HOW WORKING IRREGULAR HOURS AFFECTS JOB SATISFACTION: A MODERATED MEDIATION STUDY
by
Willem Bruining
University of Groningen Faculty of Economics and Business MSc Human Resource Management
August, 2012
Aquamarijnstraat 497 9743 PM Groningen
+31 62410819 S1530399@student.rug.nl
Supervisor:
dr. D.B. (Dennis) Veltrop
2 ABSTRACT
Drawing on the data from the publicly available European Social Survey Round 5 – 2010, the present study explored the nature of the association between irregular working hours and job satisfaction. That is, this study posits that the indirect effect of irregular working hours on job satisfaction may be either positive or negative. Specifically, task variety was hypothesized to positively mediate the link between irregular working hours and job satisfaction, whereas work-life imbalance was hypothesized to negatively mediate the link. Furthermore, autonomy was hypothesized to moderate the indirect effects. Using a sample of 1,829 respondents, this study confirmed the mediation hypotheses. Autonomy, however, solely moderated the indirect effect of irregular working hours on job satisfaction through work-life imbalance. That is, the indirect effect of work-life imbalance was weaker for low levels of autonomy.
Keywords: irregular working hours, job satisfaction, task variety, work-life (im)balance, autonomy.
3 INTRODUCTION
Since the 1930’s, the construct of job satisfaction has attracted considerable attention from both scholars and practitioners for a variety of reasons. To begin with, job satisfaction is relevant for academics interested in identifying the job dimensions that impact on a worker’s utility (Hackman & Lawler, 1971; Hackman & Oldham, 1976). Secondly, job satisfaction is relevant for scholars and practitioners interested in organizational outcomes, such as performance (Lawler & Porter, 1967; Locke, 1970), absenteeism (Hackman & Lawler, 1971), and turnover (Hackman & Oldham, 1976; Martin, Sinclair, Lelchook, Wittmer, & Charles, 2012). Thirdly, the antecedents and consequences of the job satisfaction construct do not merely pertain to contexts of work. Spillover effects of job satisfaction on family and other contexts of life – and vice versa – are assumed to exist (Ilies, Wilson, & Wagner, 2009; Qu &
Zhao, 2012). Thus, both work and other contexts of life can potentially influence reports of worker satisfaction.
Research on the linkage between flexible work schedules and work-related criteria reported that flexible work schedules were associated with increased levels of job satisfaction (Baltes, Briggs, Huff, Wright, & Neuman, 1999). That is, increased flexibility offers the worker a possibility to incorporate employment into his/her personal life, which in turn results in to a favorable attitude toward the work (McNall, Masuda, & Nicklin, 2010). Hence, workers have increasingly sought greater flexibility in scheduling. Further, economic difficulties experienced by advanced economies as well as technological change have also driven the need for flexible work schedules (Brewster, Mayne, & Tregaskis, 1997). In accordance, a growth in the usage of flexible work schedules has been observed across Europe (Tregaskis, Brewster, Mayne, & Hegewisch, 1998). As aforementioned, research has demonstrated that a link between flexible work schedules and job satisfaction exists.
Nonetheless, due to ambiguous findings the nature of the linkage remains unclear.
With this assertion in mind, the present study anticipates that intervening mechanisms influence the aforementioned linkage, either positively or negatively. In accordance with work adjustment theory, the present study posits that irregular working hours will increase what Hackman and Lawler (1971) termed variety, which refers to the variety of activities and skills required in carrying out the work. That is, flexible work schedules tend to govern more complex jobs (Sims & Szilagyi, 1976). In turn, variety is expected to increase job satisfaction
4 (Hackman & Lawler, 1971; Brief & Aldag, 1975; Hackman & Oldham, 1976). In addition, the present study also anticipates that irregular working hours will increase the inability to balance work and family, thereby reducing job satisfaction (Maxwell & McDougall, 2004).
Considering the aforementioned, the indirect effect of working irregular hours on job satisfaction may be either positive or negative. Furthermore, the present study proposes that
‘autonomy’ may act as an important moderator in this respect (Hackman & Lawler, 1971), by enhancing the linkage between irregular working hours and job satisfaction through variety (Dodd & Ganster, 1996). In a similar vein, autonomy is expected to mitigate the negative effects of the linkage between irregular working hours and job satisfaction through work-life imbalance.
The purpose of this thesis is threefold. First, to explore the effect of irregular working hours on job satisfaction. Second, to examine the mediating roles of task variety and work-life imbalance for the relationship between irregular working hours and job satisfaction. Third and final, to investigate the moderating role of autonomy for the relationship between irregular working hours and job satisfaction through task variety and work-life imbalance. Thus, the present study extends the literature on job satisfaction by incorporating underlying mechanisms from multiple contexts to clarify the relationship between irregular working hours and job satisfaction.
THEORY AND HYPOTHESES DEVELOPMENT
Job Satisfaction
Scholars have viewed job satisfaction as “the pleasurable emotional state resulting from the appraisal of one's job as achieving or facilitating the achievement of one's job values” (Locke, 1969: p. 316). Invariably then, job dissatisfaction is “the unpleasurable emotional state resulting from the appraisal of one's job as frustrating or blocking the attainment of one's job values or as entailing disvalues” (Locke, 1969: p. 316). Past research has found a significant relationship between job satisfaction and performance (Lawler &
Porter, 1967; Locke, 1970; Judge, Thoresen, Bono, & Patton, 2001). Early empirical research conducted by Lawler and Porter (1967) and Locke (1970) identified job satisfaction as a product of performance. Locke (1969), however, noted that performance might also be a
5 product of job satisfaction, albeit indirectly. In a similar vein, Lawler and Porter (1967) indicated that the relationship between the two constructs might be reciprocal. Further, research showed a negative relationship between job satisfaction and turnover intentions (Mobley, 1977; Delfgaauw, 2006; Yang, 2010). Mobley (1977) demonstrated that job dissatisfaction led to withdrawal behavior, such as passive job behavior, absenteeism, and turnover. Furthermore, Delfgaauw (2006) found that satisfaction with job domains not only determined whether a worker searched for another job, but also where. Thus, job satisfaction has been associated with several essential organizational outcomes. In particular, high levels of job satisfaction have been associated with positive organizational outcomes, whereas low levels of job satisfaction have been associated with negative organizational outcomes.
