J Bus Psychol (2012) 27:305–316

DOI 10.1007/s10869-011-9244-3

Job Autonomy as a Predictor of Mental Well-Being:

The Moderating Role of Quality-Competitive Environment
Rhokeun Park • Deborah Searcy

Published online: 28 August 2011

Ó Springer Science+Business Media, LLC 2011

Abstract market in particular should provide their employees with

Purpose This study examines the moderating role of more job autonomy.
quality-competitive environment on the relationships Originality/Value This is one of the first studies that
between job autonomy and employees’ mental well-being investigated the vertical fit between job autonomy and
and organizational commitment. It also investigates the organizational contexts while focusing on individual
mediating role of organizational commitment on the rela- employees’ outcomes (attitudes and mental well-being).
tionship between job autonomy and mental well-being. The results were obtained by data from a nationally rep-
Design/Methodology/Approach The proposed hypotheses resentative sample, allowing us to generalize the results.
were tested by hierarchical linear modeling using an Additionally, since the dataset was collected from multiple
archival dataset from the Workplace Employment Rela- sources, self-report and common-method biases are
tions Survey, which was conducted in Great Britain in 2004 minimized.
and 2005 (12,836 employees and 1,190 managers).
Findings This study found that quality-competitive Keywords Job autonomy
Organizational commitment

environment moderated the relationships between job Mental well-being
Quality-competitive environment

autonomy and mental well-being and between job auton- Mediated moderation
Exchange theory

omy and organizational commitment. That is, job auton- Demand-control model
omy was more strongly related to mental well-being and
organizational commitment in more quality-competitive
organizations. The results also indicated that this modera- When employees are given job autonomy, they can more
tion was partially mediated by organizational commitment. effectively perform producing and servicing activities by
Implications Because job autonomy is related to more efficiently utilizing their knowledge, skills, and
employees’ mental well-being and organizational com- abilities. Job autonomy refers to the degree to which
mitment, organizations need to provide their employees employees are provided with the freedom and discretion to
with job autonomy. More importantly, because these determine how to conduct their job (Dodd and Ganster
positive relationships are stronger in quality-competitive 1996; Shirom et al. 2006), and often has positive rela-
companies, organizations in a highly quality-competitive tionships with employee attitudes and organizational per-
formance (Ahuja et al. 2007; Hackman and Oldham 1976;
Shirom et al. 2006). Research on job autonomy has taken
R. Park (&)
two different approaches: autonomous workgroup and job
College of Business, Kwangwoon University,
26 Kwangwoon-gil, Nowon-gu, Seoul 139-701, South Korea characteristics (Wall et al. 1986). The former is based on
e-mail: [email protected] the socio-technical systems approach to work design and
provides a team or group with autonomy. In contrast, the
D. Searcy
latter approach is based on the job characteristics model
Robert H. Smith School of Business, University of Maryland,
College Park, MD, USA and is oriented toward individuals (Hackman and Oldham
e-mail: [email protected] 1976; Kemp et al. 1983; Wall et al. 1986). Langfred (2000)

123
306 J Bus Psychol (2012) 27:305–316

termed the former as group autonomy and the latter as refers to the degree to which an organization competes on
individual autonomy. This article limits its scope to indi- the basis of the quality of products or services it provides
vidual autonomy, because the extent to which individuals (Kinicki and Kreitner 2008). According to recent scholar-
perceive such autonomy may differ even within the same ship, quality-based competition is one of the most impor-
workgroup. tant challenges which current companies should address
Extant research has shown that job autonomy is posi- (Kinicki and Kreitner 2008).
tively related to job satisfaction (Dodd and Ganster 1996; In addition, although mental well-being is an important
Schulz et al. 1995; Spector 1986), organizational commit- variable that has close associations with employee attitudes,
ment (Ahuja et al. 2007; Wall et al. 1986; Spector 1986), behavior, and productivity (Cooper and Cartwright 1996;
mental health (Bond et al. 2008; Bond and Flaxman 2006; Ganster and Schaubroeck 1991; Wright and Hobfoll 2004),
Vigoda-Gadot 2007), and performance (Dodd and Ganster there have been relatively few studies that have investigated
1996; Shirom et al. 2006), and is negatively related to the underlying mechanisms through which job autonomy
turnover intentions (Ahuja et al. 2007; Spector 1986). It affects mental well-being (an exception is the work by
seems, however, that the beneficial relationships of job Schulz et al. 1995). This study focuses on organizational
autonomy with such employee outcomes do not work in all commitment as the key mediating mechanism that links job
organizations. For instance, Parker (2003) found in a quasi- autonomy to mental well-being. Organizational commit-
experimental field study of an UK-based company that job ment has been shown to be negatively related to stress (e.g.,
autonomy did not have significant correlations with orga- Begley and Cazjka 1993; Nikolaou and Tsaousis 2002; Siu
nizational commitment, job anxiety, and job depression. 2002; Warr 1990). Therefore, the other objective of this
Likewise, Moore (2000) indicated in a study for technology article is to examine through what mechanisms job auton-
professionals that job autonomy did not have the significant omy is associated with employees’ mental well-being. The
relationships with work exhaustion and turnover intention. findings of this article provide managerial implications on
Furthermore, Eby et al. (1999) found via a meta-analysis what particular types of organizations should provide
that, while significant, the effect size of job autonomy on employees with more job autonomy, as well as how job
organizational commitment was small. autonomy affects employee outcomes.
These inconsistent findings suggest that the relationship
between job autonomy and employee outcomes may be
contingent on the environment in which organizations are Theory and Hypotheses
positioned. There have been, however, few empirical
studies investigating the characteristics of organizations in While there is no single definition or conceptualization of
which job autonomy is more effective in improving mental well-being (Kasl 1973; Strupp and Hadley 1977;
employee outcomes (for an exception, see Man and Lam Warr 1987), for the purposes of this study we follow the
2003). Although Dodd and Ganster (1996) found a positive conceptualization as set forth by Warr (1987). He proposed
interaction between job autonomy, skill variety, and feed- a multi-faceted definition of mental health that consists of
back on job satisfaction and job performance, skill variety, five major dimensions: affective well-being, competence,
and feedback are characteristics of the job itself. Thus, autonomy, aspiration, and integrated functioning. Warr
there is need to investigate the vertical fit between job (1987) furthered this analysis by dichotomizing mental
autonomy and organizational factors, in addition to the well-being into two elements: context-free and context-
horizontal fit within the core job characteristics, which has specific well-being. In our study we are only concerned
been studied previously. Vertical fit represents the con- with the context-specific mental well-being, that is, the
gruence of HR practices with the organization’s charac- mental well-being resulting from an employee’s work
teristics, while horizontal fit refers to the internal experiences.
consistency within the organization’s HR practices (Baird Karasek’s (1979) demand-control model proposes that
and Meshoulam 1988; Delery and Doty 1996). Therefore, workers with low decision latitude and high psychological
one main objective of this article is to explore the mod- workload demands face the greatest risk to physical and
erating role that competitive environment at the organiza- mental health from stress. Job decision latitude consists of
tion level plays in the relationships between job autonomy both the decision authority available to the worker (or job
and employee outcomes at the individual level. autonomy) and the ability to use skills on the job (or skill
In particular, this article argues that job autonomy may discretion). Karasek (1979) asserted that if employees must
be more effective in the quality-competitive environment suppress their desires because of low decision authority,
because companies in such environments need more input the unreleased energy may remain internally as mental
from their employees for high quality products or services strain. This model implies that job autonomy can reduce
(Youndt et al. 1996). A quality-competitive environment mental strain in the workplace by providing employees

