Full text of DOI:10.1080/02678370120791

This article was downloaded by: [Ams/Girona*barri Lib]
On: 17 October 2014, At: 03:27
Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954
Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH,
UK
Work & Stress: An International
Journal of Work, Health &
Organisations
Publication details, including instructions for authors
and subscription information:
http://www.tandfonline.com/loi/twst20
The association between
job skill discretion, decision
authority and burnout
Yvonne Rafferty , Ronald Friend & Paul A. Landsbergis
Published online: 26 Nov 2010.
To cite this article: Yvonne Rafferty , Ronald Friend & Paul A. Landsbergis (2001) The
association between job skill discretion, decision authority and burnout, Work & Stress:
An International Journal of Work, Health & Organisations, 15:1, 73-85
To link to this article: http://dx.doi.org/10.1080/02678370120791
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the
information (the “Content”) contained in the publications on our platform.
However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or
suitability for any purpose of the Content. Any opinions and views expressed
in this publication are the opinions and views of the authors, and are not the
views of or endorsed by Taylor & Francis. The accuracy of the Content should
not be relied upon and should be independently verified with primary sources
of information. Taylor and Francis shall not be liable for any losses, actions,
claims, proceedings, demands, costs, expenses, damages, and other liabilities
whatsoever or howsoever caused arising directly or indirectly in connection
with, in relation to or arising out of the use of the Content.
This article may be used for research, teaching, and private study purposes.
Any substantial or systematic reproduction, redistribution, reselling, loan, sub-
licensing, systematic supply, or distribution in any form to anyone is expressly

forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions

&
, 2001,
15,
1, 73±85
no.
work stress
vol.
The association between job skill discretion,
decision authority and burnout
YVONNE RAFFERTY*
Psychology Department, Pace University, 41 Park Row, New York, NY 10038, USA
RONALD FRIEND
Psychology Department, State University of New York at Stony Brook, Stony Brook,
NY 11794, USA
and PAUL A. LANDSBERGIS
Cardiovascular Institute, Box 1030, Mount Sinai Medical Centre, One Gustave L. Levy
Place, New York, NY 10029, USA
Keywords: Job strain; Job demands; Job control; Burnout.
This paper reports on the relationship between dimensions of control (skill discretion and decision
authority) and burnout (emotional exhaustion, depersonalization, and reduced personal accomplish-
ment) among 164 human service workers. It examines the diVerential in¯ uence of job demands,
control (skill discretion and decision authority) and social support (supervisor, co-workers, others)
on each burnout dimension. Then it examines the moderating eVects of higher skill discretion,
higher decision authority, and higher social support on burnout. Low skill discretion was found to
be associated with high emotional exhaustion and depersonalization and low personal accomplishment.
The eVects of decision authority were not statistically signi®cant. High job demands were associated
with high emotional exhaustion only. Social support (supervisor, co-worker, and others) was not
associated with burnout when demographic variables and job characteristics were controlled for.
Neither dimension of control moderated the impact of high job demands on burnout. Social support
did not moderate the impact of high demands, low skill discretion, or low decision authority on any
%
burnout dimension. The full model explained 44 of the variance in emotional exhaustion, 25 in
depersonalization, and 42 in personal accomplishment. Despite its limitations, the study suggests
%
%
that the Job Demand-Control model may provide a useful theoretical foundation for the study of
burnout, but that the control dimensions need to be evaluated independently since they appear to
be diVerentially related to the burnout dimensions.
1. Introduction
Maslach’s (1982) conceptualization of burnout as a syndrome of emotional exhaustion,
depersonalization, and reduced personal accomplishment is widely cited (Kop, Euwema,
and Schaufeli, 1999; Lee, and Ashforth, 1990; Schaufeli, Maslach, and Marek, 1993;
*Author for correspondence. e-mail: YRaVerty@Pace.edu
Work & Stress ISSN 0267-8373 print/ISSN 1464-5335 online 2001 Taylor & Francis Ltd
http://www.tandf.co.uk/journals/
DOI: 10.1080/02678370110064627

