Abstract
BACKGROUND:
Work-family conflict (WFC) is an inter-role conflict, which suggests that fulfilling expectations of family roles makes it difficult to satisfy expectations of work roles, and vice versa. Living an academic life includes balancing multiple work demands and family responsibilities, which may generate WFC for many faculty members. Researchers have emphasized the need for further studies of how faculty integrate work and family demands.
OBJECTIVE:
This study explores WFC among Iranian faculty. We examine relationships among work hours, time spent with family, work-interference with family (WIF), family-interference with work (FIW), and job satisfaction.
METHODS:
Faculty members from 25 Iranian public universities completed a questionnaire. Structural equation modeling was used to test hypotheses in a single model.
RESULTS:
Findings suggest a positive relationship between faculty weekly work hours and WIF, and between time spent with family and FIW. WIF correlated negatively with job satisfaction, and work hours correlated positively with job satisfaction. Time spent with family and FIW had no influence on job satisfaction, and spouse employment moderated the relationship between WIF and job satisfaction.
CONCLUSIONS:
Findings have implications for human resources and organizational development professionals seeking insight into how faculty members and other knowledge workers experience work-family interrelationships.
Introduction
Academic careers require simultaneous fulfillment of multiple roles – teacher, researcher, and service provider [1–3]. As teachers, faculty members are expected to devote time to class preparation, be present in class and during office hours, and provide feedback to students [4]. As researchers, they are also required to publish in peer-reviewed and top-tier journals, especially for tenure-track positions [2]. As service providers, academicians must participate in administrative meetings and serve in diverse committees [5]. Many experience high stress due to obligations regarding fulfilling multiple responsibilities [6].
Research across several countries suggests faculty work has become more demanding in recent decades [2]. “The modern academic workplace is characterized by student diversity, new technologies, a shift in emphasis toward the learner, expanding faculty workloads, and a new labor market for faculty” [5]. Transitions in higher education have changed expectations from faculty, and extended workloads and professional responsibilities in many directions [2, 5]. It has become increasingly challenging for faculty members to balance work demands and family responsibilities. In the past few decades, growth of dual-earner and single-parent families, and increasing numbers of employees with child and eldercare responsibilities, denote that a significant number of employed individuals, including faculty members, must find ways to combine professional tasks with personal responsibilities [7]. Approximately one-third of U.S. employees are satisfied with how they juggle work and family [8]. Researchers with various research interests have studied how faculty members handle work and family demands [9–16]. For example, a few studies focus on women faculty members and how they tackle tenure requirements and childcare responsibilities [17–19], or how work-family conflict(WFC) among female faculty members differs from male counterparts [20–24].
Researchers have also identified what causes WFC among faculty members. For example, work-related stress [25] and time pressure [26] are among sources that yield WFC. Researchers have also demonstrated negative impacts of WFC on faculty health, satisfaction, and work outcomes [27]. However, there is a dearth of knowledge concerning work-family issues of faculty members in non-western contexts [28, 29]. This study examines relationships among work hours, time spent with family, work interference with family (WIF), family interference with work (FIW), and job satisfaction for Iranian faculty. This study is first to address WFC among Iranian faculty, which contributes to literature on WFC in non-western contexts, especially developing countries.
Context of the study
Iran has more than 1,500 four-year universities, with approximately 66,100 faculty members (30% women) and 4,042,564 students at both public and private universities [30]. Major public universities, the focus of this study, are research institutions. Faculty positions are prestigious in Iran, and it is highly competitive to be hired as a faculty member at a research university. Similar to higher education in the United States and throughout Europe, most Iranian universities require faculty postions’candidates to have a PhD, publications in peer-reviewed national or international journals, presentations at conferences, and teaching experience. Most faculty members are recruited tentatively. After two years, if new hires meet requirements regarding publications and performance, they are hired permanently. Major ranks for Iranian faculty include assistant, associate, and full professors. Clinical and researchfaculty positions are available at most universities. New hires usually start at the assistant level where they fulfill academic requirements and garner votes from their programs, colleges, or universities to move to the next rank.
