THE MEDIATING ROLE OF PSYCHOLOGICAL AND SOCIAL RESOURCES
Master thesis, MSc, specialization Human Resource Management University of Groningen, Facility of Economics and Business
July 1, 2018 LISET MOL Student number: 3266826 Hendrikstraat 5 9724 NA Groningen Tel: +31653621277 e-mail: e.mol.2@student.rug.nl Supervisor/ university H. J. van de Brake.
SHIFT WORK AND WORK-RELATED STRESS:
THE MEDIATING ROLE OF PSYCHOLOGICAL AND SOCIAL RESOURCES
ABSTRACT
Shift work has become a common work practice for organizations that operate 24 hours a day. Previous research has linked shift work with negative psychological well-being and disturbance of the social lives of shift workers. This study examines the consequences of shift work on work stress from a conservation of resources (COR) perspective. COR theory proposes that people experience stress due to the threat of losing or the actual loss of resources. To identify those resources, I will draw on self-efficacy theory and social capital theory. The present research proposes that the positive indirect relationship between shift work and work stress is mediated by decreases in an individual employee’s self-efficacy (i.e., a key psychological resource) and lower friendship network centrality (i.e. as an important social resource). I tested my conceptual model using archival and survey data from a Dutch hospital (N = 287). The results demonstrate that shift work was positively associated with friendship network centrality and self-efficacy was negatively associated with work stress. I did not find support for the positive indirect relationship between shift work and work stress, mediated by self-efficacy and friendship network centrality. These findings offer new
theoretical insights into the consequences of shift work and identify a mechanism for work stress.
SHIFT WORK AND WORK RELATED STRESS:
THE MEDIATING ROLE OF PSYCHOLOGICAL AND SOCIAL RESOURCES Over 38% of the employees in EU member states work in shifts to cover industries that are operating 24 hours a day (McMenamin, 2007). Generally, shift work can be defined as work that employees perform in morning, evening, and night shifts (McMenamin, 2007). Shift workers do not work according to the standard working schedule that consists of day-shifts (8 hours a day) from Monday to Friday, but instead rotate in day-shifts (Wallace, 2000; Presser, 1995). Shift work has become increasingly common in, for example, the military, hospitality, retail, manufacturing, transportation, police, and healthcare industries
(Blachowics & Letizia, 2006; Lee, Mc Cann & Messenger, 2007).
It is well established that shift work has detrimental consequences for the well-being of employees. Shift work has been linked to fatigue, digestive disturbances, and obesity (Åkerstedt, 2003; Dunham, 1976; Gomez-Parra et al., 2016). The reason behind these problems among shift workers is the conflict between irregular working hours and the
individual’s biological clock, which disturbs the sleeping rhythm of shift workers (Gamble et al., 2011). In addition, research suggests that shift work is related to work stress. Due to their nonstandard working schedule, shift workers miss important parts of their family and social lives (e.g. anniversaries or graduations; Cordova, Brandford & Stone, 2004; Shen & Dicker 2008; Winwood, Winefield & Lushington, 2006). Shift workers cannot fulfil the needs of their relations, because they have to work or recover from working in shifts. This causes work related stress among shift workers (Baker et al, 2000; Shen & Dicker, 2008). From an
organizational perspective, work stress among shift workers is linked with absenteeism, turnover, and low job satisfaction (Cordova, Brandford & Stone, 2004; EU, 2003; Smith and Wederburn, 1998; Tepas, Paley and Popkin, 1997; Wedderburn, 1996).
often investigated consequence of shift work (Cordova, Brandford & Stone, 2004; Shen & Dicker 2008; Winwood, Winefield & Lushington, 2006). However, there is limited research which focuses on explaining the underlying mechanisms of this relationship. I believe a better understanding of those mechanisms will clarify why shift workers suffer from work stress. This study examines the relationship between shift work and work stress from a COR theory perspective, a commonly used perspective for predicting work-related stress. (Alarcon,
Edwards & Menke, 2011; Wright & Hobfoll, 2004). The basic principle of COR theory is that people have access to certain resources and when they lose or when there is a chance of losing those resources, it will cause work stress (Hobfoll, 1989). Resources are objects, personal characteristics, conditions, and energies that an individual values and they serve as means for the achievement of those objects, personal characteristics, conditions, and energies (Hobfoll, 1989). I suggest that working in shifts can contribute to the threat of losing and the actual loss of psychological and social resources.
Previous research examined the impact of shift work on psychological and social resources. For example, shift work is related to poor psychological well being (Dean & Lin, 1977; Pisariski et al., 2009) and disturbance in a shift workers’ social relations (Baker et al., 2000; Shen & Dicker, 2008). COR theory identifies those resources as important sources whose loss results in work related stress (Hobfoll, 1989). In this study, I will draw on self-efficacy theory (Bandura, 1977) and social capital theory (Klein, Lin, Saltz & Mayer, 2004; Lin, 1999) to identify self-efficacy and friendship network centrality as key psychological and social resources. Hence, I suggest that shift work is positively and indirectly related to work stress, through the threat of losing and the actual loss of self-efficacy and friendship network centrality.
interventions to solve those issues successfully (Bandura, 1977; Snyder & Lopez, 2009). I suggest that shift workers have less self-efficacy, because the feelings of fatigue influence their perceptions about their performance successes (Bandura, 1977; 1982). The threat of losing or the actual loss of self-efficacy influence a shift workers’ level of work stress (Hobfoll, 1989). Therefore, I propose that self-efficacy mediates the relationship between shift work and work stress.
