Internal Factors Affect Change of Leadership in Organization

The current issue and full text archive of this journal is available at www. emeraldinsight. com/1753-8351. htm IJWHM 2,1 Work-related health attributions: their impact on work attitudes ? ? Sara Goransson, Katharina Naswall and Magnus Sverke 6 Department of Psychology, Stockholm University, Stockholm, Sweden Abstract Purpose – The purpose of this study is to introduce the concept of work-related health attributions and investigate the effects of such perceptions as well as of health status on work-related attitudes and turnover intentions.Design/methodology/approach – Building on attribution theory, the study tests the assumption that negative work-related health attributions impair employee work-related attitudes and intentions, and moderate the relation between health status and work-related attitudes. Cross-sectional questionnaire data from 785 Swedish retail white-collar workers are collected to test these assumptions by utilizing moderated regression analyses.Find ings – The results show that negative work-related health attributions are related to lower levels of job satisfaction and organizational commitment as well as higher levels of turnover intention, even after controlling for demographics, work climate variables, and mental distress. Further, the signi? cant interaction between attributions and mental distress indicates that it makes a difference for employees’ turnover intentions if an individual with high mental distress attributes it to work or not.Practical implications – Work-related health attributions should be taken into account in order to avoid impaired levels of employee work motivation. The measure introduced renders it possible to identify and help those individuals who believe that work affects their health negatively. Originality/value – The results underscore the relevance of how individuals think their health is affected by their work, and contributes to the understanding of how health stat us relates to work-related attitudes.Since the measure of work-related health attributions is easily administered it is also valuable for practitioners working with employee health and attitudes. Keywords Personal health, Job satisfaction, Employee turnover, Sick leave, Sweden Paper type Research paper International Journal of Workplace Health Management Vol. 2 No. 1, 2009 pp. 6-21 q Emerald Group Publishing Limited 1753-8351 DOI 10. 1108/17538350910945974 1. Introduction The high rates of sick leave and the evidence regarding the effects of work on employee health in general have received increased attention in both research and media as of ? ate (Catalan Matamoros et al. , 2007). This increased attention, along with the importance of work and health to individuals, is likely to have contributed to an increased interest among employees in how work actually affects their health (Harding and Hikspoors, 1995; Harpaz, 2002; Kallenberg and Larsson, 2000). The Third European Study on Wor king Conditions indicates, for instance, that 27 percent of employees believe that their health and safety are at risk because of their work (European Foundation, 2001).Further, a Swedish study on persons on long-term sick-leave found that 66 percent of those asked named work as the underlying cause for ? their sick-leave (Goransson et al. , 2002). Moreover, Ettner and Grzywacz (2001) found that individuals’ perceptions of how work affects health tend to differ between different work situations, even after dispositional differences were controlled for. None of these studies, however, have studied how employee attitudes and behavioral intentions may be affected by such a negative view of how work affects health.We introduce the concept of work-related health attributions, which we argue may add to the understanding of the factors that contribute to variation in job satisfaction, organizational commitment, and the will to stay with the organization. Whereas numerous studies hav e documented ill-health at work to be related to low job satisfaction (Bradley and Cartwright, 2002; Fairbrother and Warn, 2001; Judge and Watanabe, 1993; Spector, 1997; Wright and Bonett, 2007), low organizational commitment (Leong et al. 1996; Mathieu and Zajac, 1990), as well as turnover cognitions and behaviors (Houkes et al. , 2001; Lee, 1988; Wright and Bonett, 2007), a closer examination of these studies reveals that there are differences in the strength of the association between ill-health and various outcomes depending on the measures used, but also between studies using the same measures. We propose that work-related health attributions can be an important part of the mechanism determining the relationship between ill-health and various outcomes.It is likely that an individual who is not feeling well and who attributes this to factors at work will be less satis? ed with work, less committed, and more prone to leave the organization, compared to an individual with same lev el of ill-health but who does not associate their health status with factors at work. The purpose of the present study is to investigate whether negative work-related health attributions are separate from ill-health, and if the concept contributes to the understanding of negative attitudes towards the job and the organization, and turnover intentions. . 1. Work-related health attributions Work-related health attributions concern the degree to which the individual considers working conditions to be a strong contributing factor to her health status. The perceived contribution of work to health can be positive, negative, or none at all. We propose that work-related health attributions represent an important factor in explaining the link between health and work-related attitudes.By taking work-related health attributions into account it is possible to study differences between those who experience ill-health and attribute this to working conditions and others with ill-health which they do not believe is caused by work. According to attribution theory, people tend to look for causes for events and situations they experience (Michotte, 1963; Weiner, 1985; Stratton, 1997), especially when the outcome is negative, such as in the case of ill-health.Work-related health attributions, then, can be viewed as a type of attribution, since they entail the individual’s explanation of events (in this case how work in? uences health). Furthermore, work-related health attributions