The impact of stereotypes and supervisor perceptions of employee work–family conflict on job performance ratings

Abstract

We draw on gender role theory to examine the relationships among employee-rated work–family conflict, supervisor perceptions of employee work–family conflict, employee gender and supervisor-rated job performance. We found that the relationship between employee-rated work–family conflict and supervisor perceptions of employee conflict varied based on both employee gender and the direction of conflict under consideration. Specifically, the relationship between the two rating sources (employee and supervisor) was stronger for male employees when conflict was considered. However, the relationship between the two rating sources was stronger for female employees when family-to-work conflict was considered. Supervisor perceptions of employee work–family conflict were negatively related to employee job performance ratings. More generally, we found support for a moderated mediation model such that the relationship between employee-rated work–family conflict and job performance was mediated by supervisor perceptions of employee work–family conflict, and the effect was moderated by employee gender. Implications for research and practice are discussed.

Keywords

gender job performance stereotypes work–family conflict

In recent decades, the topic of work–family conflict has become a familiar theme in scholarly publications and popular press. Greenhaus and Beutell (1985: 77) stated that work–family conflict occurs when ‘participation in the work (family) role is made more difficult by virtue of participation in the family (work) role.’ Past research has distinguished between two directions of work–family conflict based on where the conflict is initiated and where it is received (Frone et al., 1992; Gutek et al., 1991): work-to-family conflict (WFC) and family-to-work conflict (FWC). WFC refers to demands in the work domain making it difficult for individuals to meet their responsibilities at home. FWC refers to demands in the family domain making it difficult for individuals to meet their responsibilities at work.¹

Both empirical studies and meta-analytic reviews have uncovered detrimental effects associated with both directions of work–family conflict, although it’s worth noting that most of these findings rely on self-reported variables – such as job satisfaction, organizational commitment, self-rated job performance, or turnover intention – as outcomes of work–family conflict (see Casper et al., 2007, for a critical review of this literature). These findings are problematic for two reasons. First, the relationship between work–family conflict and self-reported outcome variables may potentially be inflated owing to common method variance. Second, to make a stronger business case for organizational resource investment into creating a family-friendly work environment, empirical research needs to demonstrate a potent link between employee work–family conflict and organizationally-valued outcomes such as supervisor-rated employee performance (Wang et al., 2013).

In response to these criticisms, recent research has started to correlate employee work–family conflict with supervisor-rated performance. However, a perusal of the literature has uncovered a number of limitations. First, theoretically it is unclear whether WFC, FWC, or both, should be related to the performance variable. The cross-domain theory of work–family conflict posits that since work is the receiving domain of conflict, FWC, rather than WFC, should be related to employee work performance (Frone et al., 1992). In contrast, the matching-domain hypothesis posits that since conflict expresses its primary effects on the domain from which the conflict originates, WFC, rather than FWC, should be related to employee work performance (Amstad et al., 2011). Empirical support for both hypotheses is far from conclusive. For example, lending empirical support to the cross-domain hypothesis, Li et al. (2013) found that FWC was related to employee work performance but WFC was not (see similar results reported by Witt and Carlson, 2006). In contrast, Muse and Pichler (2011) offered support to the matching-domain hypothesis by showing that it was WFC rather than FWC that was significantly related to performance. In other studies, neither WFC nor FWC was significantly related to supervisor-rated work performance (Odle-Dusseau et al., 2012; Wang et al., 2013). These conflicting findings lead to the somewhat premature conclusion that ‘there is limited evidence today that work–family conflict is a robust predictor of job performance’ (Muse and Pichler, 2011: 664).

Second, the magnitude of the relationships between work–family conflict and employee performance rated by supervisor varies across studies. A recent meta-analysis (Hoobler et al., 2010) shows that although both FWC and WFC were significantly and negatively related to manager-rated performance, there was considerable variance in these relationships that was not accounted for (as evidenced by the significant Q_w). As such, these researchers argue that ‘an interesting next step for researchers will be to explore the boundary conditions of our models of the impact of WFC and FWC’ (Hoobler et al., 2010: 492; see Allen et al., 2000; Kossek and Ozeki, 1999, for other meta-analytic mixed findings).

Third, more research is needed to understand the process through which employee work–family conflict is related to work performance as rated by their supervisor. Past research tends to draw on the stress/strain/resource availability frameworks to account for this relationship, arguing that conflict causes stress and reduces the supply of resources thereby resulting in lower performance (Muse and Pichler, 2011; Shih et al., 2010). However, the mediator has rarely been examined empirically. In the only study we are aware of, Netemeyer et al. (2005) found that although the relationship between WFC and supervisor-rated performance was mediated through job stress, the relationship between FWC and supervisor-rated performance was not. The understanding of the pathway through which employee work–family conflict is related to their work performance as rated by their supervisor has important implications. If employee strain accounts for this relationship then organizational interventions need only target the employee by increasing their abilities to cope with work–family conflict so that the effects of conflict on performance can be minimized. However, if the relationship cannot be fully explained by the employee’s stress reactions, then organizational interventions targeting the employee may not achieve this desired outcome.

Taken together, our understanding of the relationship between employee work–family conflict and supervisor-rated employee job performance is incomplete based on available research. Contrary to contemporary thinking that suggests that conflict reduces the supply of resources thereby leading to lower supervisor-rated employee performance (Hoobler et al., 2009), we argue for a different theoretical mechanism. In particular, gender role theory (Eagly, 1987) predicts that supervisors’ stereotypes bias their perceptions of the work–family conflict that their male and female employees experience. These biases impact men and women differently, depending upon the direction of the work–family conflict. It is important to note that we are not suggesting that conflict, per se, lowers or raises performance. Rather, we posit that gender stereotypes impact supervisor perceptions of employee work–family conflict and their resulting job performance ratings.

Our framework is depicted in Figure 1. Working from right to left, we propose that supervisors tend to rate their subordinates’ job performance lower when they believe that these subordinates are experiencing work–family conflict. That is, because supervisors believe that conflicted employees are less effective or motivated at work, they assign these employees lower performance ratings compared with the ratings assigned to those perceived to be less influenced by conflict. We then identify employees’ own experience of work–family conflict as the source of supervisor perceptions of employee conflict, but argue that the particulars of these effects vary for men and women. As we explain below, men are viewed as being more likely to experience WFC. Thus, male employees and their supervisor are more likely to agree on the level of WFC that the employees experience. Conversely, women are viewed as being more likely to experience FWC. Thus, female employees and their supervisor are more likely to agree on the level of FWC that the employees experience. In all, we examine a moderated mediation model such that supervisor perceptions of employee work–family conflict will bridge the relationship between employee-rated work–family conflict and supervisor-rated job performance, and this mediating relationship will be moderated by employee gender.

Figure 1.

Hypothesized model.

We attempt to make the following contributions to the literature. First, by drawing on gender role theory (Eagly, 1987), we show that a missing link in past research on the relationship between employee work–family conflict and supervisor-rated employee performance is the supervisor’s perceptions of employee work–family conflict. Our model posits that the extent to which the supervisor perceives their employees’ work–family conflict is influenced at least in part by the conflict that these employees experience and such perceptions have important downstream effects on their evaluation of employee performance. This view suggests that when the role of supervisor perceptions of employee work–family conflict is considered, both directions of employee work–family conflict can relate to supervisor-rated performance (Amstad et al., 2011; Ford et al., 2007; Frone et al., 1992; Shockley and Singla, 2011). However, we qualify these theoretical arguments by identifying employee gender as a moderating mechanism, showing that WFC impacts male employees’ performance whereas FWC impacts female employees’ performance. Therefore, we not only add more clarity to the conceptual debate on which direction of conflict should be related to employee performance but also explain the situation under which the relationship is more likely to emerge.

