Page 1
Comparing three alternative types of employment withpermanent full-time work: How do employment contractand perceived job conditions relate to health complaints?
CLAUDIA BERNHARD-OETTEL1, MAGNUS SVERKE2 &
HANS DE WITTE3
1National Institute for Working Life, Stockholm, Sweden, 2Department of Psychology, Stockholm
University, Sweden, 3Department of Psychology, Katholieke Universiteit Leuven, Belgium
AbstractPrevious research has found that alternative employment arrangements are associated with bothimpaired and improved well-being. Since such inconsistencies are likely to derive from the type ofemployment contract as well as the characteristics of the job, this paper compares permanent full-timework with forms of alternative employment (permanent part-time, fixed-term and on-call work) inorder to investigate how different employment contracts and perceptions of job conditions relate toindividual well-being. This study contributes to the literature by addressing several questions.Different forms of alternative employment are distinguished and individual background character-istics that might be intertwined with the employment contract are controlled for. Moreover, the scopeof this study extends to the effects of perceived job conditions, and possible interactive effects withtype of employment are tested. Analyses of questionnaire data from 954 Swedish healthcare workersshow that perceptions of the job (job insecurity, job control and demands), but not the type ofemployment contract, predicted health complaints. However, type of employment interacted withperceptions of job insecurity, in that insecurity was associated with impaired well-being amongpermanent full-time workers, while no relationship was found for on-call or core part-time employees.Despite the absence of interactions between employment contract and job demands or job control, itcan be concluded that knowledge about the relationship between alternative employment arrange-ments and the well-being of workers can be enhanced when the combined effects of employmentcontract and job conditions are studied.
Keywords: Alternative employment, temporary work, part-time work, job control, job demands, job
insecurity, health complaints
Introduction
Alternative forms of employment, such as temporary and part-time work, deviate from
traditional ongoing full-time employment concerning the nature and future existence of the
job as well as in the number and distribution of working hours (Reilly, 1998). As a
consequence, they are often assumed to be associated with impaired well-being. However,
the findings of previous research are inconclusive. Some studies, in which temporary or
part-time workers have been found to report more health complaints than those in
Correspondence: Claudia Bernhard-Oettel, National Institute for Working Life, 113 91 Stockholm, Sweden;
e-mail: [email protected]
ISSN 0267-8373 print/ISSN 1464-5335 online # 2005 Taylor & Francis
DOI: 10.1080/02678370500408723
Work & Stress, October-December 2005; 19(4): 301�/318
Page 2
permanent employment (Benach, Gimeno, & Benavides, 2002; Martens, Nijhuis, Van
Boxtel, & Knottnerus, 1999), support the assumption of detrimental effects of alternative
employment arrangements. In contrast, other studies report better psychological well-being
and fewer health complaints among employees in alternative types of employment (Paoli &
Merllie, 2001; Sverke, Gallagher, & Hellgren, 2000), while yet other studies find no clear
differences (Claes et al., 2002). There may be several explanations to this mixed pattern of
results.
First of all, the majority of studies examining consequences of alternative employment
contracts fail to control for individual and work-related background variables (Virtanen,
M., 2003). Second, research in this area typically does not take into account the
heterogeneity of alternative contracts. Recent research, however, illustrates the necessity
of differentiating between different types of alternative employment arrangements in order
to better understand the consequences for employee well-being (Aronsson, Gustafsson, &
Dallner, 2002; Benavides, Benach, Diez-Roux, & Roman, 2000; Organization for
Economic Co-operation and Development [OECD], 2002). A third possible reason for
the mixed research findings is that the job characteristics of the contract holder are often
overlooked. A growing number of studies suggest that there may be important differences in
job conditions and perceptions between various types of alternative employment, for
instance concerning job insecurity (De Witte & Naswall, 2003; Virtanen, Vahtera,
Kivimaki, & Pentii, 2002), but also concerning job demands and control (Goudswaard &
Andries, 2002). Disadvantageous combinations of job characteristics and job insecurity,
which have been found to be associated with health problems (Strazdins, D’Souza, Lim,
Broom, & Rodgers, 2004), may be more characteristic of some forms of employment than
others (Wikman, Andersson, & Bastin, 1998).
The overarching aim of this paper is to contribute to the understanding of the
relationship between employment contracts and employee well-being. More specifically,
we investigate the direct and interactive effects of employment contract status and perceived
job conditions on individual health complaints, after controlling for individual background
characteristics.
Alternative employment and well-being
Although full-time, open-ended employment contracts still constitute the majority, part-
time and temporary work have become important alternative employment options in most
industrialized countries (OECD, 2002; Sparrow & Cooper, 2003). Alternative employment
can take various forms, with the most common types including fixed-term employment
(e.g. for the duration of a project or as a temporary replacement), on-call employment (e.g.
filling short-term vacancies for a limited number of hours or days) and temporary agency
work (where workers are employed by the agency but temporarily ‘hired out’ to a client
firm) among others (Aronsson, 1999; McLean Parks, Kidder, & Gallagher, 1998; OECD,
2002). Part-time work may be organized on a temporary or*/particularly in Europe*/
permanent basis with benefits comparable to full-time core staff (Krausz, Sagie, &
Bidermann, 2000; OECD, 2002). Thus, permanent part-time workers are yet another
group to be distinguished from ongoing (full-time) and temporary employment (part-time
or full-time).
Despite the fact that many forms of alternative employment exist, they have fre-
quently been studied as a single homogeneous category compared to permanent full-
time employment, which might partly explain the contradictory findings. For example,
some studies report more job dissatisfaction, mental distress and somatic complaints
302 C. Bernhard-Oettel et al.
Page 3
among temporary workers (Kinnunen & Natti, 1994; Klein Hesselink & Van Vuuren,
1999), while others have reported less mental distress (Letourneux, 1998), fewer somatic
complaints (Sverke et al., 2000), fewer chronic diseases and better self-rated health
among temporary workers compared to permanent staff (Virtanen, Kivimaki, Elovainio,
Vahtera, & Cooper, 2001). Along similar lines, the results are contradictory for part-
time work, with some studies reporting no differences between part-time and full-time
workers (Krausz et al., 2000), while others report higher subjective well-being and job
satisfaction among part-time workers compared to full-time employees (Conway & Briner,
2002).