Moreover, several studies have examined the consequences for shift workers’ informal relationships outside of organizations (Baker et al., 2000; Shen & Dicker, 2008), but little research has been done on the informal relations of shift workers within organizations. Social capital theory focuses on social relations within organizations (e.g. friendship network
centrality; Klein et al., 2004; Lin, 1999). It has been recognized that friendship network centrality has a prominent role in perspectives about interpersonal relationships (Lin, 1999, Klein et al., 2004). I suggest that working in shifts influences the amount of friendly relations among shift workers, because a shift workers’ working schedule affects the amount of social contact between shift workers and their colleagues, and therefore the quality of their
relationships. The threat of loosing or the actual loss of friendly relationships causes work stress among shift workers (Hobfoll, 1989). Therefore, I propose that friendship network centrality mediates the relationship between shift work and work stress.
organizations is a better understanding of the occurrence of work stress among shift workers. Organizations can apply this research to reduce the amount of work stress among their shift workers, and therefore reduce the amount of absenteeism and turnover rates and increase the job satisfaction of their employees (Baker, Fletcher & Dawson, 2000; Caporale, 2005; Shen & Dicker, 2008; Wedderburn, 1996). --- Figure 1 here --- THEORETICAL BACKGROUND Shift work
Current literatures define shift work as work that is performed beyond conventional working hours (i.e. 8 hours a day, from Monday to Friday; Wallace, 2000; Presser, 1995). Organizations develop a shift work system to cover the hours of their operations (Harrington, 2001). This shift work system consists of successive shifts whereby employees (individually or in teams) take over and continue each other’s work (Saksvik-Lehouillier et al., 2012; Costa, 2003).
organization occurs in various forms.
The shift work system of an organisation demands a degree of flexibility of their shift workers. Shift workers who are rotating between shifts do not have a routine in their working schedule (Jamal, 1981). There is no border between their work and leisure time and they constantly have to adapt to their changing working schedule (Costa, 2003). In contrast, shift workers who are working according a fixed working schedule (i.e. standard one kind of shift) have more routine in their working schedule. Those shift workers do not have to adapt
because their working schedule barely changed. Because of those differences, this study refers to shift work as shift work variety. Shift workers who rotate in different kind of shifts
(including night shifts), and therefore do not have a fixed working schedule, have a high level of shift work variety. Shift workers who are working one kind of shift or only working during conventional working hours, have a low level of shift work variety.
Current literature refers to shift work as a dichotomous variable which separates shift workers from regular day workers (i.e. non shift workers). For example, Drake, Roehrs, Richardson, Walsh and Roth (2004), asked their participants to select the category that best describes their working schedule of the last two weeks. The categories were: ‘regular day shifts’, ‘regular night shifts’, ‘regular evening shifts’, ‘rotating shifts’ or ‘not working/retired’. They considered those who are working in ‘regular night shifts’ of ‘rotation shift’ as shift workers. Those participants were compared with regular day workers. The participants who are working in ‘regular evening shifts’ and those who are ‘not working/retired’ were excluded from the research (Drake et al., 2004). This example shows that current research measures shift work as a dichotomous variable which distinguishes shift workers from non-shift workers.
then becomes possible to include all kinds of variations of shift work, and to make
distinctions between shift workers who have a low or high variety of shifts. I suggest that shift workers who have a low variety of shift work (e.g., when they solely working night shifts), and therefore do not have to adapt to their changing working schedule, suffer less from the adverse consequences of shift work. In contrast, shift workers with a high variety of shifts, constantly have to adapt to their changing schedule and therefore suffer more from the adverse consequences. By measuring shift work as a dimension, it is possible to make a distinction within those employees who work in shifts.
To conclude, this study refers to shift work as shift work variety. Shift work variety depends on the arrangement of working hours and the rotation in shifts. If shift workers rotate in different kind of shifts (including night shift), they experience a high level of shift work variety. In contrast, shift workers experience a low level of shift work variety if they standard work a morning, evening or night shift.
Conservation of Resources Theory
This study focuses on the predictors of work stress among shift workers from a COR theory perspective. The basic principle is that “people strive to retain, protect, and build resources and that what is threatening to them is the potential or actual loss of these valued resources” (Hobfoll, 1989, p. 516). I define work stress as a reaction to the environment when there is a perception of losing, actual loss of the lack of gain of resources that are valued by an individual (Hobfoll, 1989).
The loss of resources has always been a dominant concept in research about stress (Dohrenwend et al., 1978; Holmes & Rahe, 1967; Sarason, Johnson & Siegel, 1978). The threat of losing or the actual loss of resources is related to work stress due to the dimension of ‘undesirability’. Losing a resource that an individual values causes stress because it is
Individuals instinctually strive to protect and enhance themselves (Pearlin et al., 1981). When individuals lose their resources, they cannot protect or enhance themselves which causes stress (Hobfoll, 1989). I believe that shift work plays a critical role in this regard. Due to their unconventional working schedule, shift workers are threatening to lose their resources. COR theory recognizes psychological and social resources as important resources whose loss results in work related stress (Hobfoll, 1989). I suggest that shift workers with a high variety of shifts suffer more from the threat and actual loss of psychological and social resources, due to their constantly changing working schedule. In contrast, I suggest that shift workers with a low variety of shift work suffer less from the threat and actual loss of
psychological and social resources, due to the fact that their working schedule is more fixed. Shift work and psychological resource
Bandura’s (1977) self-efficacy theory suggests that individuals have strong cognitive capabilities, which makes individuals capable of self-observation and self-regulation
(Bandura, 1986, 1997; Barone, Maddux, & Snyder, 1997). This means that individuals are able to analyse and evaluate their own behaviour and are able to set goals and regulate their behaviour in obtaining those goals. Hence, those cognitive abilities influence an individuals’ level of self-efficacy (Snyder & Lopez, 2009). Self-efficacy can be defined as having
a mean for the achievement of positive cognitions, motivations and behaviours (Hobfoll, 1989).