Second, our model also answers the call for more research into why employee work–family conflict is related to their supervisor-rated performance (Witt and Carlson, 2006). Past research tends to use an employee-centric approach to explain this relationship, arguing that employee conflict impairs their performance owing to increased employee stress and depleted resource supplies (Li et al., 2013; Shih et al., 2010). This employee-centric perspective implies that the conflict-performance link can be deactivated by ‘fixing’ the employee (reduce their stress reaction as a result of experienced work–family conflict). We add more nuance to our understanding of this relationship by offering a supervisor-centric perspective. This perspective implies that the conflict-performance link can be alternatively deactivated by ‘fixing’ the supervisor (reduce their stereotype in order to change the way they evaluate those employees perceived to experience conflict). In addition, past research has suggested that supervisor perceptions of employee FWC may explain why female employees are disadvantaged in the workplace (Hoobler et al., 2009). Drawing on gender role theory, we argue that both male and female employees may be disadvantaged as a result of their supervisor’s employee work–family conflict perceptions that are driven by gender role-congruent stereotypes.

Third, several recent studies have examined the important role of supervisor perceptions of employee work–family conflict (Carlson et al., 2008; Hoobler et al., 2009), showing that employees’ FWC was related to their supervisor’s perceptions of employee FWC and that these perceptions had important career implications for these workers. While these investigations are informative, they are also incomplete, leaving out the other direction of work–family conflict. This oversight is problematic because to the extent that employee WFC and FWC are two distinct constructs, it follows that supervisor perceptions of employee WFC may also be distinct from their perceptions of employee FWC in their relationships with antecedents and outcome variables. In addition, the correlations between supervisors’ perceptions of employee FWC and employee self-rated FWC as reported in these studies were rather modest, suggesting the need to understand why supervisors and their employees sometimes converge on conflict perceptions and why sometimes such perceptions diverge. We plan to advance our understanding of this construct in two important ways. First, we fill the gap in the literature by including supervisor perceptions of both employee WFC and FWC and examining their antecedents and effects. Second, we also answer the question of when employees and their supervisors agree on the level of work–family conflict that the employees experience by identifying employee gender as an important boundary condition.

Supervisor perceptions of employee work–family conflict and job performance ratings

Consistent with previous research, we define the term ‘family’ broadly, considering it ‘a larger, nonwork entity that includes responsibilities towards spouses, children, nonmarried partners, and home life in general’ (Netemeyer et al., 2005: 130). Work refers to the engagement of physical and mental activities in order to meet one’s responsibilities in a paid/unpaid job. Most extant work–family conflict studies rely on employees, or to a much lesser extent their spouses (e.g. Grandey et al., 2005), to provide ratings of employee work–family conflict. However, recent research has suggested that since supervisors are responsible for evaluating their employees’ job performance, the extent to which they can accurately judge the level of work–family conflict that their employees experience becomes important. That is, supervisors’ evaluations of employee work–family conflict may influence subsequent performance ratings and other assessments with important career implications, such as promotability and job-person fit (Carlson et al., 2008; Hoobler et al., 2009).

This prediction is based on the notion of ‘work devotion schema’ (Blair-Loy, 2003), which represents a cultural assumption that work should be prioritized over other life pursuits. According to this schema, work ‘deserves undivided and intensive allegiance’ and should be ‘the central focus of one’s life’ (Williams et al., 2013: 211). Individuals who fail to meet the requirements of the ‘work devotion schema’ may risk negative performance consequences (Hoobler et al., 2009; Williams, 2001) whereas those that embody the schema are viewed as ‘ideal workers’ who are given favorable jobs and opportunities (Blair-Loy, 2004). This schema has become widely institutionalized as many organizations are structured based on the unlimited availability assumption of their employees and deeply ingrained into the personnel management process (Vandello et al., 2013). Managers, who are organizations’ institutional gatekeepers, may evaluate their employees based on whether these employees uphold the schema. There is some preliminary evidence in support of this argument. Past research has shown that employees who ask for more flexibility at work – presumably to reconcile competing demands from work and family – receive lower performance ratings (Cohen and Single, 2001). Therefore, managers who believe that their employees experience work–family conflict may question these employees’ abilities to live up to the ‘work devotion schema,’ thereby leading them to assign a lower performance rating to these employees.

Drawing from the cultural ‘work devotion schema,’ managers may presume that both WFC and FWC represent potential distractions from employees’ abilities to live up to the schema that requires ‘single-minded, emotional intensity’ (Blair-Loy, 2004: 293) and ‘long hours, hard work, emotional involvement, financial investment, and restrictions on geographical freedom’ (Blair-Loy, 2004: 294). As such, when managers perceive that their employees experience a high level of work–family conflict (WFC or FWC), they may come to believe that these employees are preoccupied by matters unrelated to work thereby hindering their abilities to effectively meet work requirements. Supervisors may also believe that these employees’ job performance levels may sag as a result of coping with work–family conflict (Witt and Carlson, 2006). Indeed, Carlson et al. (2008) and Hoobler et al. (2009) both found negative relationships between supervisor perceptions of employee FWC and employee job performance. Neither of these studies, however, included WFC. Although WFC represents a situation under which work demands take precedence over family demands, supervisors may believe that the WFC that their employees are perceived to experience may lead to stress that rebounds back to the work role and reduces their performance. Past research suggests that individuals blame work for causing WFC (Amstad et al., 2011), which can be viewed by the supervisor as a signal that the employee is not living up to the ‘work devotion schema.’ These arguments lead us to make the following predictions:

Hypothesis 1a: Employee WFC as rated by supervisor is negatively associated with employee performance as rated by supervisor.

Hypothesis 1b: Employee FWC as rated by supervisor is negatively associated with employee performance as rated by supervisor.

Gender stereotypes and supervisor perceptions of work–family conflict

Supervisor perceptions of their employee’s display of WFC and FWC can be taken as an indication that the subordinate is unable to manage work and home responsibilities. This could be the case even if the worker is in fact performing adequately. If this is so, then it is incumbent upon our model to explain where supervisors derive their judgments of employee work–family conflict. We propose a two-step process, whereby the employee initially experiences some sort of conflict and, subsequently, this conflict comes to the attention of the supervisor. When the conflict is recognized by the supervisor it will harm job performance ratings. Put differently, supervisory performance ratings will be lower when the employee and the manager agree that some type of conflict is present. If the supervisor does not recognize the conflict that the employee experiences, then the conflict will not impact supervisory performance ratings.

Given this consideration, our model suggests that supervisors and subordinates may sometimes (but not always) agree as to the amount of conflict the employee is experiencing. To the best of our knowledge, two published studies have included both supervisor ratings and employee self-ratings of employee work–family conflict (see Carlson et al., 2008; Hoobler et al., 2009). Both studies reported relatively weak correlations between the two sources. This is loosely consistent, or at least not inconsistent, with our present view. That is, the reported weak correlations raise the question of when these ratings would converge with each other. To address this empirical and conceptual gap, we draw on gender role theory (Eagly, 1987; Eagly et al., 2000) to explain when the two sources should converge and when they should not.

Gender role theory

Gender role theory posits that men and women traditionally occupy different social roles (Eagly, 1987; Eagly et al., 2000). Men are typically assigned the responsibility of being the primary wage earner, whereas women are typically assigned the responsibility of managing domestic activities. Closely associated with the gender division of social roles is the existence of gender stereotypes (Heilman, 2012). Men are stereotyped to possess agentic qualities such as assertiveness, competitiveness and independence. These attributes are often associated with career-promoting behaviors at work, such as spending more time on work responsibilities at the expense of home demands. Conversely, women are stereotyped to possess communal qualities such as warmth, kindness and cooperativeness (Burgess and Borgida, 1999; Heilman, 2001). These attributes are often associated with family-oriented behaviors, such as spending time on family responsibilities and household chores.