A review of the literature suggests that scholars have devoted much attention to the antecedents of job satisfaction (e.g., Locke, 1969; Hackman & Lawler, 1971). In particular, scholars have placed emphasis on job characteristics as the determinants of job satisfaction.
(Lawler & Hall, 1970; Hackman & Lawler, 1971; Hackman & Oldham, 1976; Brief & Aldag, 1975; Loher, Noe, Moeller, & Fitzgerald, 1985). Hackman and Lawler (1971) advance a set of four core dimensions, i.e. variety, autonomy, task identity, and feedback. Research has shown that the core job dimensions are positively related to personal and work outcomes (Lawler & Hall, 1970; Hackman & Lawler, 1971; Hackman & Oldham, 1976). That is, jobs high on variety, autonomy, task identity, and feedback, lead to higher levels of job satisfaction. Recent research conducted by Skalli, Theodossiou, and Vasileiou (2008), however, incorporated a broader set of job characteristics, such as pay, the work itself, and working hours. They proposed that job satisfaction reflects the combination of partial satisfactions related to various features (i.e. characteristics) of a job (Skalli et al., 2008). Thus, the job values set forth by Locke (1969) are analogous to the partial satisfactions suggested by Skalli et al. (2008). Moreover, Skalli et al. (2008) found that across Europe, despite the importance of other facets (e.g., type of work, working conditions), working times significantly affected measures of overall job satisfaction.
Research on job (re)design has predominantly focused on the benefits associated with the enhancement of the aforementioned core job dimensions. Literature related to the field of psychology, however, has considered some of the potential effects of the core job dimensions on mental health. Nonetheless, the overall verdict remains the same. For instance, Kelloway and Barling (1991) reported that mental health issues mainly resulted from role stress (i.e., role ambiguity, role conflict). Characteristics of the job (i.e., autonomy, task variety,
6 feedback), on the other hand, were negatively associated with mental health issues – through personal accomplishments at work and work satisfaction (Kelloway & Barling, 1991). Thus, it seems that the absence, rather than the presence, of the job characteristics is related to ill health (Pousette & Hanse, 2002).
As aforementioned, scholars have indicated that working times significantly affect worker satisfaction. Nonetheless, research into the relationship between irregular working hours and job satisfaction has yielded ambiguous results. Some scholars have argued that the introduction of flexible working practices offer workers opportunities to fully utilize their skills and talents (e.g., Hackman & Lawler, 1971; Pierce & Newstrom, 1980). Others, however, have suggested that workers face the difficult task of balancing work and family life, following the introduction of irregular working hours (e.g., Tausig & Fenwick, 2001;
Adkins & Premeaux, 2012). Therefore, the present study posits that both positive and negative indirect effects may underlie the linkage between irregular working hours and job satisfaction.
Irregular Working Hours and Job Satisfaction
Over the past few decades flexible working hours have gained in prominence, with over 25 percent of the United States’ working population now being able to vary their schedules (Beers, 2000). Further, research indicates that similar growth can be observed in Europe (Brewster, Mayne, & Tregaskis, 1997). As a case in point, a 69 percent increase in the organizational use of part-time work in the Netherlands, during the 1993-96 period, has been reported. Additionally, the data reported a mere 1 percent decrease in the organizational use of part-time work during the same period (Tregaskis, Brewster, Mayne, & Hegewisch, 1998).
Scholars suggest that alternative work schedules are “schedules that do not fit the fixed 8 hour day, 40 hour week definition” (Baltes, Briggs, Huff, Wright, & Neuman, 1999: p. 497).
Examples of frequently cited alternative work schedules include the compressed workweek, flexible working hours (e.g., flextime, gliding schedule), part-time employment (Baltes et al., 1999; Beers, 2000), non-permanent employment (i.e., temporary and/or fixed term), and subcontracting (Brewster et al., 1997; Tregaskis et al., 1998). Baltes et al. (1999) proposed that shiftwork does not always adhere to the aforementioned definition, and is therefore omitted. Others, however, have stressed the relevance of shiftwork as an alternative work schedule, arguing that the practice has spread to non-traditional industries (e.g., telephone sales, banking) (Brewster et al., 1997; Tregaskis et al., 1998).
7 Irregular working patterns such as the compressed workweek and flextime, have been positively associated with several work-related criteria (Baltes, Briggs, Huff, Wright, &
Neuman, 1999). In particular, Orpen (1981) reported higher levels of satisfaction among workers following the introduction of flextime. Although Orpen (1981) did not uncover the exact reasons for the observed increases in satisfaction, McNall, Masuda, & Nicklin (2010) suggested that control over the scheduling of the work may be of significant importance.
Further, research conducted by Wotruba (1990) showed that part-time salespeople had higher levels of job satisfaction than full-time salespeople. A finding that Wotruba (1990) contributed to the fact that part-timers were less involved in organizational functioning, and thus had less opportunity to accumulate dissatisfying experiences. Accordingly, the following hypothesis is proposed:
Hypothesis 1. Irregular working hours positively affects job satisfaction.
The Mediating Role of Task Variety
Beers (2000) established “that flexible working hours are most common among workers in executive, administrative, and managerial occupations” (p. 34). The logic underlying this finding holds that the nature of the work dictates the schedule an occupation adheres to. Further, research found that the higher the level of the job, the higher the score on the core job dimensions set forth earlier (Sims & Szilagyi, 1976). Variety, in particular, was found to be significantly higher at higher occupational levels (Sims & Szilagyi, 1976).
Hackman and Oldham (1976) defined task variety as “the degree to which a job requires a variety of different activities in carrying out the work, which involve the use of a number of different skills and talents of the person” (p. 257). Hence, flexible work schedules are expected to govern high variety jobs, which in turn lead to higher levels of job satisfaction.
Hackman and Oldham (1976) suggested that various forms of variety would provide the worker with a perceived sense of meaningfulness, with regard to the job (s)he is performing. Consequently, the worker is challenged to use the skills and abilities (s)he values, to complete the job (Hackman & Lawler, 1971). Hence, if a worker succeeds at a job (s)he perceives as meaningful (i.e., challenging), (s)he will experience a greater sense of satisfaction.