Faculty job requirements and responsibilities include teaching, research, and service provision, and staff members are required to work 40 hours per week. The number of credit hours faculty members teach depends on the college and department’s needs, but according to higher education regulations in Iran, clinical faculty are required to teach 8 to 13 credit hours weekly. Research faculty members are required to devote 24 to 37 hours to research per week, but since student populations are large, these faculty members teach several hours beyond their required loads, at the expense of research hours.
During the current decade, a number of social and demographic trends are transforming boundaries between work and family in Iran, altering society from traditional to modern [31]. For example, the number of women attaining higher education degrees is increasing, and full-time employment among women is growing across the country. In 2010, 11% of Iranian women held full-time jobs, and 62% of university students were female [32]. These transitions led to changes regarding work and family for individuals in various occupations, including university faculty members.
Theoretical framework
Several scholars have examined work-family interaction from various theoretical lenses. For example, Greenhaus and Beutell [33], guided by role theory [34], refer to WFC as the intersection of work and family. They regard WFC as an inter-role conflict, arguing that fulfilling expectations of family roles makes it difficult to satisfy expectations of work roles, and vice versa. Spillover theory [35], which examines spillover between work and family domains, and work-family border theory [36], which considers individuals as border-crossers who transit the two spheres daily, are among theories that explain interactions between work and family.
The conflict perspective, suggesting demands related to one domain interfere with participation in another, dominates the literature. This study is based on role-conflict theory [34]. Based on the role incompatibility conceptualization of WFC, Greenhaus and Beutell [33] identify two directions for work-family conflict: work interference with family (WIF) and family interference with work (FIW). They distinguish between strain-, time-, and behavior-based conflicts. Work and family roles can become incompatible because time, strain, and behavior from work domains interfere with family, or because time, strain, and behavior from family domains interfere with work [37]. The focus on the time-based conceptualization of WFC [33], assumes that time spent on activities in one role cannot be devoted to activities in another. For example, in work domain, faculty members who work long hours or have rigid work schedule requirements are more likely to experience WFC. In a family domain, having young children to care for, having an employed spouse, and having a large family contribute to WFC.
Work-family conflict among faculty
Work hours and WIF
Empirical literature identifies variables that contribute to WIF or FIW, both from work and family domains [33]. Time-based variables such as work-hours, inflexible work schedules, and shiftwork are studied prominently as antecedents of work interference with family [38]. Many studies examine the direct effects of work hours on WFC, supporting a positive relationship between hours of work and work-family conflict [39]. One meta-analysis demonstrated a strong correlation (0.26) between time spent at work and WIF [38], and another found a strong path coefficient (0.20) between work time demands and WIF [40]. Researchers also found that time spent at work moderates the relationships between WIF and family-related outcomes such as family satisfaction and strain [41].
Empirical research also incorporates faculty work overload and extensive work hours [12]. Jacobs and Winslow [12] use a national database to study academic job time pressures in the United States. Analyzing data from 11,162 faculty members at four-year institutions, they show that on average, U.S. faculty members across all academic statuses and including both genders work over 50 hours per week, and more than one-third work over 60 hours. Faculty members working over 60 hours weekly were highly successful at publishing papers and books [11]. O’Laughlin and Bischoff [42] study 184 tenured academicians employed in full-time positions at four-year universities and colleges, primarily in the United States, finding that both tenured and pre-tenured professors’ work hours exceed 50 hours per week. Extended literature includes findings from other countries such as the United Kingdom [43], Australia [44], Norway [2] Quebec [10], and Indonesia [26], which confirm long hours of work for faculty positions. Given these findings, we hypothesize that work hours correlate positively with WIF among faculty members.
H1: Iranian faculty members’ weekly work hours correlate positively with WIF.
Hours spent with family and FIW
Family interference with work arises when time devoted to requirements of family roles makes fulfillment of work-role requirements difficult [33]. The literature suggests a relationship between time spent on family responsibilities and WFC. Byron’s [38] meta-analytic review reveals that the more time is spent on family, housework, childcare, and other non-work-related activities, the more FIW is experienced (p = 0.21). Similarly, Michel et al.’s [40] meta-analysis found a correlation (0.15) between family time demand and FIW.