To understand how gender stereotypes may influence the extent to which supervisor and employee agree on the amount of work–family conflict that the employee experiences, it is instructive to first examine how stereotypes influence information processing (Devine, 1995; Heilman, 2012; Rudman and Glick, 2008). Past research has suggested that stereotypes exist as a schema defined as a ‘cognitive structure that represents organized knowledge about a given concept or type of stimulus’ (Fiske and Taylor, 1984: 140). Schema biases information processing in a way that favors the confirmation of pre-existing belief. Specifically, three mechanisms have been proposed to explain how this process operates: attention, interpretation and information recall.

The first mechanism is attention. Ruble and Stangor (1986: 228) argued that ‘schema helps determine what information will be attended to and what information will be ignored.’ Individuals tend to direct their attention to information that is consistent with their stereotypes (Bodenhausen, 1988; Fiske and Neuberg, 1990; Johnson and Judd, 1983; Macrae et al., 1994). Since individuals selectively pay attention to information through the lens shaped by gender stereotypes, supervisors may pay attention to different types of behaviors exhibited by their male and female employees. Specifically, gender stereotype would direct the supervisor to notice when male employees engage in behaviors consistent with a breadwinner role, such as coming to work early or staying late (Eagly and Karau, 2002). In contrast, gender stereotype would direct the supervisor to notice when female employees engage in behaviors consistent with a homemaker role, such as taking time off from work to take the child to the doctor’s office or stepping out from a meeting for a personal phone call (Allen and Russell, 1999; Butler and Skattebo, 2004; Rudman and Mescher, 2013). These arguments suggest that when a male employee experiences WFC, his supervisor is more likely to be aware of it. Conversely, his supervisor may be less likely to be aware of the male employee’s FWC owing to less attentional resources being devoted to it. For similar reasons, a supervisor may be more likely to become aware of a female employee’s FWC but less aware of her WFC.

The second mechanism is interpretation. Interpretation of information is guided by stereotypes, which create an initial expectation. Information that is inconsistent with the stereotype is more likely to be explained away so that the stereotype can be preserved. Given this, supervisors are likely to interpret information in a manner consistent with their gender stereotype. When male employees make personal sacrifices for their work (WFC), such behaviors make sense to their supervisor because they are consistent with the male gender role. When male employees’ work is influenced by their family obligations (FWC), however, such behaviors are more difficult to interpret because they contradict the male gender stereotype. Similarly, supervisors are more likely to interpret female employees’ efforts to meet family obligations (FWC) as a confirmation of the female stereotype, whereas information suggesting that female employees struggle to meet family requirements owing to work (WFC) does not converge with the stereotypic association between women and the domestic role (Eagly and Mladinic, 1994)

The final mechanism is information recall. Fyock and Stangor (1994: 332) stated that ‘people are more likely to remember information that supports their stereotype than to remember that which disconfirms them. Such a “consistency bias” has frequently been discussed as a mechanism by which stereotypes are maintained.’ Supervisors may tend to recall instances when male employees’ work compromises their availability at home (WFC). For example, if a male worker mentions missing a daughter’s athletic event, then this conversation will likely be stored in memory. Conversely, when female employees’ family needs spill into their work domain (FWC), the information is also likely to be remembered. For example, the instance when a female employee misses work owing to the sickness of her child may be more likely to be remembered by her supervisor than when she misses a family trip owing to a looming deadline at work. Male employees’ FWC and female employees’ WFC do not conform to expectations resulting from gender stereotypes, making them more difficult to recall by supervisors.

It is important to note that past research has shown that individuals may sometimes show memory bias favoring stereotype-inconsistent information. Stangor and McMillan (1992: 42) argued that ‘because behavioral instances that disconfirm one’s expectations are particularly salient, they may be processed more extensively or in a more detailed manner in comparison with expectancy-congruent material.’ Therefore, supervisors may also attend to instances of male employees’ FWC and female employees’ WFC, both of which are inconsistent with gender stereotypes. There are two reasons to question the validity of this argument.

First, although past research has shown that perceivers may preferentially notice or recall information inconsistent with their stereotypes, these findings are more likely to be observed when the stereotypes are relatively weak or experimentally manufactured (Fyock and Stangor, 1994). Gender, in contrast, is a much more robust and flexible category of which people tend to develop strong a priori expectations (Fiske and Neuberg, 1990). Ruble and Stangor (1986: 254) stated that ‘one of the most striking effects in the gender literature is the overwhelming indication of better memory for gender-consistent versus inconsistent information.’ Since male employees’ FWC and female employees’ WFC are inconsistent with their gender stereotype, they are less likely to be noticed or recalled, thereby reducing their correlation with supervisor perceptions of such conflicts.

Second, gender stereotypes are not only descriptive but also prescriptive. Individuals who fail to live up to their gender stereotype (such as when men exhibit FWC or when women exhibit WFC) may face backlash for not acting like their type. For example, past research has shown that men who use flexible work arrangements are viewed as less likely to succeed relative to those who do not (Almer et al., 2004). Rudman et al. (2012: 212) stated that ‘when typical actors fear backlash, [actors] may strategically hide their deviance to avoid social rejection.’ These arguments suggest that male (female) employees may ‘closet’ their FWC (WFC) from their supervisor making it difficult for the supervisor to correctly discern the level of conflict these employees experience. As such, the relationship between FWC (WFC) and supervisor-perceived employee FWC (WFC) will be depressed for male (female) employees owing to the supervisor’s lack of access to conflict information hidden from them by their employees.

Taken together, considering the effects of gender stereotypes on individuals’ information processing suggests that supervisors will have more accurate information when male employees are experiencing WFC and when female employees are experiencing FWC, but less accurate information when male employees are experiencing FWC and when female employees are experiencing WFC. Based on these arguments, we predict that supervisors will be more attentive to, and subsequently more aware of, subordinates’ work–family conflict when it is gender-congruent according to traditional gender roles:

Hypothesis 2a: Gender moderates the relationship between employee-rated and supervisor-rated employee WFC, such that the relationship is stronger for male employees and weaker for female employees.

Hypothesis 2b: Gender moderates the relationship between employee-rated and supervisor-rated employee FWC, such that the relationship is stronger for female employees and weaker for male employees.

On the whole, our model suggests that supervisors are apt to identify the direction of work–family conflict that is consistent with traditional gender roles: WFC for male employees and FWC for female employees. Once the conflict is detected, supervisory ratings of performance are lowered (see Hypotheses 1a–1b). Therefore, we propose a model in which supervisor perceptions of employee WFC/FWC mediate the relationship between employee-rated WFC/FWC and supervisor-rated performance. As discussed, these effects are conditional on gender as a moderator. These processes lead to a first-stage model of conditional indirect effects (Preacher et al., 2007). As such, we predict:

Hypothesis 3a: Gender moderates the indirect effects of employee-rated WFC on supervisor-rated performance through supervisor-rated employee WFC, such that these effects are more pronounced for male employees as opposed to female employees.

Hypothesis 3b: Gender moderates the indirect effects of employee-rated FWC on supervisor-rated performance through supervisor-rated employee FWC, such that these effects are more pronounced for female employees as opposed to male employees.

Pilot study and present investigation

As described above and in Figure 1, our theoretical model posits that men and women will receive different performance ratings because, owing to gender stereotypes, supervisors are more likely to believe that WFC is typically associated with men and FWC is typically associated with women. However, we were concerned that having supervisors rate both male and female employees’ work–family conflict while assessing their gender stereotypes could introduce demand characteristics into the main study. To address this problem, we independently investigated stereotyping. Specifically, to test whether gender stereotype may influence perceived work–family conflict as a function of employee gender, we conducted a pilot study using a sample of working adults recruited from Amazon Mechanical Turk (N = 122. 52% were men, 54% were married/with a partner, and 78% were employed at the time of the study). Participants were randomly assigned to one of two experimental conditions where they would judge whether various statements representing either WFC or FWC were more descriptive of men or women. Consistent with our expectation, participants who read the WFC statements were more likely to believe that these statements described a man, whereas those who read the FWC statements were more likely to believe that these statements described a woman. These findings support our argument that people’s perceptions of employee work–family conflict are in part driven by their stereotype. Details of the pilot study can be obtained from the first author upon request. Given these supportive results, we then turned to the main study, which is described next.