Considering the aforementioned, irregular working hours is expected to indirectly enhance job satisfaction by providing the worker with variety. This notion is in line with the
8 theory of work adjustment (Pierce & Newstrom, 1980); that is, (1) the more closely a worker's abilities correspond with the requirements of the job, the more likely it is that the worker will perform the job well, and (2) the more closely the rewards of the job correspond to the values (i.e., needs) a workers seeks to satisfy, the more likely it is that the worker will perceive the job as satisfying. Hence, the degree of flexibility of a schedule affects the degree of variety associated with a job, which in turn affects measures of satisfaction. Consistent with the aforementioned findings, the following hypotheses are proposed:
Hypothesis 2. Irregular working hours positively affects task variety.
Hypothesis 3. Task variety positively affects job satisfaction.
Hypothesis 4. Task variety mediates the relationship between irregular working hours and job satisfaction.
The Mediating Role of Work-Life Imbalance
Darcy, McCarthy, Hill, and Grady (2012) define work-life balance as “the general term to describe organizational initiatives aimed at enhancing employee experience of work and non-work domains” (p. 112). Many organizations have implemented a variety of initiatives to address work-family conflict issues (Anderson, Coffey, & Byerly, 2002). Some of these initiatives have taken the form of flexible work arrangements, and have been associated with increased levels of control over work and family issues (McNall, Masuda, &
Nicklin, 2010). Pierce and Newstrom (1980), for instance, suggest that flexible work arrangements can reduce stress regarding instances of late arrival at work, resulting from balancing work and family. In a similar vein, Hughes and Parkes (2007) found that work-time control moderated the relationship between work hours and work-family interference.
Nonetheless, empirical research conducted by Tausig and Fenwick (2001) has shown that flexible work schedules, including part-time employment, have not led to a greater sense of work-life balance. For instance, workers adhering to standard Monday to Friday, daytime schedules have reported more balance between work and family than workers adhering to flexible work schedules (Tausig & Fenwick, 2001). Tausig and Fenwick (2001) and McNall et al. (2010) indicated that the nature of the flexible work arrangement affected a worker's perceived sense of work-life balance. Furthermore, research has shown that unsocial working hours and irregular working patterns associated with shiftwork and nightwork are inversely related to personal and work outcomes (Furnham & Hughes, 1999; Nabe-Nielsen, Kecklund,
9 Ingre, Skotte, Diderichsen, & Garde, 2010). In a similar vein, Martin, Sinclair, Lelchook, Wittmer, and Charles, (2012) established that workers on nonday shifts, and nightwork in particular, are at greater risk of turnover. One possible explanation for these negative effects may lie in the fact that shiftwork and nightwork often require workers to work overtime.
Scholars indicated that as the amount of hours a worker spends at work increases, conflict between the worker’s work and family lives increases as well (Adkins & Premeaux, 2012). In particular, working involuntary overtime has been associated with work life imbalance and low satisfaction (Beckers, van der Linden, Smulders, Kompier, Taris & Geurts, 2008).
Nonetheless, shiftwork and nightwork are only prevalent in those occupations and industries that provide services that are needed at all hours (Beers, 2000).
Recent research conducted by Anderson et al. (2002) reported that conflict resulting from work-life imbalance is associated with job satisfaction. In particular, work-to-family conflict related negatively to job satisfaction. In a similar vein, Zhao, Qu, and Chiselli and Qu and Zhao (2012) found that work-to-family conflict moderated the relationship between life satisfaction and job satisfaction. Thus, less work-family conflict is likely to produce a positive spillover effect of life satisfaction on job satisfaction. Further, Beauregard and Henry (2009) suggested that work-life balance practices might influence organizational performance through a series of alternative paths, such as improved productivity and reduced turnover.
In their meta-analytic examination of the linkage between family-friendly work environments and work/family conflict, Mesmer-Magnus and Viswesvaran (2006) reported that flexibility of the work schedule does not significantly affect work/family conflict. They suggested that workers who perceive the work/family culture in their organization as negative may choose not to use flexible work practices in fear of retaliation (Mesmer-Magnus &
Viswesvaran, 2006). Hence, the availability of such practices may not be a sufficient condition for balancing work and family. Consequently, a failure to balance work and life may then result in dissatisfaction with the job. On the basis of these findings the following hypotheses are proposed:
Hypothesis 5. Irregular working hours positively affects work-life imbalance.
Hypothesis 6. Work-life imbalance negatively affects job satisfaction.
Hypothesis 7. Work-life imbalance mediates the relationship between irregular working hours and job satisfaction.
10 The Moderating Role of Autonomy
A study of employee reactions to job characteristics by Hackman and Lawler (1971) suggested that motivating jobs should provide the worker with a sense of personal responsibility for outcomes of the work. Hackman and Lawler (1971) indicated that the autonomy dimension accurately represents “the degree to which workers feel personal responsibility for their work” (p. 263). Autonomy has been defined as “the degree to which the job provides substantial freedom, independence and discretion to the individual in scheduling the work and in determining the procedures to be used in carrying it out”
(Hackman & Oldham, 1976: p. 258). Dunnette, Campbell, and Hakel (1967) reported that job content dimensions, such as achievement, responsibility, and recognition are associated with both satisfaction and dissatisfaction. Thus, the outcomes of the work are the direct result of the worker's actions, and therefore contribute to an enhanced sense of satisfaction or dissatisfaction.
As mentioned in the previous sections, flexible work schedules have been associated with more control (i.e., autonomy). In particular, flexible work schedules are expected to lead to increases in job satisfaction by affecting perceptions of job aspects, such as autonomy (Baltes et al., 1999). However, it is also expected that autonomy affects the linkage between flexible work schedules and job satisfaction. As a case in point, much relevant research (Hackman & Lawler, 1971; Brief & Aldag, 1975; Dunham, 1976; Hackman & Oldham, 1976) has empirically confirmed the association between autonomy and variety. Moreover, Dodd and Ganster (1996) found that autonomy interacted with variety; that is, “in a high variety task, increased autonomy led to increased satisfaction, while in a low variety task, increased autonomy had a negligible effect on satisfaction” (p. 329). They suggested that in a complex (i.e., challenging) job, gains (e.g., performance, satisfaction) are maximized when workers are able to change the “pacing, method, and order of doing the work” (Dodd & Ganster, 1996: p.