123
J Bus Psychol (2012) 27:305–316 307

with the ability to manage stressful situations. Bond et al. which a company faces quality competition in a product or
(2008) found that job autonomy was effective in enhancing service market. Quality was ranked as one of the top two
employees’ mental well-being, especially for employees concerns (the other was customer service) in a survey of
who had higher levels of psychological flexibility. Ahuja managers from a range of countries (Kinicki and Kreitner
et al. (2007) reported that job autonomy was effective in 2008). Nonetheless, quality as a competitive environ-
reducing work exhaustion by providing IT road workers ment has received limited empirical scholarship (Sage and
with freedom and flexibility to manage their own work- Kalyan 2006).
loads. Likewise, Vigoda-Gadot (2007) found in a sample of Job autonomy granted to employees may have differen-
school teachers from Israel that job autonomy had negative tial impacts on employee outcomes depending on the degree
relationships with job stress and burnout. to which the organization competes on the basis of quality
In previous studies, job autonomy also has been found to of products or services. Organizations that compete on
be an essential predictor of organizational commitment quality emphasize improving the quality of their products or
(e.g., Ahuja et al. 2007; Wall et al. 1986; Spector 1986). services in order to differentiate themselves from the
Organizational commitment has emerged as central to the competitors (Neal et al. 2005). They have to compete by
study of work-related attitudes and behavior (Allen et al. offering better quality products than the competitors; this
2003; Allen and Meyer 1990; Eisenberger et al. 2001; forces the company to adopt the ‘‘differentiation’’ strategy
Mathieu and Zajac 1990; Rhoades et al. 2001). When a of high quality products and services, as defined by Porter
company grants employees job autonomy, employees may (1980). The companies adopting a differentiation strategy
consider these freedoms and discretions as indications that have to distinguish their products or services from their
the company respects them and values their inputs. In other competitors by better quality and continuous innovation
words, discretionary actions such as granting more job (Neal et al. 2005). Skill acquisition and development are
autonomy can be interpreted as organizational support by essential to a successful quality strategy (Neal et al. 2005;
employees (Rhoades and Eisenberger 2002; Eisenberger Youndt et al. 1996). The creation of high quality products
et al. 1999). Based on the principle of reciprocity, social requires that workers are highly skilled in product innova-
exchange theory posits that people reciprocate commen- tions, and highly skilled workers need high discretion in
surately with what they have received from the other party their jobs. Therefore, companies in highly quality-compet-
(Brockner et al. 2004). This theory suggests that employees itive markets need to provide employees with more job
feel an obligation to reciprocate the support they receive autonomy to take competitive advantage in terms of product
from the organization and make their best effort toward the or service quality. It is for this reason that job autonomy fits
organization’s goals (Park et al. 2010; Rhoades and well with organizational climates emphasizing high quality.
Eisenberger 2002; Wayne et al. 1997). Thus, job autonomy Osterman (1994) found that employee involvement prac-
may improve employees’ commitment in social exchange tices (teams, job rotation, quality circles, and total quality
relationships with the organization. Eby et al. (1999) noted management) were more frequently adopted in companies
in their meta-analysis that job autonomy was associated with a quality strategy. This finding implies that organiza-
with affective organizational commitment, although the tions need employees’ input to improve product quality.
effect size was small. Since product or service quality can be improved by
more employee inputs (Neal et al. 2005; Youndt et al.
A Moderating Role of Quality-Competitive 1996), companies in highly quality-competitive markets
Environment may have a competitive advantage by providing employees
with more job autonomy. Job autonomy can create a source
Scholars have examined the roles of individual differences of sustainable competitive advantage when it is adequately
and external forces in Karasek’s model. For example, Van aligned with a competitive strategy (Huselid 1995). On the
Yperen and Hagedoorn (2003) investigated the interaction other hand, even if employees in less quality-competitive
effect between job autonomy and job demand on fatigue companies are granted job autonomy, they are less likely to
and found that job autonomy reduced fatigue in highly have the opportunity to utilize job-related discretion and
demanding jobs. There have been, however, few studies authority because intense efforts to improve quality are less
that examined the moderating role of the organizational required in such companies.
context on the relationship between job autonomy and Job autonomy has been shown to be of greater necessity
employee attitudes and mental health. That is, while job for employees in more quality-competitive companies than
autonomy may be positively related to mental well-being employees in less quality-competitive companies (Neal
both directly and via organizational commitment, the et al. 2005; Youndt et al. 1996). Therefore, when job
relationships are likely to differ depending on the firm- autonomy is granted to them, the level of mental well-
specific context. One important dimension is the extent to being may be more greatly increased for employees in