Recent research suggests that alternative forms of employment may differ along a core�/
periphery continuum with, for example, probationary employment being closer to the core
permanent employees than on-call or seasonal work (Aronsson et al., 2002). With this
perspective, it is also presumed that employment contracts that are more similar to
permanent full-time work will also have similar consequences for well-being. Taking such a
core�/periphery perspective, it may be that different access to fringe benefits (Gustafsson,
Kenjoh, & Wetzels, 2001) and differences in the amount and reliability of income streams
(Beard & Edwards, 1995) influence the well-being of individuals*/particularly in the case
of shorter assignments and high levels of economic dependency on a job (Aronsson,
Dallner, & Lindh, 2000).
Other employment circumstances influencing well-being may be the scheduling of
working time (Martens et al., 1999), prevalence of unfavourable working hours (Paoli,
1996) or duration of service (Sverke et al., 2000). Two recent empirical analyses clearly
demonstrate that different health problems may be more prevalent in certain alternative
forms of employment than others (Aronsson et al., 2002; Benavides et al., 2000). For
example, the study by Aronsson et al. (2002) revealed that discomfort prior to work was
typically reported by substitutes as well as on-call workers; however, while the risk for
stomach complaints was higher among substitutes, back/neck pain was more characteristic
of on-call workers.
Apart from the differences in terms of employment, alternative work arrangements are
spread over different sectors and professions (Aronsson et al., 2002; Davis-Blake & Uzzi,
1993). Hence, alternative forms of employment are found in a variety of jobs that differ
greatly in content and qualification requirements, ranging from manual, low-skilled work to
jobs for the most educated members of the workforce (e.g. a consultant or researcher),
lasting from less than a year to more than 5 years (OECD, 2002). Hence, effects of
employment contract status on health and well-being are difficult to identify if job
conditions are not disentangled from employment conditions (Goudswaard & de Nanteuil,
2000).
Job conditions and well-being
As Goudswaard and Andries (2002) note, the unfavourable job conditions believed
to characterize alternative work arrangements may be reflected in characteristics of
the job, such as limited possibilities to exert influence at work or the terms of employment,
for instance the security of a job and its future prospects. Indeed, empirical studies
often find alternative employment to be associated with poorer job characteristics
(Goudswaard & Andries, 2002; OECD, 2002) and higher job insecurity (Kinnunen &
Natti, 1994; Sverke et al., 2000). However, such a general conclusion may represent
an oversimplification. First, as noted in the previous section, alternative work arrange-
ments are not a single homogeneous group. Indeed, research suggests that poorer
Employment contracts and health complaints 303
Page 4
job conditions and deviations from conditions in ongoing full-time employment might
be more typical the more peripheral the employment contract (Aronsson et al., 2002;
Wikman et al., 1998). Second, even individuals in comparable employment forms and
positions can perceive job insecurity and job characteristics rather differently, depending on
how a given situation (e.g. employment contract) is interpreted in terms of the future
existence of one’s job or how earlier expectations are met (De Witte & Naswall, 2003;
Virtanen, P. et al., 2002). Hence, not only may different forms of alternative employment
differ in health outcomes, but perceptions of the job may also have an additional influence
on well-being.
The psychological concept of job insecurity refers to subjective concerns about the
continuation of a job (Hartley, Jacobsson, Klandermans, & Van Vuuren, 1991). Feelings of
job insecurity can be considered as a psychosocial risk factor, and the results of a recent
meta-analytic study indicate detrimental consequences for physical and mental health
(Sverke, Hellgren, & Naswall, 2002). However, the majority of previous studies collected
data from permanent employees, for whom such experiences involve a fundamental change
in their perceived position in the labour market (Virtanen, P. et al., 2002). In contrast, for
temporary workers the termination of the job is integrated in their type of employment,
which may lead to less negative reactions. Indeed, two recent studies suggest that the
association between job insecurity and outcomes may depend on the context of the
contractual agreement, as distress increased and job satisfaction diminished significantly
among permanent employees who felt insecure, while temporary workers were unaffected
by perceptions of job insecurity (De Witte & Naswall, 2003; Virtanen, P. et al., 2002).
However, since both these studies differentiated simply between permanent and temporary
workers, it remains unclear whether this conclusion applies to different kinds of alternative
employment. Although job insecurity appears to be related to contractual status, it is not a
perception shared by all individuals in alternative employment arrangements. For instance,
25% of European temporary workers appear to consider their job to be secure (Letourneux,
1998), and part-time workers tend to express fewer feelings of job insecurity than
temporary workers (Sverke et al., 2000). This again illustrates the necessity of differentiat-
ing alternative employment types and of considering subjectivity in perceptions of job
insecurity.
Associations between employment contract and perceived job characteristics in terms of
job control and demands have been investigated in several empirical studies (Aronsson et
al., 2002; Goudswaard & Andries, 2002; Wikman, 2002). Data from the third European
survey on working conditions suggest that high-strain jobs, that is, a combination of high
job demands and low control, are more pronounced among temporary workers than
permanent employees (Goudswaard & Andries, 2002). Similarly, part-time work is often
experienced as low control but, in contrast to temporary work, also low demand, which
results in high monotony and thus passive work (Corral & Isusi, 2004). Again, however,
more can be learnt by a consideration of the heterogeneity of alternative work arrange-
ments. Research contrasting different types of temporary contracts with permanent
employment has found workers in project-based employment to report most influence
and control over their work, whereas substitute employees were in an intermediate position
and on-call workers experienced the fewest possibilities of asserting control (Aronsson et
al., 2002; Wikman, 2002). These results fit well with the findings of Wikman et al. (1998),
who categorized different employment forms using a demand�/control index according to
Karasek’s (1979) model.
304 C. Bernhard-Oettel et al.
Page 5
The demand�/control model (Karasek, 1979) has been tested in a large number of
studies. Consistent with the model’s assumptions, these studies indicate that psychological
demands tend to be associated with impaired health, increased sick leave and reduced
well-being (Richter et al., 2000), whereas control appears to be related to positive
psychophysiological consequences (Morrison, Payne, & Wall, 2003). An empirical test of
reciprocal relationships (de Jonge et al., 2001) found strong evidence for lagged causal
effects of demands and control on well-being and health. A recent review of high quality
longitudinal studies, however, provides only modest support for the strain hypothesis,
according to which control acts as a moderator buffering the negative influences of demands
(de Lange, Taris, Kompier, Houtman, & Bongers, 2003). Notably, however, the body
of research on job characteristics has been conducted in the context of permanent
employment.
In summary, the combined effects of perceived job characteristics and the features of
various contractual arrangements have received only limited research attention (Aronsson &
Goransson, 1999). Yet there are reasons to assume that the interaction between type of
employment contract and perceived job conditions may have important implications.