focus on how the individual evaluates the contribution of work to her current or future health status, implying that even if the individual currently does not experience any ill-health due to work, she may see a risk of work affecting her health negatively if the situation does not change.Drawing on theories on social representation (e. g. Moscovici, 2000), we also suggest that beliefs are formed from individuals’ experiences as well as socially created and shared ideas and knowledge in gr oups. Such knowledge may consist of health complaints and sickness absence among people surrounding an individual, as well as the way work and its in? uence on health is discussed in research and media (see, for ? example, Catalan Matamoros et al. , 2007), and is important in shaping the individual’s perceptions of the impact of working conditions on health.Thus, negative work-related health attributions may arise both from an individual’s own experiences of a negative impact of work on health and from the individual’s awareness of a connection between Work-related health attributions 7 IJWHM 2,1 8 work and ill-health in the organization and society. The social aspects in the formation of attributions have been observed in previous research such as, for instance, a study where it was found that norms and social debate affected the overall attitude towards absenteeism in organizations (Van Vuuren et al. 2003). At this point, one important distinction should be ma de. We consider work-related health attributions to be separate from actual ill-health at work. Actual ill-health is a general concept and takes a person’s current health status and well-being into account, including such factors as problems with social functioning, sleeping problems, low self-esteem, and depression. Work-related health attributions, on the other hand, are work-focused and entail to what extent individuals perceive that work in? ences health, and do not have to be connected to actual symptoms or health status. Work-related health attributions can, in other words, be negative in spite of a person being in good health. Conversely, an individual can have symptoms of ill-health but believe that work has nothing to do with these symptoms. This discrepancy between health status and work-related health attributions may be due to early signs of health problems, or a conviction that the current work situation is not bene? ial to one’s health in the long run, bu t it is ultimately based on the individual’s evaluation of the situation. Figure 1 summarizes the relation between ill-health and work-related health attributions, and how they are expected to relate to work attitudes and turnover intentions. 1. 2. Consequences of work-related health attributions Expectancies and beliefs, of which work-related health attributions are one example, tend to affect attitudes and behaviors (Ajzen and Fishbein, 1980; Holland et al. , 2002; Weiner, 1985).In addition, the value placed on an object of interest plays an important part in shaping the attitudes toward that object (Ajzen and Fishbein, 1980). Since health is important and highly valued by most individuals, we expect that individuals with negative work-related health attributions also hold more negative attitudes towards work, such as lower job satisfaction and organizational commitment, and higher levels of turnover intention, compared to individuals with less negative work-related health attributions.In all employment relationships, it is the employer who is responsible for providing a decent work climate, preventing health risks at work, and making efforts to keep the employees healthy (Walters, 2002). Employees expect to stay healthy at work because of labor legislation which emphasizes the employers’ responsibility for occupational safety and health. Employers who take measures to safeguard the work environment show that they value their employees and care about their well-being, and reinforce the Figure 1. The relation between ill-health and work-related ealth attributions and their impact on work-related attitudes employees’ expectations of fair treatment in the employee-employer relationship. In response to such support from the organization, employees may develop more positive attitudes toward the job and increase their loyalty towards the organization (see Robinson et al. , 1994; Shore and Wayne, 1993). However, employees with negative work-rel ated health attributions probably hold their employer more or less responsible and exhibit strong negative reactions toward the organization (cf.Robinson and Morrison, 2000; Shore and Tetrick, 1994). Such reactions can include feelings of anger, disappointment, or resentment toward the party perceived as being responsible, which can be manifested as lowered satisfaction with the job and decreased commitment to the organization. Moreover, it has been argued that employees (especially those most quali? ed) tend to leave the company when they experience unfavorable working conditions (e. g. Hirschman, 1970; Pfeffer, 1998).The same might apply to individuals high on negative work-related health attributions; those who have the option may very well leave the organization if the employer does not take steps to remedy the conditions perceived to be threatening the individual’s health. We suggest that work-related health attributions should be taken into account when studying the rel ation between employee health and attitudes toward the organization, especially given that previous research on this relation has found differing results.However, the studies conducted consistently indicate that employee health may have a negative impact on job satisfaction, commitment and the intention to remain with the organization (see, for example, Leong et al. , 1996; Wright and Bonett, 2007). We believe that work-related health attributions make up an important part of the mechanism between employee health and its potential outcomes. 1. 