Results on job demand is negatively related to the self-employer’s work-related stress
Descriptive statistics – means and standard deviations and bivariate correlations – are presented in
Table 15 for the whole dataset (N= 3,743). Results for Hypothesis 1a and Hypothesis 1b are
displayed in Table 14. The direct and interaction effect results on Job Anxiety and Job Depression,
which pertain to Hypotheses 2 to 6, are presented in Tables 16 and 17 separately.
Multicollinearity was checked, with results showing that the largest variance inflation factor was 1.3,
below the value of 10 that is commonly viewed as problematic (Neter et al., 1996). Thus,
multicollinearity is not a major threat to the integrity of the results.
Hypothesis 1a proposes that the self-employed (solo self-employed and self-employed with hiring
employees) experience less work-related stress than individuals in wage work. Hypothesis 1b
proposes that the self-employed (solo self-employed and self-employed with hiring employees)
experience higher work-related stress than individuals in wage work. This study applies the t-test on
two different work-related stresses: job anxiety and job depression in two different comparing groups
(the SEE vs Employees, and the Solo SE vs Employees) based on before- and after-matching
samples. As shown in Table 14, for the SEE vs Employees, before matching the significance of
difference is not strong on job anxiety and job depression (p>0.05), but after matching, the
significance of differences of job depression presented is stronger, with the SEE having significantly
higher level of job depression (p<0.05). For the Solo SE vs Employees, based on before- and after-
matching samples, the Solo SE has significantly lower levels of work-related stress (both with
respect to job anxiety and job depression) (p<0.01). Thus, hypothesis 1a is supported in the Solo SE
vs Employees group, not supported in the SEE vs Employees comparing group. Hypothesis 1b is
tenable in the SEE vs Employees comparing group rather than the Solo SE vs Employees group.
Regarding the following hypotheses:
Hypothesis 2a: job demand is negatively related to the self-employed’s work-related stress
Hypothesis 2b: the self-employed (baseline is employees) moderate the relationship between job
demand and work-related stress
127
Hypothesis 2c: the self-employed with hiring employees (baseline is the solo self-employed)
moderate the relationship between job demand and work-related stress,
The results are shown in Table 16 and Table 17. The results indicate that job demands are closely
associated with job anxiety (p<0.05) and job depression (p<0.01) among all groups. Hence,
Hypothesis 2a is supported. For both job depression and job anxiety, the interaction term job
demand*SE is positively significant, which means the relationship between job demand and work-
related stress is stronger in the self-employee group (p<0.01) Hence, Hypothesis 2b is supported as
well. Moreover, the job demand was significant in both SEE and solos SE group (p<0.01), and the
significance level is the same. Thus Hypothesis 2c is not supported.
The results of testing the following hypotheses:
Hypothesis 3a: job controls are negatively related to the self-employed’s work-related stress
Hypothesis 3b: the self-employed (baseline is employees) moderate the relationship between job
control and work-related stress
Hypothesis 3c: the self-employed with hiring employees (baseline is the solo self-employed)
moderate the relationship between job control and work-related stress,
Firstly, job control exerted a significant impact on job anxiety (p<0.05) and job depression (p<0.05)
within both SEE and solo SE groups, thus Hypothesis 3a is tenable. Job Control*SE was not
significant in the whole dataset. Thus, the results does not offer support for Hypothesis 3b. Finally,
the differences of the significance of job control between SEE and solo SE was not showed in
regressions both on job depression (SEE: p<0.05, SE: p<0.05 ) and on job anxiety (SEE p<0.05; SE:
p<0.05 ). Thus, the results do not support the moderation effect of SEE on the relationship between
job control and job depression, and Hypothesis 3c is only partly supported.
In light of:
Hypothesis 4a: social support is negatively related to the self-employed’s work-related stress
128
Hypothesis 4b: the self-employed (baseline is employees) moderate the relationship between social
support and work-related stress, and
Hypothesis 4c: the self-employed with hiring employees (baseline is the solo self-employed)
moderate the relationship between social support and work-related stress,
The results displayed that the social support did not have significant impacts on both job anxiety
(p>0.05) and job depression (p>0.05) among the SEE, but has significant impact within solo SE
groups on job depression (p<0.01). Thus, the results partly support Hypothesis 4a. Also, the results
showed that the relationship between social support and work-related stress is more significant in the
self-employed group (social support* the self-employed: p<0.05). Thus, Hypothesis 4b has
supportive evidence. As the significances of social support are different between the SEE and the
solo SE groups, Thus, Hypothesis 4c is supported.
