Psychological Reports, 2012, 110, 2, 547-554.

© Psychological Reports 2012

Demographic and occupational

CORRELATES of workaholisM1

Toon W. Taris, Ilona van BeeK,  and Wilmar B. Schaufeli

Utrecht University

Summary.—Drawing on a convenience sample of 9,160 Dutch employees,

the present study examined whether commonly held ideas about the associations
between demographic, professional, and occupational characteristics and worka-
holism would be observed. For example, it is sometimes assumed that managers
are more likely to display workaholic tendencies than others. Analysis of variance
was used to relate workaholism scores (measured as the combination of working
excessively and working compulsively) to participant age, sex, employment status
(self-employed or not), profession, and occupational sector. Relatively high average
scores on workaholism were obtained by workers in the agriculture, construction,
communication, consultancy, and commerce/trade sectors, as well as managers
and higher professionals. Low scores were found for those in the public administra-
tion and services industry sectors, and for nurses, social workers, and paramedics.
The other characteristics were not or only weakly related to workaholism.

A body of research has addressed the correlates of workaholism—the

compulsion or the uncontrollable need to work incessantly (Oates, 1971).
Previous findings show that workaholism is often associated with adverse
outcomes. For example, workaholics are likely to experience work–home
conflict (Bonebright, Clay, & Ankenmann, 2000), report higher stress and
health complaints than others (Shimazu, Schaufeli, & Taris, 2010), and ex-
perience low satisfaction with work (Burke & MacDermid, 1999). Final-
ly, workaholism has been associated with personality traits such as per-
fectionism and neuroticism (Clark, Lelchook, & Taylor, 2010). Although
many of us have specific ideas about the characteristics of the “typical”
workaholic (for example, managers are often believed to be more likely
to display workaholic behaviors, such as working excessively long hours;
cf. Brett & Stroh, 2003), the demographic and occupational profile of the
“typical” workaholic has not yet been characterized. Therefore, at present
it is unknown whether and to what extent common stereotypes concern-
ing the characteristics of workaholics actually hold up. In this study, the
demographic and occupational correlates of workaholism were examined.
Method
Participants
The sample included 9,160 Dutch workers (56% men; M age = 38.0
1
Address correspondence to Dr T. Taris, Department of Work and Organizational Psychol-
ogy, Utrecht University, P.O. Box 80.140, NL-3508 TC Utrecht, The Netherlands or e-mail
([email protected]).

DOI 10.2466/03.09.17.PR0.110.2.547-554 ISSN 0033-2941

548 T. W. Taris, et al.

yr., SD = 10.4) who participated in nine independent studies on worka-

holism during 2001–2008. Six of these (overall N = 3,987, 51% overall re-
sponse rate) were conducted as part of regular health audits conducted
by various Dutch occupational health services. One study (N = 4,767, re-
sponse rate unknown) was conducted among the readers of a popular
Dutch psychology magazine. This magazine placed an Internet link on
its web site that could be clicked by those interested in participating in an
online survey on work and well-being. The remaining studies (two stud-
ies, total N = 406, overall response rate unknown) were conducted as part
of two master’s theses. Parts of the overall data set were used in other
publications (e.g., Schaufeli, Taris, & Bakker, 2008; Schaufeli, Taris, & van
Rhenen, 2008; Schaufeli, Bakker, van der Heijden, & Prins, 2009; Shimazu,
et al., 2010). The overall data set included in the present manuscript has
not been used in previous research. For the current study, its character-
istics were compared to those of the general Dutch working population
of 16 years and older using figures of the Dutch census (Centraal Bureau
voor de Statistiek, 2009). This comparison showed that the sample was
roughly comparable to the Dutch population in terms of gender (sam-
ple: 45% female; population: 51% female) and age (sample: M age = 38 yr.;
population: M age = 40 yr.). About 38% of the sample was employed in
the health and social work sector (as compared to 12% of the population),
whereas workers in most other sectors were slightly underrepresented.
Thus, the sample is not representative of the Dutch labor market in terms
of the sectors included, meaning that the findings should be interpreted
with caution.
Measures
Workaholism was measured with the Dutch Work Addiction Scale
(Schaufeli, Shimazu, & Taris, 2009). In line with current theorizing (e.g.,
Scott, Moore, & Miceli, 1997; Shimazu & Schaufeli, 2009), this instrument
consists of two subscales. The most obvious characteristic of workaholics
is they work excessively hard and beyond what is required. This behav-
ioral aspect of workaholism is measured using the Control tendency scale
of Robinson’s Work Addiction Risk test (1999; relabeled Working Exces-
sively: nine items, including “I seem to be in a hurry and racing against
the clock” and “I find myself continuing to work after my coworkers have
called it quits,” 1: Almost never, 2: Sometimes, 3: Often, 4: Almost always;
α = .81). As workers may work not only because of workaholism, but for a
variety of reasons, this subscale is complemented with a subscale tapping
the cognitive/compulsive aspect of workaholism. This aspect is measured
with Spence and Robbins’s Drive scale (1992; labeled Working compul-
sively: seven items, including “I feel that there’s something inside me that
drives me to work hard” and “It is important to me to work hard, even if
 Correlates of workaholism 549

