Job Demand-Control (JDC) model on wellbeing
In 1979, Robert Karasek introduced a seminal model that outlines the impact of adverse job
characteristics on health and wellbeing the Job Demand-Control (JDC) model. Karasek (1979)
identified job demands and job control as essential job characteristics influencing work-related stress.
‘High stress’ jobs are those with a combination of high job demand and low levels of job control.
High-job demand with a high level of control would not be associated with stress because these are
active jobs which allow the individual to develop proactive behaviours that can increase motivation
to perform and learn (Karasek, 1979). ‘Passive jobs’ however, are characterised by low demand and
low control and are considered to be dissatisfying. According to Fox et al. (1993, p. 290) ‘when
employees adapt to low-control and low-demand situations, they tend to find it difficult to make
sound judgments and address the problems, and challenges that they may be confronted with.’ The
following years, social support was integrated into the model as a further fundamental characteristic
of the work environment, which be used to reduce stress in working environment, after that, named
the Job Demand-Control-(Support) (JDCS) model (Johnson and Hall, 1988; Johnson, Hall, and
Theorell, 1989 ).
Thus The JDCS model has three components: job demands, job control, and social support (Karasek
and Theorell, 1990). Job demands are originally defined as ‘psychological stressors involved in
accomplishing the workload’ (Karasek and Theorell, 1990:291). Job control (originally decision
latitude) is the extent to which an employee has authority to make decisions and utilise skills
concerning the job, while social support is characterised by helpful relations with supervisors and
coworkers. According to hypotheses suggested in the models, the JDC/JDCS model contains two
107
kinds of relationship between JDC/JDCS dimensions and job-related wellbeing: 1) the additive effect
of demands, control, and social support on general psychological wellbeing and 2) the interactive
effects predicted by the buffer hypotheses of the JDC/JDCS model: interaction between demands and
controls which affect wellbeing, and the interactive effect of social support on the workplace as a
third dimension.
With reviewing the previous literature by applying the JDCS model, the additive effects are
consistently found when sufficiently large samples are employed. For example, reviewing of 20
years of empirical research using Karasek’s model confirmed that high demand and low control work
environments are associated with lower psychological wellbeing and job satisfaction, burnout and
other forms of psychological distress (Van der Doef and Maes, 1999), and significantly impact on
employee wellbeing (Noblet, 2003). An early study by Marshall, Barnett and Sayer (1997) involving
600 manufacturing and services industries in the United States found that job demands significantly
affect workers’ psychological distress.
Secondly, turning towards to the interactive effects, the empirical status of the interactive hypotheses is
less conclusive: Only weak empirical support for multiplicative effects has been obtained to date. For
example, Van der Doef and Maes (1999) report that out of 31 studies that examined the moderating effect
of job control on the relationship between job demands and wellbeing, only 5 partially supported the
buffering hypothesis of the JDC model. For instance, similar to Pelfrene et al. (2002) who did not find
evidence for buffering effect of job control on the relationship between job demands and psychological
distress, neither Pomaki and Anagnostopoulou (2003) nor Rasku and Kinnunnen (2003) found buffering
effect on teachers’ wellness outcomes. Testing the buffer hypothesis of the JDC model,
Niedhammer et al. (2008) also did not find evidence of the interaction between job demands and job
108
control on health outcomes in self-reported health, sickness absence and work injury among French
workers. This issue was recently addressed in an important theoretical article by Taris (2006).
Interpreting the body of evidence presented by van der Doef and Maes (1999), Taris raised the
question whether the interactive hypothesis is a ‘‘zombie theory’’ that should have died from lack of
empirical evidence but persists in theoretical debate and empirical research. Therefore, this study
aims to provide more empirical evidence to answer the Taris’s question.
Job demand is typically operationalised in terms of quantitative aspects such as workload and time
pressure (Karasek, 1989; van der Doef and Maes, 1999). However, role conflicts, as well as physical
and emotional demands, are also frequently employed to measure job demands recently.
(Gunnarsson, 2010). Within entrepreneurship, most research on the characteristics of the self-
employed found that they report high job demands and a high workload (Stephan and Roesler, 2010).
Working conditions in micro-enterprises often entail working long, irregular and arduous hours with
a great deal of time pressure and heavy workloads (Lindstrom et al., 2000). These factors might have
a negative influence on work-related stress.
Therefore, H2a: job demands are positively related to the self-employed’s work-related stress
The second job characteristic, job control (also termed decision latitude), refers to the extent to which a
person is capable of controlling their tasks and general work activity. Self-employment and psychosocial
working conditions in micro-enterprises often mean close relationships, flexibility and control: these
factors can facilitate a balance between work and family responsibilities, reduce stress and promote good
health. Several studies show that the self-employed have very high decision authority, control how work
is organised and control how resources are distributed at their workplace,
109
as they often own their enterprise (Hundley, 2001; Stephan and Roesler, 2010). From the entrepreneur’s
perspective, on the aspect of job control, objectively, entrepreneurs have very high decision authority as
they own their enterprise and control how work is organised and how resources (e.g., time, money,
assets) are distributed at their workplace (Rau et al. 2008). Based on the finding that high job control is
beneficial for employees’ health and wellbeing (de Lange et al., 2003), Thus it is expected that
entrepreneurs experience better health compared, when they report higher job control.
Hence, H3a: job controls are negatively related to the self-employed’s work-related stress
As mentioned above, the JDC model was extended by integrating social support in the workplace as
a third dimension. It has been suggested that social support at work, the positive or helpful social
interaction available from management and co-workers, could be a significant factor in the ethology
of stress for IS professionals (Thong and Yap, 2000). A sound support may appear to improve coping
of work-related stress.(Johnson and Hall, 1988). Considerable researches have indicated that both
work and non-work related social support reduces, or buffers the adverse impact of exposure to
work-related job stress, it has been suggested that such support can also be counterproductive to
psychological wellbeing (Dollard et al., 2000).
Therefore, H4a: social support is negatively related to the self-employed’s work-related stress
According to the buffer hypothesis of the JDCS model, social support moderates the impacts of job
demand and job control of work-related stress. However, a limited number of researches have conducted
study on this moderating effect. Results addressing the moderating effect of social support were
inconsistent (e.g., Chay, 1993; Rahim, 1996). Some researches support the moderating effects of
110
social support. For example, Chay (1993) found support for the moderating effect such that those
individuals with high social support were little affected by decrements in job discretion, whereas
the psychological wellbeing of those with low social support was affected by low job discretion.
Moreover, both Riolli and Savicki (2003) and Thong and Yap (2000) have posited that social
support can provide a moderating effect in the stress sequence for information system
professionals. Rahim (1996), on the other hand, did not find support for the moderating effect of
social support for either entrepreneurs or managers. Cohen and Wills (1985) suggested that one
reason for this lack of consistency regarding the moderating effect of social support is that
research has not necessarily matched the source of support with the domain of stress. However,
to explore this issue with new evidence, this study address the perspective of JDCS model to
assume
H5a: social support moderates the relationship between job demands and the self-employed’s
work-related stress
H6a: social support moderates the relationship between job controls and the self-employed’s
work-related stress