123
308 J Bus Psychol (2012) 27:305–316

more quality-competitive companies than their counter- Tsaousis (2002) succinctly argued, ‘‘employees feel less
parts in less quality-competitive companies. In other distressed when they feel that their organization values them
words, employees in more quality-competitive companies and is committed to them, rather than the opposite (p. 337).’’
may reduce their mental strain by utilizing job autonomy Meyer and Herscovitch (2001) confirmed Warr’s (1990)
more flexibly in coping with quality improvement claim that positive emotions, which constitute the core of
demands. Likewise, when job autonomy is granted to affective commitment, exert a direct influence on mental
employees in quality-focused companies, they are more strain and well-being. A panel study by Begley and Cazjka
likely to appreciate it and thus be more committed to their (1993) and a quasi-experimental study by Parker (2003)
companies than their counterparts in less quality-competi- found that organizational commitment is a predictor of
tive companies. Das et al. (2000) found that companies mental well-being rather than the reverse. More recently,
facing strong quality competition depended much more on Galais and Moser (2009) found in a longitudinal study that
high involvement work practices. organizational commitment of temporary workers towards
Therefore, based on these arguments and previous the client organization affects their mental well-being.
findings, we propose that the quality-competitive environ- On the basis of the results of longitudinal and experi-
ment of an organization will moderate the relationships mental studies above, we establish organizational com-
between job autonomy and mental well-being and organi- mitment as a predictor of mental well-being, rather than as
zational commitment. a consequence. By combining these results with Hypothe-
sis 2, which specifies the interaction between job autonomy
Hypothesis 1 Quality-competitive environment will
and quality-competitive environments on organizational
moderate the relationship between job autonomy and
commitment, we hypothesize as follows:
mental well-being. That is, job autonomy will be more
strongly related to mental well-being in more quality- Hypothesis 3 Organizational commitment will mediate
competitive organizations. the interaction between job autonomy and quality compe-
tition on mental well-being.
Hypothesis 2 Quality-competitive environment will
moderate the relationship between job autonomy and Figure 1 presents a conceptual model based on the
organizational commitment. That is, job autonomy will be hypotheses of this study. We propose that job autonomy
more strongly related to organizational commitment in has stronger relationships with mental health and organi-
more quality-competitive organizations. zational commitment when it is provided to employees of
more quality-competitive companies. Organizational com-
mitment mediates the interaction between job autonomy
Organizational Commitment as a Mediator and quality competition on mental health. We test this
conceptual model using mediated moderation as suggested
Organizational commitment has been shown to be nega- by Muller et al. (2005).
tively related to job strain and to be positively related to
mental well-being (e.g., Begley and Cazjka 1993; Galais
and Moser 2009; Parker 2003). Employees committed to Method
organizational goals will feel less stress even in negative
situations, as compared with less committed employees. In Sample
other words, the more committed that employees are to
their company and the more loyalty they feel, the more To investigate the hypotheses of this article, we used the
easily they will accept stressful situations. Thus, more 2004 Workplace Employment Relations Survey (WERS),
committed employees will be more mentally healthy than
less committed employees. Level-2: Quality competition
Yet, there have been two approaches to the relationship
between organizational commitment and mental well-being: ---------------------------------------------------------------------------------------------------------
organizational commitment as a consequence of mental Level-1: Organizational commitment
well-being and organizational commitment as a predictor of H-1 H-2
mental well-being. Ahuja et al. (2007) argued that when
work exhaustion increased, organizational commitment of H-3

employees declined because they felt that the company did

not take care of them. Similarly, Hakanen et al. (2006) found Job autonomy Mental well-being
that burnout in workplaces led to a lower level of organi-
zational commitment. On the other hand, Nikolaou and Fig. 1 A conceptual-mediated moderation model

123
J Bus Psychol (2012) 27:305–316 309

which covered a broad range of human resource practices by Van Yperen and Snijders (2000). The response format
across almost every sector of the economy in Great Britain. for the present study consisted of 4-point scales ranging
The 2004 WERS data were collected from managers, from 1 (a lot) to 4 (none). The coefficient alpha of these
employee representatives, and employees in private and items was .84. These items were reversed and then aver-
public sector companies with 5 or more employees in 2004 aged into a composite score with higher scores indicating
and 2005 (Department of Trade and Industry 2005). that employees perceive greater job autonomy.
This article used both the interview from managers and
the survey from employees. Because the variables used in Quality-Competitive Environment
this article are based on multiple sources, self-report and
common-method biases could be minimized (Donaldson In our study, quality competition was assessed at the
and Grant-Vallone 2002). The managers were randomly establishment level by asking the managers the following:
selected among senior persons responsible for employment ‘‘To what extent the demand for your product or service
relations or personnel matters and interviewed face-to-face. depends upon offering better quality than your competi-
The employee participants were randomly selected up to 25 tors?’’ Although most methodologists recommend the use
individuals in each workplace and they were surveyed of multiple-item measures, single-item measures are
using a self-completion form. The employee survey was acceptable if they are concrete (Bergkvist and Rossiter
collected from 1,965 establishments, which was 85.6% of 2007; Rohland et al. 2004). This variable was measured by
the establishments in which managers were interviewed. A a 5-point scale from 1 (not depend at all) to 5 (depend
total of 2,295 managers and 22,451 employees were heavily). Higher scores indicate that companies were more
involved in the data collection. The overall response rate quality-competitive in product or service markets.
was 64% in the manager interview and 61% in the
employee survey (Department of Trade and Industry 2005).
The number of employees surveyed in each establishment Mental Well-Being
ranged from 1 to 25. Establishments where only one
employee was involved in the survey accounted for 0.19% This construct was measured at the individual level by
of the establishments in the sample (the least frequent three items, ‘‘Thinking of the past few weeks, how much of
mode), and establishments where 20 employees were the time has your job made you feel each of the following?
involved accounted for 7.22% of the sample (the most tense, worried, and uneasy.’’ These items are a shortened
frequent mode). The establishments with single employee- version of Warr’s (1990) anxiety-contentment measure,
respondents were excluded for a multilevel analysis. which is one of the key indicators of mental well-being.
Fifty-three percent of the respondents were male, and The response format in the present study consisted of
the majority of respondents (94%) were White. Their 5-point scales ranging from 1 (all of the time) to 5 (never),
average age was approximately 40 years and they earned so higher scores indicate better mental well-being. The
about £17,000 per year. Manufacturing companies coefficient alpha of these items was .84. Mental well-being
accounted for 18% of the sample, electric and transporta- was measured by averages of individual responses to the
tion companies 15%, sales companies 25%, business ser- three items.
vice companies 22%, and public service companies 20%.
Most of companies (97%) were in private sector, and the Organizational Commitment
average company employed 414 people (SD = 950).
This construct was measured at the individual level by three
Measures items, ‘‘To what extent do you agree or disagree with the
following statements about working here? I share many of
Perceived Job Autonomy the values of my organization; I feel loyal to my organi-
zation; and I am proud to tell people who I work for.’’ These
At the individual level, the predictor variable was the items are concerned with affective organizational commit-
extent to which employees perceive job autonomy over ment and are often used in studies using the Organizational
their work. This construct was measured by the average of Commitment Scale (e.g., Mowday et al. 1979). These items
individual responses to the four items, ‘‘In general, how were measured on 5-point scales ranging from 1 (strongly
much influence do you have over the following? what tasks agree) to 5 (strongly disagree). The coefficient alpha of
you do in your job; the pace at which you work; how you these items was .90. These items were reversed and aver-
do your work; and the order in which you carry out tasks.’’ aged into a composite score with higher scores indicating
These are similar to the items that were used and validated greater organizational commitment.