I do not enjoy what I am doing,” 1: Almost never, 2: Sometimes, 3: Often,

4: Almost always; α = .84). It is high scores on both subscales that signifies
workaholism, in that it is the inner compulsion to work hard that turns
working excessively into workaholism (Schaufeli, Shimazu, et al., 2009).
The correlation among the two subscales was .64 (p < .001), providing fur-
ther justification for combining these two aspects of workaholism. De-
mographic and occupational characteristics were measured by five items
tapping participants’ sex, age, labor market sector, profession, and em-
ployment status (self-employed or not), respectively.
Statistical Analysis
Univariate analysis of variance (ANOVA) was conducted to examine
workaholism across the study variables. As the sample was very large,
even practically irrelevant differences among groups may become statis-
tically significant. Moreover, the exploratory nature of the study may in-
crease the risk of capitalization on chance. To offset these issues to some
extent, a more rigid alpha level of p < .001 was employed rather than the
more commonly used 5% threshold. Moreover, effect sizes (η2 and, where
applicable, Cohen’s d) are provided for each analysis; effect sizes of η2 < .01
(1% of variance accounted for) or below will be considered irrelevant from
a practical point of view and are not further discussed.
Results
Men (M = 2.19, SD = .50) did not differ significantly from women
(M = 2.21, SD = .50) in their scores on workaholism (F1,9114 = 3.45, p > .001;
η2 < 0.001, Cohen’s d = 0.04). Older employees (age > 45 yr.) obtained on
average slightly lower scores on workaholism (M = 2.15, SD = 0.51) than
younger employees (age < 30 yr.; M = 2.23, SD = 0.50) and middle-aged
employees (M = 2.22, SD = 0.49; F2,9113 = 20.10, p < .001). However, although
this effect was statistically significant, with an η2 of only 0.004, its prac-
tical relevance is negligible. Further, although the self-employed ob-
tained higher mean scores on workaholism (M = 2.25, SD = 0.55) than oth-
ers (M = 2.19, SD = 0.49), this difference was not significant (F1,9114 = 9.66,
p > .001; η2 = 0.001, Cohen’s d = 0.12).
Table 1 shows that different labor market sectors varied significantly
in terms of the average level of workaholism of workers in these sectors
(F14,9143 = 19.25, p < .001; η2 = 0.03). Table 1 provides the means and standard
deviations on workaholism as a function of occupational sector. Work-
ers in the communication, consultancy, and commerce/trade sectors ob-
tained, on average, higher scores on workaholism than others, with mean
scores in the top five of high-scoring sectors. However, workers in the
agricultural and construction sectors were also in this top five. Inspec-
tion of the scores on the two factors constituting the overall workaholism
550 T. W. Taris, et al.