With respect to the hypothesis proposing the moderating effect of social support and the differences
of this moderating effect between the two different comparing groups (The SEE vs Employees, and
the Solo SE vs Employees) the results are presented to test the following hypotheses:
H5a: social support moderates the relationship between job demands and the self-employed’s work-
related stress
H5b: the self-employed (baseline is employees) moderate the relationship between the interaction of
social support*job demands and work-related stress, and
H5c: the self-employed with hiring employees (baseline is the solo self-employed) moderate the
relationship between the interaction of social support*job demands and work-related stress,
The interaction term social support*job demands are only significant in the model of the regression on
job depression (p<0.05) among the self-employed rather than the employees. Moreover, the interaction
term social support*job demands are more significant (p<0.05) in a regression on job depression in the
solo self-employed group when compared with SEE group. Therefore, H5b and H5c were also partly
supported. The moderating effect on job demand and job depression was stronger in the self-employed
129
group when it was compared with employees, and it was also stronger in the Solo self-employed
group when it was compared with the solo self-employed.
The last sets of hypotheses are:
H6a: social support moderates the relationship between job controls and the self-employed’s work-
related stress
H6b: the self-employed (baseline is employees) moderate the relationship between the interaction of
social support*job control and work-related stress, and
H6c: the self-employed with hiring employees (baseline is the solo self-employed) moderate the
relationship between the interaction of social support*job control and work-related stress.
From Tables 16 and 17, the results indicate that social support did not significantly moderate the
relationship between job control and work-related stress (regression on job anxiety: p>.05); regression on
job depression: p>0.05). Moreover, this moderating term did not the present significance of variety
between the self-employed and employees. Moreover, the interaction term social support*job control is
not significant among the SEE and solo self-employed in the model of regression on both job anxiety and
job depression (p>0.05). Therefore, results do not provide support for H6c.
130
Table 16 Hierarchy regressions on Job Anxiety
Whole Employees Solo SE SEE
M1 M2 M3
M4 M5
M6 M7 M8 M9
Age 0.00 0.00 0.00 -0.01 -0.01 -0.01 -0.01 0.00 0.00
Sex 0.27 *** 0.28 *** 0.29 *** 0.17 0.18 -0.27 -0.26 0.23 0.13
Single -0.04 -0.04 -0.04 -0.05 -0.03 -0.04 -0.02 -0.71 -0.71
Education 0.12 *** 0.12 *** 0.12 *** 0.11 ** 0.11 ** 0.08 * 0.08 * 0.22 * 0.23 *
Physical Health -0.02 *** -0.02 *** -0.02 *** -0.02 ** -0.02 ** -0.01 -0.01 -0.02 -0.02
Mental Health -0.11 *** -0.11 *** -0.11 *** -0.12 *** -0.12 *** -0.09 *** -0.09 *** -0.16 *** -0.16 ***
log(Job Income) -0.22 *** -0.22 *** -0.22 *** -0.13 ** -0.13 ** -0.06 -0.06 -0.07 -0.07
Self-employed -0.44 *** -0.44 *** -0.44 ***
Job Demand 0.02 *** 0.02 *** 0.02 *** 0.01 ** 0.01 ** 0.02 *** 0.02 *** 0.03 *** 0.03 ***
Job Control -0.03 *** -0.03 *** -0.03 *** -0.05 ** -0.05 ** -0.04 * -0.04 * -0.11 * -0.13 *
Social Support -0.01 -0.01 -0.01 -0.03 -0.03 -0.09 ** -0.10 ** -0.09 -0.10
Job Demand* SE -0.01 *
Job Control*SE 0.03
Social Support*SE 0.01 *
JD*Social Support -0.00 -0.00 -0.04 ** -0.06 *
JC*Social Support 0.00 0.00 0.02 0.01
R2 0.16 0.16 0.17 0.19 0.20 0.13 0.14 0.26 0.28
F 204.4 140.40 126.70 30.28 21.22 13.97 10.40 7.40 5.53
P-Value 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
*** p<.001; ** p<.0.01; * p<.05
131
Table 17 Hierarchy regressions on Job Depression
Whole (Before Matched) Employees(After Matched)
Solo Self-employed
SEE
M10 M11 M12
M13 M14
M15
M16
M17
M18
Age
0.00 0.00
0.00 0.01
0.01
0.01 0.01 0.01 0.01
Sex
0.21
***
0.21
***
0.21 *** 0.08
0.09
0.20 * 0.18
*
0.26 0.30
Single
-0.02 -0.02 -0.02
-0.05
-0.01 -0.21 -0.18 -0.28 -0.30
Education
0.08
***
0.09
***
0.09 *** 0.07 *
0.07
*
0.02 0.02 0.15
*
0.15
*
Physical Health -0.02 *** -0.02
***
-0.02 ***
-0.02
***
-0.02 *** -0.02
**
-0.02
**
-0.01 -0.01
Mental Health -0.12 *** -0.12
***
-0.12 ***
-0.11
***
-0.11 *** -0.09
***
-0.09 *** -0.17
***
-0.17 ***
log(Job Income)
-0.18 *** -0.18
***
-0.18 ***
-0.10
**
-0.11
**
-0.04 -0.04 -0.03 -0.03
The self-employed -0.47 *** -0.47
***
-0.47 ***
Job Demand
0.02
***
0.01
***
0.01 *** 0.01
***
0.01
*** 0.01
***
0.01 *** 0.02
**
0.02
**
Job Control
-0.04 *** -0.04
***
-0.04 ***
-0.04
**
-0.03
**
-0.03 * -0.04
*
-0.01
*
-0.01
*
Social Support
-0.03 -0.03 -0.03
-0.02
-0.02 -0.04 * -0.04
*
-0.03 -0.03
Job Demand*SE
-0.01
*
Job Control*SE
0.01
Social Support*SE
0.01
JD*Social Support -0.00 * -0.00 -0.00 -0.00
JC*Social Support
0.00 0.00
0.01 0.03
R2
0.23 0.24
0.25 0.22
0.23
0.17 0.18 0.36 0.37
F
312.30 214.50 176.40
36.27
25.71
20.49 14.71 11.79
8.20
P-Value
0.00 0.00
0.00 0.00
0.00
0.00 0.00 0.00 0.00
*** p<.001; ** p<.0.01; * p<.05
Par stands for Partner supports; Fri stands for Friend supports; Fam stands for Family supports; JD stands for Job Demand; JC stands for Job Control
132
3.5 Conclusion and Discussion
Who experience higher level of work-related stress? The self-employed or employees?