Peripheral employment status may interfere with an individual’s capability to cope with
poor job conditions and may impair health considerably. Likewise, demands and control
may yield diverging effects in different contractual arrangements, sometimes even contra-
dicting the theoretically-based assumptions. For some temporary workers, high demands
may represent an opportunity to show one’s skills in order to secure renewed or permanent
employment (Sverke et al., 2000). Due to insufficient levels of information and training
(Aronsson et al., 2002; Corral & Isusi, 2004; Goudswaard & de Nanteuil, 2000), workers in
certain forms of alternative employment may perceive high job control and decision latitude
as threatening and stressful. In other words, there is evidence that type of employment
contract may alter the relationship between job characteristics and employee health and
well-being.
The present study
This paper analyses the direct and interactive effects of employment contract status and
perceived job conditions on employee health complaints in an integrative study. In doing so,
the study extends previous research in several ways. First, adopting a core�/periphery
perspective, we distinguish several alternative forms of employment (part-time work, fixed-
term contracts, and on-call employment) and contrast them with permanent full-time
employment in order to study relationships between employment contract and health
complaints in greater detail. Second, in addition to employment contract status we
investigate perceived job conditions, focusing on job insecurity as well as perceptions of job
control and demands in different employment forms. Beside descriptive associations, we
study the effects of employment contract and subjective perceptions of job conditions in a
combined analysis, thus enabling evaluation of their relative impact on health complaints.
Finally, taking the analysis a step further, we investigate whether job perceptions have the
same effect on individual health complaints in permanent full-time employment compared
to different groups of alternative work arrangements. As individual background character-
istics are often intertwined with employment contract, job conditions and health, we control
in the analyses for age, gender, marital status, professional qualification, organizational
tenure and working hours.
Employment contracts and health complaints 305
Page 6
Method
Setting and sample
The study is based on questionnaire data collected among the total workforce of two
Swedish hospitals. Questionnaires were mailed by post to the homes of all 2455 employees,
accompanied by a letter explaining the aim of the study, that participation was voluntary,
and assuring that responses were confidential. A total of 1505 participants returned the
questionnaires in the self-addressed envelopes provided, corresponding to a response rate of
61%. Listwise deletion of missing data resulted in an effective sample of 954 individuals
who responded to all the questions that served as the input for the analysis. Of these, 560
(59%) were on permanent full-time contracts, 275 (29%) were permanent part-time
employees, 83 (9%) worked on a temporary basis on fixed-term contracts, and 36 (4%)
were temporary employees working on-call. With respect to their profession, 13% worked
as physicians, 64% as nurses and 23% in administrative positions or as support staff (e.g.
laboratory assistant). The sample consisted of a large majority of women (82%) with a
mean age of 42.5 years. Almost three out of four participants were cohabiting (73%). On
average, the sampled employees worked 36 hours per week and had an organizational
tenure of 11 years.
Measures
Employment contract. Various responses to questions about the employment contract led to the
construction of three categorical dummy variables: permanent part-time (1�/part-time
workers, 0�/other), fixed-term (1�/fixed-term, 0�/other) and on-call (1�/on-call, 0�/
other) with permanent full-time workers as the comparison group (Aiken & West, 1991).
Job conditions. Job insecurity was measured by three items based on Ashford, Lee, and Bobko
(1989). These items refer to a cognitive dimension of job insecurity (e.g. ‘I may be laid off
permanently’) and were assessed on a 5-point Likert scale ranging from 1 (strongly
disagree) to 5 (strongly agree). The a coefficient was .76. Demands were assessed using
three items (e.g. ‘I often have too much to do in my job’) from a scale measuring role
overload (Beehr, Walsh, & Taber, 1976). The a coefficient was .81. Control was assessed
using a 4-item scale (Sverke & Sjoberg, 1994) based on Hackman and Oldham (1975) and
Walsh, Taber, and Beehr (1980) to capture the amount of influence and control over one’s
job (e.g. ‘I have satisfactory influence over decisions concerning my job’). The scale
demonstrated satisfactory internal consistency (a�/ .81). Demands and control were also
assessed on a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree).
Results of confirmatory factor analyses showed that the presupposed three-factor model
provided a satisfactory fit to data [x2 (32)�/378.05; p B/.001; RMSEA�/0.09; AGFI�/
0.90] and clearly outperformed rivalling one- or two-factor models, thus supporting the
existence of three distinct job conditions.
Health complaints. Mental distress was measured using the General Health Questionnaire
(Goldberg, 1979), which consists of 12 questions where respondents indicated how often
they had experienced certain symptoms (0�/never, 3�/always) over the past weeks. The
scale, which is deemed to be a good indicator of non-psychiatric disorders, demonstrated
adequate reliability (a�/.82). House and Rizzo’s (1972) scale was used to assess job-induced
tension . The seven items (e.g. ‘Problems associated with my job have kept me awake at
306 C. Bernhard-Oettel et al.
Page 7
night’), scored on 5-point Likert scales, yielded a reliability estimate of .84. Again,
confirmatory factor analysis suggested distinct constructs. The results indicated that the
two health complaints measures reflect two distinct factors [x2 (151)�/1830.75; p B/.001;
RMSEA�/0.11; AGFI�/0.80], and the two-factor solution provided a significantly better
fit to data than a one-factor model.
Individual background characteristics. Age and organizational tenure were measured in years.
Weekly working hours were assessed by respondents giving their average working hours per
week. Gender (1�/female, 0�/male) and partner (1�/married or cohabiting, 0�/living
alone) were measured as dummy variables. Two categorical variables were constructed
representing the most frequent occupational groups: physicians (1�/physicians, 0�/other)
and nurses (1�/nurses, 0�/other).
Descriptive statistics (means, standard deviations and intercorrelations) for the variables
are presented in Table I. Continuous variables were screened for normality and the majority
of variables’ distributions were normally shaped. Job insecurity showed a tendency towards
positive skewness and the number of working hours was found to have a positive kurtosis,
indicating a distribution that is slightly too peaked.
Statistical treatment
In order to compare how the different employment contracts differed in the variables under
study, three multivariate analyses of variance (MANOVA) were performed on individual
background characteristics, job conditions and health indicators. Before the use of
MANOVA, Box’s M tests of homogeneity of variance-covariance matrices were conducted.
