EXAMINING THE EFFECTS OF WORK-FAMILY CONFLICT.
AN EXPLORATORY STUDY ON SICKNESS PRESENTEEISM
CRISTIANA CĂTĂLINA CICEI
National School of Political Studies and Public Administration,
College of Communication and Public Relations
Abstract
This study explores the relationship between work-family conflict and attendance
dynamics, having in view the case of sickness presenteeism. We presume that the negative
interference of work with family can predict if employees attend work when ill. The
research was conducted on a convenience sample of composed of 126 employees (M=41.6,
SD=5.50), working in public and private Romanian organizations. The majority of the
participants were from dual-earner couples and over a half had dependent care
responsibilities. Work-family conflict was assessed using 9 items measuring the
interference between work and family from Carlson, Kacmar and Williams (2000)
multidimensional WFC scale and one item measure of sickness presenteeism (Aronsson,
Gustafsson, & Dallner, 2000). Results indicated that work-family conflict predicted
sickness presenteeism, but in a small extent. The generalization of results must be made
with cautions, due to the small sample size, and other research limitations, including the
cross-sectional design of the study, sample composition and the use of self-report measures.
Implications and future research directions are discussed.
Keywords: work-family conflict, sickness presenteeism, health, attendance dynamics.
1. THEORETICAL FRAMEWORK
According to Carlson, Kacmar and Williams (2000) work–family conflict
(WFC) represents a topic that is intensively investigated in contemporary
organizational behavior research, for practitioners, as well as academics, the
management of the two domains’ interface gaining its centrality (Rothbard &
Dumas, 2006). Amstad, Meier, Fasel, Elfering, and Semmer (2011) indicate that
work–family conflict is frequently perceived as a potential stressor that impacts
well-being and behavior.
*
Corresponding author: Cristiana Cătălina Cicei
Email: [email protected]
4
Starting from the canonical definition introduced by Greenhaus and Beutell
(1985), Bellavia and Frone (2005) assert the bidirectional nature of WFC; thus, two
directions can be traced: family interfering with work (family-work conflict) and
work interfering with family (work-family conflict) (Bellavia & Frone, 2005).
Also, as Brough and O’Driscoll (2005) indicate, three distinct forms of WFC were
introduced by Greenhaus and Beutell (1985), namely time, strain and behaviourbased conflict (Brough & O’Driscoll, 2005).
According to Frone (2003), the possible consequences of WFC have been
examined in an increasing number of studies. Allen, Herst, Bruck and Sutton
(2000) introduced an effective taxonomy that distinguished between: work-related
(e.g. job satisfaction and performance, commitment, turnover and absenteeism),
non-work related (such as family, marital, or life satisfaction), and stress related
(including somatic symptoms, general psychological strain, burnout) work-family
conflict outcomes.
Focusing on the last class of effects, meta-analytic studies indicate a weighted
mean correlation of .29 between WFC and physical symptoms or somatic
complaints (Allen et al., 2000), and of .28 between WIF and health problems
(Amstad et al., 2011). Work-family conflict can also be associated with healthrelated attendance dynamics, in their study, Jansen, Kant, van Amelsvoort,
Kristensen, Swaen and Nijhuis (2006) indicating that WFC is linked to sickness
absenteeism.
Although it represents a recent subject of research, information regarding its
antecedents being scarce (Böckerman & Laukkanen, 2010), sickness presenteeism
(SP) can be considered a topic with arising popularity (Biron & Saksvik, 2009).
According to Hemp (2004), although less visible, presenteeism seems to imply
higher costs than absenteeism. Also, Hansen and Andersen (2008) provide
evidences that sickness presenteeism can have the same occurrence as sickness
absenteeism.
Johns (2010, p. 521) defines sickness presenteeism as „attending work while
ill”, suggesting to embed work attitudes and experiences (Johns, 2010, p. 535), in
the SP research.
In their empirical study, Aronsson, Gustafsson, and Dallner (2000) indicate
that middle-aged employees and parents can experience increased sickness
presenteeism, women exhibiting mildly elevated SP rates than men (Aronsson et
al., 2000). Böckerman and Laukkanen (2010) provide evidence that among others,
regular overtime, working full-time, and more than 48 hours/week can enhance SP.