123
310 J Bus Psychol (2012) 27:305–316

Control Variables allowing us to partition the total variance in the two con-
structs into within- and between-establishment compo-
To enhance the generalizability of the results, we con- nents. The intraclass correlations, ICC(1), showed that
trolled for several variables likely to influence the two variances of 13.6% of organizational commitment and
dependent variables. The control variables at the individual 8.2% of mental well-being resided between establishments.
level were gender, race, age, and wages. Male employees These variances between establishments for the two con-
and racial majorities were coded 1, and 0 otherwise. Age structs were significant and not-trivial; therefore, multilevel
was coded as 9 categories (from less than 17 to 65 or analysis was justified.
more), and wages as 14 categories from £50 or less to £871 The hypotheses of the present study were tested by a
or more per week. Control variables at the establishment mediated moderation model, as proposed by Muller et al.
level were comprised of establishment size, industry, and (2005). They suggested four conditions to demonstrate
sector. Establishments were categorized into 4-size (from mediated moderation: (1) the overall effect of the inde-
less than 50 to 1,000 or more) and 5-industry groups on the pendent variable on the dependent variable should depend
basis of Standard Industrial Classification: manufacturing, on the moderator; (2) the effect of the independent vari-
electric and transportation, sales, business services, and able on the mediator should depend on the moderator;
public services. Private-sector establishments were coded 1 (3) the mediator should have an effect on the dependent
and public-sector establishments were coded 0. variable; and (4) the moderation of the residual direct
effect of the independent variable should be lower than
Analytic Strategy the moderation of the overall effect of the independent
variable.
We conducted a confirmatory factor analysis to evaluate the
construct validity of the three constructs (job autonomy,
organizational commitment, and mental well-being) using Results
LISREL 8.50, in which 1-factor, 2-factor, and 3-factor
models were tested. The 2-factor model consisted of job Table 1 displays the means and standard deviations of each
autonomy on one hand, and organizational commitment and variable and correlations between variables used in the
mental well-being on the other hand. The 3-factor model fit analyses, separately at the individual and establishment
the data better than the 1-factor and 2-factor models levels. Job autonomy has significant correlations with
(Dv(3)2 = 52,610, p \ .01; Dv(2)2 = 25,670, p \ .01, respec- outcome variables, mental well-being (r = .16, p \ .01),
tively). The goodness-of-fit indexes for the 3-factor model and organizational commitment (r = .32, p \ .01). Orga-
also indicated a good fit with the data (RMSEA = .05; nizational commitment is significantly related to mental
SRMR = .03, NFI = .98, CFI = .98, GFI = .99); the well-being (r = .25, p \ .01).
1-factor and the 2-factor models showed unacceptable fit Table 2 presents the results of HLM analyses to exam-
indexes. The 3-factor model showed evidence of convergent ine the mediated moderation hypotheses of this study.
validity, since all the indicators had reasonably large load- These equations are the slopes-as-outcomes models, where
ings on each corresponding factor. There is also evidence of cross-level interaction terms as well as the individual and
discriminant validity, since correlations among factors (.10 organizational levels of variables were included as
to .33) are only moderate (Kline 1998). predictors (Hofmann 1997; Raudenbush and Bryk 2002).
Since employees are nested within establishments, we Model 1 evaluates the moderating role of quality compe-
employed hierarchical linear modeling (HLM; Bryk and tition in the relationship between job autonomy and mental
Raudenbush 1992). When a dataset is multilevel-struc- well-being, Model 2 assesses the moderating role of quality
tured, as in this study, using ordinary least squares competition in the relationship between job autonomy and
regression models may lead to an overestimation of the organizational commitment, and Model 3 examines the
parameters. HLM can overcome this problem related to mediating role of organizational commitment.
multilevel data by simultaneously estimating equations for Applying the four conditions for the mediated modera-
both individual and establishment effects (Raudenbush and tion model, as suggested by Muller et al. (2005), Model 1
Bryk 2002). Following the recommendation of Hofmann shows that the quality-competitive environment acted as a
and Gavin (1998), we used group mean centering for all moderator in the relationship between job autonomy and
individual level predictors except for the dummy variables. mental well-being (b = .03, p \ .01). The cross-level
To examine the appropriateness of multilevel analysis, interaction accounted for 4% of variances in the level-1
we tested a null model. This model provides answers about slope. In other words, job autonomy was more strongly
whether the two constructs (organizational commitment related to mental well-being of employees in the companies
and mental well-being) vary across establishments by positioned in more quality-competitive environment than