Significant results for the variables of job conditions and individual background
characteristics indicate that smaller samples produced larger variances and covariances in
these variables. Therefore, Pillai’s trace was used to evaluate multivariate significances in
the MANOVA analyses on job conditions and individual background characteristics. The
MANOVAs were supplemented with univariate F-tests and post-hoc Scheffe tests. For
individual background characteristics assessed as categorical variables, x2 tests were used to
test for differences between employment forms.
For the predictive analysis of the direct and interactive effects of employment contract
types and job conditions on health complaints, a hierarchical regression analysis was used
for each of the dependent variables (mental distress and job-induced tension). In order to
ensure that assumptions for the use of multiple regressions are met, the data was screened
for outliers, multicollinearity and tolerance levels. In each of these regressions, individual
background variables were first entered into the model to control for their influence on the
outcomes. In the second step, the different types of employment contract were included in
the regression equation to find out whether differences in employment contracts explained
variance in the health-related outcomes, while controlling for individual background
characteristics. In step 3, job insecurity, job control and demands were added to the model.
In steps 4, 5 and 6, the interaction terms between employment contracts (part-time, fixed-
term, on-call) and job perceptions (job insecurity, job control, job demands) were included
in the model separately in order to test which of them had any additional significant effect.
The interaction terms were created with the cross-product of the variables. Following the
recommendations by Aiken and West (1991), continuous predictor variables were centred
before the cross-products were calculated in order to avoid artificial multicollinearity.
Employment contracts and health complaints 307
Page 8
Table I. Correlations, means and standard deviations for all variables, and a reliabilities for scales.
Variable 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 M SD a
Individual background characteristics
1 Age 1.00 42.49 10.19 �/
2 Gender
(female)
.01 1.00 0.82 .38 �/
3 Partner .02 .03 1.00 0.73 .44 �/
4 Physician .12* �/.45* .08* 1.00 0.13 .34 �/
5 Nurse �/.18* .26* .00 �/.52* 1.00 0.64 .48 �/
6 Weekly
working hours
.09* �/.20* .00 .42* �/.29* 1.00 36.34 8.64 �/
7 Organizational
tenure
.53* .15* .03 �/.12* .05 .01 1.00 11.38 8.99 �/
Employment contract
8 Part-time .02 .16* .13* �/.20* .19* �/.49* .06* 1.00 0.28 .45 �/
9 Fixed-term �/.27* .00 �/.05 .12* �/.06* .07* �/.27* �/.21* 1.00 0.10 .30 �/
10 On-call �/.12* �/.06* �/.06* �/.07* .02 �/.34* �/.12* �/.13* �/.07* 1.00 0.04 .20 �/
Job conditions
11 Job insecurity �/.07* �/.03 �/.05 �/.03 �/.01 .�/07* �/.03 �/.09* .30* .17* 1.00 1.75 .99 .76
12 Job control .11* .02 .06* �/.04 �/.10* .06* .09* �/.02 �/.09* �/.09* �/.16* 1.00 3.30 .90 .81
13 Job demands �/.04 �/.01 �/.01 .11* �/.08* .12* �/.09* �/.04 �/.02 �/.07* .05 �/.30* 1.00 3.66 1.00 .81
Health complaints
14 Mental distress
(GHQ)
.06* �/.04 �/.11* �/.03 �/.08* -�/.04 .03 �/.01 .05 .05 .23* �/.33* .31* 1.00 1.74 .39 .82
15 Job-induced
tension
�/.06* �/.10* �/.01 .12* �/.10* .15* �/.06* �/.06* .08* �/.08* .19* �/.28* .53* .56* 1.00 2.63 .91 .84
*pB/ .05.
Scale range: 0�/1 (variables 2�/5, 8�/10 [for these variables the mean value symbolizes the proportion scoring 1]), 1�/5 (variables 11�/13, 15), 0�/3 (variable 14), years
(variables 1, 7), hours/week (variable 6).
308
C.
Bern
hard
-Oettel
etal.
Page 9
Results
Characteristics of different employment contracts
Table II presents mean levels for job conditions and health complaints as well as means/
frequencies for individual background characteristics for full-time, part-time, fixed-term
and on-call workers. It also reports the results of the tests for differences between
employment contracts.
For job conditions, an overall multivariate effect was found (Multivariate F
(df�/9,2850)�/18.58, p B/.001). Univariate F-tests revealed differences in job insecurity as
well as job control and job demands. Scheffe post-hoc tests showed that the level of job
insecurity was higher for fixed-term and on-call workers than for part-time and core full-
time employees. Core full-time workers were also found to have higher levels of control and
demands compared to employees contracted on an on-call basis.
There was also a significant overall difference between employment contracts in health
complaints (Multivariate F (df�/6,1900)�/4.21, p B/.001). Univariate F-tests revealed no
significant differences in mental distress among employment groups, whereas there was a
difference in job-induced tension. The post-hoc tests revealed that job-induced tension was
higher for fixed-term workers than for employees on an on-call contract.
In terms of individual background characteristics, there was a significant multivariate
difference between the employment groups in the MANOVA conducted on age, weekly
working hours and organizational tenure (Multivariate F (df�/9,2850)�/67.23, p B/.001),
and the univariate follow-up tests revealed differences in all three variables. Scheffe post-hoc
tests showed that core full-time workers resembled part-time workers in terms of age, while
the other two groups were on average younger. Core full-time workers reported the highest
number of working hours per week, followed by fixed-term workers, but this number was
significantly reduced for part-time workers and lowest for the group of on-call workers.
Organizational tenure was higher for core full-time and part-time workers compared to
fixed-term and on-call employees. Chi-square tests performed for categorical variables
revealed significant differences in all variables. In terms of the distribution across
occupations, employees on fixed-term contracts were most comparable to full-time workers
while part-time workers resembled on-call contracted workers. The percentage of women
was highest among part-time workers, as was also the proportion of those living with a
partner.
Predicting health complaints
The results of the regression analyses predicting mental distress and job-induced tension are
presented in Table III. In Step 1, all individual background variables except for weekly
working hours and organizational tenure displayed a small but significant association with
mental distress (R2�/.03, pB/ .001). The second step, in which the employment contracts
were added, represented a small increase in explained variance (R2change�/.01; p B/.05), but
none of the contract variables reached significance. In Step 3 all job conditions evidenced
significant effects on mental distress and increased the explained variance to .24 (pB/ .001).