In addition, research indicated that high job demands can encompass in time
presenteeism (Demerouti, Le Blanc, Bakker, Schaufeli, & Hox, 2009).
Having in view these findings, we can observe that SP can be associated with
demographic and work-related factors that are pellicular to work-family conflict.
Among other work experiences, Johns (2011) examined work-family conflict
and the connections with sickness presenteeism. WFC predicted subjective (β=.23,
5
p <.01) and objective rates (β=.18, p <.05) of SP (Johns, 2011, p. 491), and also
productivity loss measured with Work Limitations Questionnaire (β=.16, p <.01)
(Johns, 2011, p. 493). The conflict between family and work domains was related
to the productivity loss measured by Standford Presenteeism Scale and Work
Limitations Questionnaire (Johns, 2011).
2. OBJECTIVE AND HYPOTHESES
Starting from this theoretical background, the objective of this paper is
exploratory, examining the link between work-family conflict and sickness
presenteeism. We thus explore whether SP is prevalent among employees
experiencing WFC. The main hypothesis of the study is that work-family conflict
predicts sickness presenteeism. We consider that employees reporting WFC will be
present at work despite illness. According to Aronsson et al. (2000), we measured
sickness presenteeism as employees’ own assessment regarding their attendance
rate at work, although being sick.
3. METHOD
3.1. PARTICIPANTS
The study is based on a convenience sample, 126 employees from Romanian
organizations participating at the current research. 78.6% were women and 21.4%
men, with ages ranging from 27 to 56 years (M=41.6, SD=5.50), mainly from the
educational, human resources, communication and banking sectors.
All the participants were married, 88.1% being from dual-earner couples
(with an employed spouse), and 53.2 % having at least one child. Also, 13.5% had
an elder dependent in their care. All employees were working full-time, 32.5% in
the public sector and 67.5% from the private sector, 29.4% having a leadership
position.
3.2. INSTRUMENTS
Work-family conflict was assessed using the Carlson, Kacmar and Williams
(2000) multidimensional measure. The scale is composed of 18 items that tap the
two directions and the three forms of conflict, rated on a five point scale ranging
from „strongly disagree” to „strongly agree”. The scale was translated and adapted
in Romanian by Şulea, Vîrgă and Galben (2010). For the current study, the 9 items
measuring work interference with family were selected, the internal consistency of
the scale being of α=.87.
We measured presenteeism, following Aronsson et al. study (2000), with a
single question: „Has it happened in the previous 12 months that you have gone to
6
work despite feeling that you really should have taken sick leave due to your state
of health?”(p. 504). The responses were made on a four point scale, ranging from
„No, never” to „Yes, more than 5 times” (Aronsson et al., 2000). We dichotomized
the responses in view of conducting logistic regression.
3.3. PROCEDURE
The employees were contacted and informed about the topic of the research;
those who agreed to participate filled in an online survey regarding work-family
conflict that contained the WFC and SP measures. They were briefed regarding the
anonymity and confidentiality of the results. The data were collected and a logistic
regression was conducted using SPSS 16.00. However, we must mention that,
having in view the statistical properties of logistic regression, the generalization of
the results must be made with cautions. More precisely, logistic regression requires
a much larger sample size than linear regression (Popa, 2010, p. 187), the number
of participants ranging from 50 to at least 10 subjects/predictor, some authors
considering the sample size to be of a minimum of 150 subjects (Popa, 2010, p.
187).
4. RESULTS
In order to examine if work-family conflict predicts sickness presenteeism,
logistic regression analysis was performed, the dependent variable being
dichotomized. We followed Aronsson et al. (2000, p. 504) coding procedure in
order to examine the occurrence of sickness presenteeism (0 = No, never/Yes,
once, 1 = Yes, 2–5 times/Yes, more than 5). For work-family conflict, means and
standard deviations were calculated (M=3.08, SD=.77).
The regression model tested was significant (2 =17.55, p<.001), the
Nagelkerke’s R² coefficient of .175 indicating however, a weak prediction, 17.5%
of the variance of sickness presenteeism being explained by work-family conflict
(Table 1).
Table 1. Model Summary
Step
1
-2 Log likelihood
Cox & Snell R Square
Nagelkerke R Square
153.931
.130
.175
The Hosmer & Lemeshow test (2 =5.69, p>.68) indicates that the prediction
model fits the research data (Table 2).