Shift workers have to work when individuals with a conventional working schedule do not have to work. If a shift workers’ working schedule consists of a high variety of shifts, it is in conflict with their sleep rhythm (Åkerstedt, 2003; Jamal, 1981;). This means that their working schedule is not consistent with their biological clock. For example, if a shift worker has to work a morning shift, they have to get up earlier than his or her biological clock tells them to wake up. Besides the disturbance of their sleep rhythm, shift workers have a poorer quality of sleep (Åkerstedt & Torsvall, 1981; Kripke, 2004; Suzaki et al., 2004). Several studies have examined this conflict and found evidence that it causes a feeling of fatigue (Åkerstedt, 2003; Comperatore & Krueger, 1990; Saksvik-Lehouillier et al., 2012;Wilson, 2002). The feeling of fatigue is linked with several consequences such as decreased alertness during work hours (Tucker, Smith, Macdonald & Folkard, 1988), reduced productivity, and fatigue-related accidents at work (Åkerstedt, 2003; Cordova, Brandford, Stone, 2016; Suzaki et al., 2004).
Gold and colleagues (1992) also found evidence for the relationship between shift work and increasing number of mistakes committed at work. They conducted a study among nurses working in shifts at a hospital. They found evidence that nurses who work in rotating shifts (i.e. high variety of shift work) were getting less sleep and were more likely to fall asleep during their work. Those nurses suffered from lapses of attention and increased reaction time during their work and reported that they more likely commit medical errors while working (Gold et al., 1992). Thus, nurses with a high variety of shift work committed more accidents when they perform their tasks at work.
Bandura’s (1977) self-efficacy theory suggests that performance accomplishments are the most powerful sources of an individuals’ self-efficacy. If individuals perceive their performance is a success, their confidence in their ability to succeed at a task increases. If they perceive their performance is a failure, their expectations about their abilities, and therefore their confidence in their abilities decreases (Bandura, 1977). Therefore, I suggest that shift workers with a high variety of shifts have less self-efficacy compared to shift workers with a low variety of shifts.
Following this theory, I suggest that shift workers with a high shift work variety are getting less sleep and have a lower quality of sleep. They experience a feeling of fatigue while the performance of their duties at work, which influences their alertness, productivity and chances of accidents at work. Due to the self-observation and self-regulation abilities, individuals are able to evaluate their own behaviour and to set goals and regulate their behaviour in obtaining those goals (Snyder & Lopez, 2009). Therefore, shift workers should evaluate their own behaviour and notice that they are less alert while working. They perceive that their productivity is decreased and they easily could make mistakes during their work. They likely doubt their abilities to perform their tasks properly and therefore their confidence in their abilities will decrease. In contrast, shift workers with a low shift work variety are getting more sleep and suffer less from fatigue. They are more likely to perceive they are performing well and therefore they do not doubt their abilities. They experience a higher level of self-efficacy.
Therefore, shift workers with a high variety of shifts are threatening to lose or actual loose the psychological resource self-efficacy. I develop the following hypothesis:
Hypothesis 1. Shift work is negatively related to self-efficacy. Psychological resource and work stress
Lazarus and Folkmans’ (1984) ‘stress, appraisal and coping’ theory argues that individuals suffer from stress when they believe they do not have the resources to handle difficult situations. The demands those situations require are evaluated as a ‘threat’ or a ‘challenge’. Personal beliefs (e.g. self-efficacy) are a crucial resource when individuals evaluate difficult situations (Lazarus and Folkman, 1984). Individuals with a high level of self-efficacy are more likely to evaluate a difficult situation as a challenge, because they are confident about their abilities to handle the situation (Bandura, 2007; Chemers, Hu, and Garcia, 2001; Lazarus and Folkman, 1984; Pintrich and De Groot, 1990). Those individuals do not perceive the situation as a ‘threat’, and it is therefore not stressful (Lazarus and Folkman, 1984). According to this theory, individuals who evaluate a difficult situation as a challenge are able to use an effective coping strategy. Thus, they are able to manage their tasks during those situations. Individuals with a low level of self-efficacy are more likely to evaluate difficult situations as threatening and therefore stressful. They are less confident about their abilities to handle difficult situations. Hence, they are more likely to experience stress (Bandura, 2007).
Therefore, I suggest that the threat of and actual loss of self-efficacy is related to work stress. I hypothesize:
Hypothesis 2. Self-efficacy is negatively related to work stress. Shift work and social resources
Social capital theory focuses on the informal relationships between individuals in an organization, also called a social network (Lin, 1999; Klein et al., 2004). Individuals invest in their social network (i.e. contact with their colleagues) and in return they expect to receive certain resources, such as information, social support, social credentials or reinforcements (Lin, 2017). One of the most important parts of a social network, and thus a social resource, is friendship network centrality (Klein et al., 2004). Benefits of having friendly relationships with colleagues are having access to information, influence and more positive rating of a colleagues’ performance (Balwin, Bedell & Johnson, 1997; Brass, 1984; Ibarra & Andrews, 1993). Hence, friendship network centrality is a social recourse because it serves as a mean for the achievement of, for example, social support and access to information (Hobfoll, 1989; Ferh, 2004; Klein et al., 2004).
Friendships are defined as relationships between individuals that are based on affection and comradery (Baldwin et al., 1997). Individuals invest in their friendly relationships with their colleagues and in return they expect to receive resources such as social support, comfort, companionship and fun (Fehr, 2004). Individuals perceive someone as a friend if this person has similar interests, values and attitudes (Sprecher & Regan, 2002). Being friends with colleagues also means socialising with each other in leisure time (Klein et al., 2004). Friendship network centrality is measured through indegree centrality, an essential tool to analyse social networks (Wey, Blumstein, Shen & Jordan, 2008). An individual’s social network captures all the connections of an individual and the position of those
network, based on their position in a social network (Freeman, 1978; Wey, Blumstein, Shen & Jordan, 2008). Indegree centrality refers to the number of social connection an individual receives (Wey, Blumstein, Shen & Jordan, 2008). Thus, how many individuals nominate a certain individual in their social network. Centrality in friendship networks refers to what extend individuals in a social network identify a certain individual as their friend (Kilduff & Krackhardt, 1994). An individual has a high level of friendship network centrality if most of his or her colleagues identify this certain individual as their friend. In contrast, a low level of friendship network centrally means that none or a few individuals in a network identify a certain individual as their friend.