Both job insecurity and job demands were positively related to mental distress, while the
association was negative for job control. In steps 4 to 6, which tested interactions between
employment contract types and each of the perceived job conditions, only the interaction
terms between employment contracts and job insecurity (Step 4) increased the overall
explained variance of the model significantly (R2change�/.02; p B/.001). In comparison with
the core full-time workers, there was a difference in the relationship between job insecurity
Employment contracts and health complaints 309
Page 10
Table II. Individual background characteristics, job conditions, general health and job induced tension of workers with alternative versus core full time employment.
Employment contract
Variable Full-time [1] Part-time [2] Fixed-term [3] On-call [4] Total Chi square Univariate F Means comparison
Job conditions
Job insecurity 1.60 1.62 2.77 2.64 1.75 �/ 51.24*** 1B/3, 4; 2B/3, 4
Job control 3.39 3.27 3.03 3.00 3.31 �/ 5.65*** 1�/4
Job demands 3.72 3.61 3.56 3.26 3.66 �/ 3.17* 1�/4
Health complaints
Mental distress 0.73 0.74 0.82 0.82 0.74 �/ 1.82 n.s.
Job-induced tension 2.65 2.54 2.86 2.30 2.63 �/ 4.39** 3�/4
Individual background characteristics
Age (mean) 44.2 42.7 34.2 36.5 42.5 �/ 32.14** 1�/3, 4; 2�/3, 4
Gender (% female) 78.6 91.6 79.5 66.7 82.0 27.81*** �/ �/
Partner (% cohabiting) 70.0 82.5 68.7 66.7 73.4 16.87*** �/ �/
Physician% 17.0 2.2 25.3 2.8 12.9 51.02*** �/ �/
Nurse% 58.2 78.9 55.4 72.2 64.5 38.51*** �/ �/
Weekly working hours (mean) 40.4 29.6 37.4 22.2 36.3 �/ 220.73*** 1�/2, 3, 4; 3�/2, 4; 2�/4
Organizational tenure (years) 12.5 12.2 3.7 6.1 11.4 �/ 30.48*** 1�/3, 4; 2�/3, 4
*p 5/.05; **p 5/.01; ***p 5/.001.
310
C.
Bern
hard
-Oettel
etal.
Page 11
and mental distress only for employees in on-call arrangements, but not for part-time or
fixed-term workers. As can be seen in Figure 1, permanent full-time workers reported
higher mental distress when they perceived high job insecurity (t�/ 5.72, p B/.001), whereas
on-call workers’ mental distress was not dependent on the level of perceived insecurity (t�/
0.34, p �/.05). The results remained the same when the interaction terms were added in a
different order.
0.2
0.4
0.6
0.8
low job insecurity high job insecurity
Full time On-call
0
1
Figure 1. Interaction of perceived job insecurity and type of contract (core full-time and on-call) on mental distress
(scale 0�/3).
Table III. Predicting mental distress and job induced tension: standardized regression coefficients (N�/954).
Mental distress b last step Job-induced tension b last step
Step 1: Individual background characteristics
Age .10** �/.05
Gender (female) �/.07* �/.08**
Partner �/.08** .00
Physician �/.16*** �/.04
Nurse �/.13*** �/.06
Weekly working hours �/.02 .09*
Organizational tenure .04 .04
R2change .03*** .04***
Step 2: Employment contract
Part-time contract .01 .01
Fixed-term contract �/.02 .00
On-call contract .05 �/.02
R2change .01* n.s.
Step 3: Job conditions
Job insecurity .18*** .19***
Job control �/.27*** �/.11**
Job demands .28*** .51***
R2change .20*** .30***
Step 4: Employment x Job Insecurity
Part-time�/Job insecurity �/.05 �/.07*
Fixed-term�/Job insecurity .07 .01
On-call�/Job insecurity �/.11** �/.08*
R2change .02*** .01*
Step 5: Employment�/Job control
R2change n.s. n.s
Step 6: Employment�/Job demands
R2change n.s. n.s.
*p 5/.05; **p 5/.01; ***p 5/.001.
Employment contracts and health complaints 311
Page 12
Among the individual background variables, only (female) gender and the number of
weekly working hours were positively related to job-induced tension (R2change�/.04; p B/
.001). With the effect of individual background variables held constant, there was no
significant increase in the explained variance when the types of employment contract were
entered in Step 2. All perceived job conditions entered in the third step were significantly
associated with job-induced tension, and the proportion of explained variance increased to
34% (p B/.001). Job demands and job insecurity were positively associated with job-induced
tension, whereas the effect of job control was negative. Again, in steps 4 to 6, only the
interactions between forms of employment contract and job insecurity (Step 4) explained
additional variance in the outcome (R2change�/.01; pB/ .05). Both part-time employees and
on-call workers, but not fixed-term workers, displayed associations between job insecurity
and job-induced tension that differed from full-time permanent workers. As can be seen in
Figure 2, job-induced tension increased significantly for permanent full-time workers when
they perceived high levels of job insecurity (t�/4.66, p B/.001). For core part-time (t�/ 1.52,
p �/.05) and on-call workers (t�/ 0.05, p �/.05), the level of perceived job insecurity did not
affect job-induced tension. These results were unaffected by the order in which interaction
terms were added.
Since there may be a risk of multicollinearity between working hours, organizational
tenure and profession, with employment contract reducing the impact of the predictor
variable, several regressions were run, leaving out one or all of these control factors. In
general, these supplementary analyses produced similar results, with the same pattern of
significant predictors.
Discussion
Previous research on alternative work arrangements and well-being highlights the
importance of considering individual background characteristics of the contract holders
and, in addition, of going beyond a simple distinction between permanent work and a
uniform category of alternative employment contracts. Furthermore, research suggests that
the understanding of the consequences of alternative employment is enhanced by taking
into account other explanatory effects that possibly may be intertwined with the type of
employment contract, among which perceived job conditions are believed to have a major
potential impact. Thus, the overall objective of this study was to investigate the direct and
2.4
2.6
2.8
3
2low job insecurity high job insecurity
Full time
Part time
On-call
2.2
Figure 2. Interaction of perceived job insecurity and type of contract (core full-time, core part-time and on-call) on
job induced tension (scale 1�/5).
312 C. Bernhard-Oettel et al.
Page 13
interactive effects of employment contract status and job conditions on employee health
complaints once individual background characteristics were controlled for.
In order to broaden the understanding of the characteristics of different types of
employment contract, we initially compared three categories of alternative employment
(part-time, fixed-term and on-call contracts) with permanent employment. Consistent with
previous research, we found some support for a core�/periphery distinction, but only for job
insecurity as well as the distribution of age and organizational tenure (cf. results reported by
Sverke et al., 2000, and OECD, 2002). Fixed-term workers were found to be rather similar
to core full-time employees and clearly different from both part-time and on-call workers.
