical Multiple Regression Analysis for Psychological Well-Being
Variables entered
Steps
1 2 3 4 5
Step 1 Job complexity (C) .05 .05 .05 .18� .17�
Job autonomy (A) .33��� .36��� .36��� .27�� .26��
Step 2 C2 �.03 �.03 �.13† �.15†
A2 .08 .08 .07 .03 Step 3
C � A .00 �.03 �.04 Step 4
C � A2 �.31�� �.25†
C2 � A .20† .22†
Step 5 C2 � A2 .09
Overall R2 .11��� .11��� .11��� .16��� .26���
�R2 .01 .00 .04�� .00
Note. The displayed coefficients are beta weights at each step. † p � .10. � p � .05. �� p � .01. ��� p � .001.
244 CHUNG-YAN
T hi
s do
cu m
en t i
s co
py ri
gh te
d by
th e
A m
er ic
an P
sy ch
ol og
ic al
A ss
oc ia
tio n
or o
ne o
f i ts
a lli
ed p
ub lis
he rs
. T
hi s
ar tic
le is
in te
nd ed
s ol
el y
fo r t
he p
er so
na l u
se o
f t he
in di
vi du
al u
se r a
nd is
n ot
to b
e di
ss em
in at
ed b
ro ad
ly .
Only when job autonomy rises to match the increas- ing complexity of the work are these negative con- sequences offset.
Job autonomy, however, was found to be a positive feature of work and did not show a similar reversing of the relationships with the outcomes that was seen with job complexity. Nevertheless, in the presence of very high job complexity, there was a point at which increasing job autonomy did not see a positive impact on the outcomes and the influence of job autonomy plateaued (i.e., the associations with the outcomes were reduced). Overall, the worst outcomes (i.e., low
job satisfaction, high turnover intentions, low psy- chological well-being) occur when job complexity is high and job autonomy is low.
Beginning with the work enrichment movement, it has long been held that job complexity is a positive feature of work and this study supports this view. How- ever, this study also adds the qualification that other job characteristics must be considered to properly assess the influence of job complexity on workers. When several work characteristics are considered together, they are often done so in an additive fashion, rather than con- sidering that some characteristics might work in tandem
−2
−1
0
1
−2 −1
0 1
2
1
2
3
4
5
6
7
Job AutonomyJob Complexity
J o
b S
a ti
s fa
c ti
o n
A
−2 −1 0 1 22 1
2
3
4
5
6
7
Job Complexity
J o
b S
a ti
s fa
c ti
o n
High Mean Low
Job Autonomy
−2 −1 0 1 2 1
2
3
4
5
6
7
Job Autonomy
J o
b S
a ti
s fa
c ti
o n
High Mean Low
Job Complexity
B C
Figure 1. Interaction between job complexity and job autonomy on job satisfaction.
245JOB COMPLEXITY AND AUTONOMY
T hi
s do
cu m
en t i
s co
py ri
gh te
d by
th e
A m
er ic
an P
sy ch
ol og
ic al
A ss
oc ia
tio n
or o
ne o
f i ts
a lli
ed p
ub lis
he rs
. T
hi s
ar tic
le is
in te
nd ed
s ol
el y
fo r t
he p
er so
na l u
se o
f t he
in di
vi du
al u
se r a
nd is
n ot
to b
e di
ss em
in at
ed b
ro ad
ly .
and, therefore, the possibility that the benefits of job complexity are only fully realized in the presence of job autonomy is not assessed.
The nonlinear findings of this investigation, and specifically the finding that job complexity can be perceived by workers as both a positive and a nega- tive feature of work, has implications for research on challenge and hindrance stressors (e.g., Podsakoff, LePine, & LePine, 2007). According to this two- dimensional framework, stressors can have positive or negative effects on retention-related variables (e.g., job satisfaction, commitment, and turnover in- tentions) depending on whether they provide chal- lenges and opportunities for personal growth or
whether they present obstacles to personal growth and task accomplishment. This study indicates that there may be job characteristics or “stressors” that do not fit into either category of a challenge or hin- drance, but instead their influence depends on third variables such as other job characteristics or individ- ual differences.