To what extend a certain individual is identified as a friend in their own social network depends on the strength of their relationships (i.e. centrality; Lin, 1999). Granovetter’s (1973) identifies friendly relationships in an individual’s social network as valuable relationships. Friendly relationships offer social support and help individuals cope with stressful situations (Van der Doef & Meas, 1999; Morrison, 2002). Friendly relationships arise when individuals have connections with individuals who are similar to them in terms of experience and
perspectives. Their connection becomes stronger when similar individuals frequently interact with each other, because they become even more similar (Marsden & Friedkin, 1993). If individuals have less frequent and intense contact with each other, then they see each other as distant colleagues (Krackhardt, 1992). Distant colleagues are less likely to give each other social support (Van der Doef & Meas, 1999).
schedule is constantly changing and every individual has a different schedule, there is no guarantee that shift workers frequently have to work with each other. Hence, there is less frequent and intense contact among the shift workers and their colleagues. Also, it is less likely that they discover similarities among each other and therefore they do not easily develop friendly relationships. Even when they discover similarities among each other it is harder for them to have frequently contact because of a varied work schedule. Therefore, it is harder for them to develop a friendly relationship. To conclude, shift workers with a high variety of shifts may occupy less central positions within their friendship network. In contrast to the above, I suggest that shift workers with a more fixed working schedule (i.e. low variety of shift work) have a higher level of friendship network centrality. Their working schedule does not change frequently, so they often work with the same colleagues and therefore have more opportunities for intense contact. Hence, they likely discover similarities and therefore more easily develop a friendship. Therefore, shift workers with a low level of shift work variety may occupy more central positions within their
friendship networks.
support (Granovetter, 1973). Thus, shift workers with a high variety of shift work actual loose social resource friendship network centrality. I hypothesize:
Hypothesis 3. Shift work is negatively related to friendship network centrality. Social resources and work stress
Johson and Ellen’s’ (1988) theory of ‘demand- control- support model’ argues that social support, obtained through friendly relationships, plays an important role in the occur of work stress among individuals. The lack of social support from co-workers and supervisors causes stress, because individuals have no relationships where they can rely on during stressful situations (Lin, 1999). It is hard for them to deal with those stressful situations because they cannot count on support from their social network. As mentioned before, friendship offer social support and help individuals to cope with stressful situations. On the other hand, distant relationships are less useful during stressful situations (Van der Doef & Meas, 1999). When individuals possess friendly relationships in their social network, they are more likely to achieve social support during stressful situations and therefore experience less stress.
Following Lazarus and Folkmans’ (1984) theory about ‘stress, appraisal and coping’, I suggest that individuals who have friendly relationships with colleagues, believe that they have the resources to handle a difficult situation. They would evaluate the situation as a challenge because social support from their friends may protect them. Therefore, they
experience less stress. If an individual’s social network does not include friendships, he or she experience more stress because he or she will evaluate the difficult situation as a threat. Those individuals do not have someone to rely on and therefore, a difficult situation can cause more damage (Lazarus & DeLongs, 1983; Lazarus & Folkman, 1984). They experience more stress (Lazarus & Folkman, 1984).
are threating to lose or actual loose friendship network centrality (Hobfoll, 1989). Those individuals who perceive that they have friends are able to protect and enhance themselves during difficult situations (Perlean et al., 1981). They feel less threatening because they have relationships where they can rely on and support them. But, when there is a threat of losing or the actual loss of social resources, individuals can no longer protect and enhance themselves in stressful situations. This situation is undesirable and therefore it causes stress (Mueller, Edwards & Yarvis, 1977; Thoits, 1983; Vinokur & Selzer, 1975). To conclude, I suggest that the loss of social resource is related with work stress. I hypothesize:
Hypothesis 4. Friendship network centrality is negatively related to work stress. The Mediating Role of Psychological and Social Resources
Taken together the arguments of hypothesis 1 and 2, I cast self-efficacy as mediating mechanism for work stress among shift workers. Increasing the variation in shift work may lower shift workers’ level of self-efficacy, thus, may indirectly associate with higher level of work stress. I therefore hypothesize an indirect relationship between shift work and work stress, with the loss of the psychological resource self-efficacy as one of the key mediating mechanisms.
Hypothesis 5a. There is a positive indirect relationship between shift work and work stress, mediated by self-efficacy.
Hypothesis 5b. There is a positive indirect relationship between shift work and work stress, mediated by friendship network centrality.
METHOD
Data and procedure
I tested my hypotheses in a sample of employees of a hospital located in the South of the Netherlands. I collected data from two independent sources. First, I used a survey. The survey was translated to Dutch by using a double-blind back translation procedure. The participants were contacted via their email and were asked to fill out the online survey. Their participation was voluntary. Before the participants started with the survey, they had to select if they wanted to participate. The participants had three weeks to fill in the survey. After two weeks that the survey had been opened, the participants received a reminder email. Second, I used archival data from the three months’ prior the start of the survey. I received the archival data from the hospital’s HR department to determine the variables gender, age, tenure, workload (FTE) and variety of shift work. By hand of the working schedules of the
participants, I calculated the variety of shift work of each participant. After matching the two datasets, their personal information was anonymized.
Sample
I obtained archival data for 429 employees of the hospital. Those employees worked at three different departments and all those departments had different duties and responsibilities. Employees who did not respond to the survey where omitted from the sample. I retained a usable sample of 287 employees (response rate = 67%).
anonymity. Measures
Shift work. Shift work was measured through archival data of employees’ working schedules, using data from the three months before the start of the survey. The working schedules of the participants distinguished between night and day shifts. Besides, the hospital made use of stand-by shift where the employees have to wait at home until they are called to come to work. Those stand-by duties were included as relevant shifts because the employees were not able to spend their time on leisure. I calculated the variety of shift work of the participants by using the Blau’s (1977) index of heterogeneity. The formula for this index is 1 - ∑ Pi² (e.g., Richard, 2000), where P is the proportion of the total amount of hours that the participants had worked, and i is the number of different shifts the participants had worked in. For example, when a participant has an index of 0, the participant only worked day or night shift. When the index is .50, the participant had worked, for example, one half of their working hours during the day and one half of their working hours in a nightshift.