123
J Bus Psychol (2012) 27:305–316 311

Table 1 Mean, standard deviation, and correlations

Mean SD 1 2 3 4 5 6 7

Level-1
1. Autonomy 3.14 .77
2. Gender .53 .50 .00
3. Race .94 .23 -.01 .01
4. Age 5.42 1.46 .09** .09** .07**
5. Wage 8.32 3.30 .21** .41** .04** .18**
6. Commitment 3.63 .84 .32** -.07** -.03** .07** .07**
7. Mental well-being 3.58 .87 .16** -.02 .02** .04** -.20** .25**
Level-2
1. Quality competition 4.05 1.10
2. Size 1.87 .97 .04
3. Sector .97 .17 .07* -.10**
4. Manufacturing .18 .38 .04* .20** .06*
5. Electric .15 .36 -.08** .11** -.25** -.20**
6. Sales .25 .43 -.02 -.17** .09** -.27** -.24**
7. Business service .22 .41 .04 -.01 .07** -.25** -.23** -.31**
8. Public service .20 .40 .01 -.10** -.01 -.23** -.21** -.29** -.27**
N = 12,846 (employees), 1,190 (establishments)
* p \ .05, ** p \ .01

Table 2 Results of HLM

Well-being Commitment Well-being
analysis to examine the
Model 1 Model 2 Model 3
mediated moderation
hypotheses Intercept 3.35 (.07)** 3.50 (.08)** 3.33 (.07)**
Level-1
Job autonomy .15 (.01)** .28 (.01)** .11 (.01)**
Gender .08 (.02)** -.07 (.02)** .09 (.02)**
Race .13 (.04)** -.11 (.03)** .15 (.04)**
Age .03 (.01)** .03 (.01)** .03 (.01)**
Wages -.09 (.00)** .03 (.00)** -.09 (.00)**
Commitment .16 (.01)**
Level-2
Quality competition -.00 (.01) .00 (.01) -.00 (.01)
Size -.00 (.01) -.05 (.01)** -.00 (.01)
Sector .06 (.06) .16 (.06)** .05 (.06)
Electric .01 (.03) .08 (.04)* .00 (.03)
N = 12,846 (employees), 1,190 Sales .10 (.03)** .10 (.04)** .11 (.03)**
(establishments) Business service -.13 (.03)** .10 (.04)** -.13 (.03)**
Parenthesis: robust standard Public service .09 (.03)** .33 (.04)** .09 (.03)**
errors Cross-level interaction
All entries are unstandardized Job autonomy 9 quality competition .03 (.01)** .02 (.01)* .02 (.01)*
regression coefficients
Model deviance 31,926 29,838 31,682
* p \ .05, ** p \ .01

in less quality-competitive environment, which supports interaction accounted for 4% of variances in the
Hypothesis 1. level-1 slope. This result confirms that job autonomy is
Model 2 of Table 2 shows that quality competition more strongly related to organizational commitment
moderated the relationship between job autonomy and orga- in more quality-competitive organizations, supporting
nizational commitment (b = .02, p \ .05). The cross-level Hypothesis 2.

123
312 J Bus Psychol (2012) 27:305–316

In Model 3, organizational commitment had a signifi-

cant relationship with mental well-being, after controlling
for the interaction term between job autonomy and quality
competition. Finally, the interaction between job autonomy
and quality competition on mental well-being was reduced
when organizational commitment was controlled (in Model
3 as compared with in Model 1). A Sobel test confirmed
that this reduction was significant (z value of 1.98,
p \ .05). These results support Hypothesis 3, which spec-
ifies a mediating role of organizational commitment in the
interaction between job autonomy and quality competition
on mental well-being.
The findings shown in Table 2 satisfy all of the condi-
tions for mediated moderation that Muller et al. (2005) Fig. 3 The moderating role of quality competition in the relationship
proposed. In addition, since the interaction term between between autonomy and organizational commitment
job autonomy and quality competition in Model 3 was still
significant, organizational commitment played a role as a relationship with job autonomy regardless of the extent of
partial mediator. In sum, job autonomy was more strongly quality competition, the relationship was more remarkable
related to mental well-being and organizational commit- in establishments facing more quality competition. Fig-
ment in quality-competitive organizations. Meanwhile, ure 3 shows that job autonomy had a differential relation-
organizational commitment mediated the moderation ship with organizational commitment according to the
between job autonomy and quality competition on mental extent of quality competition. While organizational com-
well-being. mitment was positively related to job autonomy in both
To interpret the findings more visually, we illustrated types of establishments, the relationship was greater when
the moderating roles of quality-competition in the rela- job autonomy was provided to employees in more quality-
tionships between job autonomy and mental well-being competitive establishments than those in less quality-
(Fig. 2) and between job autonomy and organizational competitive establishments.
commitment (Fig. 3). Employees were divided into two Yet, the unresolved issue in the above analyses is
groups, a group who perceived less (mean - one SD) job whether job autonomy has significant relationships with
autonomy and a group who perceived more (mean ? one those dependent variables even in less quality-competitive
SD) job autonomy, and the sample establishments were organizations. To answer to this question, we examined,
divided into less and more quality-competitive groups following the recommendations of Preacher et al. (2004),
(Aiken and West 1991). the significance of the simple slopes of those relationships
Figure 2 illustrates the moderating role of quality in less quality-competitive establishments. Job autonomy
competition in the relationship between job autonomy and was significantly related with mental well-being and
mental well-being. While mental well-being had a positive organizational commitment (b = .23, t = 5.94, p \ .01;
b = .33, t = 9.03, p \ .01, respectively) in less quality-
competitive establishments as well.

Discussion

Implications

This study is unique in that it bridges the gap between the

macro and micro approaches through multilevel analyses
and it investigates the vertical fit between job autonomy
and organizational contexts while focusing on individual
employees’ outcomes. Results from this study imply that
job autonomy may generally improve employees’ mental
well-being and organizational commitment. Employees
Fig. 2 The moderating role of quality competition in the relationship who have been given more autonomy by their organization
between autonomy and mental well-being are generally more committed and better off mentally.