As mentioned, it was unexpected that at a high level of autonomy, job complexity is related to turnover in- tentions in an inverted-U pattern, such that turnover intentions is higher at moderate levels of complexity than at low and high levels of complexity (see Figure 2b). At extremely high levels of autonomy (i.e., be- yond � 1 SD), a similar finding is found for job satis-
−2 −1 0 1 22 1
2
3
4
5
Job Complexity
T u
rn o
v e
r
High Mean Low
Job Autonomy
−2 −1 0 1 2 1
2
3
4
5
Job Autonomy
T u
rn o
v e
r
High Mean Low
Job Complexity
B C
−2 −1
0 1
22 −2
−1 0
1 2
1
2
3
4
55
Job AutonomyJob Complexity
T u
rn o
v e
r
A
Figure 2. Interaction between job complexity and job autonomy on turnover intentions.
246 CHUNG-YAN
T hi
s do
cu m
en t i
s co
py ri
gh te
d by
th e
A m
er ic
an P
sy ch
ol og
ic al
A ss
oc ia
tio n
or o
ne o
f i ts
a lli
ed p
ub lis
he rs
. T
hi s
ar tic
le is
in te
nd ed
s ol
el y
fo r t
he p
er so
na l u
se o
f t he
in di
vi du
al u
se r a
nd is
n ot
to b
e di
ss em
in at
ed b
ro ad
ly .
faction and emotional well-being. Overall, across the three outcome variables, the best outcomes occur when job complexity is low and job autonomy is high. It seems that not only is highly complex and autonomous work desirable by workers, but so too is low complex- ity, highly autonomous work. This again reinforces the conclusion that job autonomy is a desirable feature of work in its own right, even when the work is not challenging. Thus, although these findings do support the idea of matching job resources (i.e., high complexity, high job autonomy), it is clear that such a match is not
necessary for low job complexity. The matching of job resources is also likely to be far more complex as there are many more job elements than just these two. Job autonomy—according to the JCM—also match the in- ternal needs of workers in that they lead to a positive impact on critical psychological states.
Practical Implications
Given the numerous methods and considerations that could possibly go into performing the duties
−2
−1
0
1
2 −2 −1
0 1
2
1
2
3
4
5
6
7
Job Autonomy Job Complexity
P s y c h
o lo
g ic
a l W
e ll −
B e in
g
A
−2 −1 0 1 22 1
2
3
4
5
6
7
Job Complexity
P s y c h
o lo
g ic
a l W
e ll −
B e in
g
High Mean Low
Job Autonomy
−2 −1 0 1 2 1
2
3
4
5
6
7
Job Autonomy
P s
y c
h o
lo g
ic a
l W
e ll
− B
e in
g
High Mean Low
Job Complexity
B C
Figure 3. Interaction between job complexity and job autonomy on psychological well- being.
247JOB COMPLEXITY AND AUTONOMY
T hi
s do
cu m
en t i
s co
py ri
gh te
d by
th e
A m
er ic
an P
sy ch
ol og
ic al
A ss
oc ia
tio n
or o
ne o
f i ts
a lli
ed p
ub lis
he rs
. T
hi s
ar tic
le is
in te
nd ed
s ol
el y
fo r t
he p
er so
na l u
se o
f t he
in di
vi du
al u
se r a
nd is
n ot
to b
e di
ss em
in at
ed b
ro ad
ly .
inherent in complex jobs, a greater amount of auton- omy must be given to workers engaged in these types of jobs. If workers were required to seek approval for every action and decision they made, they would be inefficient and may even miss time-sensitive oppor- tunities. They would also feel constrained by the necessity to seek approval and this might negatively impact innovation as employees would seek to ac- complish their goals through conventional means in order to avoid delays. Thus, for organizations to achieve desired outcomes through work design initi- atives, they must consider how work characteris- tics— both those that can change and those that will remain in place—interact together to facilitate or hinder workers’ efforts.
Strengths and Limitations
The current study was designed to build upon past work design and job complexity research by also incorporating work stress theory. Borrowing from the theory underlying the DC model, this study demon- strates the importance of investigating the synergistic relationships among work characteristics and not just their additive influence on important work outcomes. Furthermore, interactions and nonlinear relationships are not often investigated simultaneously, possibly masking important findings or resulting in inconsis- tent findings of interactions and curvilinearity.