Psychological resources. I measured the psychological resource self-efficacy (Parker, 1998) with 6 items on a 7-Likert scale (1 = strongly disagree; 7 = strongly agree or 1 = strongly agree; 7 = strongly disagree). Example items were: “I feel confident analysing a long-term problem to find a solution” and “I feel confident helping to set targets/goals in my work area” (α > .80). A higher score indicates a high level of self-efficacy.
were the names from colleagues who were working at the same department as the participant. I asked them “Which colleagues do you see as a good friend?” Hence, the participants had to mark those who they perceived as a good friend. The variable was compute in proportions where 1 = all colleagues of a certain department selected an individual as their friend; 0 = none colleague selected an individual as his or her friend. If a participant scored, for example, a proportion of 0.8 then 80% of the colleagues of a certain individuals selected that individual as their friend.
Work stress. I measured the amount of work stress among shift workers with a burnout scale (Koppes, Vroome, Bossche, 2011) with 5-items on a 7-Likert scale (1 = strongly disagree; 7 = strongly agree or 1 = strongly agree; 7 = strongly disagree). Example items were: “I feel emotionally exhausted by work” and “I feel empty at the end of my workday” (α > .80).
Control variables. This study included several control variables. First control variable was gender. According to Moulton (1980), women are more likely to experience stress than men because when women gain new resources they often have to deal with more new
(Glaser, Tatum, Nebeker and Sorenson, 1999). I tested all hypothesis with and without the control variables. The results and conclusion remain the same after I conducted both analyses. Data analysis
I conducted a regression analysis using model 4 of Hayes (2009) to examine the hypothesized relationships as displayed in Figure 1. Prior to the analyses, I standardized the predictors gender, age and tenure. Besides, I standardized the shift work measure.
RESULTS
Descriptive Statistics
Table 1 presents means, standard deviations, and bivariate correlations for all variables of this study. Contrary to my expectations, I did not found a negative correlation between shift work and self-efficacy (r = .01, p >.05). I did found preliminary support for a positive correlation between shift work and friendship network centrality (r = .13, p <.05). In addition, I found a correlation between self-efficacy and work stress (r = -.33, p <.01). There was no correlation between friendship network centrality and work stress (r = -.03, p >.05).
Considering the control variables, gender was positively related with self-efficacy (r = .12, p <.05), but not with friendship network centrality and work stress. Autonomy was significantly related with selfefficacy (r = .35, p <.01), friendship network centrality (r = -.13, p <.05) and work stress (r = -.24, p <.01). As expected, age and tenure are strongly correlated (r = .72, p < .01), although age and tenure are not correlated with self-efficacy, friendship network centrality and work stress. FTE was not related with self-efficacy, friendship network centrality and work stress.
Hypotheses Testing
As shown in Table 2, I found no support for the predicted negative association
between an employee’s variation in shift work and self-efficacy suggested in Hypothesis 1 (B = .02, p > .05). Therefore, Hypothesis 1 was not supported. In addition, I found support for a negative relationship between self-efficacy and work stress (B = -.46, p < .01), even after considering the control variables. Hence, Hypothesis 2 was supported.
Hypothesis 3 predicted the negative association between shift work and friendship network centrality. As illustrated in Table 3, I found evidence for a positive relationship between shift work and friendship network centrality (B = .01, p < .05). This finding is in contradiction with my theoretical prediction and therefore I do not support Hypothesis 3. I will return to these unexpected finding in the Discussion section.
Hypothesis 4 predicted the negative association between friendship network centrality and work stress. I found no support that there is a relationship (B = -2.18, p > .05). In
addition, Hypothesis 4 is not supported.
Hypothesis 5a predicted the indirect relationship between shift work and work stress through self-efficacy. As shown in Table 2, I found no support for this indirect relationship (95% CI = -.07 to .04). Hence, hypothesis 5a is not supported. Finally, Hypothesis 5b predicted the indirect relationship between shift work and work stress through friendship network centrality. As illustrated in Table 3, I did not found support for the indirect relationship (95% CI = -.05 to -.01). Thus, Hypothesis 5b is not supported.
--- Insert Table 2, Table 3 here
Summary
Scholars has found that over 38% of the employees in EU member states work in shifts (McMenamin, 2007). Although previous research linked shift work to work stress (Cordova, Brandford & Stone, 2004; Shen & Dicker 2008; Winwood, Winefield & Lushington, 2006), most research did not focus on the underlying mechanism of this relationship. Therefore, it is unclear why shift workers suffer from work stress. The present research examines this indirect relationship between shift work and work stress from a COR perspective. Hence, I aimed to provide new insights why shift workers suffer from work related stress.
I developed and tested a conceptual model that draws on the self-efficacy theory and the social capital theory to specify shift workers psychological and social resources (i.e. self-efficacy and network friendship centrality). Unexpectedly, I found no evidence for the positive indirect relationship between shift work and work stress mediated by decreases in an individual’s level of self-efficacy and lower friendship network centrality. In contrary with my expectations, the results show a positive association between shift work and friendship network centrality. If employees work in a high variety of shifts, they occupy more central positions within their friendship networks. In addition, the results show evidence for a negative relationship between self-efficacy and work stress. If an individual has a high level of self-efficacy, he or she suffers less from work stress. In contrast, individuals with a low level of self-efficacy suffer more from work stress.
Theoretical Implications
has shown that shift work could be measured as a continuous dimension on which employees score a lower or higher score based on their variety in shifts. This way of measuring shift work deserves further consideration in occupational shift work research, because it makes it possible to include all kinds of variations of shift work. Besides, future research could make a distinction within those employees who are working with a high or low level of variation of shifts. Present study shows that a continuous dimension of shift work is significantly related with an important social consequence of shift work namely, friendship network centrality. Future research could use this way of measuring shift work to examine possible new consequences or reconsider current consequences of shift work.