Men and women faculty members spend similar hours at work, and share the same family role orientations [20, 45]. Having children lessens both male and female faculty work hours slightly [12], but female professors spend 12 more hours on family demands in comparison to men [16], experience more strain when they do more housework, and have young children and childcare issues [20]. The literature supports a positive relationship between faculty family demands and FIW. Caring for children and elderly dependents increases work and family role strains among 785 faculty and staff of a public university in the western United States. [46]. Higher parental demands also relate to higher FIW among two samples of New Zealand and Malaysian faculty [47]. There is a gap in the literature concerning the relationship between family time demands and FIW among faculty members. The authors assume that time devoted to one role domain (e.g., family) cannot be devoted to another (e.g., work); therefore hypothesize that faculty members who spend more hours with their families experience more FIW.
H2: Iranian faculty members’ hours spent with family correlates positively with FIW.
WIF and job satisfaction
Although faculty members struggle with juggling professional and personal responsibilities, “the intrinsic aspects of work seem to be the critical motivator for many” [5]. Most faculty members are attracted inherently to academic work, providing a base for job satisfaction [48]. From extant literature, faculty members are generally satisfied with their jobs [48] and the flexibility academic positions offer. However, work-life imbalance increases job dissatisfaction among faculty members with less secure employment statuses [49], so they face difficulties balancing work and family responsibilities [50]. There is a dearth of knowledge regarding the impact of WFC on faculty job satisfaction. This article explores the impact of both WIF and FIW on job satisfaction. The literature supports the impact of work domain stressors on job dissatisfaction, and it is reasonable to expect stressors from a family domain to influence job satisfaction [51, 52]. The authors further hypothesize that WIF and FIW both correlate with job satisfaction.
H3: WIF and job satisfaction correlate negatively.
H4: FIW and job satisfaction correlate positively.
Work hours, time spent with family, and job satisfaction
Another line of research suggests both male and female faculty members are dissatisfied with their workloads [11]. Extensive workloads, long work hours, and high stress influence both job and family satisfaction negatively [52]. Results of a meta-analytic review demonstrate that work time demands relate, directly and indirectly through WIF, to job satisfaction [40]. Thus, we expect that longer work hours impact job satisfaction negatively.
H5: Iranian faculty work hours and job satisfaction correlate negatively.
Ford et al. [52] argue, “Stressors and demands within one role create conflict with other roles, which in turn reduce satisfaction within those other roles” (p. 59). Work stressors not only increase the possibility of job dissatisfaction, but also influence satisfaction with family and personal life; family stressors and demand increase the chances of losing job satisfaction. Ford et al. [52] found a path coefficient of –0.24 between work hours and job satisfaction, and a coefficient of 0.14 between family hours and job satisfaction. In this study, job satisfaction is a dependent variable, and it is noteworthy to examine the potential impacts of family hours on job satisfaction. Therefore, the authors not only hypothesize that longer work hours increase dissatisfaction with a job, but also examine whether longer hours spent with family relate to lower job satisfaction.
H6: Iranian faculty members’ time spent with family correlates positively with job satisfaction.
Demographic variables
Demographics such as gender, age, marital status, and parenthood are often included in WFC studies to be controlled or examined as moderators [38, 52]. Regarding gender, the literature suggests gender-roles contribute to the experience of higher WFC. “[T]hough gender roles are beginning to converge, women are still thought to assume the majority of family responsibilities despite their increasing work domain roles” [40, p. 700]. The literature also suggests that married individuals with children and who have extensive family responsibilities have the potential to grapple with what causes WFC [40]. Demographics determine the roles individuals play in their lives. Based on role-conflict theory, roles define expectations and pressures with which one must cope. For example, a woman’s roles determine expectations she must satisfy in those roles, which influence the magnitude of the relationship between time spent at or time devoted to family on WIF or FIW [40]. Faculty work-family literature studies the impact of demographics on WFC [12, 50]. A disparity exists for women faculty members concerning accomplishing work and family responsibilities [19–21, 24]. Murray, Tremaine, and Fountaine [19] suggest that lack of family responsibilities facilitates work-family management for women. The authors examine demographic moderation in our model.
H7: Demographics, including gender, marital status, having children, and spouse employment, moderate effects in the model of WFC among Iranian faculty members.