Method

Participants were executive MBA students who were employed full time at the time of the study. This study was incorporated into a career/personal developmental program to provide students with feedback pertaining to their strengths and weaknesses in many aspects of their work and life. Participants were instructed to complete a survey themselves and then forward an email to their supervisor who would complete a separate survey by clicking a link embedded in the email. At the end of the semester, participants received aggregate feedback based on the information gathered from the self and the supervisor surveys.

All of the students in the program (N = 119) participated in the study. In addition, 113 supervisors completed the supervisory survey, resulting in a matched sample of 113 dyads (95%). The average age of the employees was 33.73 (SD = 6.93) and 50% of them were women. These participants held a wide variety of positions (such as account manager, project manager, regional operational director, insurance operations manager) across many different industries (such as healthcare, finance, entertainment, transportation). They worked on average 43.87 (SD = 12.20) hours per week. Their average organizational tenure was 5.18 years (SD = 4.12). The average age of the supervisors was 45.35 (SD = 10.97) and 41% of them were women. The supervisors typically held senior positions such as human resources director, plant manager and sales manager.

Measures

Gender

Employees indicate their gender as either male (1) or female (0).

All the measures described below were rated on a five-point scale ranging from 1 (strongly disagree) to 5 (strongly agree).

Employee-rated work-to-family conflict

Time-based WFC was measured with a three-item scale developed by Carlson et al. (2000). A sample item of the measure is ‘My work keeps me from my family activities more than I would like.’

Employee-rated family-to-work conflict

Time-based FWC was also measured with a three-item scale by Carlson et al. (2000). A sample item of the measure is ‘The time I spend on family responsibilities often interferes with my work responsibilities.’

Supervisor perceptions of employee work-to-family conflict

We used the same three-item scale for supervisor ratings as the one used for measuring employee-rated WFC (Carlson et al., 2000), except the focus of the items was changed from the self to the employee. For example, an item of the measure is ‘This employee’s work keeps him/her from family activities more than he/she would like.’

Supervisor perceptions of employee family-to-work conflict

We used the same three-item scale for measuring supervisor ratings as we did for employee-rated FWC (Carlson et al., 2000), except that the focus of the items was changed from the self to the employee. For example, an item of the measure is ‘The time this employee spends on family responsibilities often interferes with his/her work responsibilities.’

Supervisor-rated job performance

Performance was measured with a seven-item scale developed by Williams and Anderson (1991). A sample item of the measure was ‘This employee adequately completes assigned duties.’

Control variables

We controlled for supervisor’s gender (male = 1; female = 0) as past research has shown that supervisor’s gender may impact the amount of family support provided to their employees (Li and Bagger, 2011). We also controlled for the number of hours that employees worked every week and the employment status of their spouse (spouse works = 1; spouse doesn’t work = 0) since both of these variables may impact the conflict that employees experience between work and family.

Results

Means, standard deviations and inter-correlations of study variables are reported in Table 1.

Table 1.

Means, standard deviations, reliability coefficients and inter-correlations.

Variable	M	SD	1	2	3	4	5	6	7	8	9
1. Supervisor gender	.59	.49
2. Employee work hours	43.87	12.20	.16
3. Spouse employment status	.80	.40	−.17	−.05
4. Employee gender	.50	.50	.14	.18	−.27**
5. Employee-rated WFC	2.87	1.05	.08	.57**	−.14	.27**	(.86)
6. Employee-rated FWC	2.10	.71	−.02	.11	.12	−.03	.30**	(.71)
7. Supervisor-rated WFC	2.23	.85	.20*	.26**	.00	.11	.40**	.16	(.90)
8. Supervisor-rated FWC	1.59	.63	.21*	.08	.06	.06	.21*	.19*	.39**	(.87)
9. Supervisor-rated job performance	4.59	.41	.02	.02	.05	−.14	−.10	−.03	−.22*	−.46**	(.76)

n = 113. M = Mean, SD = standard deviation. Gender: men = 1, women = 0. Spouse employment status: 1 = Yes, 0 = No.

WFC = work-to-family conflict, FWC = family-to-work conflict. Coefficient alpha reliabilities are reported in parentheses on the diagonal of the matrix.

p < .05.

p < .01.

Hypotheses 1a–1b: Main effects of supervisor-rated work–family conflict and family-to-work conflict on job performance

Hypothesis 1 predicted that supervisor perceptions of employee WFC (1a) and FWC (1b) would be related to their performance ratings of their employee. We used hierarchical regression analyses to test this hypothesis and standardized the variables prior to entering them into the equation. Results of these analyses are presented in Table 2. Both supervisor perceptions of employee WFC (β = –.25, p < .05) and FWC (β = –.48, p < .01) were significant predictors of supervisor-rated job performance after the control variables were accounted for. Therefore, Hypotheses 1a–1b were supported.

Table 2.

Regression analysis: Supervisor perceptions of employee work-to-family conflict/family-to-work conflict on supervisor-rated performance.

Variable	Model 1	Model 2	Model 1	Model 2
	β	β	β	β
Supervisor gender	.04	.08	.04	.14
Employee work hours	−.01	.05	−.01	.03
Spouse employment status	.04	.05	.04	.09
Supervisor perceptions of employee WFC		−.25*
Supervisor perceptions of employee FWC				−.48**
R ²	.00	.06	.00	.23
Delta R²		.06		.23

n = 113. WFC = work-to-family conflict, FWC = family-to-work conflict.

p < .05.

p < .01.

Hypotheses 2a–2b: Moderating effects of employee gender

Hypothesis 2a stated that gender would moderate the relationship between employee-rated WFC and supervisor perceptions of employee WFC, such that the relationship would be stronger for male employees and weaker for female employees. We tested this prediction with moderated regression analysis, standardizing the variables before their entry into the equation. As shown in Table 3, the interaction was significant (β = .28, p < .05). We also plotted the interaction to understand the interactive effects for males and females separately. As shown in Figure 2, the correlation was stronger for male employees (β = .52, p < .05) than for female employees (β = .18, NS). These results support Hypothesis 2a. Hypothesis 2b stated that gender would moderate the relationship between employee-rated FWC and supervisor perceptions of employee FWC, such that the relationship would be stronger for female employees and weaker for male employees. As shown in Table 4, the interaction was significant (β = –.26, p < .05). The correlation was weaker for male employees (β = –.02, NS) than for female employees (β = .22, p < .05, see Figure 3). These results support Hypothesis 2b.

Table 3.

Regression analysis: Interactive effects of employee-rated work-to-family conflict and gender on supervisor perceptions of employee work-to-family conflict.

Variable	Model 1	Model 2	Model 3
	β	β	β
Supervisor gender	.17	.18	.16
Employee work hours	.23*	.01	−.06
Spouse employment status	.04	.08	.09
Employee-rated WFC		.39**	.21
Employee gender		.00	.02
Employee-rated WFC × Employee gender			.28*
R ²	.09	.19	.22
Delta R²		.10	.03

n = 113. WFC = work-to-family conflict.

p < .05.

p < .01.

Figure 2.

Interaction of employee-rated work-to-family conflict and employee gender on supervisor perceptions of employee work-to-family conflict.

Table 4.

Regression analysis: Interactive effects of employee-rated family-to-work conflict and gender on supervisor perceptions of employee family-to-work conflict.