Table 1
Means and Standard Deviations of Workaholism as
a Function of Occupational Sector and Profession

Sector N Workaholism Working Working

Fa = 19.25, Compulsivelyc Excessivelyc
η2 = 0.03 Fa = 4.04, Fa = 5.55,
η2 = 0.01 η2 = 0.01
M SD M SD M SD
Agriculture 39 2.55 .57 2.48 .66 2.61 .59
Construction 111 2.38 .49 2.34 .61 2.49 .50
Communication 664 2.35 .44 2.15 .54 2.51 .43
Consultancy 1,098 2.32 .49 2.17 .59 2.45 .49
Commerce/Trade 212 2.27 .57 2.19 .64 2.34 .59
Banking and insurance 150 2.26 .54 2.11 .59 2.37 .56
Transport 59 2.26 .51 2.11 .61 2.37 .51
Culture 155 2.22 .55 2.13 .64 2.29 .56
Education 640 2.21 .52 2.07 .62 2.32 .51
Industry 634 2.20 .55 2.24 .64 2.16 .61
Health and social work 3,522 2.17 .47 1.99 .57 2.31 .48
Law enforcement, military 434 2.16 .52 2.03 .60 2.27 .55
Other 167 2.12 .46 1.89 .53 2.30 .48
Public administration 587 2.09 .46 1.94 .55 2.20 .47
Services 686 2.08 .54 2.12 .64 2.05 .56
Profession Fb = 19.02 Fb = 3.68, Fb = 15.55,
η2 = 0.03 η2 = 0.01 η2 = 0.03
Manager 1,280 2.35 .47 2.14 .56 2.51 .47
Higher professional 610 2.34 .47 2.19 .56 2.45 .47
Sales 163 2.26 .56 2.21 .63 2.30 .58
Artist 106 2.24 .50 2.13 .57 2.33 .51
Blue collar 150 2.24 .58 2.16 .67 2.30 .61
Teacher, primary school 102 2.23 .48 2.13 .58 2.30 .48
Medical resident 2,119 2.22 .44 2.02 .57 2.37 .43
Pink collar 152 2.20 .61 2.20 .66 2.20 .64
White collar 1,223 2.19 .49 2.03 .59 2.32 .50
Teacher, secondary school 305 2.18 .57 2.06 .68 2.28 .54
Executive 505 2.16 .51 2.01 .59 2.29 .55
Physician 166 2.14 .51 1.94 .59 2.31 .53
Lower professional 257 2.10 .49 1.98 .58 2.20 .49
Paramedic 305 2.06 .46 1.90 .56 2.20 .48
Social worker 383 2.03 .49 1.90 .56 2.14 .51
Nurse 225 2.02 .50 1.93 .56 2.11 .54
Note.—aThis F value has (14, 9143) degrees of freedom. bThis F value has (15, 8045) degrees
of freedom. cThe multivariate Fs for the effects of Sector and Profession were significant,
FSector(28, 14,432) = 4.01, p < .001; η2 = 0.01, and FProfession(30, 14,432) = 10.4, p < .001, η2 = 0.02.
*p < .001.
 Correlates of workaholism 551