343). Hence, only then can workers fully utilize their skills and talents. Consistent with the aforementioned findings, the following hypothesis is proposed:
Hypothesis 8. Autonomy moderates the relationship between task variety and job satisfaction, such that the positive indirect effect of irregular working hours on job satisfaction through task variety is enhanced.
Andresen, Domsch, and Cascorbi (2007) found that stressful working conditions (e.g., working unsocial hours, irregular working patterns) led to stress associated with the inability
11 to balance work and family, which in turn led to a reduction in job satisfaction. Andresen et al. (2007), however, also indicated that factors associated with control (e.g., involvement and participation, co-determination) reduced the strain resulting from work-life imbalance. Thus, in addition to interacting with variety, autonomy is also expected to interact with work-life imbalance. Accordingly, the following hypothesis is proposed:
Hypothesis 9. Autonomy moderates the relationship between work-life imbalance and job satisfaction, such that the negative indirect effect of irregular working hours on job satisfaction through work-life imbalance is mitigated.
Figure 1. The hypothesized conceptual scheme.
METHODS
Sample and Data Collection
For the purpose of the current study, a questionnaire (the European Social Survey) consisting of approximately 120 items was used. The ESS, currently in its fifth round, consists of a core module, and a set of rotating modules. Whereas the core module remains stable, monitoring a range of social variables, the rotating modules change over time. The ESS collects data from over thirty European nations, allowing for cross-country comparisons.
To ensure high methodological standards were met, National Coordinators of the ESS were provided a set of measures regarding field procedures. The aim of these measures was
Irregular working hours
Autonomy
Job satisfaction Work-life
imbalance Task variety
12 twofold: (1) to ensure a high response rate (i.e., 70%), and (2) to capture a sample that closely represented the country’s total population.
For the purpose of the current study, the country file specific to the Netherlands was used. The dataset included the responses of 1,829 respondents. The data were collected in 2010. In March 2012 a 2.0 revision of the data was released. Subsequently a set of measures, discussed in more detail later on in this section, were defined. Consequently, a sample of 784 respondents yielding valid responses remained.
Measures
Irregular working hours (the independent variable), work-life imbalance, and task variety (the mediators), autonomy (the moderator), and job satisfaction (the dependent variable).
Irregular working hours. A three-item scale was used to measure irregular working hours. Respondents were asked how often their work involved (1) working evenings or nights, (2) having to work overtime at short notice, and (3) working at weekends (1 = “never,” 7 =
“every day”). The Cronbach’s alpha coefficient for the items was .67.
Work-life imbalance. Work-life imbalance was measured using six items.
Respondents were asked how often they (1) kept worrying about work problems when they were not working, (2) felt too tired after work to enjoy things they liked to do at home, (3) found that their work prevented them from giving the time they wanted to their partner or family, (4) found that their partner or family got fed up with the pressures of their jobs, (5) found that their family responsibilities prevented them from giving time they should to their job, and (6) found it difficult to concentrate on work because of their family responsibilities (1
= “never,” 5 = “always”). The Cronbach’s alpha was .76.
Task variety. I measured task variety using a two-item scale developed for this study (1 = “not at all true,” 4 = “very true”). The two items that were extracted to develop the scale were, “There is a lot of variety in my work” and “Job requires that I keep learning new things”. The Cronbach’s alpha coefficient was .66.
Autonomy. I used a two-item scale to ask respondents the extent to which they perceived to be in control in their current/past job(s) (1 = “I have/had no influence,” 10 = “I have/had complete control”). The items were, “Decide how daily work is/was organized” and
13
“Choose or change your pace of work”. The Cronbach’s alpha coefficient for the items was .81.
Job satisfaction. Respondents were asked to evaluate their current job using a two- item scale. The scale consists of two items that aim to measure an individual’s overall job satisfaction. The two items were, “Satisfaction in main job” (1 = “extremely dissatisfied,” 10
= “extremely satisfied”) and “Enjoy working in current job even if I did not need the money”
(1 = “disagree strongly,” 5 = “agree strongly”). The Cronbach’s alpha was .54.
Control variables. Because of the potential effects of various demographic variables on the hypothesized conceptual model, I controlled for, gender, age, and education level.
Measures of these demographic variables were obtained as part of the ESS questionnaire.
Gender was dummy-coded (1 = “male,” 0 = “female”). Age was calculated (i.e., year of birth was self-reported). Education level was measured on a seventeen-point Likert scale (1 = “did not finish primary school,” 17 = “doctors degree”).
Data Analyses
Analysis of the dataset (i.e., the 2.0 country file specific to the Netherlands) was carried out using the statistical software package SPSS. For the purpose of the current study, the aforementioned measures, with exception of the control variables, were standardized to:
(1) eliminate potential inconsistencies arising from the use of variables measured on different scales, and (2) to prevent multicollinearity in respect to the interaction terms.
A bivariate correlation test was carried out to explore the relationship between job satisfaction and the four job characteristics; the control variables were included as well.
Simple and multiple linear regression analyses were then used to further examine the relationships proposed in figure 1. To investigate the indirect effects, I used the method of bootstrapping. Bootstrapping circumvents the limitations of some statistical methods, such as the Sobel test, that make assumptions about the shape of sampling distributions (e.g., normality) (Kenny, 2012). To that end, I utilized a macro for SPSS written by Preacher and Hayes (2004). Finally, the moderated mediation hypotheses were assessed using a macro for SPSS written by Preacher, Rucker, and Hayes (2007).
14 RESULTS
Descriptive Statistics
Table 1 presents the means, standard deviations, and zero-order Pearson correlations of all the key variables. An examination of the correlations reveals that irregular working hours (r = .09, p < .05), task variety (r = .30, p < .01), and autonomy (r = .15, p < .01) were positively related to job satisfaction. Conversely, work-life imbalance was inversely related to job satisfaction (r = -.16, p < .01). A further investigation of the results indicated that gender was inversely related to job satisfaction (r = -.07, p < .05); suggesting that women perceive higher levels of job satisfaction then men.