74
Y. RaVerty et al.
Soderfeldt, Soderfeldt, Warg, and Ohlson, 1996). Emotional exhaustion refers to the
depletion of one’s emotional resources and has been linked to such psychological constructs
as tension, anxiety, physical fatigue, and insomnia. Depersonalization refers to a negative
and detached response to clients. Reduced personal accomplishment represents an aspect
of personal eYcacy and refers to a low level of perceived competence and successful
achievement in one’s work (Maslach, 1993). Research on the factorial structure of the
Maslach Burnout Inventory indicates that personal accomplishment is distinct from the
other two dimensions and that emotional exhaustion and depersonalization are linked (Lee,
and Ashforth, 1996; Soderfeldt et al., 1996; Taris, Schreurs, and Schaufeli, 1999). Research
also suggests that the three dimensions of burnout might be related to diVerent job
characteristics (Cordes, and Dougherty, 1993; Janssen, Schaufeli, and Houkes, 1999;
Maslach, 1993).
1.1. The Job Demand-Control model and burnout
One conceptual model guiding research on burnout is the Job Demand-Control ( JDC)
model (Karasek, 1979), which has been expanded to include social support ( Johnson, and
Hall, 1988). The JDC model predicts that adverse psychological and physiological reactions
are aVected by the combined impact of two structural conditions of the workplace: high
demands (workload pressures) and low control (skill discretion plus decision authority).
Social support from supervisors and co-workers may reduce the eVects of job strain
(Karasek, and Theorell, 1990). A review of research between 1979 and 1997 on the JDC
model and psychological well-being (e.g. anxiety, depression, job satisfaction, burnout)
concluded that there is only minimal evidence that control and/or support moderate the
negative impact of high job demands (Van der Doef, and Maes, 1999). None of the
burnout studies reviewed (n=4) found evidence that control moderates the negative impact
of high job demands (Landsbergis, 1988; Melamed, Kushnir, and Meir, 1991; Nijhuis, and
Smulders, 1996; Sonnentag, Brodbeck, Heinbokel, and Stolte, 1994). The two studies that
examined the buVering eVects of social support also refuted the interaction hypothesis
(Melamed et al., 1991; Nijhuis, and Smulders, 1996). Several additional studies have applied
the JDC model to burnout dimensions in recent years (de Jonge, Janssen, and van Breukelen,
1996; de Rijk, Le Blanc, Schaufeli, and de Jonge, 1998; Schreurs, and Taris, 1998; Taris
et al., 1999). As with the earlier studies, these studies do not indicate that control or support
moderate the impact of high demands on burnout. They do, however, suggest that job
demands and control may be diVerentially related to each burnout dimension. They also
suggest that skill discretion and decision authority may not be equally important in
predicting and explaining burnout. Finally, a recent examination of the factorial structure
of the three variables comprising the JDC model demonstrated that a three-factor model
( job demands, skill discretion, and decision authority) ®tted the data better than a
two-factor model ( job demands and decision latitude) (Schreurs, and Taris, 1998).
In summary, although there is considerable research with the JDC model, few studies
have applied the model to burnout. In addition, the majority of the burnout studies are
not comprehensive, in that all of the variables in the model and each burnout dimension
are systematically included and examined. Nonetheless, their ®ndings suggest several conclu-
sions. First, it does not appear that the negative impact of high job demands on burnout
can be moderated by high levels of control (de Jonge, and Kompier, 1997; Schreurs, and
Taris, 1998; Van der Doef, and Maes, 1999). Second, job demands appear to be a stronger
predictor than control for emotional exhaustion, and a weaker predictor than control
for depersonalization and reduced personal accomplishment. Higher job demands were