Based on the hypotheses discussed above, Fig. 1 shows the model tested in this study.
Methods
Data collection and sample
The population included faculty members at 25 public research universities in Iran (from 18 provinces). Since a comprehensive list of the population’s e-mail addresses was not available, we collected public data from the universities’ websites. In most cases, each faculty member had an online profile that included her/his email. We sent a cover letter and the link to our online survey to all retrieved emails. Three hundred ninety eight participants completed the questionnaire; the majority was male (84.6% ). Approximately 29% of the respondents worked in engineering and design departments, 27.4% in social and behavioral sciences, 26.7% in natural and formal sciences, 15.3% in health science, and 4.3% in art. Most faculty members held a PhD (82.9% ), 9.2% held an MA/S, and 0.3% held a BA. Most respondents were 30 to 45 years old (… % ), 91.1% were married, and 49.65% of participants’ spouses were employed. The majority had one (38.5% ) or no children (25.3% ). The mean age of their children was 10.38 (SD = 7.28), and 51.1% had children under 10 years old. Most were not responsible for elderly care (83.5% ).
For the missing values, a missing data pattern test was conducted. Little’s missing completely at random (MCAR) test identified the data as MCAR (p > 0.05). Full information maximum likelihood (FIML) was used to handle missing data because the method offers a practical advantage when using a statistical package for structural equation modeling (SEM) for MCAR data [53].
Measures
Work hours
The primary variable in this domain was faculty work hours. To capture a better estimate of faculty weekly work hours, we did not ask participants to report a general estimate. Instead, respondents identified how many hours they spent on the following tasks: teaching, student advising, administrative meetings, research and publication activities, review and feedback (e.g., thesis and dissertation), administration, and projects external to the university. The final report of faculty work hours was weekly.
Hours spent with family
Participants reported an average of how many hours they spent with their families daily.
Work-family conflict
Both directions of WFC (i.e., WIF and FIW) were measured using Gutek et al.’s [54] measure. The measure includes four items targeting WIF (e.g., “After work, I come home too tired to do some of the things I’d like to do.”), and four items measuring FIW (e.g., “My personal life takes up time that I’d like to spend at work.”). A five-point Likert scale ranging from strongly disagree to strongly agree was used; higher scores represent greater WFC. Researchers [55] have used these measures previously, suggesting that the Farsi version of the scales conveys the meaning intended in the English version. Cronbach’s alpha coefficients were 0.85 for WIF and 0.76 for FIW.
Job satisfaction
The Farsi translation of Smith et al.’s [56] measure to assess job satisfaction was used. The measure included five items scored on five-point Likert scale, ranging from strongly disagree to strongly agree; higher scores represent higher job satisfaction. Back-translation was used to ensure clarity. Cronbach’s alpha coefficient for this scale was 0.71.
Analysis
Since self-report data collected using a one-time questionnaire was used, common method bias might exist in the data. To test the extent of common method effects, two approaches were employed: Harman’s single-factor test and single-method-factor test [57]. All numbers of factors were identified in an exploratory factor analysis. During confirmatory factor analysis, average factor loading estimates for the common factor was 0.08, which suggests less than 1% of variance was explained by a common method factor in the measurement model. Given results from both tests, we concluded there was little influence of common method variance in the data.
Maximum likelihood estimation was used during SEM for both the measurement and structural models. In a test of the measurement model, we focused on model-fit indices and factor loading coefficients, and model-fit indices and path coefficients were the foci of a structural model test. Fit criteria suggested by Hair, Black, Babin, Anderson, and Tatham [58]: CFI >0.90; IFI >0.90; and RMSEA <0.08 were used. Standardized factor loading coefficients for CFA were expected to be significant (p < 0.05) and greater than 0.3 [59]. Multi-group path analyses were conducted to test influences of gender, marital status, having children, and spouse employment. Constrained and unconstrained models were compared for each demographic variable using the chi-square test.