7
Table 2. Hosmer & Lemeshow test
Step
Chi-square
df
Sig.
1
5.692
8
.682
The classification table presents the correspondence between the observed
values of the criterion and those predicted by the model (Table 3). As it can be
observed, the total percentage of right classification is of 70.6% (52.8% for the
absence of sickness presenteeism and 83.6% for SP occurrence).
Table 3. Classification model
Observed
Step 0
Predicted
Sickness presenteeism Percentage Correct
No
Yes
Sickness presenteeism
No
28
25
52.8
Yes
12
61
83.6
Overall Percentage
70.6
Results indicate that WFC can be considered a predictor of SP (Wald=14.52,
p<.001, OR=2.94, 95%CI=1.69-5.14).
Thus, a logistic regression analysis was conducted to predict sickness
presenteeism using work-family conflict as a predictor. The test of the full model
against a constant only model was significant, indicating that the predictor reliably
distinguished between the occurrence and absence of sickness presenteeism (
17.55, p<.001 with df=1).
Nagelkerke’s R² of .175 indicated a weak relationship between prediction and
grouping. Overall prediction success was 70.6% (52.8% for the absence of SP and
83.6% for the occurrence of SP). The Wald criterion demonstrated that workfamily conflict made a significant contribution to prediction (Wald=14.52, p
<.001). EXP(B) value indicates that when work-family conflict is raised by one
unit, the odds ratio is 2.94 times as large, and therefore employees are
approximately 3 times more likely to come to work when sick.
8
5. CONCLUSIONS
The current study is an exploratory one, examining the role of work-family
conflict in predicting sickness presenteeism. The hypothesis of the study was
supported; employees that experienced interference between work and family
reported coming to work despite illness. Thus, the results highlight the impact of
work-family conflict on health-related attendance dynamics (Johns, 2011).
Starting from these results, a number of practical implications can be traced.
First, for example, by providing family-friendly benefits as flexible working
arrangements, which can enhance control in dealing with various work and family
demands (Allen, 2001), organizations can reduce deleterious „presenteeism
cultures” (Johns, 2010) and foster „family-supportive environments” (Allen, 2001).
Also, by redesigning job demands as Demerouti et al. (2009, p. 64) suggest,
presenteeism, but also work-family conflict, can be decreased.
However, we must notice that the explained variance in the logistic regression
indicates a weak prediction and the percent of right classification can be improved,
when examining the absence of SP. Thus, some strong limitations of the research
must be highlighted.
The first weakness relies in the sample size, which can be considered unfitted
for logistic regression (Popa, 2010). Thus, the results of the study must be
interpreted with cautions. Secondly, according to Geurts and Demerouti (2003) a
large majority of WFC studies are based on cross-sectional design, and self-report
measures. The current study does not surpass this methodological weakness, and
we suggest that further studies should adopt a longitudinal design and preferably
based on objective reports.
Also, as Geurts and Demerouti (2003) draw attention, the samples used in
WFC research are traditionally composed of women and white-collar employees.
The structure of the current sample reflects this limitation, and more studies should
be conducted having in view more balanced and occupationally diverse sampling.
Other limitations rely in not controlling for the effects of general health, which was
strongly associated with presenteeism (Johns, 2011) and measuring presenteeism
with a single, self-report, item.
Future research directions might consist in examining the relationship
between WFC and SP having in view gender and occupational differences, but also
the forms of WFC (especially strain-based and time-based conflict).
Also, focusing on the suggestion that presenteeism and absenteeism should be
examined concurrently (Johns, 2010, p. 534), and having in view that family-work
conflict was related to the number of days of absenteeism, but also with
presenteeism productivity loss (Johns, 2011, p. 494), future studies could explore
the role of work-family conflict and family-work conflict in predicting sickness
absenteeism and sickness presenteeism.
Received at: 23.12.2011, Accepted for publication on: 14.01.2012
9
6. REFERENCES
Allen, T. D., Herst, D. E. H., Bruck, C. S., & Sutton, M. (2000). Consequences
associated with work to family conflict: A review and agenda for future research. Journal
of Occupational Health and Psychology, 5(2), 278-308.