The Job DemandÐcontrol model and burnout
75
associated with emotional exhaustion in each of the four studies where this relationship
was examined (de Jonge et al., 1996; de Rijk et al., 1998; Landsbergis, 1988; Taris et al.,
1999) and with higher depersonalization in only one (Taris et al., 1999) of the three studies
where it was explored (de Rijk et al., 1998; Landsbergis, 1988; Taris et al., 1999). None
of the four studies that examined the relationship between job demands and personal
accomplishment identi®ed a signi®cant relationship (de Jonge et al., 1996; de Rijk et al.,
1998; Landsbergis, 1988; Taris et al., 1999). Third, control appears to be associated with
each burnout dimension, although the relationships between control and burnout are less
consistent than for job demands, and particularly so for emotional exhaustion and deperson-
alization. Inconsistent ®ndings may be related to the diVerent ways in which control has
been de®ned and suggest that future research on the JDC model should diVerentiate
between dimensions of control. Emotional exhaustion has been signi®cantly associated
with higher control when de®ned as decision latitude (skill discretion plus decision author-
ity) (Landsbergis, 1988) and as autonomy (de Jonge et al., 1996). The ®ndings have been
inconsistent when it has been de®ned as decision authority: signi®cant in one study (Taris
et al., 1999) but not in the other (de Rijk et al., 1998). It was not related to skill discretion
in the one study that examined this relationship (Taris et al., 1999). Higher depersonalization
was associated with lower control when it was de®ned as decision latitude (skill discretion
plus decision authority) (Landsbergis, 1988) and as skill discretion (Taris et al., 1999). The
®ndings were inconsistent when it was de®ned as decision authority: signi®cant in one
study (Taris et al., 1999) but not in the other (de Rijk et al., 1998). Reduced personal
accomplishment has been consistently associated with lower control. This relationship held
when control was de®ned as decision latitude (skill discretion plus decision authority)
(Landsbergis, 1988), as skill discretion (Taris et al., 1999), and as decision authority (Taris
et al., 1999).
A ®nal conclusion concerns the role of social support and how it may moderate the
impact of high demands and low control on burnout (de Jonge, and Kompier, 1997; Van
der Doef, and Maes, 1999). While several studies have identi®ed a relationship between
social support and burnout dimensions ( Janssen et al., 1999; Lee, and Ashforth, 1996;
Leiter, 1991, 1993), there is little, if any, support for a three-way interaction between
demands, control, and support (de Jonge et al., 1996; Van der Doef, and Maes, 1999).
Social support, however, may be more eVective as a buVer of high job demands when the
type of support matches the demands of the job (Van der Doef, and Maes, 1999). It is also
possible that skill discretion and decision authority interact diVerently with social support
in buVering the eVects of high demands on burnout dimensions. This issue, however, has
not yet been examined.
2. Hypotheses
The following three hypotheses are proposed.
(1) Hypothesis 1. Job demands and control are diVerentially associated with each burnout
dimension (emotional exhaustion, depersonalization, reduced personal accomplish-
ment). Emotional exhaustion is associated primarily with high job demands and, to
a lesser degree, with low control. Depersonalization is associated primarily with low
control and, to a lesser degree, with high job demands. Reduced personal accom-
plishment is associated with low control. For each burnout dimension, the two
components comprising decision latitude (skill discretion and decision authority)
will be comparable in their predictive ability.

76
Y. RaVerty et al.
(2) Hypothesis 2. Social support (supervisor, co-worker, and others) is negatively associ-
ated with each burnout dimension. It predicts each burnout dimension after job
demands and control are taken into consideration.
(3) Hypothesis 3. Job demands, control (skill discretion and decision authority), and
social support (supervisor, co-workers, others) interact in explaining burnout. Thus,
it is predicted that (1) at higher levels of job demands, workers with lower control
or lower support will experience higher burnout than workers with higher control;
(2) at lower levels of work control, workers with lower support will experience
higher burnout than workers with higher support; and (3) the highest degree of
burnout (high emotional exhaustion and depersonalization and low personal accom-
plishment) will be among those with high job demands, low control, and low
support.
3. Method
3.1. Participants and procedures
All 304 workers involved in direct human service and employed by a New York City
guidance and employment service agency were surveyed. They included speech, art,
occupational, recreational, and rehabilitation therapists, social workers, psychologists, and
case aides. They were at risk of burnout since they worked in emotionally demanding jobs
with mentally challenged clients. Almost half (49%) worked in developmental day treatment
programs. The remainder worked in intermediate care residences (7%), or job training
programs providing hands-on individualized instruction (44%). A volunteer at each site
distributed questionnaires to each worker. Workers placed their completed questionnaires
in an attached envelope and returned them to a prearranged location on-site. To ensure
anonymity, each worker was assigned a code number, which was known only to the
investigators. Questionnaires were returned by 188 workers (62%); 24 were excluded
because of missing data. Thus, the analyses were based on responses from 164 workers.
The majority of the sample (75%) was female, 47% were Caucasian, and the median age
was 30 years (M=33.5 years, SD=10.6 years). Overall, 33% reported having less than an
undergraduate degree, 36% had a college degree, and 31% had a masters degree or higher.
To determine if respondents were representative of the overall sample, these two groups
were compared by type of program, gender, and education. No signi®cant diVerences
were identi®ed.
3.2. Measures
Burnout was assessed by the Maslach Burnout Inventory (MBI), which consists of 22 items
representing emotional exhaustion (9 items), depersonalization (5 items), and reduced
personal accomplishment (8 items) (Maslach, and Jackson, 1981, 1986). Items are scored
on a 7-point rating scale ranging from `never’ (0) to `daily’ (6). The Job Content
Questionnaire (Karasek, 1985) was used to assess job demands (5 items), skill discretion (6
items) and decision authority (3 items). Social support was evaluated by four items (e.g.
good listener, there to rely on when help is needed) (Caplan, Cobb, French, Harrison,
and Pinneau, 1980). Respondents indicated to what extent they agreed with each item and
answered each question with regard to their supervisor, co-workers, and spouse, friends,
or relatives. Each of these measures has demonstrated acceptable levels of reliability. Five
demographic variables (age, gender, race/ethnicity, education, marital status) were included
as control variables.