Job complexity was also more clearly defined and operationalized in this study. Measures of job com- plexity and job autonomy were evaluated separately rather than combined into an overall measure of job scope, and the measure used had superior psycho- metric properties to the JDS. This might also explain why there were no significant correlations between job complexity and the criterion variables even though past research on job scope would suggest that there should be. Furthermore, careful attention was paid to the operationalization of job complexity so as to provide a better theoretical link to job autonomy. Although autonomy has been studied a great deal in the context of the DC model, and the two approaches seem superficially similar, studies of the DC model predominantly operationalize job demands as work- load, which is fundamentally different from com- plexity. Thus, past research that looks at the interac- tion between demands and control cannot be used to assess interactions with complex work.
In terms of the study’s limitations, cross-sectional data were used and, therefore, conclusions about cau- sation cannot be made. However, the study was founded upon long-established theories and a large
body of suggestive empirical findings, which give more confidence that the causal linkages are correctly specified. Nevertheless, experimental and longitudi- nal research is necessary to establish causation.
The study sample was diverse, composed of people from different occupations and educational/training requirements. They also spanned the range of com- plexity and autonomy levels. A sample of this type is necessary to find the complex curvilinear effects pro- posed and found in this study. Nevertheless, the data were collected though self-report measures, raising the possibility that common method variance (Pod- sakoff, MacKenzie, Lee, & Podsakoff, 2003) may have affected the results. However, although com- mon method variance may artificially inflate bivariate correlations (cf. Spector & Brannick, 1995), they are unlikely to create artifactual interactions and may in fact attenuate true interactions (Evans, 1985). If any- thing, common method variance would have the more likely effect of suppressing the interactive find- ings in this study. Spector (2006) also argues that problems associated with common method variance are largely overstated.
Future Research
The current study concentrated on the features of the workplace rather than the psychological re- sponses to those features. However, work design theories like the JCM also focus on the psychological experience of work. As such, more attention needs to be paid to the underlying psychological mechanisms by which work design impacts work-related out- comes and worker well-being. Although theory de- velopment and research continue to progress in this area (e.g., Champoux, 1992; Grant, 2008; Pierce, Jussila, & Cummings, 2009; Renn & Vandenberg, 1995), this research should be extended to include the interactive effects of different work characteristics on psychological factors.
Job complexity and autonomy (referred to as con- trol) are also key components of Frese, Garst, and Fay’s (2007) theoretical model of personal initiative. This model is based on the concept of reciprocal determinism (Bandura, 1997) and posits that work characteristics should affect personal initiative through control orientation (i.e., the belief that one is in control of important work issues). So too, control orientation should affect work characteristics through personal initiative. Along similar lines, Meier, Sem- mer, Elfering, and Jacobshagen (2008) investigated control beliefs (i.e., internal locus of control and self-efficacy) and found that the buffering impact of
248 CHUNG-YAN
T hi
s do
cu m
en t i
s co
py ri
gh te
d by
th e
A m
er ic
an P
sy ch
ol og
ic al
A ss
oc ia
tio n
or o
ne o
f i ts
a lli
ed p
ub lis
he rs
. T
hi s
ar tic
le is
in te
nd ed
s ol
el y
fo r t
he p
er so
na l u
se o
f t he
in di
vi du
al u
se r a
nd is
n ot
to b
e di
ss em
in at
ed b
ro ad
ly .
job control on the stressor-strain relationship de- pended on the level of one’s control beliefs. Taken together, it is likely that job control is perceived differently depending on individuals’ perceptions of themselves and their perceptions of their capacity to exercise influence. This, in turn, could affect the extent to which individuals make use of the job control they have. Thus, the interaction between job complexity and autonomy could be further moder- ated by control beliefs and other individual differ- ences. Future studies that consider how individual differences affect the interactions between different job characteristics (and vice versa) would advance the literature in both work design and work stress (e.g., DC model).