Results
A confirmatory factor analysis was performed to examine whether observed data fit the measurement model. Goodness of fit indices suggested that a model with all parameters fit the data satisfactorily (CFI = 0.92; IFI = 0.93; and RMSEA = 0.072). Although the chi-square for model fitness revealed discrepancy between the sample and fitted covariance matrices (χ2 = 186.17; df = 62; p < 0.05), the test is sensitive to multivariate deviations and sample size, suggesting poor fit [59]. All standardized factor-loading coefficients were greater than 0.3. Means, standard deviations, and correlation estimates of all variables are shown in Table 1. Weekly hours spent on thefollowing tasks to calculate weekly work hours for each participant: teaching, student advising, administrative meetings, research and publications, review and feedback (e.g., thesis and dissertation), administration (if holding any), and projects external to the university were summated. The mean of weekly work hours was 37.44, and the mean of daily hours spent with family was between 1 and 2 hours and 2 and 3 hours. WIF correlated with work hours by week (r = 0.18; p < 0. 05) and job satisfaction (r = –0.24; p < 0.01), and FIW correlated with hours spent with family (r = 0.13; p < 0.05) and job satisfaction (r = –0.11; p < 0.05).
A structural model was used to test overall fit and the paths in the model. The fit indices suggested that the model explained relationships satisfactorily (χ2 = 228.72; df = 85; p < 0.05; CFI = 0.92; IFI = 0.93; RMSEA = 0.06). Standardized path coefficient (SPC) estimates are shown in Fig. 2. The SPC estimate between work hours and WIF was positive (β= 0.20; p < 0.05). The association between work hours and job satisfaction was also positive (β= 0.21; p < 0.05), and the correlation between WIF and job satisfaction was negative (β= –0.28; p < 0.05). The SPC estimate between hours spent with family and FIW was positive (β= 0.14; p < 0.05), but there was no relationship between hours spent with family and job satisfaction (β= 0.10; p > 0.05) and between FIW and job satisfaction (β= –0.08; p > 0.05). The correlation between exogenous variables (i.e., work hours and hours spent with family) was negative (β= –0.11; p < 0.05). The squared multiple correlation for job satisfaction was 0.12.
To examine whether data were influenced by gender, marital status, spouse employment, and having children, multiple-group analyses were conducted. Shown in Table 2, comparisons between constrained and unconstrained models suggested no differences among models for gender, marital status, and having children. Only for spouse employment the constrained and unconstrained models were significantly different regarding both measurement and structural weights. The small sample for female, single, and no-children groups might have influenced results. To test where disparities existed among path coefficients between the model with having a spouse employed and one without employment, pairwise parameter comparisons were used. A difference was found only between WIF and job satisfaction (z = 2.31; p < 0.05). The beta for the model with employed spouses was –0.33, and –0.09 for the non-employment model.
Discussion
This study focuses on work-family conflict (WFC) and job satisfaction among Iranian faculty members, shedding light on how allocation of time between work and family relates to work and family conflict, and how time devoted to work and time pressure from family influence job satisfaction. SEM, which allowed us to test relationships between research variables in a single model was used. The study reports interesting findings of work-family interaction of faculty members in a non-western cultural context, and contributes to literature on work and family among facultymembers.
One finding relates to work hours of faculty members. Faculty members reported how much time they spend weekly on a range of work-related responsibilities, including teaching, advising students, research and publications, administration, review and feedback. The mean for weekly work hours summated from the work activities was 37.44. The literature reports a variety of averages for faculty members working hours, and the percent of the time allocated to categories of activities. A study of academicians in 13 countries reports an average of 47.2 hours of work per week [1]. The mean of weekly work hours for our sample was lower than the average found in many western and developed countries. However, as Bentley and Kyvik [1] argue, the percent of time allocated to categories depends on country, region, and the university policies. Since faculty remuneration is low in Iran in comparison to western countries, some faculty members might need to work at a second job (e.g., teaching at multiple universities, working on projects outside the university, consulting, etc.) to fulfill financial needs. This might influence the time they are able to allocate to academic responsibilities. Another explanation, as Pal [60] discusses, is the culture of accepting long work hours. As Pal [60] argues some cultures may dictate long work hours and individuals accept it as the norm. However, some cultures, for example our study’s culture, may accept below-average work hours.