3. The present studyThe purpose of the present study was to investigate the role of work-related health attributions in the context of mental distress and work-related attitudes (job satisfaction and organizational commitment) and behavior (turnover intention). We propose that low employee well-being and negative work-related health attributions, respectively, are associated with lower levels of job satisfaction and organizational commitment, and higher levels of turnover intention.We also propose that work-related health attributions moderate the relation between mental distress and work-related outcomes, in the sense that the negative effect of ill-health on work-related attitudes is stronger among individuals who hold negative work-related health attributions than among those with more neutral work-related health attributions. Since work-related health attributions can be considered a rather new empirical construct, we analyze its conceptual relation to mental distress before going on to testing our propositions.Research has shown that characteristics of the work environment are important in shaping attitudes such as job satisfaction and organizational commitment (Fairbrother and Warn, 2001; Parker et al. , 2003), and turnover intention (Griffeth et al. , 2000; Lambert et al. , 2001). In order not to overestimate the effect of work-related health attributions on outcome variables, we control for factors describing the climate at work (job autonomy, quantitative role overload, qualitative role overload, workgroup cohesiveness, and job challenge), in addition to demographic characteristics (gender, age and education).The research model is graphically represented in Figure 2). Work-related health attributions 9 IJWHM 2,1 10 Figure 2. Research model 2. Method 2. 1. Participants and procedure Data were collected within the framework of a project investigating how to attract, develop and retain white-collar workers in the Swedish retail and wholesale sectors. A random sample of 1,589 individuals was drawn from the membership roster of the Union of Commercial Salaried Employees (HTF), which is af? liated with the Swedish Confederation of Professional Employees (TCO).The HTF, which is the union representing white-collar employees in this particular sector, has a unionization rate of approximately 80 percent (Kjellberg, 2001). Questionnaires were mailed to the members’ homes during the spring of 2 002. A cover letter explained the general aim of the study and included information about compensation for participation in the study (a gift certi? cate was raf? ed off among the respondents) and that participation was voluntary. Participants returned their questionnaires in pre-addressed, postage-paid envelopes. Two follow-up mailings were administered to increase the response rate, one with a new questionnaire.The response rate after two follow-ups was 52 percent (n ? 829). After listwise deletion of missing data, the sample comprised 785 persons. The mean age of the participants was 44 years (SD ? 11), women made up 54 percent of the sample, 14 percent had a university exam, and the average organizational tenure was 11 years (SD ? 10). 2. 2. Measures Table I presents the correlations, mean values, standard deviations, and reliability estimates (Cronbach’s alpha) for all study variables. In general, the measures exhibited adequate reliability (alpha exceeding 0. 70), with the exception of qualitative role overload (alpha ? . 59). 2. 2. 1. Demographics. Gender (1 ? woman, 0 ? man) and education (1 ? university degree, 0 ? lower education) were assessed as dichotomous variables. Age was measured in years. 2 Mean SD Minimum Maximum Alpha 0. 54 0. 50 0 1 – 43. 57 1. 92 21 65 – Gender (woman) – Age 2 0. 06 – Education (University) 0. 01 2 0. 04 Job autonomy 2 0. 15 0. 13 Quantitative role overload 2 0. 08 0. 00 Qualitative role overload 2 0. 09 0. 05 Work group cohesiveness 2 0. 04 0. 04 Job challenge 2 0. 17 0. 15 Negative work-related health attributions 0. 01 2 0. 10 Mental distress 0. 12 2 0. 10 Job satisfaction 2 0. 03 0. 15Organizational commitment 2 0. 04 0. 23 Turnover intention 2 0. 03 2 0. 20 1 0. 14 0. 35 0 1 – – 2 0. 05 0. 00 2 0. 09 2 0. 02 2 0. 03 0. 02 2 0. 04 2 0. 08 2 0. 11 0. 08 3 5 6 7 8 9 10 11 12 3. 54 0. 90 1 5 0. 78 3. 44 0. 91 1 5 0. 78 2. 22 0. 78 1 5 0. 59 3. 74 0. 96 1 5 0. 84 3. 57 0. 8 7 1 5 0. 74 2. 52 1. 08 1 5 0. 84 9. 19 5. 56 0 32 0. 88 3. 74 0. 91 1 5 0. 88 3. 08 0. 94 1 5 0. 71 – 2 0. 13 – 2 0. 08 0. 31 – 0. 32 2 0. 09 2 0. 13 – 0. 38 0. 14 0. 13 0. 35 – 2 0. 40 0. 42 0. 35 2 0. 35 2 0. 20 – 2 0. 37 0. 28 0. 27 2 0. 34 2 0. 26 0. 56 – 0. 63 2 0. 09 2 0. 11 0. 47 0. 55 2 0. 52 2 0. 49 – 0. 48 2 0. 02 0. 03 0. 8 0. 53 2 0. 34 2 0. 32 0. 71 – 2 0. 39 0. 10 0. 10 2 0. 35 2 0. 35 0. 50 0. 40 2 0. 65 2 0. 56 4 Notes: – Not applicable; for r . 0. 06, p , 0. 05; r . 0. 08, p , 0. 01; r . 0. 10, p , 0. 001 (n = 785) 1 2 3 4 5 6 7 8 9 10 11 12 13 Variable 2. 04 1. 06 1 5 0. 80 – 13 Work-related health attributions 11 Table I. Correlations and descriptive statistics for the variables in the analysis IJWHM 2,1 12 Table II. Results of con? rmatory factor analysis of work-related health attributions and mental distress 2. 2. 2. Work climate. All climate variables were measured using a ? ve-point Likert scale (1 ? strongly disagree; 5 ? trongly agree). Job autonomy was assessed with a ? three-item index (Sverke and Sjoberg, 1994), including items (e. g. â€Å"I have enough freedom as to how I do my work†) drawn from Hackman and Oldham (1975) and Walsh et al. (1980). Quantitative role overload was measured using a three-item scale (Beehr et al. , 1976), consisting of items such as â€Å"I often have too much to do in my job†. Qualitative role overload was measured with three items (e. g. â€Å"I consider my responsibilities to be unreasonable;† Sverke et al. , 1999). Three items drawn from Nystedt (1992) were used to assess work group cohesiveness (e. g. Members stick together in my work group†). Job challenge was measured with a three-item scale (e. g. â€Å"I’m learning new things all the time in my job†) developed by Hellgren et al. (1997). 2. 2. 3. Health-related variables. We used the short version of the General Health Questionn aire (GHQ-12) to measure mental distress (Goldberg and Williams, 1988). The 12 items (e. g. â€Å"Have you been feeling unhappy and depressed during the last two weeks? †) were scored on a scale ranging from 0 (never/hardly ever) to 3 (always/almost always). Work-related health attributions were measured with a three-item scale developed for the present study.The items (â€Å"I believe that my work affects my health in a negative way†; â€Å"I think I can continue to work as I do now and remain healthy in the long run† [reverse coded]; â€Å"If I had another job my health would probably be better†) were scored on a ? ve-point Likert scale (1 ? strongly disagree; 5 ? strongly agree). The scale measuring work-related health attributions demonstrated a satisfactory internal consistency (Cronbach’s alpha ? 