Part-time workers shared some characteristics with those on full-time contracts (e.g.
relatively positive job perceptions) but were similar to on-call workers in other respects (e.g.
a low number of weekly working hours). As argued in previous research (Aronsson et al.,
2002; Bernhard & Sverke, 2003), these results underline important distinctions among
different types of alternative employment. However, the differences and similarities
obtained in the present data are not clear-cut and challenge the idea of a core�/periphery
perspective.
This complex pattern of similarities and differences between forms of contract may
explain why it remains unclear whether and how health and well-being are related to
differences in employment contracts. When individual background characteristics were held
constant, contract status was unrelated to job-induced tension and only marginally
contributed to the variance explained in mental distress. The fact that these results held
up also when background characteristics that might be related to employment contract
status (i.e. tenure, position or working hours) were excluded from the model calculation
lends further credibility to our findings. Thus, a rather minimal association between
contract and well-being was found (for similar findings, see Krausz et al., 2000), even
though alternative employment types were differentiated. This finding contradicts the
suggested core�/periphery continuum, according to which permanent workers should have
the best and on-call workers the worst conditions of health and well-being (Aronsson et al.,
2002). It is possible that different types of employment contract do not automatically reflect
differences in employment conditions along a core�/periphery continuum. For example,
organizational tenure was initially assumed to be much shorter in contracts localized at the
more peripheral end of the continuum (Sverke et al., 2000), but tenure appeared to be
relatively long in all contracts of this sample. Hence, merely comparing forms of
employment contracts may not reveal much information about differences or similarities
in employee well-being if the specifics of employment terms are not studied.
This study also argued that subjective perceptions of one’s job conditions might be
another possible major explanation of well-being in different employment situations.
Indeed, in agreement with previous research, we found impaired mental health and higher
levels of job-induced tension to be associated with higher perceived job demands and
experiences of lower job control (de Lange et al., 2003) as well as higher levels of job
insecurity (Sverke et al., 2002). All three perceptions increased the proportion of explained
variance in health complaints markedly, and thus our results support the conclusion that
subjective perceptions of job conditions, compared to objective conditions (i.e. contract
status), are stronger predictors of employee health and well-being (for a similar conclusion,
see, for example, Krausz et al., 2000).
This study further examined the potential interaction effects between type of employ-
ment contract and perceived job conditions on health complaints. Type of employment
contract was not found to alter the effects of job demands or job control on employee well-
Employment contracts and health complaints 313
Page 14
being. In line with earlier studies (De Witte & Naswall, 2003; Virtanen et al., 2002),
however, significant interaction effects were found between types of employment contract
and job insecurity. Fixed-term workers reacted just like core full-time employees, with
impaired well-being when experiencing high job insecurity, while health complaints of
on-call workers and job-induced tension of core part-time as well as on-call workers
were unrelated to job insecurity. Thus, negative effects of job insecurity on health
complaints appear to depend on employment circumstances, and the relevance of
distinguishing different alternative employment contracts becomes obvious. However, the
present results challenge previous findings that job insecurity affects only permanent
workers because it is not part of their employment contract (De Witte & Naswall, 2003;
Virtanen et al., 2002).
Theories of balance of the psychological contract between employer and employee in
terms of efforts and rewards may be added here (Blau, 1964; Siegrist, 1996), suggesting
that if future rewards are promised by the employer or assumed by the employee (e.g. a
prolonged or permanent contract), sacrifices (e.g. high job insecurity) may be acceptable for
employees in the short term, if the balance is re-adjusted afterwards. The individual motives
to take up alternative employment may influence the relevance of job security, since the
effect of job insecurity may, for example, be altered for individuals with lower work
involvement (Hellgren, 2003). In terms of the absence of interaction effects between
employment contract types and perceived control and demands, this exploratory study
cannot draw a firm conclusion on whether this finding shows the non-existence of such an
effect. Since there are no earlier studies for comparison, it may be equally plausible that
relations between perceived job conditions and well-being are altered only for certain
categories of temporary workers (e.g. individuals who have an intermittent relationship with
the organization) or under certain circumstances (e.g. during periods when employers make
decisions about contract renewal).
Limitations
As with other research, there may be limitations to our study that influence its conclusions.
First, although a number of individual background characteristics were controlled for, other
individual as well as work-related variables might be of interest, e.g. household income,
number of children and work involvement. An interesting effect to be addressed in future
research concerns the nature of the relations between background characteristics and
outcomes. Albeit not present in this study, it is plausible that associations between, for
instance, working hours and health outcomes may be U-shaped, such that too few as well as
too many working hours may be associated with health problems. Second, only three types
of alternative work arrangements were studied and compared to permanent full-time work.
Despite the fact that our distinction between different forms of alternative employment
extends the focus of many previous comparative analyses on employment contracts and
well-being, the results presented here do not allow generalization to all forms of alternative
employment. A third limitation is that we studied only certain aspects of job perceptions,
although previous research also shows that factors such as volition (Krausz, Brandwein, &
Fox, 1995), social support (Isaksson & Bellaagh, 2002) and psychological contract
perceptions (De Cuyper, Isaksson, & De Witte, 2005) may be of importance for employees’
health and well-being. Moreover, although the empirical support for interactive effects
between employment contract types and perceived job conditions was limited, it is
conceivable that employment contract interacts with other perceptions of employment,
job conditions or aspects of choice.
314 C. Bernhard-Oettel et al.
Page 15
Another potential limitation to the present findings concerns sample restrictions. The
current study was restricted to healthcare workers, who may share certain role character-
istics (Sverke et al., 2000) and whose perceptions may be related to circumstances in the
labour market for caring occupations (Isaksson, Bernhard, & Gustafsson, 2003). Expand-
ing the analyses to other organizational circumstances and sector settings would be fruitful,
since several research studies hint at problems of generalization across occupations or
sectors, which may use different alternative contracts with job conditions tailored to their
specific requirements (Cohen, Haberfeld, & Ferber, 1993). Another sample specific that
may have influenced the effects of perceived job conditions and employment circumstances
is the rather high organizational tenure of temporary employees in our study.
Yet another limitation can be seen in the use of cross-sectional data, making causal
inferences more difficult. Although our results, for the most part, are in line with theoretical
propositions and earlier research, it appears to be important to replicate the study
longitudinally, for instance to control for overestimation of the effects of job perceptions.