As stated at the outset, the nature of work is continually changing. Today, these changes are oc- curring at a more rapid pace and work is becoming more complex in our knowledge society. Employees are required to accomplish more work with fewer resources. The need for workers who can and will adapt to their work environment is nothing new, and proper selection and retention of top talent will con- tinue to be a priority for organizations in the foresee- able future. However, employers will also need to attend to the structural, organizational and procedural features of work to ensure that unnecessary con- straints are not impeding important discretionary be- haviors workers must enact to adapt and thrive in their workplace.
References
Baltes, B. B., Bauer, C. C., Bajdo, L. M., & Parker, C. P. (2002). The use of multitrait-multimethod data for de- tecting nonlinear relationships: The case of psychological climate and job satisfaction. Journal of Business and Psychology, 17, 3–17.
Bandura, A. (1997). Self-efficacy: The exercise of control. New York: Freeman.
Bowling, N. A., & Hammond, G. D. (2008). A meta- analytic examination of the construct validity of the Michigan Organizational Assessment Questionnaire Job Satisfaction Subscale. Journal of Vocational Behavior, 73, 63–77.
Bozeman, D. P., & Perrewé, P. L. (2001). The effect of item content overlap on organizational commitment question- naire—turnover cognitions relationships. Journal of Ap- plied Psychology, 86, 161–173.
Brief, A. P., Burke, M. J., George, J. M., Robinson, B. S., & Webster, J. (1988). Should negative affectivity remain an unmeasured variable in the study of job stress? Journal of Applied Psychology, 73, 193–198.
Cammann, C., Fichman, M., Jenkins, D., & Klesh, J. (1983). Assessing the attitudes and perceptions of orga- nizational members. In S. Seashore, E. Lawler, P. Mirvis, & C. Cammann (Eds.), Assessing organizational change:
A guide to methods, measures and practices. New York: Wiley.
Campbell, D. J. (1988). Task complexity: A review and analysis. Academy of Management Review, 13, 40 –52.
Campbell, D. J., & Gingrich, K. F. (1986). The interactive effects of task complexity and participation on task per- formance: A field experiment. Organizational Behavior and Human Decision Processes, 38, 162–180.
Campion, M. A. (1988). Interdisciplinary approaches to job design: A constructive replication with extensions. Jour- nal of Applied Psychology, 73, 467– 481.
Champoux, J. E. (1980). A three sample test of some extensions to the job characteristics model of work mo- tivation. Academy of Management Journal, 23, 466 – 478.
Champoux, J. E. (1992). A multivariate analysis of curvi- linear relationships among job scope, work context sat- isfactions, and affective outcomes. Human Relations, 45, 87–111.
Clegg, C., & Wall, T. (1990). The relationship between simplified jobs and mental health: A replication study. Journal of Occupational Psychology, 63, 289 –295.
Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2003). Applied multiple regression/correlation analysis for the behavioral sciences (3rd ed.). Mahwah, NJ: Erlbaum.
Cortina, J. M. (1993). Interaction, nonlinearity, and multi- collinearity: Implications for multiple regression. Jour- nal of Management, 19, 915–922.
de Jonge, J., & Kompier, M. A. J. (1997). A critical review of the demand-control-support model from a work psy- chological perspective. International Journal of Stress Management, 4, 235–258.
de Jonge, J., & Schaufeli, W. B. (1998). Job characteristics and employee well-being: A test of Warr’s vitamin model in health care workers using structural equation modeling. Journal of Organizational Behavior, 19, 387– 407.
de Lange, A. H., Taris, T. W., Kompier, M. A. J., Houtman, I. L. D., & Bongers, P. M. (2003). The very best of the millennium: Longitudinal research and the demand- control (-support) model. Journal of Occupational Health Psychology, 8, 282–305.
Evans, M. G. (1985). A Monte Carlo study of the effects of correlated method variance in moderated multiple regres- sion analysis. Organizational Behavior and Human De- cision Processes, 36, 305–323.
Fletcher, B. C., & Jones, F. (1993). A refutation of Karasek’s demand-discretion model of occupational stress with a range of dependent measures. Journal of Organizational Behavior, 14, 319 –330.