0. 74), and we went on to investigate whether work-related health attributions and mental distress represented two distinct constructs. This was tes ted using con? rmatory factor analysis.Three subscales with four GHQ items in every scale were constructed (see Mathieu and Farr, 1991) creating three parallel indices. The GHQ subscales and the items measuring work-related health attributions were then subjected to the con? rmatory factor ? ? analysis procedures in Lisrel 8 (Joreskog and Sorbom, 1996). The chi-square value did not indicate a perfect ? t between the two-factor model and the data (see Table II), but since the chi-square test is sensitive to sample size (Bentler and Bonett, 1980), we also relied on other indicators to determine model ? t. The adjusted goodness-of-? t index ? ? AGFI; Joreskog and Sorbom, 1996), the normed ? t index (NFI; Bentler and Bonett, 1980), the Akaike measure (AIC; Akaike, 1987), the standardized root mean square ? ? residual (SRMR; Joreskog and Sorbom, 1996) and the root mean square error of approximation (RMSEA; Browne and Cudeck, 1993) indicated that the two-factor model provided an improveme nt in ? t as compared to a model that tested for a single factor. Thus, our results indicate that work-related health attributions and mental distress represent two distinct constructs. The factor loadings ranged from 0. 78 to 0. 81 Model df x2 RMSEA SRMR AGFI AIC NFI Ddf Dx 2 Null 1 factor factors 15 9 8 2,851. 07* 426. 74* 7. 11* 0. 49 0. 24 0. 10 0. 42 0. 09 0. 04 0. 23 0. 64 0. 92 2,863. 07 450. 74 96. 11 0. 00 0. 83 0. 97 – 6 1 – 2,424. 33* 356. 63* Notes: * p , 0. 05; – not applicable for mental distress, and from 0. 73 to 0. 76 for work-related health attributions, indicating good local ? t. The inter-factor correlation was 0. 68. 2. 2. 4. Work attitudes. All attitude variables were measured using a ? ve-point Likert scale (1 ? strongly disagree; 5 ? strongly agree). Job satisfaction was assessed with a three-item measure (e. g. â€Å"I am satis? ed with my job†) developed by Hellgren et al. (1999) on the basis of Bray? ld and Rothe (1951). Organiz ational commitment was measured using three items (e. g. â€Å"The company means a lot to me personally†) from Allen and Meyer’s (1990) scale re? ecting the affective dimension of commitment. Three items were used to assess turnover intention (e. g. â€Å"I am actively looking for other jobs†). The scale is based on items from Lyons (1971) and Cammann et al. (1979), and ? modi? ed and translated to Swedish by Sjoberg and Sverke (2000). 2. 3. Analyses Three moderated hierarchical regressions were conducted with job satisfaction, organizational commitment, and turnover intention, respectively, as dependent variables.The interaction term between mental distress and work-related health attributions was formed by centering the predictors and calculating the product term, following the procedure described by Cohen et al. (2003). The control variables were entered in the ? rst (demographics) and second (work climate) steps. The main effects of mental distress and work- related health attributions were entered in the third and fourth step, respectively, while the interaction term was entered in the ? fth and last step. 3. Results Table III contains the results of the hierarchical multiple regression analyses.The demographic variables (gender, age, and education) explained 3 percent of the variance in job satisfaction. When the work climate variables were added in the second step, the total amount of explained variance increased substantially to 58 percent, and an additional 2 percent was explained by mental distress in the third step. In the fourth step, when negative work-related health attributions were entered, a total of 62 percent of the variance was explained (DR 2 ? 0. 02). However, the interaction term between mental distress and work-related health attributions (Step 5) did not contribute signi? antly to the explained variance. The regression coef? cients from the last step showed that women and those without a university degree were more satis? ed with their job. Furthermore, autonomy, quantitative overload, work group cohesiveness, and job challenge were positively related to job satisfaction. Of most importance to the present study was the ? nding that both mental distress and negative work-related health attributions were negatively related to job satisfaction and that they remained signi? cant predictors of job satisfaction when demographics and work climate variables were controlled for.In the next analysis, where organizational commitment was the dependent variable, the demographics explained 7 percent, the climate variables another 36 percent, and mental distress added 1 percent to the explained variance. Negative work-related health attributions added another 1 percent to the explained variance, whereas, again, the interaction effect was found to be non-signi? cant. Taken together, the model variables explained 43 percent of the variation in organizational commitment. All three demographics evidenced signi? cant effects in the last step; being a woman, beingWork-related health attributions 13 IJWHM 2,1 14 Table III. Results of hierarchical multiple regression analyses (standardized regression coef? cients from the last step) Dependent variable Step 1 Gender (woman) Age Education (university) DR 2 Step 2 Job autonomy Quantitative role overload Qualitative role overload Work group cohesiveness Job challenge DR 2 Step 3 Mental distress DR 2 Step 4 Negative work-related health attributions DR 2 Step 5 Mental distress* Negative work-related health ttributions DR 2 Model R 2 adjusted Job satisfaction Organizational commitment Turnover intention 0. 2*** 0. 02 2 0. 06** 0. 03*** 0. 10** 0. 14*** 2 0. 08** 0. 07* 2 0. 13*** 2 0. 11*** 0. 07* 0. 05*** 0. 36*** 0. 05* 2 0. 01 0. 14*** 0. 31*** 0. 55*** 0. 24*** 0. 04 0. 06 0. 12*** 0. 34*** 0. 36*** 2 0. 13** 2 0. 08* 2 0. 04 2 0. 09** 2 0. 16*** 0. 24*** 2 0. 11*** 0. 02*** 2 0. 05 0. 01**** 0. 11** 0. 04*** 2 0. 20*** 0. 02*** 2 0. 12*** 0. 01* ** 0. 35*** 0. 07*** 2 0. 04 0. 00 0. 63*** 2 0. 00 0. 00 0. 43*** 0. 06* 0. 00* 0. 39*** Notes: * p , 0. 05; ** p , 0. 01; *** p , 0. 