The Job DemandÐcontrol model and burnout
77
3.3. Data analyses
Hierarchical linear regression analysis was conducted to determine whether characteristics
of the worker, job demands, control, and social support were predictive of burnout. This
enabled an assessment of the independent eVects of each variable block (net of other blocks)
on burnout. For each burnout dimension, variable blocks were entered in the following
order: (1) demographic characteristics; (2) job demands; (3) skill discretion and decision
authority; and (4) social support from supervisor, co-workers, and others. On a ®nal step,
the three-way interaction, and each possible two-way interaction, were entered and
individually examined to determine their level of signi®cance.
4. Results
4.1. Descriptive statistics and correlation coeYcients
Scale means, standard deviations, and reliability coeYcients (table 1) indicate that there was
suYcient variability within the sample to provide a fair test of the model and that the
reliabilities of the subscales were acceptable (ranging from .69 to .90) (Nunnaly, 1987).
Reports of burnout ranged from 0±54 for emotional exhaustion (M=23.02, SD=12.44),
from 0±30 for depersonalization (M=6.58, SD=6.24), and from 0±48 for personal
accomplishment (M=35.15, SD=8.53). These scores are consistent with those obtained
by Maslach, and Jackson (1986) for emotional exhaustion (M=24.08, SD=11.88) and
personal accomplishment (M=36.01, SD=6.93). The depersonalization scores are slightly
lower (M=9.40, SD=6.90). Reports of job characteristics were consistent with other
populations (Karasek, Brisson, Kawakami, Houtman, Bongers, and Amick, 1998).
The zero-order correlation coeYcients for the independent and dependent variables
and each of the control variables included in the model (table 1) indicate that personal
accomplishment is distinct from the other two dimensions, while emotional exhaustion
and depersonalization are highly correlated (r=.58). These data also indicate that some job
characteristics and demographic variables were signi®cantly correlated. Lower skill discretion
was associated with having less education; lower decision authority was associated with
being female, having less education, and being unmarried. Higher co-worker support was
associated with higher education; higher support from others was associated with being
married. Job demands and control were not related although both were signi®cantly
associated with social support.
The results of three hierarchical regression analyses, which provide a detailed examina-
tion of the eVects of job characteristics and demographic variables on each burnout
dimension, are presented in table 2 and discussed below.
4.2. Tests of hypotheses
Hypothesis 1 was partially supported. Job demands and control were diVerentially associated
with each burnout dimension. Skill discretion was a stronger and more consistent predictor
of each burnout dimension than decision authority. As shown in table 1, higher job
demands were signi®cantly correlated with higher emotional exhaustion (r=.52) and higher
depersonalization (r=.27), and not with reduced personal accomplishment. Lower skill
discretion was signi®cantly correlated with higher emotional exhaustion (r=Õ.28), higher
depersonalization (r=Õ.22), and reduced personal accomplishment (r=.37). Lower
decision authority, in contrast, was signi®cantly correlated with only higher emotional
exhaustion (r=Õ.17).
As shown in table 2, job demands accounted for a signi®cant portion of the variability