Further, in this study we chose to limit our analysis to measures of job-related and mental
health complaints. However, a challenging question is to what extent employment
conditions as well as perceptions of job conditions constitute health risks and how they
are associated with diagnosed diseases, which would constitute extra difficulties for
transition into permanent or full-time employment.
Concluding remarks
Despite these limitations, this study enhances scientific understanding of forms of
alternative employment, and adds to previous research in several ways. First, an effort
was made to differentiate between forms of alternative employment. The results show that
alternative employment contracts are associated with important differences, thus indicating
that a basic distinction between core and alternative work forms may be an over-
simplification. However, whether the proposed categorization into a core�/periphery
perspective (Aronsson et al., 2002) is helpful in depicting differences of employment and
job conditions of different contract types remains an open question. There are several
dimensions according to which differences between contracts can emerge, and further
research on other contract types and other branches is needed to evaluate how employment
contracts are related to health.
Second, the results show the benefit of complementing studies on employment contract
effects with the effects of job perceptions. Subjective perceptions of job conditions are
important in understanding consequences for health, but they can furthermore have
combined effects with the form of contract, which may alter their meaning for employees’
well-being in certain employment circumstances. Here, perhaps, practitioners and
researchers can gain more knowledge by analysing precisely what employment conditions
are tied to the contracts under investigation, particularly since variations in forms of
employment are almost infinite and not invariant across sectorial and national boundaries
(McLean Parks et al., 1998).
In conclusion, integrating objective as well as subjective aspects relating to employment
and jobs in analyses of future research seems to be highly necessary in order to shed more
light on the question of how alternative employment arrangements may affect individual
well-being and health.
Employment contracts and health complaints 315
Page 16
Acknowledgements
The research presented here was financed by the Swedish Council for Working Life and
Social Research and Stockholm County Council. The authors thank Peter Richter for
comments on a previous version of the manuscript.
References
Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting interactions . Newbury Park, CA: Sage.
Aronsson, G. (1999). Contingent work and health and safety. Work, Employment & Society, 13 , 439�/459.
Aronsson, G., Dallner, M., & Lindh, T. (2000). Flexibla inkomster och fasta utgifter [Flexible income and fixed
costs]. Arbete och Halsa , 20 , 1�/26.
Aronsson, G., & Goransson, S. (1999). Permanent employment but not in a preferred occupation: Psychological
and medical aspects, research implications. Journal of Organisational Health Psychology, 4 , 152�/163.
Aronsson, G., Gustafsson, K., & Dallner, M. (2002). Work environment and health in different types of temporary
jobs. European Journal of Work and Organizational Psychology, 11 , 151�/175.
Ashford, S. J., Lee, C., & Bobko, P. (1989). Content, causes and consequences of job insecurity: A theory-based
measure and substantive test. Academy of Management Journal , 32 , 803�/829.
Beard, K. M., & Edwards, J. R. (1995). Employees at risk: Contingent work and the psychological experience of
contingent workers. In C. L. Cooper, & D. M. Rousseau (Eds), Trends in organizational behaviour (pp. 109�/
126). Chichester, UK: John Wiley.
Beehr, T. A., Walsh, J. T., & Taber, T. D. (1976). Relationship of stress to individually and organizationally valued
states: Higher order needs as a moderator. Journal of Applied Psychology, 61 , 41�/47.
Benach, J., Gimeno, D., & Benavides, F. G. (2002). Types of employment and health in the European Union . Dublin:
European Foundation for the Improvement of Living and Working Conditions.
Benavides, F. G., Benach, J., Diez-Roux, A. V., & Roman, C. (2000). How do types of employment relate to health
indicators? Findings from the second European Survey on Working Conditions. Journal of Epidemiology and
Community Health , 54 , 494�/501.
Bernhard, C., & Sverke, M. (2003). Work attitudes, role stress and health among different types of contingent
workers in the Swedish health care sector. Research and Practice in Human Resource Management , 11 , 1�/16.
Blau, P. M. (1964). Exchange and power in social life . New York: Wiley.
Claes, R., De Witte, H., Schalk, R., Guest, D., Isaksson, K., Krausz, M., et al. (2002). Het psychologisch contract
van vaste en tijdelijke werknemers (The psychological contract of permanent and temporary workers). Gedrag en
Organisatie , 15 , 436�/455.
Cohen, S., Haberfeld, L., & Ferber, G. (1993). Temporary help service workers: Employment characteristics and
age determination. Industrial Relations , 32 , 272�/287.
Conway, N., & Briner, R. B. (2002). Full-time versus part-time employees: Understanding the links between work
status, the psychological contract, and attitudes. Journal of Vocational Behavior , 61 , 279�/301.
Corral, A., & Isusi, I. (2004). Part-time work in Europe . Dublin: European Foundation for the Improvement of
Living and Working Conditions.
Davis-Blake, A., & Uzzi, B. (1993). Determinants of employment externalisation: A study of temporary workers
and independent contractors. Administrative Science Quarterly , 38 , 195�/223.
De Cuyper, N., Isaksson, K., & De Witte, H. (Eds). (2005). Employment contracts and well-being among European
workers. Review of the literature . Aldershot: Ashgate.
de Jonge, J., Dorman, C., Janssen, P. P. M., Dollard, M. F., Landeweerd, J. A., & Nijhuis, F. J. N. (2001). Testing
reciprocal relationships between job characteristics and psychological well-being: A cross lagged structural
equation model. Journal of Occupational and Organizational Psychology, 74 , 29�/46.
de Lange, A. H., Taris, T. W., Kompier, M. A. J., Houtman, I., & Bongers, P. M. (2003). ‘The very best of the
Millenium’: Longitudinal research and the demand-control-(support) model. Journal of Occupational Health
Psychology, 8 , 282�/305.
De Witte, H., & Naswall, K. (2003). Objective versus subjective job insecurity: Consequences of temporary work
for job satisfaction and organizational commitment in four European countries. Economic and Industrial
Democracy, 24 , 149�/188.
Goldberg, D. (1979). Manual of the General Health Questionnaire . Windsor: National Foundation for Educational
Research.
Goudswaard, A., & Andries, F. (2002). Employment status and working conditions . Luxembourg: Office for Official
Publications of the European Communities.
316 C. Bernhard-Oettel et al.
Page 17
Goudswaard, A., & de Nanteuil, M. (2000). Flexibility and working conditions: A European bibliographical review.
Dublin: European Foundation for the Improvement of Living and Working Conditions.