001 older, and having a lower education were all associated with higher levels of organizational commitment.All climate variables except quantitative overload were positively related to organizational commitment. Mental distress was unrelated to organizational commitment, whereas negative work-related health attributions were negatively related to the criterion. The demographic variables explained 5 percent of the variance in turnover intention in the ? rst step and the work climate variables in Step 2 added 24 percent to the variance explained. Mental distress in Step 3 added 4 percent and negative work-related health attributions another 7 percent in the next step.In the last step, when the interaction term was included, the variance explained increased by 6 units and the total model explained 39 percent. The interaction term was signi? cant; indiv iduals high in mental distress and reporting negative work-related health attributions exhibited the strongest intentions to quit whereas those with low levels of distress and less negative work-related health attributions expressed the strongest willingness to stay within the organization. This indicates that it makes a difference for the intentions to turnover if an individual with high mental distress attributes it to work or not.There were small effects of the demographic variables, indicating that men, younger persons, and individuals with a university degree had stronger intentions to leave the organization. In addition, most work climate variables, as well as mental distress (those high in mental distress reported higher levels of turnover intentions), predicted turnover intention. The strongest standardized regression coef? cient was found for the relation between negative work-related health attributions and turnover intention. 4. DiscussionDespite the increased research in terest in work-related stress and health problems, little attention has been paid to whether, or to what extent, individuals’ attributions of ill-health due to work (making negative work-related health attributions) affects attitudes towards work and the organization. Those few studies that have investigated this phenomenon (see for example Ettner and Grzywacz, 2001) have not studied how these work-related health attributions contribute to the understanding of job satisfaction, organizational commitment and turnover intention.The present study indicates not only that the construct of negative work-related health attributions is distinct from subjective mental distress, but also that it has important implications for employee attitudes and behavioral intentions. Our results show that employees who reported negative work-related health attributions also expressed less satisfaction with their job, less commitment to the organization, and stronger intentions to leave the organiza tion – even after demographics, mental distress, and factors related to the work climate were controlled for.We proposed that there would be an interaction effect between mental distress and work-related health attributions on attitudes and turnover intention, such that the relation between mental distress and the outcome variables would be stronger for those with more negative work-related health attributions than for those not attributing their mental distress to their job. This proposition received partial support. While work-related health attributions failed to moderate the effects of mental distress on job satisfaction and organizational commitment, the interaction term was signi? ant for turnover intentions. This indicates that individuals with negative work-related health attributions and mental distress were more prone to leave the organization compared to individuals with the same level of mental distress but positive work-related health attributions. Hence, whether the individual attributes ill-health to work or not appears to be important for the effect of health status on outcomes, at least in the case of the intention to leave the organization.This result indicates that the concept of negative work-related health attributions may play an important role in explaining employees’ work-related behavioral intentions, and should be taken into account when the relation between health and outcomes is investigated. Another notable ? nding is that work-related health attributions were more important than mental distress for the prediction of work attitudes and withdrawal cognitions. In contrast to mental distress, which only predicted job satisfaction and turnover intentions, work-related health attributions predicted all three outcomes.That mental distress only appears to contribute to a limited extent is contrary to previous arguments that health status is important for employee work motivation and behavioral orientations (e. g. Hom, 2002). One explanation for our ? ndings may have to do with the fact that work-related health attributions concern perceived health risks that are associated with the overall work situation, whereas measures of ill-health, such as mental distress, may concern any aspect of an individual’s life, not only factors related to the job. As long as the individual does not hold negative work-related healthWork-related health attributions 15 IJWHM 2,1 16 attributions, it is conceivable that subjective health and well-being are only marginally predictive of attitudes towards work. Negative work-related health attributions may therefore be a better predictor of work-related attitudes and behaviors, and should be included in future research on work-related outcomes. The construct of work-related health attributions can thus have great practical utility, since it in a rather simple and direct way captures how individuals perceive that their work affects their health.This is also in agreement wi th Harter et al. (2003), who suggest that researchers have conceived employee well-being too broadly and often in a way which is not intuitively actionable for managers and employees. It is also important to note that negative work-related health attributions predicted work attitudes and turnover intention after controlling not only for mental distress and demographics, but also for characteristics of the psychological work climate.Our data from Swedish white-collar workers showed that, consistent with meta-analysis ? ndings (Parker et al. , 2003), job autonomy, role overload, work group cohesiveness, and job challenge were associated with lower levels of job satisfaction and organizational commitment. The present study also goes beyond previous research by suggesting that the evident effects of work climate variables remained after subjective well-being