123
J Bus Psychol (2012) 27:305–316 313

This study also showed that organizational commitment In sum, while all companies should provide their
plays a role as a partial mediator in the relationship employees with sufficient job autonomy with no regard to
between job autonomy and mental well-being. These quality competition, more quality-competitive companies
results are consistent with Shirom et al. (2006), who found can obtain more benefits from job autonomy.
that job autonomy is a negative predictor of employee
burnout, and Ahuja et al. (2007), who found a positive Limitations and Future Research
relationship of job autonomy with organizational commit-
ment. The findings of our current study are consistent with Although the present paper found several important results,
self-determination theory, which posits that the satisfaction it has some limitations, which should be addressed by
of needs for autonomy is linked to general job satisfaction future research. First, this study cannot confirm causality
and psychological well-being (Deci and Ryan 2000; Deci among job autonomy, organizational commitment, and
et al. 1989). mental well-being because the data came from a cross-
This study also indicated that job autonomy may be sectional survey. Thus, future research is encouraged to test
particularly important for employees in quality-competitive the directions of the three variables in the mediation model
organizations. As predicted in Hypotheses 1 and 2, job of the present study using longitudinal or experimental
autonomy has stronger relationships with the two depen- data. It is meaningful, however, to mention implications
dent variables in more quality-competitive organizations, related to the causality issue in the previous studies. A
which is a topic that has never been addressed by previous meta-analysis by Eby et al. (1999) concluded that as
studies. In the job characteristics model, Dodd and Ganster employees were granted job autonomy, their organizational
(1996) found that job autonomy and task variety interact to commitment was higher. Bond et al. (2008) found in a
increase job satisfaction and performance. In other words, quasi-experimental study that job autonomy at time 1 had a
job autonomy has stronger relationships with employees’ negative relationship with psychological distress at time 2.
job satisfaction and performance when their tasks are more These studies confirm that job autonomy is a predictor of
various and complex. If a company in a highly quality- organizational commitment and mental well-being rather
competitive market requires its employees to acquire than the reverse. Regarding the causal relationship between
various skills and to conduct complex task, Dodd and organizational commitment and mental well-being, Begley
Ganster’s findings indirectly support the results of the and Cazjka (1993), in a panel study, identified organiza-
present study. tional commitment as a significant moderator of job stress.
Since organizations in highly quality-competitive mar- Moreover, Parker (2003) found in a quasi-experimental
kets can attain higher quality products or services by field study that correlations of organizational commitment
granting their employees more job autonomy, job auton- at time 1 with job anxiety and job depression at time 2 were
omy fits well with quality-competitive situations. This larger than those of job anxiety and job depression at time 1
finding is consistent with a contingency perspective, which with organizational commitment at time 2. These previous
posits that when HR practices fit well with specific orga- longitudinal and quasi-experimental studies support the
nizational characteristics, these practices are more effective idea that organizational commitment is a predictor of
(Cappelli and Neumark 2001; Delery and Doty 1996; Neal mental well-being rather than the reverse.
et al. 2005; Schuler and Jackson 1987; Youndt et al. 1996). A second limitation of the present study is that the data
Therefore, we suggest that job autonomy may be a ‘‘best of individual level variables were collected from the
practice,’’ in that it generally is related to higher mental employees via self-report. Several scholars have argued for
well-being and stronger organizational commitment the need to assess job stress and stressors objectively
regardless of organizational context, and at the same time a through supervisor or peer reports, or through job analysis
‘‘contingent practice,’’ in that those relationships can be (e.g., Aldag et al. 1981; Loher et al. 1985; Spector and Jex
greater under a specific context. 1991). Due to the sensitive nature of organizational com-
Companies under pressure of strong quality competition mitment and mental well-being, however, self-reported
can attain higher quality products or services by providing data may be more accurate. Moreover, given that there is
their employees with job autonomy. Therefore, organiza- evidence of construct validity between the variables, self-
tions competing in a highly quality-competitive market in report bias does not seem to be a serious problem in the
particular should provide employees with more job current results.
autonomy in order to improve employee outcomes. Despite Third, the moderator of the present study, quality-
this finding, less quality-competitive companies also need competitive environment, was measured by only one item.
to grant more job autonomy to their employees because Single-item measure does not allow for an examination of
they benefit from it in terms of mental well-being and reliability, and most methodologists recommend the use of
organizational commitment, as indicated in Figs. 2 and 3. multiple-item measures. Single-item measures, however,