The present research also contributes to the work stress literature. It has been documented that an increased number of employees suffer from strain during their work (Steiber & Pichler, 2015; Siegrist & Li, 2016). My findings show a negative relationship between self-efficacy and work stress. This result confirms previous research that stated that a low level of self-efficacy is linked with work related stress (Bandura, 2007; Lazarus and Folkman, 1984).
Furthermore, the present research found no evidence between friendship network centrality and work stress. This is a contradiction with previous research which found
evidence for a negative relationship between the occupation of central positions in friendship networks and the level of work stress (Lazarus & DeLongis, 1983; Lazarus & Folkman, 1984). So, some research supports my findings and others do not. I suggest that the relationship between friendship network centrality and work stress is depending of the context. For example, Dahlem, Zimet and Farley (1988) found evidence that demographics influence the way individuals cope with social support, obtained through friendship network centrality, which influence the level of work related stress. Therefore, I suggest that this relationship is depending on contingency factors such as demographics.
Limitations and Future Research
response rate to examine the relationship between shift work and work stress and the mediating role of friendship network centrality.
Furthermore, I examined the hypotheses in a sample of employees that all worked at the same organization in one country, the Netherlands. This limits the generalizability of my results because it is possible that other organizational settings determine the findings of the present research. If my conceptual model was tested in a different organizational setting my findings could have shown different results. For example, the relationship between shift work and friendship network centrality could be different when it was examined within a transport company with truck drivers. Because truck drivers often work individually, it could be harder for them to develop friendly relationship with their colleagues. Therefore, the generalizability of my results to other organizational setting remains open. Future research should replicate the conceptual model in other organization settings before the generalizability of the results of the present research can be justified.
Future research may benefit from examining other potential resources that could mediate the relationship between shift work and work stress (i.e. from a COR-perspective). The present research did not find an indirect relationship between shift work and work stress mediated by self-efficacy and friendship network centrality. Several studies found evidence for a relationship between the threat of losing and the actual loss of resources with work stress during different organizational setting and working conditions (Grandey & Cropanzano, 1999; Hobfoll & Shirom, 2001). Therefore, I suggest that future research should examine other potential resources that are influenced by shift work, for example the amount of
perceived workload (Spector & Jex, 1998) or the amount of autonomy to determine your own schedule (Bakker & Demerouti, 2007; Karasek, 1979) and the level of work-life balance (Grandey & Cropanzano, 1999).
Even though previous research has suggested that working in shifts might influence the psychological well-being of shift workers (Dean & Lin, 1977; Gordon, Cleary, Parker,
Czeisler, 1986; Pisariski et al., 2009), the present study did not discover this relationship. The reason I did not find a relationship between those variables could be the way I measured self-efficacy. According to Bandura (2006), self-efficacy is not a universal trait but a differential set of self-belief that distinguish different realms of functioning. Thus, future research could benefit from using a measuring instrument of self-efficacy that is adjusted to the functioning and task demands of shift work to identify the perceived capabilities of shift workers. Practical Implications
It is well established that work stress among employees is negatively related with employees’ performance and health (Kahn & Byosiere, 1992) and it influences their absenteeism and turnover intentions (Gangster & Schaubroeck, 1991). Therefore, reducing the amount of work stress among employees has significant financial benefits for
organizations (Ivancevich & Matteson, 1980). The findings show that employees with a high level of self-efficacy experience less work stress. Practitioners aiming to reduce work related stress among employees could benefit from these findings by increasing the level of self-efficacy of their employees. Practitioners can do this by using professional development plans that fits the career stages of the employee, so that they can enhance their skills and
competences. In this way they can increase the confidence of the employees (Klassen & Chiu, 2010). Hereby, organizations can reduce their absenteeism and turnover rates.
Friendly relationships provide employees with social support, comfort, companionship and fun (Fehr, 2004). Research has shown that social support at work is related with a lower level of absenteeism and burnout (Russell, Altmaier, Van Velzen, 1987; Undén, 1996). Shift
relationships (i.e. social support) compared to shift workers with a low variety of shifts. This could be an important implication for practitioners managing organizations where their employees are working in shifts. Practitioners could understand and explain a part of the absenteeism and burnout rates of their employees. Based on those understandings they can come up with interventions to reduce the level of absenteeism and burnout among their employees.
CONCLUSION
The aim of this study was to identify why shift workers suffer from work stress. I found evidence that a lower level of s employee’s self-efficacy is related with increased work stress. Furthermore, I found that a high variety in shift work is related with an increase of the occupations of central positions in an individual’s friendship network. No support was found for the positive indirect relationship between shift work and work stress mediated by self-efficacy and friendship network centrality. Altogether, this research contributes to the shift work, work stress and network centrality literatures. I hope my findings will purchase future research on the detrimental consequences of shift work.
REFERENCES
Åkerstedt,T., Torsvall, L. (1981). Shift work. Shift- dependent well-being and individual differences. Ergonomics, 24, 265–273.
Åkerstedt, T. (2003). Shift work and distrubed sleep/wakefulness. Occupational Medicine, 53, 89-94.
Alarcon, G. M., Edwards, J. M., Menke, L. E. (2011). Student burnout and engagement: a test of the conservation of resources theory. The Journal of Psychology, 145(3), 211-227. Arimaru, M. Imai, M., Okawa, M., Fuijmura, T., Yamada, N. (2010). Sleep, mental health
status, and medical errors among hospital nurses in Japan. Industrial Health, 48, 811- 817.
Avey, J. B., Luthans, F., Jensen, S., M. (2009). Psychological capital: a positive resource for combating employee stress and turnover. Human Resource Management, 48(5), 677-
693.