and work-related health attributions had been taken into account.The fact that negative work-related health attributions emerged a s one of the strongest predictors of job satisfaction, organizational commitment, and turnover intention clearly indicates that models for the predicting of employee attitudes and behavior could be improved by the inclusion of work-related health attributions. The results from the present study and previous research (see for example ? European Foundation, 2001; Ettner and Grzywacz, 2001; Goransson et al. , 2002) indicate that attributions of work-related ill-health can be understood in terms of psychological contract theory (e. . Rousseau, 1989). Most employees expect that their employers will strive to provide a healthy workplace, and such an assumption is typically sustained by labor law regulations and collective bargaining agreements. If employees experience a work situation which they deem as likely to bring about ill-health, it will be in contrast to their expectations, and thus most likely be perceived as a breach of the psychological contract with the employer (see Robinson and Morrison, 2000).Perceptions of a breach of the psychological contract are likely to result in negative work-related attitudes and withdrawal behaviors (Davy et al. , 1997; Turnley and Feldman, 1999). The ? nding that negative work-related health attributions were associated with impaired job satisfaction, lower organizational commitment, and stronger turnover intention may be interpreted as the result of a perceived breach of the psychological contract. However, while psychological contract breach was not the main focus in the present study, future research may bene? from explicitly including this concept in the explanatory framework. 4. 1. Limitations and future research Although the results of the present study point to the relevance of the concept of work-related health attributions, our conclusions may be affected by a number of methodological issues. For instance, we utilized data from a single point in time, thus prohibiting the study of temporal order between negative wor k-related health attributions and its postulated consequences. However, cross-sectional research is a necessary ? rst step to empirically explore theoretical hypotheses before investments in ongitudinal data are fruitful (Spector, 1994), and two events must be found to co-vary with one another before the causal link can be explored. Furthermore, our results were based on Swedish white-collar employees in a particular service industry, and a replication of the present study using longitudinal data as well as data from different industrial sectors in different countries would be necessary before any ? rm conclusions concerning the generality of the ? ndings can be drawn. All variables were assessed using questionnaires, thus potentially making the results susceptible to mono-method bias (Campbell and Fiske, 1959).Even though meta-analytic research suggests that method variance represents less of a problem than has been assumed in the past (Crampton and Wagner, 1994), the use of other types of data, such as interviews or diaries, would contribute to a better understanding of the role played by work-related health attributions. In addition, while we controlled for a number of factors known to be related to job satisfaction (Loher et al. , 1985), organizational commitment (Mathieu and Zajac, 1990), and turnover intention (Griffeth et al. 2000), it would be relevant to take additional control variables, such as personality characteristics, into account in order not to over-estimate the effects of work-related health attributions. Finally, only three potential outcomes of negative work-related health attributions were considered. Despite the consistent associations found between negative work-related health attributions and employee attitudes and turnover intention, it might be that negative work-related health attributions are less predictive of other factors, such as job performance, safety behavior, and more importantly, occupational injuries and sick-leave. . 2. Concluding remarks Despite these potential limitations, the results of the present study clearly suggest that the investigation of employees’ negative work-related health attributions is an important avenue for future research on work and well-being. The results suggest not only that individuals evaluate how work may affect their health and well-being, but also that such perceptions may have important implications for their attitudinal and behavioral orientations toward work.The measure of negative work-related health attributions could be used as a diagnostic instrument for identifying those individuals who are more likely than others to perceive health-threatening conditions at work. The construct can be useful for employers in order to identify and help employees who experience threats to their health due to their jobs. Given that power imbalances may inhibit the direct communication of interests with an employer (Rousseau, 2001), such that more vulnerable individuals, for instance temporary employees, are less likely to share ? nformation on health risks at work (Aronsson and Goransson, 1998), the collection of information regarding work-related health attributions could be one means for employees to communicate this important information with their employer. Work-related health attributions may also constitute a more direct way of assessing the effect of work on health, and may therefore be more useful as a complement to research which gathers information on aspects of the work situation in order to predict health.An important area for future research is how individuals may take these work-related health attributions into account when making decisions regarding their Work-related health attributions 17 IJWHM 2,1 18 work situation, such as changing work tasks, work setting, or even occupation. Work-related health attributions are important for those with good health as well as for those with health complaints, as it concerns how individuals believe work will affect their health in the future as well as how health has already been affected.If such decisions are based on knowledge concerning not only employees’ health status but also their work-related health attributions, this will potentially result in healthier organizations, on which improved health and lower sick-leave rates will follow. References Ajzen, I. and