78
Y. RaVerty et al.

The Job DemandÐcontrol model and burnout
79

80
Y. RaVerty et al.
associated with emotional exhaustion, controlling for demographic variables (DR =.24).
2
Only skill discretion (and not decision authority) explained a signi®cant proportion of
additional variance after demographic variables and job demands were controlled for in the
model (DR =.13). Table 2 also indicates that job demands did not account for a signi®cant
2
portion of the variability associated with depersonalization, controlling for demographic
variables (DR =.01). The addition of skill discretion, but not decision authority, explained
2
a signi®cant proportion of additional variance (DR =.08), after demographic variables and
2
job demands were controlled for in the model. Finally, job demands did not account for
a signi®cant portion of the variability associated with personal accomplishment, controlling
for demographic variables (DR =.01). Only skill discretion, and not decision authority,
2
explained a signi®cant proportion of additional variance (DR =.20) after demographic
2
variables and job demands were controlled for in the model.
Hypothesis 2 was partially supported. Higher emotional exhaustion was signi®cantly
associated with lower support from both supervisor (r=Õ.40) and co-workers (r=Õ.17).
Higher depersonalization was associated with only lower supervisor support
(r=Õ.29). Reduced personal accomplishment was signi®cantly associated with lower
support from co-workers (r=.23) and from others (r=.22). Multiple regression analyses
(table 2) indicate that when demographic variables and job characteristics (demands and
control) are controlled for, social support is not associated with any burnout dimension.
For depersonalization, however, the addition of social support to the model resulted in
two demographic variables (age, race/ethnicity) emerging as signi®cant predictors and the
loss of skill discretion as a signi®cant predictor. This ®nding suggests that social support
may mediate the relationship between skill discretion and depersonalization.
The ®nal hypothesis focused on the moderating eVects of control and social support.
None of the hypothesized interaction eVects were supported (possibly due to the small
sample size). This ®nding was consistent when control was de®ned as skill discretion or
decision authority, and support was de®ned as supervisor support, co-worker support, or
support from others. {Owing to space limitations, these ®ndings are not presented; please
contact the authors for this information.}
5. Conclusion and discussion
The ®ndings of this study con®rm the importance of diVerentiating between skill discretion
and decision authority in research that involves the JDC model. While rarely evaluated
separately, these two subscales may be tapping two diVerent dimensions of control (Baker,
Israel, and Schurman, 1996; de Jonge et al., 1996; de Rijk et al., 1998; Kasl, 1996; Taris
et al., 1999; Wall, Jackson, Mullarkey, and Parker, 1996). In this study, skill discretion was
associated with each burnout dimension while decision authority was not. The ®ndings of
this study also point to the importance of including measures of both demands and control
in studies assessing burnout since each burnout dimension appears to be diVerentially
related to job characteristics. In this study, emotional exhaustion was more strongly
associated with higher job demands than lower control, whereas depersonalization and
reduced personal accomplishment were more strongly associated with lower control than
higher demands. It is suggested that social support (and particularly supervisor support)
may be an important variable in studies exploring the relationship between job characteristics
and burnout.
In this study, having support from one’s supervisor was signi®cantly associated with
having higher control (skill discretion and decision authority) and lower job demands. It
was also associated with reduced emotional exhaustion and depersonalization. Co-worker

The Job DemandÐcontrol model and burnout
81
support was associated with having lower demands, lower emotional exhaustion, and higher
personal accomplishment. Support outside of work was associated with higher personal
accomplishment only. These ®ndings suggest that the burnout dimensions may be diVeren-
tially related to the diVerent sources of support. It did not predict any burnout dimension
after controlling for demographic and job characteristics.
5.1. Emotional exhaustion
The full model explained 44% of the variance in emotional exhaustion. Workers with
higher levels of emotional exhaustion were more likely to experience higher job demands
(b=.50) and lower skill discretion (b=Õ.33). The relationship between emotional exhaus-
tion and higher job demands found in this study is consistent with prior research involving
the JDC model (de Jonge et al., 1996; de Rijk et al., 1998; Landsbergis, 1988; Taris et al.,
1999), as well as with research involving individual determinants of emotional exhaustion
( Janssen et al., 1999; Leiter 1991, 1993).
The higher emotional exhaustion among workers with lower skill discretion, but not
among workers with lower decision authority, is only partially consistent with prior
research. As noted earlier, only one study has examined the relationship between skill
discretion and emotional exhaustion and found it to be non-signi®cant (Taris et al., 1999).
Two studies have examined the relationship between decision authority and emotional
exhaustion. Higher emotional exhaustion was found among workers with lower decision
authority in one study (Taris et al., 1999) but not in the other (de Rijk et al., 1998). In
addition, Landsbergis (1988) found higher emotional exhaustion among workers with lower
decision latitude (skill discretion plus decision authority) while de Jonge et al. (1996) found
high emotional exhaustion among workers with low autonomy. Interestingly, Taris et al.
(1999) found that lower skill discretion was also associated with mental exhaustion (among
university staV only), job dissatisfaction, and higher intention to quit, while decision
authority was associated with job dissatisfaction only (among university staV only).
Finally, the relationship between emotional exhaustion and social support found in this
study is consistent with prior research ( Janssen et al., 1999; Lee, and Ashforth, 1996).
However, given the non-signi®cant ®ndings for social support in the regression analyses,
the relationship between social support and emotional exhaustion may act indirectly through
demographic and other job characteristics.
5.2. Depersonalization
The full model explained 25% of the variance in depersonalization. Workers with higher
levels of depersonalization were more likely to be Caucasian (b=Õ.28) and younger (b=
Õ.25). The lack of a relationship between job demands and depersonalization in this study
is consistent with some studies involving the JDC model (de Rijk et al., 1998), but not
others (Landsbergis, 1988; Taris et al., 1999). Janssen and colleagues (1999) also found that
depersonalization was not associated with job demands (i.e. work overload). The signi®cant
relationship between depersonalization and skill discretion, but not between depersonaliza-
tion and decision authority is also inconsistent with prior research. As noted earlier, higher
depersonalization was associated with lower skill discretion (Taris et al., 1999), decision
latitude (skill discretion plus decision authority) (Landsbergis, 1988). The ®ndings were
inconsistent when control was de®ned as decision authority: signi®cant in one study
(Taris et al., 1999) but not in the other (de Rijk et al., 1998). Consistent with prior
research ( Janssen et al., 1999; Lee, and Ashforth, 1996), there was a relationship between