Frese, M., Garst, H., & Fay, D. (2007). Making things happen: Reciprocal relationships between work charac- teristics and personal initiative in a four-wave longitudi- nal structural equation model. Journal of Applied Psy- chology, 92, 1084 –1102.
Frese, M., & Zapf, D. (1994). Action as the core of work psychology: A German approach. In H. C. Triandis, M. D. Dunnette, & J. M. Hough (Eds.), Handbook of industrial and organizational psychology (2nd ed., Vol. 4, pp. 271–340). Palo Alto, CA: Consulting Psycholo- gists Press.
Fried, Y., & Ferris, G. R. (1987). The validity of the job characteristics model: A review and meta-analysis. Per- sonnel Psychology, 40, 287–322.
249JOB COMPLEXITY AND AUTONOMY
T hi
s do
cu m
en t i
s co
py ri
gh te
d by
th e
A m
er ic
an P
sy ch
ol og
ic al
A ss
oc ia
tio n
or o
ne o
f i ts
a lli
ed p
ub lis
he rs
. T
hi s
ar tic
le is
in te
nd ed
s ol
el y
fo r t
he p
er so
na l u
se o
f t he
in di
vi du
al u
se r a
nd is
n ot
to b
e di
ss em
in at
ed b
ro ad
ly .
Gaillard, A. W. K. (1993). Comparing the concepts of mental load and stress. Ergonomics, 36, 991–1005.
Ganster, D. C., & Fusilier, M. R. (1989). Control in the workplace. In C. L. Cooper & I. T. Robertson (Eds.), International review of industrial and organizational psychology (Vol. 4, pp. 235–280). Chichester, England: Wiley.
Ganzach, Y. (1997). Misleading interaction and curvilinear terms. Psychological Methods, 2, 235–247.
Gardner, D. G. (1986). Activation and theory and task design: An empirical test of several new predictions. Journal of Applied Psychology, 71, 411– 418.
Grant, A. M. (2008). Designing jobs to do good: Dimen- sions and psychological consequences of prosocial job characteristics. The Journal of Positive Psychology, 3, 19 –39.
Grebner, S., Semmer, N. K., Lo Faso, L., Gut, S., Kälin, W., & Elfering, A. (2003). Working conditions, well-being, and job-related attitudes among call centre agents. Euro- pean Journal of Work and Organizational Psychology, 12, 341–365.
Hackman, J. R., & Lawler, E. E. (1971). Employee reac- tions to job characteristics. Journal of Applied Psychol- ogy, 55, 259 –286.
Hackman, J. R., & Oldham, G. R. (1975). Development of the Job Diagnostic Survey. Journal of Applied Psychol- ogy, 60, 159 –170.
Hackman, J. R., & Oldham, G. R. (1980). Work redesign. Reading, MA: Addison Wesley.
Hancock, P. A., & Ganey, H. C. N. (2003). From the inverted to the extended-U: The evolution of a law of psychology. Journal of Human Performance in Extreme Environments, 7, 5–14.
Hom, P. W., & Griffeth, R. W. (1995). Employee turnover. Cincinnati, OH: South-Western College Publishing.
Howard, A. (Ed.). (1995). Changing nature of work. San Francisco: Jossey-Bass.
Human Resources Development Canada. (1993). National occupational classification: Occupational descriptions. Ottawa, Canada: Canada Communication Group.
Humphrey, S. E., Nahrgang, J. D., & Morgeson, F. P. (2007). Integrating motivational, social, and contextual work design features: A meta-analytic summary and the- oretical extension of the work design literature. Journal of Applied Psychology, 92, 1332–1356.
Ilgen, D. R., & Hollenbeck, J. R. (1991). The structure of work: Job design and roles. In M. D. Dunnette & L. M. Hough (Eds.), Handbook of industrial and organiza- tional psychology (2nd ed., Vol. 2, pp. 165–207). Palo Alto, CA: Consulting Psychologists Press.
Inness, M., LeBlanc, M. M., & Barling, J. (2008). Psycho- logical predictors of supervisor-, peer-, subordinate-, and service-provider-targeted aggression, Journal of Applied Psychology, 93, 1401–1411.