TABLE 1
Means, Standard Deviations, Correlations, and Reliabilitiesa
Key Variables Mean s.d. 1 2 3 4 5 6 7 8
1. Respondents’ gender 1.54 0.50
2. Respondents’ age 50.42 17.50 .01
3. Respondents’ highest education
373.78 315.62 .05* -.06*
4. Irregular working hours 2.58 1.33 .04 -.12** .10** (.67)c
5. Work-life imbalance 2.38 0.77 .02 -.07 .11** .24** (.76)
6. Task variety 2.93 0.79 .02 -.03 .19** .22** .02 (.66)
7. Autonomy 6.77 2.92 .05 .01 .15** .06 .06 .18** (.81)
8. Job satisfaction 5.61 1.17 -.07* .09* -.02 .09* -.16** .30** .15** (.54)
an = 784.
bValues are Z-scores.
cCronbach’s alpha coefficients are reported on the diagonal axis.
*p < .05
**p < .01
Regression Results
A multiple regression was conducted to test Hypotheses 1, 3, 4, 6, 7, 8 and 9 following the steps suggested by Judd and Kenny (1981), and Baron and Kenny (1986). As shown in Table 2, after regressing job satisfaction on the control variables in step 1, irregular working hours was entered in step 2. Irregular working hours was found to be positively related to job satisfaction (β = 0.10, p < 0.01, ∆R2 = 0.01), lending support to Hypothesis 1, which predicts that working irregular hours is positively related to satisfaction with one’s job. In step 3, both mediating variables were entered. Table 2 shows that work-life imbalance relates negatively to job satisfaction (β = -0.18, p < 0.001, ∆R2 = 0.12), lending support to Hypothesis 6, which predicts that stress resulting from balancing work-life will decrease satisfaction with one’s
15 job. Table 2 also shows that task variety relates positively to job satisfaction (β = 0.31, p <
0.001, ∆R2 = 0.12), lending support to Hypothesis 3. In step 4, autonomy, and interactions were entered. Table 2 shows that autonomy moderates the link between work-life imbalance and job satisfaction (β = 0.09, p < 0.01, ∆R2 = 0.02), lending support to Hypothesis 9, which predicts that autonomy positively moderates the link between work-life imbalance and job satisfaction.
TABLE 2
Results of Regression Analysis for Mediation and Moderation Effects on Job Satisfactiona
Model 1 Model 2 Model 3 Model 4
Variables β t β t β t β t
Control
Respondents’ gender -0.07 -2.09* -0.08 -2.23* -0.08 -2.42* -0.09 -2.62**
Respondents’ age 0.09 2.47* 0.10 2.81** 0.10 2.85** 0.09 2.80**
Respondents’ highest education -0.02 -0.43 -0.03 -0.71 -0.06 -1.88 -0.08 -2.29*
Independent
Irregular working hours 0.10 2.90** 0.08 2.37* 0.08 2.38*
Mediators
Work-life imbalance -0.18 -5.08*** -0.17 -4.95***
Task variety 0.31 8.86*** 0.28 8.04***
Moderator
Autonomy 0.13 3.81***
Interactive effects
WLIB balance X autonomy 0.09 2.68**
Task variety X autonomy -0.03 -0.78
R2 .01 .02 .14 .17
∆R2 .01** .12*** .02***
an = 784.
*p < .05
**p < .01
***p < .001
Additionally, a simple regression was conducted to determine the nature of the relationships predicted in Hypotheses 2, and 5. As shown in Table 3A, after regressing work- life imbalance on the control variables in step 1, irregular working hours was entered in step 2. Irregular working hours was found to be positively related to work-life imbalance (β = 0.23, p < 0.001, ∆R2 = 0.05), lending support to Hypothesis 5.
16 TABLE 3A
Results of Regression Analysis for the Effects of Irregular Working Hours on Work-Life Imbalance
Model 1 Model 2
Variables β t β t
Control
Respondents’ gender 0.02 0.43 0.01 0.15
Respondents’ age -0.06 -1.78 -0.04 -1.02
Respondents’ highest education 0.10 2.94** 0.08 2.38*
Independent
Irregular working hours 0.23 6.56***
R2 .02 .07
∆R2 .05***
an = 784.
*p < .05
**p < .01
***p < .001
As shown in Table 3B, after regressing task variety on the control variables in step 1, irregular working hours was entered in step 2. Irregular working hours was found to be positively related to task variety (β = 0.20, p < 0.001, ∆R2 = 0.04), lending support to Hypothesis 2.
TABLE 3B
Results of Regression Analysis for the Effects of Irregular Working Hours on Task Varietya
Model 1 Model 2
Variables β t β t
Control
Respondents’ gender 0.02 0.47 0.01 0.23
Respondents’ age -0.03 -0.79 0.00 -0.11
Respondents’ highest education 0.19 5.47*** 0.17 5.02***
Independent
Irregular working hours 0.20 5.64***
R2 .04 .08
∆R2 .04***
an = 784.
*p < .05
**p < .01
***p < .001
Tests of Mediation
Table 4 presents additional results for Hypotheses 2, 3, 4, 5, 6 and 7. The bootstrap test for indirect effects demonstrated that in both cases the proposed indirect effect was significant, with a bootstrapped 95% CI around the indirect effect not containing zero (-.07, - .02, and .04, .09).
17 TABLE 4
Bootstrap results for indirect effect(s)
M SE LL 95% CI UL 95% CI
Effect(s)
Work-life imbalance -0.04 0.01 -.07 -.02
Task variety 0.06 0.01 .04 .09
Note. n = 784. Bootstrap sample size = 5,000. LL = lower limit; CI = confidence interval; UL = upper limit.
Tests of Moderated Mediation
To assess the conditional indirect effects depicted in Figure 1 (i.e., Hypotheses 8 and 9), I examined the conditional effect of irregular working hours on job satisfaction (through work-life imbalance, and task variety) at three values of autonomy (see Table 4): the mean (0.00), one standard deviation above the mean (1), and one standard deviation below the mean (-1). Bootstrap results indicated that two out of the three conditional indirect effects (based on moderator values at the mean and at -1 standard deviation) were negative and significantly different from zero. Thus, Hypothesis 9 was supported, such that the indirect and negative effect of irregular working hours on job satisfaction through work-life imbalance was observed when levels of autonomy were moderate to low, but not when autonomy was high.
TABLE 5
Results for conditional indirect effect(s)
Boot indirect effect
Bias corrected and accelerated LL 95% CI
Bias corrected and accelerated
UL 95% CI
Boot p
Conditional indirect effect at N-exp = M ± 1 SD Stress/WLB
-1 SD (-1.00) -0.07 -.10 -.04 .00
M (0.00) -0.04 -.07 -.02 .00
+1 SD (1.00) -0.02 -.05 .01 .13
Note. n = 784. Bootstrap sample size = 5,000. LL = lower limit; CI = confidence interval; UL = upper limit.