123
314 J Bus Psychol (2012) 27:305–316

are acceptable if they are unequivocal (Bergkvist and Begley, T. M., & Cazjka, J. M. (1993). Panel analysis of the
Rossiter 2007; Rohland et al. 2004). Future research needs moderating effects of commitment on job satisfaction, intent to
quit, and health following organizational change. Journal of
to use multiple items for a construct to improve the reli- Applied Psychology, 78(4), 552–556.
ability and validity. Bergkvist, L., & Rossiter, J. R. (2007). The predictive validity of
Finally, the organization-level variables of the 2004 multiple-item versus single-item measures of the same con-
WERS were measured from one manager in each estab- structs. Journal of Marketing Research, 44(2), 175–184.
Bond, F. W., & Flaxman, P. E. (2006). The ability of psychological
lishment. Although they were senior managers who can flexibility and job control to predict learning, job performance,
assess the overall situation of their establishments, more and mental health. Journal of Organizational Behavior Man-
raters in each establishment would be desirable to improve agement, 26, 113–130.
measurement reliability. Bond, F. W., Flaxman, P. E., & Bunce, D. (2008). The influence of
psychological flexibility on work redesign: Mediated moderation
While the survey used in this study covers a variety of of a work reorganization intervention. Journal of Applied
aspects of job autonomy including work schedule, pace, Psychology, 93(3), 645–654.
contents, and order, each of these dimensions needs to be Brockner, J., Spreitzer, G., Mishra, A., Hochwarter, W., Pepper, L., &
assessed by multiple items to improve measurement reli- Weinberg, J. (2004). Perceived control as an antidote to the
negative effects of layoffs on survivors’ organizational commit-
ability. For example, Humphrey et al. (2007) suggested ment and job performance. Administrative Science Quarterly,
multiple facets for autonomy: work scheduling autonomy, 49(1), 76–100.
work methods autonomy, and decision-making autonomy. Bryk, A. S., & Raudenbush, S. W. (1992). Hierarchical linear
By this subdivision, future research may assess the relative models. Newbury Park, CA: Sage.
Cappelli, P., & Neumark, D. (2001). Do ‘‘high-performance’’ work
contribution of each dimension of job autonomy. practices improve establishment-level outcomes? Industrial and
Since the effect of job autonomy may be contingent on Labor Relations Review, 54(4), 737–775.
organizational factors, more organizational contexts in Cooper, C. L., & Cartwright, S. (1996). Mental health and stress in
which job autonomy is more effective should be investi- the workplace: A guide for employers. Windsor: NFER-Nelson.
Das, A., Handfield, R. B., Calantone, R. J., & Ghosh, S. (2000). A
gated in the future. Also, given that interests of employees contingent view of quality management: The impact of interna-
in a quality-competitive company can be aligned with those tional competition on quality. Decision Sciences, 31(3),
of the company via job autonomy, future research may 649–690.
explore whether this interaction can be applied to other Deci, E. L., Connell, J. P., & Ryan, R. M. (1989). Self-determination
in a work organization. Journal of Applied Psychology, 74,
employee outcomes. For example, because job autonomy 580–590.
may be more effective in improving employee attitudes and Deci, E. L., & Ryan, R. M. (2000). The ‘‘what’’ and the ‘‘why’’ of
mental well-being in more quality-competitive companies, goal pursuits: Human needs and the self-determination of
employees in those companies may achieve better perfor- behavior. Psychological Inquiry, 11, 227–268.
Delery, J. E., & Doty, D. H. (1996). Modes of theorizing in strategic
mance when they are given sufficient job autonomy. human resource management: Tests of universalistic, contin-
gency, and configurational performance predictions. Academy of
Acknowledgment The present research has been conducted by the Management Journal, 39(4), 802–835.
Research Grant of Kwangwoon University in 2011. Department of Trade and Industry. (2005). Workplace employment
relations survey 2004: Introductory note to accompany data
deposit.
Dodd, N. G., & Ganster, D. C. (1996). The interactive effects of
References variety, autonomy, and feedback on attitudes and performance.
Journal of Organizational Behavior, 17, 329–347.
Ahuja, M. K., Chudoba, K. M., Kacmar, C. J., McKnight, D. H., & Donaldson, S. I., & Grant-Vallone, E. J. (2002). Understanding self-
George, J. F. (2007). It road warriors: Balancing work-family report bias in organizational behavior research. Journal of
conflict, job autonomy, and work overload to mitigate turnover Business and Psychology, 17(2), 245–260.
intentions. MIS Quarterly, 31(1), 1–17. Eby, L. T., Freeman, D. M., Rush, M. C., & Lance, C. E. (1999).
Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and Motivational bases of affective organizational commitment: A
interpreting interactions. Thousand Oaks, CA: Sage. partial test of an integrative theoretical model. Journal of
Aldag, R. J., Barr, S. H., & Brief, A. P. (1981). Measurement of Occupational and Organizational Psychology, 72, 463–483.
perceived task characteristics. Psychological Bulletin, 90(3), Eisenberger, R., Armeli, S., Rexwinkel, B., Lynch, P. D., & Roades,
415–431. L. (2001). Reciprocation of perceived organizational support.
Allen, N. J., & Meyer, J. P. (1990). The measurement and antecedents Journal of Applied Psychology, 86(1), 42–51.
of affective, continuance and normative commitment to the Eisenberger, R., Rhoades, L., & Cameron, J. (1999). Does pay for
organization. Journal of Occupational Psychology, 63(1), 1–18. performance increase or decrease perceived self-determination
Allen, D. G., Shore, L. M., & Griffeth, R. W. (2003). The role of and intrinsic motivation? Journal of Personality and Social
perceived organizational support and supportive human resource Psychology, 77, 1026–1040.
practices in the turnover process. Journal of Management, 29(1), Galais, N., & Moser, K. (2009). Organizational commitment and the
99–118. well-being of temporary agency workers: A longitudinal study.
Baird, L., & Meshoulam, I. (1988). Managing two fits of strategic Human Relations, 62(4), 589–620.
human resource management. Academy of Management Review, Ganster, D. C., & Schaubroeck, J. (1991). Work stress and employee
13, 116–128. health. Journal of Management, 17, 235–272.