Avey, J. B., Reichard, R. J., Luthans, F., Mhatre, K. H. (2011). Meta-analysis of the impact of positive psychological capital on employee attitudes, behaviors and performance. Human Resource Development Quaterly, 22(2), 127-152. Bakker, A. B., Demerouti, E. (2007). The job demands-resources model: State of
the art. Journal of managerial psychology, 22(3), 309-328.
Baldwin, T. T., Bedell, M. D., Johnson, J. L. (1997). The social fabric of a team-based M.B.A. program: network effects on student satisfaction and performance. Academy of Management Journal, 40, 1369-1397.
Bandura, A. (1977). Self-efficacy: toward a unifying theory of behavioral change. Psychological review, 84(2), 191-215.
psychologist, 37(2), 122.
Bandura, A. (1993). Perceived self-efficacy in cognitive development and
functioning. Educational psychologist, 28(2), 117-148.
Barton, J. (1994). Choosing to work at night: a moderating influence on individual tolerance to shift work. Journal of Applied Psychology, 79(3), 449-454.
Blachowicz, E., Letizia, M. (2006). The challenges of shift work. Medsurg Nursing, 15(5), 274-280.
Blau, P. M. (1977). Inequality and heterogeneity: A primitive theory of social structure (Vol.
7). New York: Free Press.
Brandes, U. (2001). A faster algorithm for betweenness centrality. Journal of mathematical
sociology, 25(2), 163-177.
Cordova, de, P. B., Brandford, M. A., Stone, P. W. (2016). Increased errors and decreased performance at night: a systematic review of the evidence concerning shift work and quality. Work, 53, 825-834.
Costa, G. (1997). The problem: shiftwork. Chronobiology International, 14(2), 89-98. Costa, G. (2003). Shift work and occupational medicine: an overview. Occupational
medicine, 53(2), 83-88.
DeLongis, A., Folkman, S., Lazarus, R. S. (1988). The impact of daily stress on health and mood: psychological and social resources as mediators. Journal of personality and
social psychology, 54(3), 486.
Drake, C. L., Roehrs, T., Richardson, G., Walsh, J. K., & Roth, T. (2004). Shift work sleep disorder: prevalence and consequences beyond that of symptomatic day
workers. Sleep, 27(8), 1453-1462.
Freeman, L. C. (1978). Centrality in social networks conceptual clarificaton. Social
networks, 1(3), 215-239.
Glaser, D. N., Tatum, B. C., Nebeker, D. M., Sorenson, R. C., Aiello, J. R. (1999). Workload and social support: Effects on performance and stress. Human
Performance, 12(2), 155-176.
Gold, D. R., Rogacz, S., Bock, N., Tosteson, T. D., Baum, T. M., Speizer, F. E., Czeisler, C. A. (1992). Rotating shift work, sleep, and accidents related to sleepiness in hospital nurses. American journal of public health, 82(7), 1011-1014.
Gomez-Parra, M., Romero-Arrieta, L., Vasquez-Trespalacios, E. M., Palacio-Jaramillo, V., Valencia-Martinez, A. (2016). Association between shift work and being overweight or obese among health care workers in a clinical setting in Medellin, Colombia. Work, 55(3), 635-642.
Gordon, N. P., Cleary, P. D., Parker, C. E., Czeisler, C. A. (1986). The prevalence and health impact of shiftwork. American journal of public health, 76(10), 1225-1228. Grandey, A. A., Cropanzano, R. (1999). The conservation of resources model applied to work-family conflict and strain. Journal of Vocational Behavior, 54, 350-370. Granovetter, M. S. (1977). The strength of weak ties. Social networks, 78(6), 347-367. Harrington, J. M. (2001). Health effects of shift work and extended hours of work. Occupational Environmental Medicine, 58, 68-72.
Hayes, A. F. (2009). Beyond Baron and Kenny: Statistical mediation analysis in the new millennium. Communication Monographs, 76(4), 408-420.
Hobfoll, S. E. (1989). Conservation of resources: a new attempt at conceptualizing stress. American Psychologist, 44(3), 513-524.
Ito, J. K., Brotheridge, C. M. (2002). Resources, coping strategies, and emotional exhaustion: a conservation of resources perspective. Journal of Vocational Behavior, 63, 490-509. Ivancevich, J. M., Matteson, M. (1980). Stress and work: A managerial perspective.
Glenview, IL: Scott, Foresman.
Jamal, M. (1981). Shift work related to job attitudes social participation and withdrawal behavior: a study of nurses and industrial workers. Personnel Psychology, 34, 535-
547.
Johnson, S., Cooper, C., Cartwright, S., Donald, I., Taylor, P., Millet, C. (2005). The experience of work-related stress across occupations. Journal of managerial
psychology, 20(2), 178-187.
Johnson, J. V., Hall, E. M. (1988). Job strain, work place social support, and cardiovascular disease: a cross-sectional study of a random sample of the Swedish working
population. American journal of public health, 78(10), 1336-1342.
Kahn, R. L., Byosiere, P. (1992). Stress in organizations. In M. D. Dunnette & L. M.
Hough (Eds.), Handbook of industrial and organizational psychology (pp. 571-650).
Palo Alto, CA, US: Consulting Psychologists Press.
Kilduff, M., Krackhardt, D. (1994). Bringing the individual back in: A structural analysis of the internal market for reputation in organizations. Academy of management
journal, 37(1), 87-108.
Klassen, R. M., Chiu, M. M. (2010). Effects on teachers' self-efficacy and job satisfaction: Teacher gender, years of experience, and job stress. Journal of
educational Psychology, 102(3), 741.
Koppes, L. L. J., Vroome, E. M. M. De, Bossche, S. N. J. van den (2011). NEA
Cohortonderzoek 2007-2008; methoden en eerste resultaten. Hoofddorp: TNO.
Krackhardt, D. (1992). The strength of strong ties: The importance of philes in organizations. Networks and organizations: Structure, from, and action, 216, 239.
Lin, N. (1999). Social networks and status attainment. Annual Review of Sociology, 25, 46-55. Lin, N. (2017). Building a network theory of social capital. In Social capital (pp. 3-28).