Variable	Model 1	Model 2	Model 3
	β	β	β
Supervisor gender	.20*	.20*	.20*
Employee work hours	.07	.03	.05
Spouse employment status	.10	.10	.12
Employee-rated FWC		.18	.36*
Employee gender		.09	.09
Employee-rated FWC × Employee gender			−.26*
R²	.05	.09	.13
Delta R²		.04	.04

n = 113. FWC = family-to-work conflict.

p < .05.

p < .01.

Figure 3.

Interaction of employee-rated family-to-work conflict and employee gender on supervisor perceptions of employee family-to-work conflict.

Hypotheses 3a–3b: Mediational effects of supervisor-rated work–family conflict and family-to-work conflict

Hypothesis 3a predicted that gender would moderate the mediating model of employee WFC, supervisor perceptions of employee WFC and supervisor-rated job performance. We examined these conditional indirect effects using the PROCESS macro for SPSS created by Hayes (2013). This macro uses a bootstrapping procedure to estimate the conditional indirect effects of an independent variable (employee WFC) through a mediator (supervisor perceptions of employee WFC) on a dependent variable (job performance) at various levels of a moderator (in this case, employee gender). In addition, the macro generates bias-corrected confidence intervals around those effect estimates. Results of these analyses are presented in Table 5. Consistent with our prediction, the indirect effect of employee-rated WFC was significant for male employees (β = –.06, p < .05), but not significant for female employees (β = –.02, NS). These results suggest support for Hypothesis 3a.

Table 5.

Analysis of conditional indirect effects of employee work-to-family conflict on performance.

Employee gender	Indirect effect estimate	95% confidence interval
		Lower	Upper
Male	–.06	–.12	–.01
Female	−.02	−.07	.01

Similarly, Hypothesis 3b predicted that gender would moderate the mediating model of employee FWC, supervisor perceptions of employee FWC and supervisor-rated job performance. We tested this prediction using the same strategy as employed for Hypothesis 3a. The results are shown in Table 6. Consistent with our prediction, the indirect effect of employee-rated FWC was not significant for male employees (β = .01, NS), but significant for female employees (β = –.07, p < .05). These results suggest support for Hypothesis 3b.

Table 6.

Analysis of conditional indirect effects of employee family-to-work conflict on performance.

Employee gender	Indirect effect estimate	95% confidence interval
		Lower	Upper
Male	.01	−.04	.06
Female	–.07	–.15	–.02

Auxiliary analyses

We conducted a series of auxiliary analyses to test a number of alternative hypotheses. First, past research has suggested that supervisor and employee gender similarity may impact how the employee is treated by the supervisor. In particular, supervisors may treat same-gender employees better than they treat different-gender employees (Li and Bagger, 2011). Though not predicted here, it is possible that supervisor–employee gender similarity moderates the relationship between employee work–family conflict and supervisor perceptions of employee work–family conflict. To test this hypothesis, we created a dichotomized variable (1 = same gender; 0 = different gender) to represent gender similarity. We then entered the control variables first (employee work hours and spouse’s work status) into the regression equation, followed by the main effects (WFC/FWC and gender similarity) and the interaction term. The interactions between employee WFC (β = .01, NS)/FWC (β = –.25, NS) and gender similarity in the prediction of supervisor conflict perceptions were not significant, suggesting that the effects observed in our study were not masked by supervisor–employee gender similarity.

Second, since supervisor perceptions of employee conflict and performance rating were measured at the same time, it is possible that the order of our model may be reversed such that employees who experience WFC/FWC may receive lower performance ratings from their supervisor who attributes the lower performance to the conflict that these employees experience. We tested this competing model that places supervisor-rated performance as the mediator and supervisor perceptions of employee WFC/FWC as the dependent variable of the model. Our supplementary analyses indicated that the indirect effects of employee WFC/FWC on supervisor perceptions of employee WFC (Male: β = .04, NS; Female: β = .01, NS)/FWC (Male: β = –.02, NS; Female: β = .04, NS), via supervisor-rated performance as a mediator, were not significant for either male or female employees.

Third, we also tested the possibility that supervisor perceptions of employee conflict may be driven by their performance ratings in a process that is gender specific. This possibility reverses our theoretically-driven causal order, suggesting that supervisor-rated performance predicts supervisor perceptions of male employees’ WFC (but not female employees’ WFC) and female employees’ FWC (but not male employees’ FWC). To ascertain this possibility, we tested a model that places supervisor-rated performance as the predictor, employee gender as the moderator, and supervisor perceptions of employee WFC/FWC as the dependent variable. Results of these additional analyses indicate that employee gender did not moderate the relationship between supervisor-rated performance and supervisor perceptions of employee WFC (β = –.02, NS) or FWC (β = .09, NS). These results provide no support for the reversed-order prediction.

Finally, it is possible that it is the supervisor’s gender, rather than the employee’s gender, that moderates the effects of conflict. We re-analysed our data using supervisor gender as the moderator and employee gender as a control variable. We found no support for this prediction. Specifically, supervisor gender did not moderate the relationship between employee WFC (β = –.07, NS) or FWC (β = –.04, NS) and supervisor-perceived employee conflict. Supervisor gender also did not moderate the relationship between employee conflict (WFC or FWC) and supervisor-rated performance (WFC: β = .12, NS; FWC: β = .17, NS). Results of these analyses can be obtained from the first author upon request.

Discussion

Although the effects of employees’ work–family conflict on attitudes and self-rated performance have been widely documented, its effects on supervisor-rated performance have been somewhat inconclusive. Using a sample of full-time employees and their supervisors, we attempted to uncover not only when but also how employee perceived work–family conflict is related to their performance rated by their supervisor. Our study underscores the important role played by supervisor perceptions of employee work–family conflict as a mediator and employee gender as a moderator of this relationship.

Theoretical implications

One important contribution made by the present study is the examination of supervisor perceptions of employee work–family conflict. Although past research has suggested that supervisor perceptions of employee work–family conflict play a critical role in determining supervisors’ evaluation of their employee, empirical investigations of this important variable have been rather rare and incomplete (see Carlson et al., 2008; Hoobler et al., 2009). We found that the relationship between work–family conflict ratings provided by these two sources was moderated by gender. Specifically, the relationship for WFC was stronger for male employees and weaker for female employees, whereas the relationship for FWC was stronger for female employees and weaker for male employees.

These results are consistent with gender role theory positing that human information processing is influenced by gender stereotypes (Fiske and Neuberg, 1990). If supervisors believe that their male employees are more likely to put their work ahead of their family owing to their role as provider, they are more likely to notice, interpret in a role-consistent fashion, and recall the extent to which these employees’ work spills into the family domain. As such, if a male employee is experiencing WFC, then the supervisor will be more likely to accurately detect it, than if he is experiencing FWC. In contrast, if supervisors believe that their female employees are more likely to put their family ahead of work owing to their domestic role, they are more likely to notice, interpret in a role-consistent fashion, and recall the extent to which these employees’ family spills into the work domain. As such, if a female employee is experiencing FWC, then the supervisor will be more likely to accurately detect it, than if she is experiencing WFC. These results are also consistent with Karniol (1990), who argued that to understand other people’s thoughts and feelings, perceivers need to possess declarative knowledge of their target. Declarative knowledge refers to knowing the ‘repeated sequences of events and actions’ by the target so that perceivers can generate a mental ‘representation that abstracts the similarities between these events,’ which they use as ‘a means of making sense of other people’s [the target’s] behaviors and their reactions to events’ (Karniol, 1990: 228). One type of such declarative knowledge is category-specific information, which means mental representations – or stereotypes – of an entire category of people of which the target is a member. In the context of the present study, supervisors make inferences of their male and female employees’ work–family conflict by relying on declarative knowledge they adduce based on the social role that these employees are stereotypically assigned at the societal level (Eagly, 1987; Eagly et al., 2000).