123
J Bus Psychol (2012) 27:305–316 315

Hackman, J. R., & Oldham, G. R. (1976). Motivation through the Osterman, P. (1994). How common is workplace transformation and
design of work: Test of a theory. Organizational Behavior and who adopts it? Industrial and Labor Relations Review, 47(2),
Human Performance, 16, 250–279. 173–188.
Hakanen, J. J., Bakker, A. B., & Schaufeli, W. B. (2006). Burnout and Park, R., Appelbaum, E., & Kruse, D. (2010). Employee involvement
work engagement among teachers. Journal of School Psychol- and group incentives in manufacturing companies: A multi-level
ogy, 43, 495–513. analysis. Human Resource Management Journal, 20(3), 227–
Hofmann, D. A. (1997). An overview of the logic and rationale of 243.
hierarchical linear models. Journal of Management, 23(6), Parker, S. K. (2003). Longitudinal of lean production on employee
723–744. outcomes and the mediating role of work characteristics. Journal
Hofmann, D. A., & Gavin, M. B. (1998). Centering decisions in of Applied Psychology, 88(4), 620–634.
hierarchical linear models: Implications for research in organi- Porter, M. E. (1980). Competitive strategy. New York: Free Press.
zations. Journal of Management, 24(5), 623–641. Preacher, K. J., Curran, P. J., & Bauer, D. J. (2004). Simple intercepts,
Humphrey, S. E., Nahrgang, J. D., & Morgeson, F. P. (2007). simple slopes, and regions of significance in HLM 2-way
Integrating motivational, social, and contextual work design interactions. http://people.ku.edu/~preacher/interact/hlm2.htm.
features: A meta-analytic summary and theoretical extention of Raudenbush, S., & Bryk, A. S. (2002). Hierarchical linear models:
the work design literature. Journal of Applied Psychology, 92(5), Applications and data analysis methods. Thousand Oaks, CA:
1332–1356. Sage.
Huselid, M. A. (1995). The impact of human resource management Rhoades, L., & Eisenberger, R. (2002). Perceived organizational
practices on turnover, productivity, and corporate financial support: A review of the literature. Journal of Applied Psychol-
performance. Academy of Management Journal, 38(3), 635–672. ogy, 87(4), 698–714.
Karasek, R. A. (1979). Job demands, job decision latitude, and mental Rhoades, L., Eisenberger, R., & Armeli, S. (2001). Affective
strain: Implications for job redesign. Administrative Science commitment to the organization: The contribution of perceived
Quarterly, 24, 285–308. organizational support. Journal of Applied Psychology, 86(5),
Kasl, S. V. (1973). Mental health and work environment: An 825–836.
examination of the evidence. Journal of Occupational Medicine, Rohland, B. M., Kruse, G. R., & Rohrer, J. E. (2004). Validation of a
15, 509–518. single-item measure of burnout against the Maslach burnout
Kemp, N. J., Wall, T. D., Clegg, C. W., & Cordery, J. L. (1983). inventory among physicians. Stress and Health, 20, 75–79.
Autonomous work groups in a Greenfield site: A comparative Sage, W., & Kalyan, D. (2006). Horses or unicorns: can paying for
study. Journal of Occupational Psychology, 56, 271–288. performance make quality competition routine? Journal of
Kinicki, A., & Kreitner, R. (2008). Organizational behavior: Key Health Politics, Policy and Law, 31(3), 531–556.
concepts, skills & best practices. Boston: McGraw-Hill Irwin. Schuler, R. S., & Jackson, S. E. (1987). Linking competitive
Kline, R. B. (1998). Principles and practice of structural equation strategies with human resource management practices. Academy
modeling. New York & London: The Guilford Press. of Management Executive, 1(3), 207–219.
Langfred, C. W. (2000). The paradox of self-management: Individual Schulz, R., Greenley, J. R., & Brown, R. (1995). Organization,
and group autonomy in work groups. Journal of Organizational management, and client effects on staff burnout. Journal of
Behavior, 21, 563–585. Health and Social Behavior, 36, 333–345.
Loher, B. T., Noe, R. A., Moeller, N. L., & Fitzgerald, M. P. (1985). Shirom, A., Nirel, N., & Vinokur, A. D. (2006). Overload, autonomy,
A meta-analysis of the relation of job characteristics to job and burnout as predictors of physicians. Journal of Occupational
satisfaction. Journal of Applied Psychology, 70, 280–289. Health Psychology, 11(4), 328–342.
Man, D. C., & Lam, S. K. (2003). The effects of job complexity and Siu, O. (2002). Occupational stressors and well-being among Chinese
autonomy on cohesiveness in collectivistic and individualistic employees: The role of organizational commitment. Applied
work groups: A cross-cultural analysis. Journal of Organiza- Psychology: An International Review, 51(4), 527–544.
tional Behavior, 24, 979–1001. Spector, P. E. (1986). Perceived control by employees: A meta-
Mathieu, J. E., & Zajac, D. M. (1990). A review and meta-analysis of analysis of studies concerning autonomy and participation at
the antecedents, correlates, and consequences of organizational work. Human Relations, 39(11), 1005–1017.
commitment. Psychological Bulletin, 108(2), 171–194. Spector, P. E., & Jex, S. M. (1991). Relations of job characteristics
Meyer, J. P., & Herscovitch, L. (2001). Commitment in the from multiple data sources with employee affect, absence,
workplace: Toward a general model. Human Resource Manage- turnover intentions, and health. Journal of Applied Psychology,
ment Review, 11, 299–326. 76(1), 46–53.
Moore, J. E. (2000). One road to turnover: An examination of work Strupp, H. H., & Hadley, S. W. (1977). A tripartite model of mental
exhaustion in technology professionals. MIS Quarterly, 24(1), health and therapeutic outcomes. American Psychologist, 32,
141–168. 187–196.
Mowday, R., Steers, R., & Porter, L. (1979). The measurement of Van Yperen, N. W., & Hagedoorn, M. (2003). Do high job demands
organizational commitment. Journal of Vocational Behavior, 14, increase intrinsic motivation or fatigue or both? The role of job
224–247. control and job social support. Academy of Management Journal,
Muller, D., Judd, C. M., & Yzerbyt, V. Y. (2005). When moderation 46(3), 339–348.
is mediated and mediation is moderated. Journal of Personality Van Yperen, N. W., & Snijders, T. A. (2000). A multilevel analysis of
and Social Psychology, 89(6), 852–863. the demands-control model: Is stress at work determined by
Neal, A., West, M. A., & Patterson, M. G. (2005). Do organizational factors at the group level or the individual level? Journal of
climate and competitive strategy moderate the relationship Occupational Health Psychology, 5(1), 182–190.
between human resource management and productivity? Journal Vigoda-Gadot, E. (2007). Redrawing the boundaries of OCB? An
of Management, 31(4), 492–512. empirical examination of compulsory extra-role behavior in the
Nikolaou, I., & Tsaousis, I. (2002). Emotional intelligence in the workplace. Journal of Business and Psychology, 21(3), 377–405.
workplace: Exploring its effects on occupational stress and Wall, T. D., Kemp, N. J., Jackson, P. R., & Clegg, C. W. (1986).
organizational commitment. The International Journal of Orga- Outcomes of autonomous workgroups: A long-term field exper-
nizational Analysis, 10(4), 327–342. iment. Academy of Management Journal, 29(2), 280–304.

123
316 J Bus Psychol (2012) 27:305–316

Warr, P. B. (1987). Work, unemployment, and mental health. Oxford: Wright, T. A., & Hobfoll, S. E. (2004). Commitment, psychological
Oxford University Press. well-being and job performance: An examination of conserva-
Warr, P. (1990). The measurement of well-being and other aspects tion of resources (COR) theory and job burnout. Journal of
of mental health. Journal of Occupational Psychology, 63, 193– Business and Management, 9(4), 389–406.
210. Youndt, M. A., Dean, J. W., & Lepak, D. P. (1996). Human resource
Wayne, S. J., Shore, L. M., & Liden, R. C. (1997). Perceived management, manufacturing strategy, and firm performance.
organizational support and leader-member exchange: A social Academy of Management Journal, 39(4), 836–866.
exchange perspective. Academy of Management Journal, 40,
82–111.

123
Copyright of Journal of Business & Psychology is the property of Springer Science & Business Media B.V. and
its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for individual use.