score (i.e., working compulsively and working excessively, Table 1) shows

that these five sectors usually also belonged to the top five of sectors on
these two measures. Apparently, workers in the sectors with high mean
workaholism scores reported, on average, slightly higher scores on both
working compulsively and excessively than workers in other sectors. The
lowest scores were found among workers in the services industry, pub-
lic administration, law enforcement/military, and health/social work sec-
tors, as well as among workers in non-specified sectors (“other”). Again,
the average scores of these sectors on working excessively/compulsively
were also among the lowest of the sectors included in Table 1.
As regards profession, it was found that the professions includ-
ed in the current research varied significantly on workaholism as well
(F15,8045 = 19.02, p < .001; η2 = 0.034). Table 1 shows managers and higher-lev-
el professionals obtained, on average, high scores on workaholism, where-
as nurses, social workers, and paramedics, on average, reported lower
workaholism. The scores on the two workaholism indicators were as ex-
pected: managers and higher professionals reported on average relatively
high scores on both working excessively and compulsively. The reverse
applied to paramedics, social workers, and nurses.
Discussion
Two findings of the present study are worth noting. The results par-
tially confirm common stereotypes about the risk groups for worka-
holism: managers and high-level professionals and those working in the
labor market sectors such as the communication, commerce/trade, and
consultancy sectors are particularly more likely to work both compulsive-
ly and excessively; the reverse applied to those in public administration.
Interestingly, the size of these effects is relatively small. Although the dif-
ferences in workaholism among these groups are statistically significant,
this is mainly due to the large sample size. The differences themselves are
not particularly impressive. Thus, although there is some justification for
commonly held ideas about typical risk groups for workaholism, the vari-
ation within these risk groups (e.g., managers) is considerable, with sub-
stantial numbers of members of these risk groups scoring below the mean
scores of low-risk groups for workaholism (cf. Table 1). These findings
contribute to the understanding of workaholism, in that they qualify ideas
about the typical risk groups for workaholism.
Several groups with high mean workaholism scores, especially con-
struction and agricultural sectors, had not previously been recognized.
This somewhat unexpected finding is not only due to the fact that work-
ers in these occupations must respond to high environmental demands,
since these groups also obtained the highest average scores on the “work-
ing compulsively” indicator of workaholism (see Table 1). Thus, workers
552 T. W. Taris, et al.

in these occupations work excessively and compulsively. It is possible that

having to meet these high environmental demands instills the compul-
sion to work hard (e.g., to feel guilty when not working) due to a sociali-
zation process. Conversely, it is conceivable workers with a workaholic
predisposition actively seek out occupations that allow them to live up
to their disposition, in that in occupations like farming and construction
work, there is always something to do—the environmental demands in
these occupations are high and will offer workaholics excellent opportu-
nities to invest too much time in their jobs. The latter position is consistent
with Benjamin Schneider’s Attraction-Selection-Attrition theory (Schnei-
der, Smith, & Goldstein, 2000). Which of these interpretations is correct
can only be tested using a longitudinal design that allows researchers to
follow the vocational trajectories of individual workers across time.
The main limitations of this study are as follows. The cross-sectional
design implies that no causal conclusions can be drawn, e.g., whether be-
ing in a particular profession “causes” workaholism, or whether worka-
holics seek professions in which they can live up to their workaholic ten-
dencies. The fact that a convenience sample was used means caution must
be taken when generalizing these findings to other members of the groups
under study. A third issue is the scores of the participants on workaholism
were on average not especially problematic, with the scores in Table 1
ranging from 2.02 to 2.55 (which corresponds with “Sometimes” to “Of-
ten” responses to the survey items). This was to be expected, since it is un-
likely that serious psychological problems will be reported by a large part
of a nonclinical population. However, this implies that at least some of the
participants in some of the study groups report high workaholism. For ex-
ample, for a normally distributed variable, a mean score of 2.55 (SD = 0.57,
cf. Table 1) means that 17% of this group reports a score of 3.12 or higher
on workaholism, corresponding with the “Almost always” response to
the survey items. Finally, the large sample size means that even very small
and practically irrelevant effects will reach statistical significance. To ad-
dress this issue, a conservative alpha level of .001 was used and consid-
ered effects accounting for less than 1% of the variance in workaholism as
practically irrelevant. Applying these procedures yielded two statistically
significant and practically relevant effects (i.e., for occupational sector and
profession), but even here the percentage of variance accounted for was
relatively low (3.4% at most). Thus, the practical relevance of the findings
should not be overstated.
In spite of these limitations, the study has some implications. Clearly,
the prototypical picture of a workaholic is too narrow, in that workahol-
ics are not only well-paid professionals and managers, but also may be
construction workers or farmers. Indeed, these occupational and demo-
 Correlates of workaholism 553