Allen, T. D. (2001). Family-supportive work environments: The role of organizational
perceptions. Journal of Vocational Behavior, 58, 414-435.
Amstad, F. T., Meier, L. L., Fasel, U., Elfering, A., & Semmer, N. K. (2011). A metaanalysis of work-family conflict and various outcomes with a special emphasis on crossdomain versus matching-domain relations. Journal of Occupational Health Psychology,
16(2), 151-169.
Aronsson, G., Gustafsson, K., & Dallner, M. (2000). Sick but yet at work. An empirical
study of sickness presenteeism. Journal of Epidemiology and Community Health, 54(7),
502-509.
Bellavia, G. M., & Frone, M. R. (2005). Work-family conflict. In J. Barling, E. K.
Kelloway, & M. R. Frone (Eds.), Handbook of work stress. (pp. 113-147).Thousand Oaks,
CA: SAGE Publications.
Biron, C., & Saksvik, P. Ø. (2009). Sickness presenteeism and attendance pressure
factors: Implications for practice. In C. L. Cooper, J. C. Quick, & M. Schabracq (Eds.),
International Handbook of Work and Health Psychology, Third Edition (pp. 77–96).
Chichester, UK: Wiley.
Böckerman, P. & Laukkanen, E. (2010). What Makes You Work while You Are Sick?
Evidence from a Survey of Workers. The European Journal of Public Health, 20(1), 43-46.
Brough, P. & O’ Driscoll, M. (2005). Work-family conflict and stress. In A.-S. G.
Antoniou & C. L. Cooper (Eds.), Research Companion to Organizational Health
Psychology (pp. 346-365). Cheltenham, UK: Edward Elgar
Carlson, D. S., Kacmar, K. M., & Williams, L. J. (2000). Construction and validation of
a multidimensional measure of work-family conflict. Journal of Vocational Behavior,
56(2), 249–276.
Demerouti, E., Blanc, P. M. Le, Bakker, A. B., Schaufeli, W.B. & Hox, J. (2009).
Present but sick: a three-wave study on job demands, presenteeism and burnout. Career
Development International, 14(1), 50-68.
Frone, M. R. (2003). Work–family balance. In J. C. Quick & L. E. Tetrick (Eds.),
Handbook of occupational health psychology (pp. 143–162). Washington, DC: American
Psychological Association.
Geurts, S.A.E. & Demerouti, E. (2003). Work/non-work interface: A review of theories
and findings. In M. Schabracq, J. Winnubst & C. L. Cooper (Eds.), Handbook of work and
health psychology (pp. 279-312).Chichester: John Wiley & Sons
Greenhaus, J. H., & Beutell, N. J. (1985). Sources of conflict between work and family
roles. Academy of Management Review, 10, 76-88.
Hansen, C. D. & Andersen, J. H. (2008). Going ill to work – What personal
circumstances, attitudes and work-related factors are associated with sickness
presenteeism? Social Science & Medicine, 67(6), 956-964.
Hemp, P. (2004). Presenteeism: At work - but out of it. Harvard Business Review,
82(10), 49–58.
10
Jansen, N. W. H., Kant I. J., van Amelsvoort, L. G. P. M., Kristensen, T. S., Swaen, G.
M., Nijhuis, F. J. (2006). Work-family conflict as a risk factor for sickness absence.
Occupational Environmental Medicine. 63(7), 488–94.
Johns, G. (2010). Presenteeism in the workplace: A review and research agenda.
Journal of Organizational Behavior, 31, 519-542.
Johns, G. (2011). Attendance dynamics at work: The antecedents and correlates of
presenteeism, absenteeism, and productivity loss. Journal of Occupational Health
Psychology, 16(4), 483-500.
Popa, M. (2010). Multivariate statistics applied in psychology. Iaşi: Polirom.
Rothbard, N. P., Dumas, T. L., (2006). Managing the work-home interface: Research
perspectives. In F. Jones, R. J. Burke, & M. Westman, (Eds.) Work-Life Balance: A
Psychological Perspective. (pp. 71- 89). New York: Psychology Press.
Şulea, C., Vîrgă, D., Galben, N. (2010). Scala Conflict Muncă-Familie: Analiza
proprietăţilor psihometrice ale versiunii în limba român. Psihologia Resurselor Umane,
8(2), 10-20.
11