Jackson, P. R., Stafford, E. M., Banks, M. H., & Warr, P. B. (1983). Unemployment and psychological distress in young people: The moderating role of employment com- mitment. Journal of Applied Psychology, 68, 525–535.
Karasek, R. (1979). Job decision latitude, job demands and mental strain: Implications for job redesign. Administra- tive Science Quarterly, 24, 285–308.
Latham, G., & Yukl, G. (1975). A review of research on the application of goal-setting in organizations. Academy of Applied Psychology, 67, 759 –768.
Man, D. C., & Lam, S. S. K. (2003). The effects of job complexity and autonomy on cohesiveness in collectiv- istic and individualistic work groups: A cross-cultural analysis. Journal of Organizational Behavior, 24, 979 – 1001.
March, J., & Simon, H. (1958). Organizations. New York: Wiley.
Meier, L. L., Semmer, N. K., Elfering, A., & Jacobshagen, N. (2008). The double meaning of control: Three-way interactions between internal resources, job control, and stressors at work. Journal of Occupational Health Psy- chology, 13, 244 –258.
Morgeson, F. P., & Campion, M. A. (2003). Work design. In W. C. Borman, D. R. Ilgen, & R. J. Klimoski (Eds.), Handbook of psychology: Industrial and organizational psychology (Vol. 12, pp. 423– 452). Hoboken, NJ: Wiley.
Morgeson, F. P., & Humphrey, S. E. (2006). The Work Design Questionnaire (WDQ): Developing and validat- ing a comprehensive measure for assessing job design and the nature of work. Journal of Applied Psychology, 91, 1321–1339.
Pearce, J. L., & Dunham, R. B. (1976). Task design: A literature review. Academy of Management Review, 1, 83–97.
Perrewé, P. L., & Ganster, D. C. (1989). The impact of job demands and behavioral control on experienced job stress. Journal of Organizational Behavior, 10, 213–229.
Perrow, C. (1970). Organizational analysis: A sociological point of view. In H. L. Tosi (Ed.), Theories of organiza- tion (pp. 128 –137). Chicago: St. Clair Press.
Piccolo, R. F., & Colquitt, J. A. (2006). Transformational leadership and job behaviors: The mediating role of core job characteristics. Academy of Management Journal, 49, 327–340.
Pierce, J. L., Jussila, I., & Cummings, A. (2009). Psycho- logical ownership within the job design context: Revision of the job characteristics model. Journal of Organiza- tional Behavior, 30, 477– 496.
Podsakoff, N. P., LePine, J. A., & LePine, M. A. (2007). Differential challenge stressor-hindrance stressor rela- tionships with job attitudes, turnover intentions, turnover, and withdrawal behavior: A meta-analysis. Journal of Applied Psychology, 92, 438 – 454.
Podsakoff, P. M., MacKenzie, S. B., Lee, J.-Y., & Podsa- koff, N. P. (2003). Common method biases in behavioral research: A critical review of the literature and recom- mended remedies. Journal of Applied Psychology, 88, 879 –903.
Renn, R. W., & Vandenberg, R. J. (1995). The critical psychological states: An underrepresented component of job characteristics model research. Journal of Manage- ment, 21, 279 –303.
Schaubroeck, J., Jones, J. R., & Xie, J. L. (2001). Individual differences in utilizing control to cope with job demands: Effects on susceptibility to infectious disease. Journal of Applied Psychology, 86, 265–278.
Schaubroeck, J., & Merritt, D. E. (1997). Divergent effects of job control on coping with work stressors: The key role of self-efficacy. Academy of Management Journal, 40, 738 –754.
Scott, W. E. (1966). Activation theory and task design. Organizational Behavior and Human Performance, 1, 3–30.
Spector, P. E. (1986). Perceived control by employees: A
250 CHUNG-YAN
T hi
s do
cu m
en t i
s co
py ri
gh te
d by
th e
A m
er ic
an P
sy ch
ol og
ic al
A ss
oc ia
tio n
or o
ne o
f i ts
a lli
ed p
ub lis
he rs
. T
hi s
ar tic
le is
in te
nd ed
s ol
el y
fo r t
he p
er so
na l u
se o
f t he
in di
vi du
al u
se r a
nd is
n ot
to b
e di
ss em
in at
ed b
ro ad
ly .
meta-analysis of studies concerning autonomy and par- ticipation at work. Human Relations, 39, 1005–1116.