Gustafsson, S., Kenjoh, E., & Wetzels, C. (2001). Employment choices and pay differences between non-standard and
standard work in Britain, Germany, Netherlands and Sweden . Amsterdam: Tinbergen Institute.
Hackman, J. R., & Oldham, G. R. (1975). Development of a job diagnostic survey. Journal of Applied Psychology,
60 , 159�/170.
Hartley, J., Jacobson, D., Klandermans, B., & Van Vuuren, T. (1991). Job insecurity. Coping with jobs at risk .
Thousand Oaks, CA: Sage.
Hellgren, J. (2003). The times they are a-changin’ . Stockholm: Stockholm University.
House, R. J., & Rizzo, J. R. (1972). Role conflict and ambiguity as critical variables in a model of organizational
behavior. Organizational Behavior and Human Performance , 7 , 467�/505.
Isaksson, K. S., & Bellaagh, K. (2002). Health problems and quitting among female ‘temps’. European Journal of
Work and Organizational Psychology, 11 , 27�/45.
Isaksson, K., Bernhard, C., & Gustafsson, K. (2003). Betydelsen av anstallningsform och tillit till ledningen for
personalens halsa inom varden [The importance of employment contract and trust in management for health
care workers’ well-being]. Arbetsmarknad & Arbetsliv, 9 , 95�/106.
Karasek, R. A. (1979). Job demands, job decision latitude, and mental strain: Implications for job redesign.
Administrative Science Quarterly, 24 , 285�/308.
Kinnunen, U., & Natti, J. (1994). Job insecurity in Finland: Antecedents and consequences. European Journal of
Work and Organizational Psychology, 4 , 297�/321.
Klein Hesselink, D. J., & Van Vuuren, T. (1999). Job flexibility and job insecurity: The Dutch case. European
Journal of Work and Organisational Psychology, 8 , 273�/293.
Krausz, M., Brandwein, T., & Fox, S. (1995). Work attitudes and emotional responses of permanent, involuntary,
and voluntary temporary help employees: An exploratory study. Applied Psychology: An International Review, 44 ,
217�/232.
Krausz, M., Sagie, A., & Bidermann, Y. (2000). Actual and preferred work schedules and scheduling control as
determinants of job-related attitudes. Journal of Vocational Behavior , 56 , 1�/11.
Letourneux, V. (1998). Precarious employment and working conditions in the European Union . Dublin: European
Foundation for the Improvement of Living and Working Conditions.
Martens, M. F. J., Nijhuis, F. J. N., van Boxtel, M. P. J., & Knottnerus, J. A. (1999). Flexible work schedules and
mental physical health. A study of a working population with non-traditional working hours. Journal of
Organizational Behavior , 20 , 35�/46.
McLean Parks, J., Kidder, D. L., & Gallagher, D. G. (1998). Fitting square pegs into round holes: Mapping the
domain of contingent work arrangements onto the psychological contract. Journal of Organizational Behavior ,
19 , 697�/730.
Morrison, D., Payne, R. L., & Wall, T. D. (2003). Is a job a viable unit of analysis? A multilevel analysis of demand-
control-support models. Journal of Occupational Health Psychology, 8 , 209�/219.
Organisation for Economic Co-operation and Development [OECD] (2002). OECD employment outlook . Paris:
OECD.
Paoli, P. (1996). Working conditions in the European Union (Summary) . Dublin: European Foundation for the
Improvement of Living and Working Conditions.
Paoli, P., & Merllie, D. (2001). Third European Survey on Working Conditions 2000 . Dublin: European Foundation
for the Improvement of Living and Working Conditions.
Reilly, P. A. (1998). Balancing flexibility �/ meeting the interests of employer and employee. European Journal of
Work and Organizational Psychology, 7 , 7�/22.
Richter, P., Hemman, E., Merboth, H., Fritz, S., Hansgen, C., & Rudolf, M. (2000). Das Erleben von
Arbeitsintensitat und Tatigkeitsspielraum �/ Entwicklung und Validierung eines Fragebogens zur orientierenden
Analyse (FIT) [Perception of work intensity and control*/development and validation of a questionaire].
Zeitschrift fuer Arbeits- und Organisationspychologie , 44 , 129�/139.
Siegrist, J. (1996). Adverse health effects of high-effort/low-reward conditions. Journal of Occupational Health
Psychology, 1 , 27�/41.
Sparrow, P., & Cooper, C. L. (2003). The employment relationship: Key challenges for HR . Oxford: Butterworth-
Heinemann.
Strazdins, L., D’Souza, R. M., Lim, L. L.-Y., Broom, D. H., & Rodgers, B. (2004). Job strain, job insecurity, and
health: Rethinking the relationship. Journal of Occupational Health Psychology, 9 , 296�/305.
Sverke, M., Gallagher, D. G., & Hellgren, J. (2000). Alternative work arrangements: Job stress, well-being, and
work attitudes among employees with different employment contracts. In K. Isaksson, L. Hogstedt, C. Eriksson,
& T. Theorell (Eds), Health effects of the new labour market (pp. 145�/167). New York: Plenum.
Employment contracts and health complaints 317
Page 18
Sverke, M., Hellgren, J., & Naswall, K. (2002). No security: A meta-analysis and review of job insecurity and its
consequences. Journal of Occupational Health Psychology, 7 , 242�/264.
Sverke, M., & Sjoberg, A. (1994). Dual commitment to company and union in Sweden: An examination of
predictors and taxonomic split methods. Economic and Industrial Democracy , 15 , 531�/564.
Virtanen, M. (2003). Temporary employment and health . Helsinki: Finnish Institute of Occupational Health.
Virtanen, M., Kivimaki, M., Elovainio, M., Vahtera, J., & Cooper, C. L. (2001). Contingent employment, health
and sickness absence. Scandinavian Journal of Work. Environment & Health , 27 , 365�/372.
Virtanen, P., Vahtera, M., Kivimaki, J., & Pentii, J. F. (2002). Employment security and health. Journal of
Epidemiological Community Health , 56 , 569�/574.
Walsh, J., Taber, T. D., & Beehr, T. A. (1980). An integrated model of perceived job characteristics. Organizational
Behavior and Human Performance , 25 , 252�/267.
Wikman, A. (2002). Temporara kontrakt och inlasningseffekter (Temporary contracts and being locked-in effects).
Arbetsliv i Omvandling , 4.
Wikman, A., Andersson, A., & Bastin, M. (1998). Nya relationer i arbetslivet (New relations in working life) .
Stockholm: National Institute for Working Life.
318 C. Bernhard-Oettel et al.