We also found support for the predicted moderated mediation models. Employee work–family conflict was related to supervisor-rated performance but this relationship was mediated by supervisor perceptions of employee work–family conflict and moderated by employee gender. These results suggest that relatively lower supervisor-rated performance, which is sometimes observed for both men and women who experience work–family conflict, may not necessarily be owing to actual performance decrements. To state the matter loosely, WFC and FWC may not lower worker effectiveness for men and women, respectively. Rather, they lower supervisory perceptions of worker effectiveness. These results are theoretically important for at least three reasons.

First, the employee-centric perspective, which has traditionally guided research in this domain, suggests that employees are responsible for the performance deficit because the work–family conflict they experience depletes the resources needed to meet work requirements (Muse and Pichler, 2011). Our supervisor-centric perspective in contrast suggests that the conflict-performance relationship may be attributable to the supervisor whose perceptions of employee conflict lead them to assign a lower performance rating to their employee. Such perceptual processes may be highly subjective making them amenable to the influence of biases and stereotypes. Hoobler and colleagues (2009) showed that women received fewer promotion opportunities than men because of their supervisor’s perceptions of these women’s FWC. We extended these findings by showing that supervisors’ gender-specific stereotypes may also impact their perceptions of male employees’ WFC and subsequent performance evaluations. Our findings suggest that quelling biased ratings may be a solution to the poorer performance ratings assigned to employees who experience work–family conflict.

Second, our supervisor-centric perspective centers on the role of the supervisor, whose perceptions potentially explain the conflict-performance relationship. If future work replicates these findings, it suggests that it is inappropriate to lay blame on the employee for performance decrements. This argument suggests that future research should examine supervisor or organization variables that create a relationship between conflict and performance. For example, what are some of the supervisor variables that may influence the extent to which supervisors attend to, recall, or interpret stereotype-related information or the extent to which they see employee work–family conflict as a performance liability? Investigating these research questions will go a long way in explaining the equivocal relationship between employee work–family conflict and supervisor-rated performance reported in the literature (Hoobler et al., 2010).

Third, there has been considerable debate on whether WFC or FWC is more closely related to supervisor-rated performance (Amstad et al., 2011; Frone et al., 1992). From our supervisor-centric perspective, we argue that the conflict-performance relationship may be more complex than previously theorized. Specifically, we show that both WFC and FWC can be related to supervisor-rated performance but that these effects appear to be gender specific and owing to stereotype. WFC appeared to harm male employees’ supervisor-rated performance because of the WFC perceived by their supervisor, whereas FWC appeared to harm female employees’ supervisor-rated performance because of the FWC perceived by their supervisor. These results suggest that both directions of conflict – WFC and FWC – are related to performance, provided that the researcher takes into account of (a) employee gender and (b) supervisor perceptions.

These results also raise an interesting psychometric question: which measure of work–family conflict should researchers place more ‘trust’ in, employee’s self-ratings or observer perceptions of employee conflict? Owing to the mixed findings reported in the literature and our own, we offer the following recommendations for future research. First, instead of trying to identify the most ‘trusted’ source of work–family conflict, future research should examine the conditions under which self-rated and other-rated work–family conflict agree and disagree. Using gender role theory to guide our research, we focus on employee gender as a moderator of conflict rating agreement/disagreement in the present study. Future research should build on our findings and use a theory-driven approach to identify other moderators at the individual level, at the dyadic level, at the organizational level, or at the cultural level. Second, future research should examine whether self- and other-ratings of conflict may differentially predict various outcome variables and whether the strength of the relationships depends on conflict type, employee characteristics, supervisor characteristics, or other variables. Third, since our study, as well as previous research (Hoobler et al., 2009), shows that supervisor perceptions of employee work–family conflict impact employment outcomes important to the employee, it is necessary that future organizational interventions be designed and implemented to increase the accuracy of such conflict perceptions. Fourth, future research should also examine how self- and other-ratings of work–family conflict may combine to impact employee outcomes. For example, does the combination of high self-rated WFC and low supervisor-rated WFC have the same impact as the combination of low self-rated WFC and high supervisor-rated WFC on employee attitudes and behaviors? Sophisticated statistical analyses using such tools as polynomial regression may help provide important insights into these questions. Fifth, future research should also examine the factors that predict other-rated work–family conflict.

Limitations and future research

A few study limitations, which also provide important avenues for future research, should be noted. First, since a cross-sectional design was used in our study, causality cannot be assumed. However, results of our auxiliary analyses suggest that supervisor-rated performance did not serve as the mediator of the model, nor was the model supported in the absence of the proposed mediator (supervisor perceptions of employee conflict). Nevertheless, the causal relationship among these variables can better be discerned when a longitudinal study that assesses all of these variables at each time period is employed.

Second, our sample consisted of mostly American, white-collar employees holding a professional position at the time of the study. Research has suggested that low-income, blue-collar workers may have working conditions, family characteristics and access to family-friendly programs that are different from white-collar workers typically used in work–family research (Griggs et al., 2013). In addition, recent research has also shown that owing to variations associated with cultural values, social-economic conditions and family/work characteristics, findings based on work–family research conducted in western cultures may not always generalize to other cultures (Spector et al., 2004, 2007). For example, Agarwala et al. (2014) found that, contrary to what is typically reported in research conducted in western countries, the relationship between supervisor support and employee WFC was not significant in India. They argued that these finding might have been caused by the unique cultural context in India where organizational members are viewed as part of the extended family whose support for family needs is not only appreciated but also expected. Future research should attempt to replicate our findings with samples from under-represented communities and cultures. We add, however, that at least some of this work might benefit research from the supervisor-centric perspective. For instance, among cultures that are less prone to gender role stereotyping, the conflict-performance relationship might be weaker. However, among cultures that are more prone to gender-role stereotyping this relationship might be stronger.

Third, Allen and colleagues (2000) argue that employee OCB, rather than their performance, is more amenable to the influence of employee work–family conflict. Employees who experience work–family conflict may still perform their required duties but may opt to reduce discretionary behaviors beneficial to the organization (Halbesleben and Bowler, 2007). If this proposition has merit, then future research should extend our model by also considering employees’ helping behaviors.

Fourth, although we use gender role theory as our theoretical framework, we did not explicitly measure supervisors’ gender role orientation in our study (Hochschild and Machung, 1989). Our fear was that doing so might build demand characteristics into our study. That said, it is known that individuals who subscribe to the traditional gender role tend to associate the work domain with men and the family domain with women. In contrast, individuals who are egalitarian tend to believe in gender equality when it comes to work and family contributions. Therefore, gender role orientation may serve as an additional moderator of our findings such that the effects may be more pronounced with supervisors who are traditional and less so with supervisors who are egalitarian. This interesting possibility should be explored in future research.

Fifth, owing to the relatively modest sample size, we were not able to use SEM to test the entire model that includes both WFC and FWC. Nevertheless, we conducted a confirmatory factor analysis to examine whether the WFC and FWC ratings reported by the employee and the supervisor were distinct from each other. The four-factor model (supervisor-rated WFC, supervisor-rated FWC, employee-rated WFC and employee-rated FWC) provided a significantly better fit to the data than a one-factor model and a two-factor model that sorted the items based on their rating source. In addition, controlling for employee-rated FWC in the WFC model and controlling for employee-rated WFC in the FWC model did not change the conclusions reported herein. These observations give us additional confidence in our results.

Finally, future research should examine the boundary conditions of our findings. For example, we posit in our study that when supervisors believe that their employees experience work–family conflict, they tend to assign them lower performance ratings because they believe that the conflict may impair work performance. This relationship, however, may be weakened in certain conditions. For example, a supervisor is less likely to assign a low performance rating to an employee perceived to experience work–family conflict if the supervisor and the employee work alongside each other (which allows the supervisor to more accurately assess the employee’s performance) or when the task the employee performs is highly analyzable making performance ratings less amenable to the influence of biases and stereotypes.