DISCUSSION
The present study aimed to address the gap in knowledge about the linkage between flexible work schedules and worker satisfaction by assessing various intervening mechanisms.
To that end, I developed a conceptual scheme that proposed that the association between working irregular hours and job satisfaction is more comprehensive than past research has
18 suggested (Baltes et al., 1999). First, I predicted that a direct link between working irregular hours and a worker’s satisfaction existed. I then proposed that both variety associated with a task or job and a failure to balance work and family acted as mediating mechanisms in the linkage between working irregular hours and job satisfaction. Lastly, I anticipated that a worker’s perceived sense of autonomy enhanced or reduced the indirect relationship between working irregular hours and job satisfaction. The outcomes of the study largely supported the hypothesized moderated mediation model, indicating that the significance of the indirect effects were at the least partially dependent on the worker’s perceived sense of autonomy.
These findings suggest that satisfaction as perceived by the worker is not merely contingent on the schedule (s)he adheres to. Rather, it appears that workers place great emphasis on the job dimensions that underlie such a schedule.
Main Findings
Irregular working hours. The first research question pertained to the direct effect of flexible work schedules on reports of worker satisfaction. As predicted, the statistical analyses established the existence of a positive association. That is, workers adhering to flexible work schedules reported higher levels of satisfaction with the work, whereas workers adhering to traditional schedules reported lower levels of satisfaction with the work. This finding seems to be consistent with prior research on the effects of flexible work schedules (e.g., Pierce, 1989;
Baltes et al., 1999). Flexible work schedules permit workers to co-determine the schedule. In particular, the scheduling of the flexible hours outside of the mandatory schedule (i.e., core hours) is left at the discretion of the worker. This sense of work-time control (i.e., autonomy) has been associated with more favorable attitudes towards the work (Baltes et al., 1999).
Task variety. The second set of research questions proposed that task variety operated as a mediating mechanism for the relationship between irregular working hours and job satisfaction. Once more, the statistical analyses confirmed the hypotheses. The results suggest that the nature of a work schedule (i.e., traditional, alternate) affects the requirements of a task or job, which in turn affects attitudes towards the work. These findings seem to be in accordance with the theory of work adjustment (Pierce & Newstrom, 1980). If the nature of the work schedule enhances (reduces) the correspondence between the abilities of the worker and the requirements of the task or job, the worker is more (less) likely to perform the work in a satisfactory manner as well as perceive the work as more (less) satisfying.
19 Work-life imbalance. Further, the third set of research questions proposed that stress resulting from a failure to balance work and family also acted as a mediating mechanism for the linkage between irregular working hours and job satisfaction. Results indicated that workers adhering to flexible work schedules experienced difficulties maintaining a balance between work and family; a finding that is consistent with previous studies on work-to-family and family-to-work conflict (Hughes & Parkes, 2007; Greenhaus, Peng, & Allen, 2012;
Adkins & Premeaux, 2012). When workers perceive a work schedule as being too demanding (e.g., unsocial working hours, irregular working patterns), their work and family identities (i.e., roles) tend to become unstable. As a consequence, both work and life satisfaction have a tendency to decrease. Furthermore, given the reciprocity of the linkage between work and life satisfaction (Qu & Zhao, 2012), either type of conflict (work-to-family, family-to-work) affects job satisfaction; either directly or indirectly.
Autonomy. The final set of research questions pertained to the moderating role of autonomy. Contrary to expectations, autonomy did not moderate the positive indirect effect of irregular working hours on job satisfaction through task variety. This finding contradicts prior research on job aspects, which suggests that certain interactions between autonomy and variety lead to more favorable attitudes towards the work (e.g., Dodd & Ganster, 1996).
Nonetheless, some occupational groups may be less susceptible to increases in autonomy than others (Pierce, 1989). For instance, professionals and managers already enjoy high levels of autonomy, and are thus less or not at all sensitive to further increases (Baltes et al., 1999).
Moderate to low levels of autonomy did, however, moderate the negative indirect effect of irregular working hours on job satisfaction through work-life imbalance. This finding seems to be consistent with prior research on work-time control (Hughes & Parkes, 2007; Beckers, van der Linden, Smulders, Kompier, Taris, & Geurts, 2008; McNall et al., 2010). That is, moderate to low levels of autonomy may act as a coping mechanism for the worker, whereas high levels of autonomy may become a burden.
Theoretical Implications
In accordance with the majority of the literature (Baltes et al., 1999), the present study found that flexible work schedules were associated with positive worker attitudes. Thus, this study corroborates many of the empirical findings of past research. However, it is by investigating the nature of the aforementioned linkage that the present study has made an attempt to extend on the literature. In doing so, this study established the presence of both
20 positive and negative intervening mechanisms. This is a significant, however, modest contribution to the literature. Specifically, since many studies have investigated the effects of flexible work schedules on job satisfaction, but only few have uncovered the reasons underlying those findings. In addition, the present study has made an attempt to address the calls of scholars for the consideration of multiple contexts (i.e., work-related, non-work- related) when assessing the effects of flexible work schedules. As Baltes et al. (1999) stated,
“as one evaluates the effects of alternative work schedules, it is important to recognize that the effect patterns obtained are context dependent” (p. 510).
Surprisingly, this study found that autonomy did not moderate the linkage between task variety and job satisfaction. Seminal works in the job design literature (Hackman &
Lawler, 1971; Hackman & Oldham, 1976) suggested that the job characteristics (i.e., variety, autonomy, task identity, feedback) interact. The product of that interaction is the so-called
‘motivating potential score’ set forth by Hackman and Oldham (1976); which is significantly influenced by the degree of autonomy. Thus, the present study extends on the literature by arguing that the effects of autonomy may be more linear than previously assumed. That is, the direct effects of autonomy on job satisfaction may have been of greater significance than its interactive effects.
Further, the present study has considered the possible effects of contexts of life on the linkage between irregular working hours and job satisfaction. Although work-life imbalance negatively intervened in the aforementioned linkage, moderate to low levels of autonomy did mitigate the effects. This suggests that control over the scheduling of the work, by itself, is not a sufficient condition for balancing work and family. Instead, controlling other aspects of the job – to some extent – as well seems to be a prerequisite for worker satisfaction. Hence, aspects of job design may also be relevant in overcoming the deficiencies associated with conflicts of work and life.