Fishbein, M. (1980), Understanding Attitudes and Predicting Social Behavior, Prentice-Hall, Englewood Cliffs, NJ. Akaike, H. (1987), â€Å"Factor analysis and AIC†, Psychometrika, Vol. 52 No. 3, pp. 317-32. Allen, N. J. and Meyer, J. P. 1990), â€Å"The measurement and antecedents of affective, continuance and normative commitment to the organization†, Journal of Occupational Psychology, Vol. 63 No. 1, pp. 1-18. ? ? ? ? Aronsson, G. and Goransson, S. (1998), â€Å"Tillfalligt Anstallda och Arbetsmiljodialogen† ? (â€Å"Contingent workers and the work environment†), Arbete och Halsa , Vol. 3, pp. 1-17. Beehr, T. , Walsh, J. T. and Taber, T. D. (1976), â€Å"Relationships of stress to individually and organizationally valued states: higher order needs as a moderator†, Journal of Applied Psychology, Vol. 61 No. 1, pp. 41-7. Bentler, P. M. and Bonett, D.G. (1980), â€Å"Signi? cance tests and goodness-of-? t in the analysis of covariance structures†, Psychological Bulletin, Vol. 88 No. 3, pp. 588-606. Bradley, J. R. and Cartwright, S. (2002), â€Å"Social support, job stress, health, and job satisfaction among nurses in the United Kingdom†, International Journal of Stress Management, Vol. 9 No. 3, pp. 163-82. Bray? eld, A. H. and Rothe, H. F. (1951), â€Å"An index of job satisfaction†, Journal of Applied Psychology, Vol. 35 No. 5, pp. 307-11. Browne, M. W. and Cudeck, R. (1993), â€Å"Alternative ways of assessing model ? t†, in Bollen, K. A. and Long, J. S. Eds), Testing Structural Equation Models, Sage, Newbury Park, CA, pp. 1 36-62. Cammann, C. , Fichman, M. , Jenkins, D. and Klesh, J. (1979), The Michigan Organizational Assessment Questionnaire, University of Michigan, Ann Arbor, MI. Campbell, D. T. and Fiske, D. W. (1959), â€Å"Convergent and discriminant validation by the multitrait-multimethod matrix†, Psychological Bulletin, Vol. 56 No. 2, pp. 81-105. ? Catalan Matamoros, D. J. , Axelsson, R. and Strid, J. (2007), â€Å"How do newspapers deal with health in Sweden? A descriptive study†, Patient Education and Counseling, Vol. 67 Nos 1-2, pp. 78-83. Cohen, J. Cohen, P. , West, S. G. and Aiken, L. S. (2003), Applied Multiple Regression/Correlation Analysis for the Behavioral Sciences, 3rd ed. , Erlbaum, Hillsdale, NJ. Crampton, S. M. and Wagner, J. A. (1994), â€Å"Percept-percept in? ation in micro-organizational research: an investigation of prevalence and effect†, Journal of Applied Psychology, Vol. 79 No. 1, pp. 67-76. Davy, J. A. , Kinicki, A. J. and Scheck, C. L. (1997), â⠂¬Å"A test of job security’s direct and mediated effects on withdrawal cognitions†, Journal of Organizational Behavior, Vol. 18 No. 4, pp. 323-49. Ettner, S. L. and Grzywacz, J. G. 2001), â€Å"Workers’ perception of how jobs affect health: a social ecological perspective†, Journal of Occupational Health Psychology, Vol. 6 No. 2, pp. 101-13. European Foundation (2001), Third European Survey on Working Conditions 2000, European Foundation for the Improvement of Living and Working Conditions, Dublin. Fairbrother, K. and Warn, J. (2001), â€Å"Workplace dimensions, stress and job satisfaction†, Journal of Managerial Psychology, Vol. 18 No. 1, pp. 8-21. Goldberg, D. and Williams, P. (1988), A User’s Guide to the General Health Questionnaire, NFER-Nelson, London. ? Goransson, S. , Aronsson, G. and Melin, B. 2002), â€Å"Vilja och Villkor – En Studie om ? Langtidssjukskrivnas Situation† (â€Å"Volition and conditions – a study about long-term ? sickness absence†), Statens Offentliga Utredningar: Handlingsplan for Okad Halsa i Arbetslivet [The Swedish Government’s Public Investigations: A Plan for Increased Health ? ? in Working Life; SOU 2002:5), Allmanna Forlaget, Stockholm. Griffeth, R. W. , Hom, P. W. and Gaertner, S. (2000), â€Å"A meta-analysis of antecedents and correlates of employee turnover: update, moderator tests, and research implications for the next millennium†, Journal of Management, Vol. 26 No. 3, pp. 463-88. Hackman, J. R. nd Oldham, G. R. (1975), â€Å"Development of the job diagnostic survey†, Journal of Applied Psychology, Vol. 60 No. 2, pp. 159-70. Harding, S. H. and Hikspoors, F. J. (1995), â€Å"New work values: in theory and in practice†, International Social Science Journal, Vol. 47, pp. 441-55. Harpaz, I. (2002), â€Å"Expressing a wish to continue or stop working as related to the meaning of work†, European Journal of Work and Organiza tional Psychology, Vol. 11 No. 2, pp. 177-98. Harter, J. K. , Schmidt, F. L. and Keyes, C. L. M. (2003), â€Å"Well-being in the workplace and its relationship to business outcomes: a review of the gallup studies†, in Keyes, L.M. and Haidt, J. (Eds), Flourishing: Positive Psychology and the Life Well-lived, American Psychological Asociation, Washington, DC, pp. 205-24. ? Hellgren, J. , Sjoberg, A. and Sverke, M. (1997), â€Å"Intention to quit: effects of job satisfaction and job perceptions†, in Avallone, F. , Arnold, J. and De Witte, K. (Eds), Feelings Work in Europe, Guerini, Milan, pp. 415-23. Hellgren, J. , Sverke, M. and Isaksson, K. (1999), â€Å"A two-dimensional approach to job insecurity: consequences for employee attitudes and well-being†, European Journal of Work and Organizational Psychology, Vol. 8 No. 2, pp. 79-195.Hirschman, A. O. (1970), Exit, Voice, and Loyalty: Responses to Decline in Firms, Organizations, and States, Harvard University Press, Cambridge, MA. Holland, R. W. , Verplanken, B. and van Knippenberg, A. (2002), â€Å"On the nature of attitude-behavior relations: the strong guide, the weak follow†, European Journal of Social Psychology, Vol. 32 No. 6, pp. 869-76. Hom, P. (2002), â€Å"The legacy of Charles Hulin’s work on turnover thinking and research†, in Brett, J. M. and Drasgow, F. (Eds), The Psychology of Work: Theoretically Based Empirical Research, Erlbaum, Mahwah, NJ, pp. 169-87. Houkes, I. Janssen, P. P. M. , de Jonge, J. and Nijhuis, F. J. N. (2001), â€Å"Speci? c relationships between work characteristics and intrinsic work motivation, burnout and turnover intention: a multi-sample analysis†, European Journal of Work and Organizational Psychology, Vol. 10 No. 4, pp. 1-23. ? ? Joreskog, K. and Sorbom, D. (1996), LISREL 8: User’s Reference Guide, Scienti? c Software International, Chicago, IL. Work-related health attributions 19 IJWHM 2,1 20 Judge, T. and Watanabe, S. (1993), â€Å"Another look at the job satisfaction – life satisfaction relationship†, Journal of Applied Psychology, Vol. 8 No. 6, pp. 938-49. ? ? Kallenberg, K. and Larsson, G. (2000), Manniskans