Practical implications

Our research shows that male employees’ WFC was related to lower performance ratings because of the WFC perceived by their supervisor, whereas female employees’ FWC was related to lower performance ratings because of the FWC perceived by their supervisor. These results underscore the important role that the supervisor plays in order to deactivate the link between employee work–family conflict and supervisor-rated performance. Instead of penalizing their employees for experiencing WFC and FWC, supervisors should focus on offering family support that allows employees to manage work and family demands. Hammer et al. (2009) identify four kinds of family-friendly supervisor supportive behavior. These include (a) emotional support (e.g. expressing concerns when employees are overwhelmed by competing work–family responsibilities), (b) role modeling (e.g. using personal experience to demonstrate how work and family can be effectively combined), (c) instrumental support (e.g. allowing an employee to take some time off in order to take care of a sick child), and (d) creative work–family management (e.g. meetings are scheduled at time slots that work for both people who have a family and people who do not). Recent meta-analytic research has shown that supervisor support is associated with lower levels of employee WFC and FWC (Kossek et al., 2011; Michel et al., 2011). These four types of support may alleviate the stress associated with employee work–family conflict. We go further, suggesting that supervisors may also benefit personally from offering support to their employees. By treating workers as actual individuals, rather than as abstract ‘women’ and ‘men,’ the tendency to stereotype might be reduced. While this possibility is speculative it is also important. Support may help the person who provides it, as it does the person who receives it.

Organizations can also create a family-friendly environment by implementing intervention programs specifically targeting supervisory staff in order to raise their awareness of work–family issues in the workplace. Recent research (Hammer et al., 2011, 2016) has recommended several supervisor training programs that focus on knowledge acquisition (learning the personal and organizational benefits of reducing employees’ work–family conflict), behavioral modifications (learning specific techniques on how to offer family support to their employees), and self-monitoring (monitoring the frequencies with which these supportive behaviors are exhibited). Empirical studies have shown that these informal, cost-effective programs have considerable promises in increasing supervisors’ family-supportiveness in the workplace.

In conclusion, our research shows that supervisor perceptions of employee work–family conflict play a critical role in explaining why employees’ work–family conflict is related to supervisor-rated performance. Therefore, it is possible that the supervisor, rather than the employee, is responsible for decreases in supervisor-rated performance as a result of the work–family conflict that the employee experiences. We challenge organizations to focus on changing how supervisors perceive their subordinate’s work–family conflict as part of their intervention effort in order to create a more family-friendly work environment.

Footnotes

Acknowledgements

We wish to thank Sherry Thatcher for helpful comments on an early draft of this manuscript. We are indebted to the Associate Editor, Samuel Aryee, and three anonymous reviewers for their constructive and thoughtful comments on our manuscript.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Notes

Andrew Li is the Williams Professor of Management at West Texas A&M University, USA. He earned his doctoral degree in management from the University of Arizona. His research focuses on the work–family interface in organizations, personality, organizational justice and team research. [Email: ali@wtamu.edu]

Jessica Bagger is Associate Professor of Management at Sacramento State, USA. Her research relates to work–family issues and organizational justice. Her research appears in Journal of Management, Human Relations, among others. She was named to the Sloan Early Career Work & Family Scholars Program where she is now Sponsoring Member. [Email: baggerj@csus.edu]

Russell Cropanzano is Professor of Management and Entrepreneurship at the University of Colorado, USA. Previously, he has published over 120 scholarly articles and chapters. Dr Cropanzano has also authored two books and edited four others. Dr Cropanzano was also a winner of the 2000 Outstanding Paper Award from the Consulting Psychology Journal, the 2007 Best Paper Award from the Academy of Management Perspectives, and the 2010 Best Paper Award from the Journal of Management. He is a past editor of Journal of Management and a fellow in the Academy of Management, the Society for Industrial/Organizational Psychology, the Southern Management Association and the Association for Psychological Science. [Email: russell.cropanzano@colorado.edu]

References

Agarwala

Arizkuren-Eleta

Del Castillo

. (2014) Influence of managerial support on work–life conflict and organizational commitment: An international comparison for India, Peru and Spain. The International Journal of Human Resource Management 25(10): 1460–1483.

Allen

Russell

(1999) Parental leave of absence: Some not so family-friendly implications. Journal of Applied Social Psychology 29(1): 166–191.

Allen

Herst

Bruck

Sutton

(2000) Consequences associated with work-to-family conflict: A review and agenda for future research. Journal of Occupational Health Psychology 5(2): 278–308.

Almer

Cohen

Single

(2004) Is it the kids or the schedule? The incremental effect of families and flexible scheduling on perceived career success. Journal of Business Ethics 54(1): 51–65.

Amstad

Meier

Fasel

. (2011) A meta-analysis of work–family conflict and various outcomes with a special emphasis on cross-domain versus matching-domain relations. Journal of Occupational Health Psychology 16(2): 151–169.

Blair-Loy

(2003) Competing Devotions: Career and Family among Women Executives. Cambridge, MA: Harvard University Press.

Blair-Loy

(2004) Work devotion and work time. In: Epstein

Kalleberg

(eds) Fighting for Time: Shifting Boundaries of Work and Social Life. New York: Russell Sage Foundation, 262–316.

Bodenhausen

(1988) Stereotypic biases in social decision making and memory: Testing process models of stereotype use. Journal of Personality and Social Psychology 55(5): 726–737.

Burgess

Borgida

(1999) Who women are, who women should be: Descriptive and prescriptive gender stereotyping in sex discrimination. Psychology, Public Policy, & Law 5(3): 665–692.

10.

Butler

Skattebo

(2004) What is acceptable for women may not be for men: The effect of family conflicts with work on job-performance ratings. Journal of Occupational and Organizational Psychology 77(4): 553–564.

11.

Carlson

Kacmar

Williams

(2000) Construction and initial validation of a multidemensional measure of work–family conflict. Journal of Vocational Behavior 56(2): 249–276.

12.

Carlson

Witt

Zivnuska

. (2008) Supervisor appraisal as the link between family-work balance and contextual performance. Journal of Business and Psychology 23(1): 37–49.

13.

Casper

Eby

Bordeaux

. (2007) A review of research methods in IO/OB work–family research. Journal of Applied Psychology 92(1): 28–43.

14.

Cohen

Single

(2001) An examination of the perceived impact of flexible work arrangements on professional opportunities in public accounting. Journal of Business Ethics 32(4): 317–328.

15.

Devine

(1995) Prejudice and out-group perception. In: Tesser

(ed.) Advanced Social Psychology. New York: McGraw Hill, 467–524.

16.

Eagly

(1987) Sex Differences in Social Behavior: A Social Role Interpretation. Hillsdale, NJ: Erlbaum.

17.

Eagly

Karau

(2002) Role congruity theory of prejudice toward female leaders. Psychological Review 109(3): 573–598.

18.

Eagly

Mladinic

(1994) Are people prejudiced against women? Some answers from research on attitudes, gender stereotypes, and judgments of competence. In: Stroebe

Hewstone

(eds) European Review of Social Psychology, Vol. 5. New York: Wiley, 1–35.

19.

Eagly

Wood

Diekman

(2000) Social role theory of sex differences and similarities: A current appraisal. In: Eckes

Trautner

(eds) The Developmental Social Psychology of Gender. Mahwah, NJ: Lawrence Erlbaum Associates Publishers, 123–174.

20.

Fiske

Neuberg

(1990) A continuum of impression formation, from category-based to individuating processes: Influences of information and motivation on attention and interpretation. In: Zanna

(ed.) Advances in Experimental Social Psychology, Vol. 23. San Diego, CA: Academic Press, 1–74.

21.

Fiske

Taylor

(1984) Social Cognition. Reading, MA: Addison-Wesley.

22.

Ford

Heinen

Langkamer

(2007) Work and family satisfaction and conflict: A meta-analysis of cross-domain relations. Journal of Applied Psychology 92(1): 57–80.