graphic characteristics were only weakly related to workaholism, suggest-

ing that research on the antecedents of workaholism should focus on en-
vironmental and personality factors instead. Further, insofar as there were
associations between occupational factors and workaholism, it is as yet
unclear whether these associations are caused by a process in which the
environment stimulates the development of workaholism (e.g., socializa-
tion processes may be important here), or whether workers with a worka-
holic tendency actively seek an environment in which they can live up to
this disposition. Again, this would seem a fruitful area for future research.
References
Bonebright, C. A., Clay, D. L., & Ankenmann, R. D. (2000) The relationship of worka-
holism with work-life conflict, life satisfaction, and purpose in life. Journal of
Counseling Psychology, 47, 469-477.
Brett, J. M., & Stroh, L. K. (2003) Working 61 plus hours per week: why do managers
do it? Journal of Applied Psychology, 88, 67-78.
Burke, R. J., & MacDermid, G. (1999)  Are workaholics job satisfied and successful in
their careers? Career Development International, 4, 277-282.
Centraal Bureau voor de Statistiek. (2009) [Structure of the Dutch labour market]. Re-
trieved on May 5th, 2011, from http://www.cbs.nl. [in Dutch]
Clark, M. A., Lelchook, A. M., & Taylor, M. L. (2010)  Beyond the Big Five: how nar-
cissism, perfectionism, and dispositional affect relate to workaholism. Personality
and Individual Differences, 48, 786-791.
Oates, W. (1971)  Confessions of a workaholic. New York: World.
Robinson, B. E. (1999) The Work Addiction Risk Test: development of a tentative
measure of workaholism. Perceptual and Motor Skills, 88, 199-210.
Schaufeli, W. B., Bakker, A. B., van der Heijden, F. M. M. A., & Prins, J. T. (2009) Worka-
holism, burnout, and well-being among junior doctors: the mediating role of role
conflict. Work and Stress, 23, 155-172.
Schaufeli, W. B., Shimazu, A., & Taris, T. W. (2009)  Being driven to work exceptionally
hard: the evaluation of a two-factor measure of workaholism in The Netherlands
and Japan. Cross-Cultural Research, 43, 320-348.
Schaufeli, W. B., Taris, T. W., & Bakker, A. B. (2008) It takes two to tango: worka-
holism is working excessively and working compulsively. In R. J. Burke & C. L.
Cooper (Eds.), The long work hours culture: causes, consequences, and choices. Bingley,
UK: Emerald. Pp. 203-226.
Schaufeli, W. B., Taris, T. W., & van Rhenen, W. (2008) Workaholism, burnout, and en-
gagement: three of a kind or three different kinds of employee well-being. Applied
Psychology: an International Review, 57, 173-203.
Schneider, B., Smith, D. B., & Goldstein, H. W. (2000) Attraction-selection-attrition:
toward a person–environment psychology of organizations. In B. W. Walsh, K.
H. Craik, & R. H. Price (Eds.), Person–environment psychology: new directions and
perspectives. (2nd ed.) Mahwah, NJ: Erlbaum. Pp. 61-86.
Scott, K. S., Moore, K. S., & Miceli, M. P. (1997) An exploration of the meaning and
consequences of workaholism: a conceptual integration and extension. Journal of
Organizational Behavior, 28, 111-136.
554 T. W. Taris, et al.

Shimazu, A., & Schaufeli, W. B. (2009) Is workaholism good or bad for employee well-
being? The distinctiveness of workaholism and work engagement among Japa-
nese employees. Industrial Health, 47, 495-502.
Shimazu, A., Schaufeli, W. B., & Taris, T. W. (2010)  How does workaholism affect
worker health and performance? The mediating role of coping. International Jour-
nal of Behavioral Medicine, 17, 154-160.
Spence, J. T., & Robbins, A. S. (1992)  Workaholism: definition, measurement, and pre-
liminary results. Journal of Personality Assessment, 58, 160-178.

Accepted March 6, 2012.

Copyright of Psychological Reports is the property of Ammons Scientific, Ltd. 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.