The results from previous studies (Andersson, 2008; Blanchflower, 2004; Harris et al.,
1999; Jamal, 2009; Patzelt and Shepherd, 2011; Baron, Franklin and Hmieleski, 2016;
Hessels, Rietveld and Zwan, 2017) do not enable consistent conclusions. Hessels,
Rietveld and Zwan (2017) suggested that this variety was due to the different stress
measurements, methods and samples applied in the research. By involving the matching
approach, dividing the self-employed group into the SEE and the Solo SE, and applying
two kinds of work-related stress: job anxiety and job depression, Results of the present
research indicate that different samples, methods and measurements can result in different
conclusions. Before matching, the work-related stress did not present differences between
the SEE and employees. Dramatically, after matching, the SEE had a significantly higher
level of job depression and job anxiety than employees. This finding is consistent with
previous research noting that the self-employed experience higher stress levels (Patzelt
and Shepherd, 2011; Baron, Franklin and Hmieleski, 2013; Hessels, Rietveld and Zwan,
2017), and at the same time, it contract to other results which suggested that the self-
employed are expected to experience lower or the same stress levels (Patzelt and
Shepherd, 2011; Baron, Franklin and Hmieleski, 2016; Hessels, Rietveld and Zwan,
2017)). However, this study also found that the solo self-employed experience
significantly lower work-related stress levels based on both before- and after-matching
samples. This is contract to previous findings that the
133
self-employed experience lower stress levels (Patzelt and Shepherd, 2011; Baron,
Franklin and Hmieleski, 2016; Hessels, Rietveld and Zwan, 2017).
The present research also sheds light on factors that contribute to the self-employed’s
work-related stress. Based on the three factors – job demand, job control and social
support, from the JDCS model – this is found that job demand, job control have a direct
predicting impact on the self-employed’s job anxiety and job depression, which is in
agreement with previous research (Hessels, Rietveld and Zwan, 2017). Moreover, the R
square value increased by 0.4, while normally 0.2 is seen as significantly contributing to
the model fit (Vaughn, 2008). Social support has direct impact on work-related stress
only within solo SE group. Morever, it has a moderating effect on the relationship
between job demand and job depression in this group, with results partially supporting the
interaction relationship in the JDCS model. Indeed, both van der Doef and Maes (1999)
and this study found unsatisfactory evidence for the interactive hypotheses of the JDCS
model. Even though the moderating term only contributed to the model fit by an increase
of 0.1, since interaction patterns are running counter to the buffer hypotheses observed in
the studies included in our analyses, we do not believe the buffer hypothesis to be a
‘zombie theory’. Moreover, this is believed that far more support for the buffer
hypothesis will be obtained in the future once the matching principle has been fully
realised in empirical tests of the JDCS model.
Our findings also offer new evidence on differences in direct and indirect impacts of
work-related stress factors between the self-employed and employees, and between
134
SEE and solo self-employed groups. For direct impacts, job control and job demand does
not differ between the self-employed and employees, or between SEE and solo SE
groups. In the comparison group of SEE vs Solo SE, social support has a stronger
connection with work-related stress within the solo SE group. On the other hand, it is
interesting to see that the relationship between the interaction term of job demand* social
support and job depression is stronger among the self-employed when compared with
employee groups. More specifically, within the self-employed, the interaction term of job
demand* social support has a stronger impact within the solo self-employed in terms of
job anxiety. On the other hand, job control*social support does not show any significant
impact among the groups.