Practical Implications
The sustained popularity of flexible work schedule arrangements suggests that businesses derive all sorts of benefits from them, including a satisfied workforce. From a practitioner’s perspective then, the decision to adopt such practices might be a straightforward one. Nonetheless, it is imperative that practitioners comprehend the mechanisms that underlie the linkage between flexible work schedules and reports of worker satisfaction for several reasons. First, for a flexible work schedule arrangement to be effective, it should correspond
21 with both the abilities of the worker and the requirements of the task or job. A highly interdependent work environment, for instance, requires a worker’s schedule to be more or less synchronized with that of his/her colleagues. Hence, the costs associated with maintaining flexible work schedules in such an environment would simply outweigh the benefits. Second, the inability of workers to balance work and family life might interfere with the relationship between flexible work schedules and worker satisfaction. If practitioners, however, succeed in creating an organizational climate that facilitates worker’s work-life balance, the negative effects associated with the interferences might be weakened. Third, in some instances giving workers more discretion may help them cope with the aforementioned work-life interferences. Nonetheless, practitioners should act cautiously when giving workers more discretion, since high levels of discretion may actually worsen the interferences.
Furthermore, workers that already enjoy high levels of discretion are not susceptible to any further increases.
Limitations and Future Research
As is the case with any study, this one is subject to limitations. A significant limitation of the present study could be the assumption of causality. The present study assumes that irregular working hours (the independent variable) affects job satisfaction (the dependent variable). However, the opposite may also be true. In a similar vein, reverse causal effects could exist between the mediating variables and the dependent variable (Kenny, 2012). One could theoretically justify the causal assumptions that underlie the conceptual scheme.
Nonetheless, one cannot conclude with absolute certainty that the causal assumptions are in fact just. A second concern could be common-source bias. When all data is collected from a single source (i.e., a common source) – the European social survey – results tend to be distorted. The results of the present study, then, may be due to the method of data collection.
Thus, without supplementary sources of data the results of the present study are not easily justifiable (Podsakoff & Organ, 1986). A third concern could be the generalizability of the present study. The dataset merely included responses from individuals located in the Netherlands. Analysis of the integrated dataset could have yielded further insights into the issue under study. Nonetheless, due to non-stratified two-stage sampling (i.e., postal delivery points, person within a household) 1,829 valid interviews were obtained (i.e., a 60 percent response rate). Thus, the inferences made in the present study are generalizable to the national context of the Netherlands. A final concern could be the quality of the interviewers. The computer-assisted personal interview (i.e., CAPI) was conducted by free-lance interviewers
22 who had no affiliation with the fieldwork organization. Nonetheless, the interviewers were extensively briefed on fieldwork procedures. Furthermore, the CAPI eliminates many of the potential inconsistencies arising from the use of multiple interviewers (Couper & Burt, 1994).
Future studies should aim to address the present studies’ weaknesses. Furthermore, it would be of interest to see how incorporating multiple job dimensions, work and other contexts, and worker attitudes and traits would affect the results of the present study. For example, how do(es) additional job characteristics (Hackman & Lawler, 1971; Hackman &
Oldham, 1976; Skalli et al., 2008), a family-friendly work environment (Mesmer-Magnus &
Viswesvaran, 2006), and/or a worker’s need for personal growth and development (Hackman
& Oldham, 1976) affect the outcomes of this study.
23 REFERENCES
Adkins, C. L., & Premeaux, S. F. 2012. Spending time: the impact of hours worked on work- family conflict. Journal of Vocational Behavior, 80: 380-389.
Anderson, S. E., Coffey, B. S., & Byerly, R. T. 2002. Formal organizational initiatives and informal workplace practices: links to work-family conflict and job-related outcomes.
Journal of Management, 28(6): 787-810.
Andresen, M., & Domsch, M. E., & Cascorbi, A. H. 2007. Working unusual hours and its relationship to job satisfaction: a study of European maritime pilots. Journal of Labor Research, 28: 714-734.
Baltes, B. B., Briggs, T. E., Huff, J. W., Wright, J. A., & Neuman, G. A. 1999. Flexible and compressed workweek schedules: a meta-analysis of their effects on work-related criteria.
Journal of Applied Psychology, 84(4): 496-513.
Baron, R. M., & Kenny, D. A. 1986. The moderator-mediator variable distinction in social psychology research: conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51: 1173-1182.
Beauregard, T. A., & Henry, L. C. 2009. Making the link between work-life balance practices and organizational performance. Human Resource Management Review, 19: 9-22.
Beckers, D. G. J., van der Linden, D., Smulders, P. G. W., Kompier, M. A. J., Taris, T. W., &
Geurts, S. A. E. 2008. Voluntary or involuntary? Control over overtime and rewards for overtime in relation to fatigue and work satisfaction. Work and Stress, 22(1): 33-50.
Beers, T. M. 2000. Flexible schedules and shiftwork: replacing the ‘9-to-5’ workday?
Monthly Labor Review, 123(6): 33-40.
Brewster, C., Mayne, L., & Tregaskis, O. 1997. Flexible working in Europe: a review of the evidence. Management International Review, 37: 85-103.
Brief, A. P., & Aldag, R. J. 1975. Employee reactions to job characteristics: a constructive replication. Journal of Applied Psychology, 60(2): 182-186.
Couper, M. P., & Burt, G. 1994. Interviewer attitudes toward computer-assisted personal interviewing (CAPI). Social Science Computer Review, 12(1): 38-54.
Darcy, C., McCarthy, A., Hill, J., & Grady, G. 2012. Work-life balance: one size fits all? An exploratory analysis of the differential effects of career stage. European Management Journal, 30: 111-120.
Delfgaauw, J. 2007. The effect of job satisfaction on job search: not just whether, but also where. Labor Economics, 14: 299-317.
24 Dodd, N. G., & Ganster, D. C. 1996. The interactive effects of variety, autonomy, and feedback on attitudes and performance. Journal of Organizational Behavior, 17: 329-347.
Dunham, R. B. 1976. The measurement and dimensionality of job characteristics. Journal of Applied Psychology, 61(4): 404-409.