Routledge.
Luthans, F., Avolio, B. J., Avey, J. B., Norman, S. M. (2007). Positive psychological capital: measurement and relationship with performance and satisfaction. Personnel
Psychology, 60, 541-572.
Luthans, F., Luthans, K. W., Luthans, B. C. (2004). Positive psychological capital: beyond human and social capital. Business Horizons, 47(1), 45-50.
Motowidlo, S. J., Packard, J. S., Manning, M. R. (1986). Occupational stress: its causes and consequences for job performance. Journal of Applied Psychology, 71(4), 618-629. Parasuraman, S., Greenhaus, J. H., Granrose, C. S. (1992). Social support, and well-being among two-career couples. Journal of Organizational Behavior, 13(4), 339-356.
Pearson, L. C., Moomaw, W. (2005). The relationship between teacher autonomy and stress, work satisfaction, empowerment, and professionalism. Educational research
quarterly, 29(1), 37.
Pisarski, A., Brook, C., Bohle, P., Gallois, C., Watson, B., Winch, S. (2009). Extending a model of shift-work tolerance. Chronobiology International, 23(6), 1363-1377. Russell, D. W., Altmaier, E., Van Velzen, D. (1987). Job-related stress, social support, and
burnout among classroom teachers. Journal of applied psychology, 72(2), 269.
Harvey, A., Costa, G., Pallesen, S. (2012). Personality factors predicting changes in shift work tolerance: a longitudinal study among working rotating shift. Work & Stress, 26(2), 142-160.
Shen, J., Dicker, B. (2008). The impacts of shiftwork on employees. The International Journal of Human Resource Management, 19(2), 392-405.
Siegrist, J., Li, J. (2016). Associations of extrinsic and intrinsic components of work stress with health: a systematic review of evidence on the effort-reward imbalance
model. International journal of environmental research and public health, 13(4), 432. Snyder, C. R., Lopez, S. J. (Eds.). (2009). Oxford handbook of positive psychology. Oxford
University Press, USA.
Spector, P. E., Jex, S. M. (1998). Development of four self-report measures of job stressors and strain: interpersonal conflict at work scale, organizational constraints scale,
quantitative workload inventory, and physical symptoms inventory. Journal of
occupational health psychology, 3(4), 356.
Steiber, N., Pichler, F. (2015). Trends in work stress and exhaustion in advanced
economies. Social Indicators Research, 121(1), 215-239.
Tausing, M., Fenwick, R. (2001). Unbinding time: alternate work schedules and work-life balance. Journal of Family and Economic Issues, 22(2), 101-119
Undén, A. L. (1996). Social support at work and its relationship to absenteeism. Work &
Stress, 10(1), 46-61.
Van der Doef, M., Maes, S. (1999). The job demand-control (-support) model and psychological well-being: a review of 20 years of empirical research. Work &
stress, 13(2), 87-114.
Issues, 22(2), 35-48.
Wey, T., Blumstein, D. T., Shen, W., Jordan, F. (2008). Social network analysis of animal behaviour: a promising tool for the study of sociality. Animal behaviour, 75(2), 333
-344.
Wright, T. A., Hobfoll, S. E. (2004). Commitment, psychological well-being and job performance: an examination of conservation of resources (COR) theory and job burnout. Journal of Business and Management, 9(4), 389-406.
Zajacova, A., Lynch, S. M., Espenshade, T. J. (2005). Self-efficacy, stress, and academic success in college. Research in higher education, 46(6), 677-706.
Zedeck, S., Jackson, S. E., Summers, E. (1983). Shift work schedules and their relationship to health, adaption, satisfaction and turnover intention. Academy of Management
Journal, 26(2), 297-310.
TABLE 1
Means, Standard Deviations, and Pearson Correlation Coefficients
Variable Mean SD 1 2 3 4 5 6 7 8 1. Workload (FTE) 1.04 3.91 2. Gender .16 .36 -.01 3. Age .06 .95 -.14* .02 4. Tenure .05 1.00 -.09 .04 .72** 5. Autonomy 3.76 1.33 -.01 .15* -.04 -.03 6. Shift work .16 .20 -.04 -.07 -.05 .09 -.06 7. Self-efficacy 4.94 .94 -.07 .12* .01 .02 .35** .01
8. Friendship network centrality .03 .04 .00 .01 .02 .20 -.13* .13* -.08
Assessing mediation of Self-efficacy in the relationship between Shift work and Work stress, controlled for Workload, Gender, Age, Tenure and Autonomy.
Predictors Dependent variables
Self-efficacy Work stress
B SE P B SE P Workload (FTE) -.02 .01 .22 -.03 .02 .08 Gender .19 .15 .19 .40 .20 .04* Age -.02 .08 .81 -.01 .11 .94 Tenure .04 .08 .65 .03 .11 .79 Autonomy .24 .04 .00** -.13 .06 .02* Shift work .02 .06 .67 .02 .08 .77 Self-efficacy -.46 .08 .00**
Indirect relationship between Shift work and Absenteeism, via Health issues
Effect 95% CI:
Lower bound
95% CI: Upper Bound
Self-efficacy -.01 -.07 .04
Assessing mediation of Friendship network centrality in the relationship between Shift work and Work stress, controlled for Workload, Gender, Age, Tenure and Autonomy.
Predictors Dependent variables
Friendship network centrality Work stress
B SE P B SE P Workload (FTE) .00 .00 .92 -.03 .02 .20 Gender .00 .00 .63 .32 .21 .13 Age .00 .00 .42 .01 .12 .96 Tenure .00 .00 .65 -.25 .06 .00** Autonomy .01 .00 .03* .02 .08 .79 Shift work .01 .00 .03* -.46 .08 .00**
Friendship network centrality -2.18 2.14 .30
Indirect relationship between Shift work and Absenteeism, via Health issues
Effect 95% CI:
Lower bound
95% CI: Upper Bound
Friendship network centrality -.01 -.05 .01