Halsa: Livsaskadning och Personlighet (Human Health: Outlook on Life and Personality), Natur och Kultur, Stockholm. Kjellberg, A. (2001), Fackliga Organisationer och Medlemmar i Dagens Sverige (Union Organizations and Union Members in Sweden Today), Arkiv, Lund. Lambert, E. G. , Hogan, N. L. and Barton, S. M. (2001), â€Å"The impact of job satisfaction on turnover intent: a test of a structural measurement model using a national sample of workers†, The Social Science Journal, Vol. 38 No. , pp. 233-50. Lee, T. W. (1988), â€Å"How job dissatisfaction leads to employee turnover†, Journal of Business & Psychology, Vol. 2 No. 3, pp. 263-71. Leong, C. S. , Furnham, A. and Cooper, C. L. (1996), â€Å"The moderating effect of organizational commitment on the occupa tional stress-outcome relationship†, Human Relations, Vol. 49 No. 10, pp. 1345-63. Loher, B. T. , Noe, R. A. , Moeller, N. L. and Fitzgerald, M. P. (1985), â€Å"A meta-analysis of the relation of job characteristics to job satisfaction†, Journal of Applied Psychology, Vol. 70 No. 2, pp. 280-9. Lyons, T. F. 1971), â€Å"Role clarity, need for clarity, satisfaction, tension, and withdrawal†, Organizational Behavior and Human Performance, Vol. 6 No. 1, pp. 99-110. Mathieu, J. E. and Farr, J. L. (1991), â€Å"Further evidence for the discriminant validity of measures of organizational commitment, job involvement, and job satisfaction†, Journal of Applied Psychology, Vol. 79 No. 1, pp. 127-33. Mathieu, J. E. and Zajac, D. M. (1990), â€Å"A review and meta-analysis of the antecedents, correlates, and consequences of organizational commitment†, Psychological Bulletin, Vol. 108 No. 2, pp. 171-94. Michotte, A.E. (1963), The Perception of Causality, Russel and Russel, New York, NY. Moscovici, S. (2000), Social Representations: Explorations in Social Psychology, Polity Press, Cambridge. ? ? ? ? Nystedt, L. (1992), Yrkesof? cerares arbetsmiljo i Armen: resultat fran Delstudie 1 om arbetsmiljo, ? ? ? fortidsavgangar och halsa (Work Environment of Army Of? cers: Results from Study 1 on Work Climate, Turnover, and Health), FOA Report C50093-5. 3, Swedish Defence Research Agency (FOA). Parker, C. P. , Baltes, B. B. , Young, S. A. , Huff, J. W. , Altmann, R. A. , Lacost, H. A. and Roberts, J. E. 2003), â€Å"Relationships between psychological climate perceptions and work outcomes: a meta-analytic review†, Journal of Organizational Behavior, Vol. 24 No. 4, pp. 389-416. Pfeffer, J. (1998), The Human Equation, Harvard Business School Press, Boston, MA. Robinson, S. L. and Morrison, W. E. (2000), â€Å"The development of psychological contract breach and violation: a longitudinal study†, Journal of Organizational Behavior, Vol. 21 No. 5, pp. 525-46. Robinson, S. L. , Kraatz, M. S. and Rousseau, D. M. (1994), â€Å"Changing obligations and the psychological contract: a longitudinal study†, Academy of Management Journal, Vol. 37 No. , pp. 137-52. Rousseau, D. M. (1989), â€Å"Psychological and implied contracts in organizations†, Employee Responsibilities and Rights Journal, Vol. 2 No. 2, pp. 121-39. Rousseau, D. M. (2001), â€Å"Schema, promise and mutuality: the building blocks of the psychological contract†, Journal of Occupational and Organizational Psychology, Vol. 74 No. 4, pp. 511-41. Shore, L. M. and Tetrick, L. E. (1994), â€Å"The psychological contract as an explanatory framework in the employment relationship†, in Cooper, C. L. and Rousseau, D. M. (Eds), Trends in Organizational Behavior, Vol. 1, Wiley, Oxford, pp. 91-109. Shore, L. M. and Wayne, S.J. (1993), â€Å"Commitment and employee behavior: comparison of affective commitment and continuance commitment with perce ived organizational support†, Journal of Applied Psychology, Vol. 78 No. 5, pp. 774-80. ? Sjoberg, A. and Sverke, M. (2000), â€Å"The interactive effect of job involvement and organizational commitment on job turnover revisited: a note on the mediating role of turnover intention†, Scandinavian Journal of Psychology, Vol. 41 No. 3, pp. 247-52. Spector, P. E. (1994), â€Å"Using self-report questionnaires in OB research: a comment on the use of a controversial method†, Journal of Organizational Behavior, Vol. 5 No. 5, pp. 385-92. Spector, P. E. (1997), Job Satisfaction: Application, Assessment, Causes and Consequences, Sage, Thousand Oaks, CA. Stratton, P. (1997), â€Å"Attributional coding of interview data: meeting the needs of long-haul passengers†, in Hayes, N. (Ed. ), Doing Qualitative Analysis in Psychology, Psychology Press, Hove, pp. 115-42. ? Sverke, M. and Sjoberg, A. (1994), â€Å"Dual commitment to company and union in Sweden: an examination of predictors and taxonomic split methods†, Economic and Industrial Democracy, Vol. 15 No. 4, pp. 531-64. ? Sverke, M. , Hellgren, J. nd Ohrming, J. (1999), â€Å"Organizational restructuring and health care ? work: a quasi-experimental study†, in Le Blanc, P. M. , Peeters, M. C. W. , Bussing, A. and Schaufeli, W. B. (Eds), Organizational Psychology and Health Care: European Contributions, Rainer Hampp Verlag, Munich, pp. 15-32. Turnley, W. H. and Feldman, D. C. (1999), â€Å"The impact of psychological contract violations on exit, voice, loyalty, and neglect†, Human Relations, Vol. 52 No. 7, pp. 895-922. Van Vuuren, T. , Smulders, P. and Smit, A. (2003), â€Å"Gender differences in absence culture: ? ? n exploratory study†, in Hellgren, J. , Naswall, K. , Sverke, M. and Soderfeldt, M. (Eds), New Organizational Challenges for Human Service Work, Rainer Hampp Verlag, Munich, pp. 175-190). Walsh, J. T. , Taber, T. D. and Beehr, T. A. (1980), â€Å"An integrat ed model of perceived job characteristics†, Organizational Behavior and Human Performance, Vol. 25 No. 2, pp. 252-67. Walters, D. (2002), â€Å"The framework directive†, in Walters, D. (Ed. ), Regulating Health and Safety Management in the European Union. A Study of the Dynamics of Change, P. I. E. – Peter Lang, Brussels.Weiner, B. (1985), â€Å"‘Spontaneous’ causal thinking†, Psychological Bulletin, Vol. 97 No. 1, pp. 74-84. Wright, T. A. and Bonett, D. G. (2007), â€Å"Job satisfaction and psychological well-being as non-additive predictors of workplace turnover†, Journal of Management, Vol. 33 No. 2, pp. 141-60. Corresponding author ? Sara Goransson can be contacted at: [email  protected] su. se To purchase reprints of this article please e-mail: reprint[email  protected] com Or visit our web site for further details: www. emeraldinsight. com/reprints Work-related health attributions 21