82
Y. RaVerty et al.
depersonalization and social support, but not when demographic and other job character-
istics were controlled for. In addition, since skill discretion was no longer signi®cantly
associated with depersonalization when social support was added to the model, it is possible
that social support may mediate the relationship between skill discretion and depersonaliza-
tion and that part of the relationship between social support and depersonalization may
act indirectly through emotional exhaustion as well as demographic and other job
characteristics.
The inconsistent ®ndings with regard to depersonalization and job demands may be
explained by the in¯ uence of emotional exhaustion, demographic characteristics, and other
job characteristics on burnout, which point to a need for longitudinal research in this area.
In this study, higher depersonalization was primarily associated with higher emotional
exhaustion and, to a lesser degree, with being Caucasian and younger. Unfortunately, the
cross-sectional nature of this study does not permit conclusions to be drawn with regard
to these ®ndings. It is possible, however, that the race of the clients was a factor (approxi-
mately half were non-Caucasian) or their disability status played a role (all were disabled).
It is also plausible that the relationships between demographic characteristics and job
characteristics may be involved, as well as other extraneous variables. Being Caucasian was
signi®cantly associated with having more education, which, in turn, was associated with
having higher skill discretion and decision authority, co-worker support, emotional exhaus-
tion, and personal accomplishment. Being younger, in contrast, was not related to either
demographic or job characteristics. In addition, younger workers were less emotionally
exhausted than older workers. Further research should assess the in¯ uence of additional
personal and situational characteristics on depersonalization.
5.3. Reduced personal accomplishment
The full model explained 42% of the variance in personal accomplishment. Workers with
reduced personal accomplishment were more likely to have lower skill discretion (b=
Õ.40), be younger (b=Õ.24), and to have less education (b=Õ.20). The ®ndings of
this study with regard to personal accomplishment are mostly consistent with prior research.
As noted earlier, personal accomplishment was not associated with job demands in each
study in which this relationship was examined within the context of the JDC model (de
Rijk et al., 1998; Landsbergis, 1988), nor in other studies involving individual determinants
of personal accomplishment ( Janssen et al., 1999). In addition, reduced personal accomplish-
ment was associated with lower control in each study where this relationship was examined:
whether it was de®ned as decision latitude (skill discretion plus decision authority)
(Landsbergis, 1988), as skill discretion (Taris et al., 1999), or as decision authority (Taris
et al., 1999). In addition, Janssen and colleagues (1999) found that reduced personal
accomplishment was signi®cantly associated with `quality of job content’ (such as having
the opportunity to develop new skills, and/or use a variety of skills). Finally, the relationship
between personal accomplishment and social support found in this study is consistent with
prior research ( Janssen et al., 1999; Lee, and Ashforth, 1996). However, since the hierarch-
ical regression analyses are non-signi®cant, it appears that the relationship between social
support and personal accomplishment may act indirectly through demographic and other
job characteristics.
The limitations of this study include a cross-sectional design (which prevents conclusions
regarding causality), a low response rate (62%), and a small sample size (n=164), which
may have been inadequate to detect existing eVects. It is also plausible that burned out
workers may be reporting higher demands, lower control, and less support. In addition,