23.

Frone

Russell

Cooper

(1992) Antecedents and outcomes of work–family conflict: Testing a model of the work–family interface. Journal of Applied Psychology 77(1): 65–78.

24.

Fyock

Stangor

(1994) The role of memory biases in stereotype maintenance. British Journal of Social Psychology 33(3): 331–343.

25.

Grandey

Cordeiro

Crouter

(2005) A longitudinal and multi-source test of the work–family conflict and job satisfaction relationship. Journal of Occupational and Organizational Psychology 78(3): 305–323.

26.

Greenhaus

Beutell

(1985) Sources of conflict between work and family roles. Academy of Management Review 10(1): 76–88.

27.

Griggs

Casper

Eby

(2013) Work, family and community support as predictors of work–family conflict: A study of low-income workers. Journal of Vocational Behavior 82(1): 59–68.

28.

Gutek

Searle

Klepa

(1991) Rational versus gender role explanations for work–family conflict. Journal of Applied Psychology 76(4): 560–568.

29.

Halbesleben

Bowler

(2007) Emotional exhaustion and job performance: The mediating role of motivation. Journal of Applied Psychology 92(1): 93–106.

30.

Hammer

Johnson

Crain

. (2016) Intervention effects on safety compliance and citizenship behaviors: Evidence from the work, family, and health study. Journal of Applied Psychology 101(2): 190–208.

31.

Hammer

Kossek

Anger

. (2011) Clarifying work–family intervention processes: The roles of work–family conflict and family-supportive supervisor behaviors. Journal of Applied Psychology 96(1): 134–150.

32.

Hammer

Kossek

Yragui

. (2009) Development and validation of a multidimensional measure of family supportive supervisor behaviors (FSSB). Journal of Management 35(4): 837–856.

33.

Hayes

(2013) Introduction to Mediation, Moderation, and Conditional Process Analysis: A Regression-Based Approach. New York: The Guilford Press.

34.

Heilman

(2001) Description and prescription: How gender stereotypes prevent women’s ascent up the organizational ladder. Journal of Social Issues 57(1): 657–674.

35.

Heilman

(2012) Gender stereotypes and workplace bias. Research in Organizational Behavior 32: 113–135.

36.

Hochschild

Machung

(1989) The Second Shift. New York: Viking.

37.

Hoobler

Wilson

(2010) Do workers who experience conflict between the work and family domains hit a ‘glass ceiling?’: A meta-analytic examination. Journal of Vocational Behavior 77(3): 481–494.

38.

Hoobler

Wayne

Lemmon

(2009) Bosses’ perceptions of family-work conflict and women’s promotability: Glass ceiling effects. Academy of Management Journal 52(5): 939–957.

39.

Johnson

Judd

(1983) Overlooking the incongruent: Categorization biases in the identification of political statements. Journal of Personality and Social Psychology 45(5): 978–996.

40.

Karniol

(1990) Reading people’s minds: A transformation rule model for predicting others’ thoughts and feelings. In: Zanna

(ed.) Advances in Experimental Social Psychology, Vol. 23. San Diego, CA: Academic Press, 211–247.

41.

Kossek

Ozeki

(1999) Bridging the work–family policy and productivity gap: A literature review. Community, Work & Family 2(1): 7–32.

42.

Kossek

Pichler

Bodner

Hammer

(2011) Workplace social support and work–family conflict: A meta-analysis clarifying the influence of general and work–family–specific supervisor and organizational support. Personnel Psychology 64(2): 289–313.

43.

Bagger

(2011) Walking in your shoes: Interactive effects of child care responsibility difference and gender similarity on supervisory family support and work-related outcomes. Group & Organization Management 36(6): 659–691.

44.

Zhang

(2013) Cross-domain effects of work–family conflict on organizational commitment and performance. Social Behavior and Personality: An International Journal 41(10): 1641–1654.

45.

Macrae

Milne

Bodenhausen

(1994) Stereotypes as energy-saving devices: A peek inside the cognitive toolbox. Journal of Personality and Social Psychology 66(1): 37–47.

46.

Michel

Kotrba

Mitchelson

. (2011) Antecedents of work–family conflict: A meta-analytic review. Journal of Organizational Behavior 32(5): 689–725.

47.

Muse

Pichler

(2011) A comparison of types of support for lower-skill workers: Evidence for the importance of family supportive supervisors. Journal of Vocational Behavior 79(3): 653–666.

48.

Netemeyer

Maxham

Pullig

(2005) Conflicts in the work–family interface: Links to job stress, customer service employee performance, and customer purchase Intent. Journal of Marketing 69(2): 130–143.

49.

Odle-Dusseau

Britt

Greene-Shortridge

(2012) Organizational work–family resources as predictors of job performance and attitudes: The process of work–family conflict and enrichment. Journal of Occupational Health Psychology 17(1): 28–40.

50.

Preacher

Rucker

Hayes

(2007) Addressing moderated mediation hypotheses: Theory, methods, and prescriptions. Multivariate Behavioral Research 42(1): 185–227.

51.

Ruble

Stangor

(1986) Stalking the elusive schema: Insights from developmental and social-psychological analyses of gender schemas. Social Cognition 4(2): 227–261.

52.

Rudman

Glick

(2008) The Social Psychology of Gender: How Power and Intimacy Shape Gender Relations. New York: Guilford Press.

53.

Rudman

Mescher

(2013) Penalizing men who request a family leave: Is flexibility stigma a femininity stigma? Journal of Social Issues 69(2): 322–340.

54.

Rudman

Moss-Racusin

Glick

Phelan

(2012) Reactions to vanguards: Advances in backlash theory. In: Devine

Plant

(eds) Advances in Experimental Social Psychology, Vol. 45. San Diego, CA: Academic Press, 167–227.

55.

Schneier

Beatty

Baird

(1986) Creating a performance management system. Training & Development Journal 40(5): 74–79.

56.

Shih

Chiang

Hsu

(2010) High involvement work system, work–family conflict, and expatriate performance – Examining Taiwanese expatriates in China. International Journal of Human Resource Management 21(11): 2013–2030.

57.

Shockley

Singla

(2011) Reconsidering work–family interactions and satisfaction: A meta-analysis. Journal of Management 37(3): 861–886.

58.

Spector

Allen

Poelmans

. (2007) Cross-national differences in relationships of work demands, job satisfaction, and turnover intentions with work–family conflict. Personnel Psychology 60(4): 805–835.

59.

Spector

Cooper

Poelmans

. (2004) A cross-national comparative study of work–family stressors, working hours, and well-being: China and Latin America versus the Anglo World. Personnel Psychology 57(1): 119–142.

60.

Stangor

McMillan

(1992) Memory for expectancy-congruent and expectancy-incongruent information: A review of the social and social developmental literatures. Psychological Bulletin 111(1): 42–61.

61.

Vandello

Hettinger

Bosson

Siddiqi

(2013) When equal isn’t really equal: The masculine dilemma of seeking work flexibility. Journal of Social Issues 69(2): 303–321.

62.

Wang

Walumbwa

Wang

Aryee

(2013) Unraveling the relationship between family-supportive supervisor and employee performance. Group & Organization Management 38(2): 258–287.

63.

Williams

(2001) Unbending Gender: Why Family and Work Conflict and What to Do About It. Oxford: Oxford University Press.

64.

Williams

Blair-Loy

Berdahl

(2013) Cultural schemas, social class, and the flexibility stigma. Journal of Social Issues 69(2): 209–234.

65.

Williams

Anderson

(1991) Job satisfaction and organizational commitment as predictors of organizational citizenship and in-role behaviors. Journal of Management 17(3): 601–617.

66.

Witt

Carlson

(2006) The work–family interface and job performance: Moderating effects of conscientiousness and perceived organizational support. Journal of Occupational Health Psychology 11(4): 343–357.