Spector, P. E. (2006). Method variance in organizational research: Truth or urban legend? Organizational Re- search Methods, 9, 221–232.
Spector, P. E., & Brannick, M. T. (1995). The nature and effects of method variance in organizational research. In C. L. Cooper, & I. T. Robertson (Eds.), International review of industrial and organizational psychology (Vol. 10, pp. 249 –274). New York: Wiley.
Spector, P. E., Fox, S., & van Katwyk, P. T. (1999). The role of negative affectivity in employee reactions to job characteristics: Bias effect or substantive effect? Journal of Occupational and Organizational Psychology, 72, 205–218.
Spector, P. E., & Jex, S. M. (1991). Relations of job char- acteristics from multiple data sources with employee affect, absence, turnover intentions, and health. Journal of Applied Psychology, 76, 46 –53.
Spector, P. E., Zapf, D., Chen, P. Y., & Frese, M. (2000). Why negative affectivity should not be controlled in job stress research: Don’t throw out the baby with the bath water. Journal of Organizational Behavior, 21, 79 –95.
Stanton, J. M., Weiss, E. W., Santuzzi, A., Kwiatkowske, A., Singh, S., Kulshrestha, A., & Edmunds, A. (Founders). (n.d.). The StudyResponse Project. Retrieved from Http:// studyresponse.syr.edu/studyresponse/index.htm
Taris, T. W. (2006). Bricks without clay: On urban myths in occupational health psychology. Work and Stress, 20, 99 –104.
Terborg, J., & Miller, H. (1978). Motivation, behavior and performance: A closer examination of goal-setting and monetary incentives. Journal of Applied Psychology, 63, 29 –39.
Thompson, E. R. (2007). Development and validation of an
internationally reliable short-form of the Positive and Negative Affect Schedule (PANAS). Journal of Cross- Cultural Psychology, 38, 227–242.
van der Doef, M. P., & Maes, S. (1999). The job demands- control (-support) model and psychological well-being: A review of 20 years of research. Work and Stress, 13, 87–114.
Wall, T. D., Jackson, P. R., Mullarkey, S., & Parker, S. K. (1996). The demands-control model of job strain: A more specific test. Journal of Occupational and Organiza- tional Psychology, 69, 153–166.
Wall, T. D., Kemp, N. J., Jackson, P. R., & Clegg, C. W. (1986). Outcomes of autonomous workgroups: A long- term field experiment. Academy of Management Journal, 29, 280 –304.
Warr, P. B. (1990). Decision latitude, job demands, and employee well-being. Work and Stress, 4, 285–294.
Warr, P. B. (1994). A conceptual framework for the study of work and mental health. Work and Stress, 8, 84 –97.
Watson, D., Pennebaker, J. W., & Folger, R. (1986). Be- yond negative affectivity: Measuring stress and satisfac- tion in the work place. Journal of Organizational Behav- ior Management, 8, 141–157.
Wood, R. E. (1986). Task complexity: Definition of the construct. Organizational Behavior and Human Decision Processes, 37, 60 – 82.
Xie, J. L., & Johns, G. (1995). Job scope and stress: Can job scope be too high? Academy of Management Journal, 38, 1288 –1309.
Received June 29, 2009 Revision received February 17, 2010
Accepted March 24, 2010 y
251JOB COMPLEXITY AND AUTONOMY
T hi
s do
cu m
en t i
s co
py ri
gh te
d by
th e
A m
er ic
an P
sy ch
ol og
ic al
A ss
oc ia
tio n
or o
ne o
f i ts
a lli
ed p
ub lis
he rs
. T
hi s
ar tic
le is
in te
nd ed
s ol
el y
fo r t
he p
er so
na l u
se o
f t he
in di
vi du
al u
se r a
nd is
n ot
to b
e di
ss em
in at
ed b
ro ad
ly .