Dunnette, M. D., Campbell, J. P., & Hakel, M. D. 1967. Factors contributing to job satisfaction and job dissatisfaction in six occupational groups. Organizational Behavior and Human Performance, 2: 143-174.
Fried, Y., & Ferris, G. R. 1987. The validity of the job characteristics model: a review and meta-analysis. Personnel Psychology, 40: 287-322.
Furnham, A., & Hughes, K. 1999. Individual difference correlates of nightwork and shift- work rotation. Personality and Individual Differences, 26: 941-959.
Greenhaus, J. H., Peng, A. C., & Allen, T. D. 2012. Relations of work identity, family identity, situational demands, and sex with employee work hours. Journal of Vocational Behavior, 80: 27-37.
Hackman, J. R., & Lawler, E. E. 1971. Employee reactions to job characteristics. Journal of Applied Psychology, 55(3): 259-286.
Hackman, J. R., & Oldham, G. R. 1976. Motivation through the design of work: test of a theory. Organizational Behavior and Human Performance, 16: 250-279.
Hughes, E. L., & Parkes, K. R. 2007. Work hours and well-being: the roles of work-time control and work-family interference. Work and Stress, 21(3): 264-278.
Ilies, R., Wilson, K. S., & Wagner, D. T. 2009. The spillover of daily job satisfaction onto employees’ family lives: the facilitating role of work-family integration. Academy of Management Journal, 52(1): 87-102.
Judd, C. M., & Kenny, D. A. 1981. Process analysis: estimating mediation in treatment evaluations. Evaluation Review, 5: 602-619.
Judge, T. A., Thoresen, C. J., Bono, J. E., & Patton, G. K. 2001. The job satisfaction-job performance relationship: a qualitative and quantitative review. Psychological Bulletin, 127(3): 376-407.
Kelloway, E. K., & Barling, J. 1991. Job characteristics, role stress and mental health.
Journal of Organizational Psychology, 64: 291-304.
Lawler, E. E., & Hall, D. T. 1970. Relationship of job characteristics to job involvement, satisfaction, and intrinsic motivation. Journal of Applied Psychology, 54(4): 305-312.
Locke, E. A. 1969. What is job satisfaction? Organizational Behavior and Human Performance, 4: 309-336.
25 Locke, E. A. 1970. Job satisfaction and job performance: a theoretical analysis.
Organizational Behavior and Human Performance, 5: 484-500.
Loher, B. T., Noe, R. A., Moeller, N. L., & Fitzgerald, M. P. 1985. A meta-analysis of the relation of job characteristics to job satisfaction. Journal of Applied Psychology, 70(2): 280- 289.
Martin, J. E., Sinclair, R. R., Lelchook, A. M., Wittmer, J. L. S., & Charles, K. E. 2012. Non- standard work schedules and retention in the entry-level hourly workforce. Journal of Occupational and Organizational Psychology, 85: 1-22.
McNall, L. A., Masuda, A. D., & Nicklin, J. M. 2010. Flexible work arrangements, job satisfaction, and turnover intentions: the mediating role of work-to-family enrichment. The Journal of Psychology, 144(1): 61-81.
Mesmer-Magnus, J. R., & Viswesvaran, C. 2006. How family-friendly work environments affect work/family conflict: a meta-analytic examination. Journal of Labor Research, 27(4):
555-574.
Mobley, W. H. 1977. Intermediate linkages in the relationship between job satisfaction and employee turnover. Journal of Applied Psychology, 62(2): 237-240.
Nabe-Nielsen, K., Kecklund, G., Ingre, M., Skotte, J., Diderichsen, F., & Garde, A. H. 2010.
The importance of individual preferences when evaluating the associations between working hours and indicators of health and well-being. Applied Ergonomics, 41: 779-786.
Origo, F., & Pagani, L. 2009. Flexicurity and job satisfaction in Europe: the importance of perceived and actual job stability for well-being at work. Labor economics, 16: 547-555.
Orpen, C. 1981. Effect of flexible working hours on employee satisfaction and performance: a field experiment. Journal of Applied Psychology, 66(1): 113-115.
Pierce, J. L, & Newstrom, J. W. 1980. Toward a conceptual clarification of employee responses to flexible working hours : a work adjustment approach. Journal of Management, 6(2): 117-134.
Podsakoff, P. M., & Organ, D. W. 1986. Self-reports in organizational research: problems and prospects. Journal of Management, 12(4): 531-544.
Pousette, A., & Hanse, J. J. 2002. Job characteristics as predictors of ill-health and sickness absenteeism in different occupational types: a multi group structural equation modeling approach. Work and Stress, 16(3): 229-250.
Qu, H., & Zhao, X. R. 2012. Employees’ work-family conflict moderating life and job satisfaction. Journal of Business Research, 65: 22-28.
Sims, H. P., & Szilagyi, A. D. 1976. Job characteristic relationships: individual and structural moderators. Organizational Behavior and Human Performance, 17: 211-230.
26 Skalli, A., Theodossiou, I., & Vasileiou, E. 2008. Jobs as Lancaster goods: facets of job satisfaction and overall job satisfaction. The Journal of Socio-Economics, 37: 1906-1920.
Tausig, M., & Fenwick, R. 2001. Unbinding time: alternate work schedules and work-life balance. Journal of Family and Economic Issues, 22(2): 101-119.
Tregaskis, O., Brewster, C., Mayne, L., & Hegewisch, A. 1998. Flexible working in Europe:
the evidence and the implications. European Journal of Work and Organizational Psychology, 7(1): 61-78.
Wotruba, T. R. 1990. Full-time vs. part-time salespeople: a comparison on job satisfaction, performance, and turnover in direct selling. International Journal of Research in Marketing, 7: 97-108.
Yang, J. T. 2010. Antecedents and consequences of job satisfaction in the hotel industry.
International Journal of Hospitality Management, 29: 609-619.
Zhao, X. R., Qu, H., & Ghiselli, R. 2011. Examining the relationship of work-family conflict to job and life satisfaction: a case of hotel sales managers. International Journal of Hospitality Management, 30: 46-54.
Dataset
ESS Round 5: European Social Survey Round 5 Data (2010). Data file edition 2.0. Norwegian Social Science Data Services, Norway – Data Archive and distributor of ESS data.