The Job DemandÐcontrol model and burnout
83
associations may potentially be aVected by self-report bias. Many unresolved questions
about the association between job characteristics and burnout remain. These include
potential intervening variables including personality characteristics (Kushnir, and Melamed,
1991), dispositional attributes such as self-esteem (Lee, and Ashforth, 1996), private self-
consciousness (Kivimaki, and Lindstrom, 1995), depression (Karasek, 1979), active coping
(de Rijk et al., 1998); locus of control (Daniels, and Guppy, 1994; Parkes, 1991), and
managerial behaviour/organizational culture (Baker et al., 1996). In addition, other work
characteristics, such as in¯ uence in the organization (Landsbergis, Schnall, Warren,
Pickering, and Schwartz, 1994) may be associated with burnout.
5.4. Conclusions
The two dimensions of control in the JDC model do not appear to be equally important
as predictors of burnout. In this study, having the capacity to use a range of skills on the
job was more strongly and consistently related to burnout than the degree of authority
given to workers to make their own decisions at work. The lack of a relationship between
decision authority and each burnout dimension was unexpected, although not entirely
inconsistent with prior research. As noted earlier, de Rijk and colleagues (1998) did not
®nd a signi®cant relationship between decision authority and emotional exhaustion. It may
be possible that this relationship was obscured by the relationship between decision authority
and demographic characteristics, which were statistically controlled for in this study. As
noted earlier, higher decision authority was reported by male workers, as well as by higher
educated workers.
The ®ndings of this study also suggest that the explanatory and predictive power of
the JDC may also be substantially improved by shifting to a more diVerentiated measurement
of the control dimension (Kasl, 1996; Wall et al., 1996). Related research con®rms the
importance of diVerentiating between diVerent dimensions of job control (Carayon, and
Zijlstra, 1999; Van der Doef, and Maes, 1999) as well as using a more focused measure of
job demands and ensuring that the measure of control assesses control over speci®c demands
that the worker experiences (Van der Doef, and Maes, 1999).
In spite of its limitations, the present study suggests that the JDC model may provide
a useful theoretical foundation for the study of burnout, but that the control dimensions
need to be evaluated independently since they appear to be diVerentially related to burnout
dimensions.
Acknowledgements
The authors would like to express their sincere gratitude to Dana Bramel, David Cross,
Joseph Schwartz, Stephen Salbod, and two anonymous reviewers for their helpful comments
on an earlier version of this manuscript.
References
Baker, E., Israel, B., Schurman, S.
&
(1996). Role of control and support in occupational stress:
An integrated model. Social Science and Medicine, 43, 1145±1159.
Caplan, R. D., Cobb, S., French, J. R. P., Harrison, R., Pinneau, S. R.
&
(1980). Job Demands
and Worker Health: Main EVects and Occupational DiVerences. Ann Arbor, MI: Institute for Social
Research, University of Michigan.
Carayon, P., Zijlstra, F
&
Studies in the USA and in The Netherlands. Work and Stress, 13, 32±48.
. (1999). Relationship between job control, work pressure and strain:

84
Cordes, C., Dougherty, T. W
Academy of Management Review, 18, 621±656.
Y. RaVerty et al.
&
. (1993). A review and an integration of research on job burnout.
Daniels, K., Guppy, A.
&
(1994). Occupational stress, social support, job control, and psychological
well-being. Human Relations, 12, 1523±1544.
de Jonge, J., Janssen, P. P. M., van Breukelen, G. J. P. (
&
1996). Testing the demands-control-
support model among health-care professionals: A structural equation model. Work & Stress,
10, 209±224.
de Jonge, J.,
Kompier, M. A. J.
(1997). A critical examination of the demand-control-support
model from a work psychological perspective. International Journal of Stress Management, 4,
&
235±258.
de Rijk, A. E., Le Blanc, P. M., Schaufeli, W. B., de Jonge, J.
& (1998). Active coping and need
for control as moderators of the job demands-control model: EVects on burnout. Journal of
Occupational and Organizational Psychology, 71, 1±8.
Janssen, P. M., Schaufeli, W. B., Houkes, I
&
. (1999). Work-related and individual determinants
of the three burnout dimensions. Work and Stress, 13, 74±86.
Johnson, J. V., Hall, E. M
&
. (1988). Job strain, work place social support, and cardiovascular
disease: A cross sectional study of a random sample of the Swedish working population.
American Journal of Public Health, 78, 1336±1342.
Karasek
, R. (1979). Job demands, job decision latitude, and mental strain: Implications for job
redesign. Administrative Science Quarterly, 24, 285±311.
Karasek
, R. (1985). Job Content Questionnaire and User’s Guide. Los Angeles, CA: University of
Southern California.
Karasek, R., Brisson, C., Kawakami, N., Houtman, I., Bongers, P., Amick, B.
&
Job Content Questionnaire ( JCQ): An instrument for internationally comparative assessments
(1998). The
of psychosocial job characteristics. Journal of Occupational Health Psychology, 3, 322±355.
Karasek, R., Theorell, T
& . (1990). Health Work. New York: Basic Books.
. (1996), The in¯ uence of the work environment on cardiovascular health: A historical,
Kasl, S. V
conceptual, and methodological perspective. Journal of Occupational Health Psychology, 1, 42±56.
Kivimaki, M., Lindstrom
&
, K. (1995). EVects of private self-consciousness and control on the
occupational stress-strain relationship. Stress Medicine, 11, 17±26.
Kop, N., Euwema, M., Shaufeli, W.
&
(1999). Burnout, job stress and violent behaviors among
Dutch police oYcers. Work and Stress, 13, 326±340.
Kushnir, T., Melamed, S. (
&
1991). Work-load, perceived control and psychological distress in
Type A/B industrial workers. Journal of Organizational Behavior, 12, 155±168.
Landsbergis,
P. A. (1988). Occupational stress among health care workers: A test of the job demands-
control model. Journal of Organizational Behavior, 9, 217±239.
Landsbergis, P., Schnall, P., Warren, K., Pickering, T.,
between ambulatory blood pressure and alternative formulations of job strain. Scandinavian
Schwartz, J.
(1994). Association
&
Journal of Work, Environment and Health, 20, 349±363.
Lee, R. T., Ashforth, B. E
&
. (1990). On the meaning of Maslach’s three dimensions of burnout.
Journal of Applied Psychology, 75, 743±747.
Lee, R. T., Ashforth, B. E.
&
(1996). A meta-analytic examination of the correlates of the three
dimensions of job burnout. Journal of Applied Psychology, 81, 123±133.
Leiter, M. P
. (1991). Coping patterns as predictors of burnout. The functions of control and escapist
coping patterns. Journal of Occupational Behavior, 12, 123±144.
Leiter,
M. P. (1993). Burnout as a developmental process; consideration of models. In
W. B. Schaufeli, C. Maslach, & T. Marek (Eds.), Professional Burnout: Recent Developments in
Theory and Research (pp. 237±250). Washington, DC: Taylor & Francis.
Maslach,
Maslach
C. (1982). Burnout: The Cost of Caring. New York: Prentice-Hall.
, C. (1993). Burnout: A multidimensional perspective. In W. B. Schaufeli, C. Maslach, &
T. Marek (Eds.), Professional Burnout: Recent Developments in Theory and Research (pp. 19±32).
Washington, DC: Taylor & Francis.
Maslach, C., Jackson, S.
&
Behavior, 2, 99±113.
(1981). The measurement of experienced burnout. Journal of Occupational
(1986). Maslach Burnout Inventory (2nd Ed.). Palo Alto, CA: Consulting
Maslach, C., Jackson, S.
&
Psychologists Press.
Melamed, S., Kushnir, T., Meir, E. I
&
and interactive eVects of perceived control and social support. Journal of Vocational Behavior,
. (1991). Attenuating the impact of job demands: Additive
39, 40±53.