Weekly teaching hours for faculty members in Iran exceeded the hours in western research universities. Our analyses suggest that the mean for teaching among Iranian faculty members is 13 hours per week. Our participants were selected from research universities; therefore the aforementioned mean is above their expected teaching hours. One reason for extended teaching hours might be an increase in the number of Iranian students in higher education in recent years [61]. Increased numbers of students result in more teaching responsibilities for faculty members, and probably more strain. Extreme cases where faculty members taught over 16 hours per week (16 participants) were observed. Future research should explore reasons Iranian faculty members differ in their use of time, and analyze their working time patterns.
Results regarding the first hypothesis suggest that consistent with the literature, a positive relationship exists between hours of work and WIF. Testing the second hypothesis, we found that faculty members who spend longer hours at family responsibilities such as taking care of children and household chores experience higher family interference with work. This finding corresponds with literature [40] from western countries. The mean for FIW was 1.86 hours per day, which isconsistent with extant studies [19] that report faculty family life does not burden work. Most of the participants were men (84.6 % ), who are regarded as the main breadwinners of families in Iran. 49.1% of participant spouses were unemployed, which suggests they had someone to assist with family responsibilities.
Testing the third and fourth hypotheses, positive relationship between WIF and job satisfaction, and no relationship between FIW and job satisfaction were found. Participants who suffered from higher WIF were less satisfied with their jobs, suggesting family-friendly programs that decrease WFI have a positive influence on employees, making them more satisfied with their jobs. Results also suggest that WIF influences job satisfaction, similar to faculty members in other countries.
Contrary to extant literature, results from the fifth hypothesis suggest Iranian faculty work hours correlate positively with job satisfaction. The literature suggests faculty members are moderately satisfied with most of their job features, but dissatisfied with work hours [43]. One reason might be that working more hours offers favorable outcomes such as increased productivity, including publications and more income. Major et al. [39] suggest that people who have strong career identities work more and might garner more satisfaction from work. Future research should explore why longer work hours among faculty members correlate with more overall job satisfaction.
Results from the sixth hypothesis suggest no relationship between time spent with family and job satisfaction. Ford et al. [51] suggest that among work hours, job stress, job involvement, and work support, job stress has the strongest effects on WFC and cross-domain satisfaction. Future research should consider the effects of stress from the family domain on faculty job satisfaction.
Our model tests the moderating effect of gender, marital status, having children, and spousal employment. Results show that having an employed spouse moderates the relationship between WFI and job satisfaction. The relationship between WIF and job satisfaction was stronger for faulty members who had an employed spouse. This might be explained by the cross-domain model Ford et al. [51] propose. They proposed that factors in the family domain influence satisfaction in the work domain. In our study, 49.6% of participant spouses were employed. It appears that an unemployed spouse lessens the strain from the family domain, which lessens the stress a person experiences at work and leads to higher work satisfaction.
Implications
This study has several implications for both human resources and organizational development professionals working with universities and faculty members in Iran and countries similar to Iran. Human resources professionals can use these findings to provide faculty members with training and development that helps them manage work and family responsibilities. One example is improving time-management skills, which allow them to use time effectively. Organizational development professionals can suggest interventions that mediate teaching responsibilities of Iranian faculty members to allow for more time spent on research. Administrators should reconsider compensation because low salaries oblige faculty members to seek external jobs. This article demonstrates that having a working spouse influences WFC. Those in managerial positions at universities should create family-friendly policies, focusing especially on dual-earning families, to help their employees integrate their work and non-work responsibilities.
Limitations
This study contributes to understanding of WFC in a non-western cultural context. However, some limitations need to be mentioned. The first relates to data collection. Our data were collected at one point of time; therefore inferences regarding causality should be made with caution. Second, we used a general measure of job satisfaction. We suggest future researchers to use scales measuring satisfaction with work-family support, satisfaction with work hours, or global job satisfaction to test results regarding relationships between WFC and job satisfaction. Although little influence of common method bias in the data was identified using statistics, this does not mean data were free from such biases. Relying on a single source can result in method biases that threaten validity of results; data from a single, self-report questionnaire always risk common method biases [62]. Therefore, future research should include alternative sources such as documents regarding work hours, multi-faceted assessments (e.g., supervisors, co-workers, and family members), and repeated measures in a longitudinal study to improve data validity.