The Job DemandÐcontrol model and burnout
85
Nijhuis, F. N.,
Smulders, P. G. W
. (1996). Die Wirkung von Arbeitsanforderungen und
&
personlichen Kontrollmoglichkeiten auf Gesundheitsbeschwerden und Febzeiten {The in¯ u-
ence of job demands and job control on health complaints and absenteeism}. Zeitschrift fur
Arbeits- und Organisationpsychologie, 40, 173±180.
Nunnaly, J. C.
(1987). Psychometric Theory. New York: Academic Press.
K. R. (1991). Locus of control as moderator: An explanation for additive versus interactive
Parkes,
®ndings in the demand-discretion model of work stress? British Journal of Psychology, 82,
291±312.
Schaufeli, W. B., Maslach, C., Marek
, T. (1993). The future of burnout. In W. B. Schaufeli,
&
C. Maslach, & T. Marek (Eds.), Professional Burnout: Recent Developments in Theory and Research
(pp. 253±259). Washington, DC: Taylor & Francis.
Schreurs, P. J. G.,
Taris, T. W.
&
(1998). Construct validity of the demand-control model: A
double cross-validation approach. Work & Stress, 12, 64±84.
Soderfeldt, M., Soderfeldt, B., Warg, L. E., Ohlson, C. G.
&
(1996). The factor structure of
the Maslach Burnout Inventory in two Swedish human service organizations. Scandinavian
Journal of Psychology, 37, 437±443.
Sonnentag, S., Brodbeck, F. C., Heinbokel, T., Stolte,W.
&
in software development teams. Journal of Occupational and Organizational Psychology, 67,
(1994). Stressor-burnout relationship
327±341.
Taris, T. W., Schreurs, P. J. G., Schaufeli, W. B
& . (1999). Construct validity of the Maslach
Burnout Inventory-General Survey: A two-sample examination of its factor structure and
correlates. Work and Stress, 13, 223±237.
Van der Doef, M., Maes, S
&
. (1999). The Job Demand-Control (-Support) Model and psychological
well-being: A review of 20 years of empirical research. Work and Stress, 13, 87±114.
Wall, T. D., Jackson, P. J., Mullarkey, S., Parker, S. K.
&
(1996). The demands-control model
of job strain: A more speci®c test. Journal of Occupational and Organizational Psychology, 69,
153±166.
Submitted September 1999.
Revised version accepted January 2001.