Abstract
Work–family conflict may give rise to different emotional reactions, depending on the causal attributions people make for the experience of work–family conflict. These emotional reactions, in turn, may result in specific behavioral reactions, that may either be adaptive or maladaptive in nature. In this essay, we advance this thesis using attribution theory and theory of emotions that specifies different behavioral outcomes associated with different attribution-induced negative emotions. We develop a model of causal attributions for work–family conflict that includes differentiated emotional reactions (e.g., guilt and shame) and both adaptive (e.g., job crafting, self-development) and maladaptive (e.g., withdrawal, aggression, turnover) behavioral outcomes of work–family conflict. We conclude the article with directions for future research, guided by the proposed model.
Over the last few decades, the composition of the workforce changed toward more dual-career families where spouses share family responsibility (Eby, Casper, Lockwood, Bordeaux, & Brinley, 2005). At the same time, the interest in the impact of work on employee well-being (e.g., Ilies, Schwind, & Heller, 2007) has considerably grown. As a consequence, the volume of research on the extent to which and how employees’ work demands and activities influence their family lives has been steadily increasing (Eby et al., 2005; Ford, Heinen, & Langkamer, 2007). Within the body of research on work and family, studies examining work-to-family conflict have been most prevalent (Eby et al., 2005; Greenhaus & Powell, 2003). Work-to-family conflict (hereafter referred to, simply, as work–family conflict) arises when demands and pressures from the work domain impede employees’ participation in their family role (Greenhaus & Beutell, 1985; Greenhaus & Powell, 2003; Gutek, Searle, & Klepa, 1991; Netemeyer, Boles, & McMurrian, 1996).
The literature on work–family conflict has made important advances over the last few decades, and much is now known about the consequences of work–family conflict for both individuals and organizations. We know, for example, that work–family conflict negatively relates to employee well-being (e.g., Allen, Herst, Bruck, & Sutton, 2000) as reflected in decreased evaluations of specific life domains (i.e., job satisfaction, marital satisfaction) as well as in decreased evaluations of one’s life in general (i.e., life satisfaction). We also know that work–family conflict, unpleasant by nature, is associated with less positive and more negative affect (see Eby, Maher, & Butts, 2010) and increased emotional strain, as reflected in higher emotional exhaustion, burnout (Eby et al., 2005; Kossek & Ozeki, 1998), and depression (Frone, Rusell, & Cooper, 1992).
However, with few notable exceptions (e.g., Judge, Ilies, & Scott, 2006; Livingston & Judge, 2008; Poposki, 2011), there is very little research on specific (differentiated) emotions as outcomes of work–family conflict. That is, most research on the topic focused on broad affective dimensions such as positive and negative affect (Ilies, Schwind, Wagner, et al., 2007) or broadly defined mood states derived by means of factor analysis (Williams & Alliger, 1994).
Given the importance of specific emotions for cognition and action (Gooty, Gavin, & Ashkanasy, 2009; Izard, 2009) and the differentiated effects of specific emotions on choice and behavior (Frijda, 1986; Roseman, Wiest, & Swartz, 1994; Tangney, 1991; Tangney, Stuewig, & Mashek, 2007), we view the paucity of research on the impact of work–family conflict on specific emotions (Eby et al., 2010) as an important gap in the literature. Moreover, the little research that did focus on specific emotions did not consider the ultimate effects of work–family conflict on employees’ decisions, actions, and behaviors, through these emotions. Livingston and Judge (2008), for example, examined the day-to-day associations between work–family conflict and guilt but did not examine whether guilt led to actions or behaviors aimed at reducing the increased work–family conflict level that caused guilt in the first place. In a similar vein, although Glavin, Schieman, and Reid (2011) examined a psychological consequence of guilt that was related to work–family interference, they focused on distress rather than on how employees actually react to the experienced guilt. Hence, little is known regarding the implications that emotional reactions to work–family conflict have for employees’ decisions, actions, and behaviors on and off the job.
We believe, however, that this is an important issue, as knowledge regarding the specific emotional and behavioral reactions to work–family conflict may shed light on how individuals cope with it and give rise to the development of interventions that stimulate adaptive rather than maladaptive responses to work–family conflict (Poposki, 2011). In order to gain this knowledge, several questions need to be addressed: Are some specific emotions experienced as a result of work–family conflict more likely to result in adaptive (i.e., aimed at restoring work–family balance) or maladaptive decisions and behaviors (i.e., withdrawal, aggression, turnover) than other emotions? Do different ways in which employees cognitively process the occurrence of work–family conflict induce different specific emotional reactions? To address these questions, we propose a model that explains how and when work–family conflict influences specific emotions and how these emotions can have different consequences for actions and behaviors relevant to balancing employees’ work and family roles. We derived our model based on Weiner’s attributional theory of motivation and emotion (e.g., Weiner, 1985) as well as theory and research suggesting that different emotions have different action tendency and behavioral consequences (e.g., Roseman et al., 1994; Tangney, 1991; Tangney et al., 2007).
Individuals’ emotional and behavioral reactions to events and outcomes in their lives are likely to depend on the attributions they make regarding the causes of these events and outcomes (e.g., Kelley & Michela, 1980; Martinko, Harvey, & Dasborough, 2011; Weiner, 1985). This suggests that the emotional consequences of work–family conflict (and thus the more distal behavioral outcomes induced by these emotional reactions) are likely to be differentiated according to what employees believe are the causes of experienced conflict. Therefore, we examine individuals’ specific emotional reactions and behavioral outcomes to work–family conflict using causal attribution theory.
The attribution theory of achievement motivation and emotion (Weiner, 1985) postulates that outcomes of events instigate two distinct affective processes. The first is a general positive or negative affective reaction based on whether the outcome is perceived of as a success or failure. These general affective reactions are considered to directly depend on the evaluation of the outcome of the event (i.e., they are outcome-dependent). General affective reactions refer to “a wide range of summative evaluative constructs, which describe how one feels about objects, events, and life experiences” (Eby et al., 2010, p. 602) and include transitory states such as moods, and more evaluative affective judgments, such as job satisfaction, life satisfaction, and subjective well-being. Attribution theory further proposes that specific emotions will be generated, depending on the causal attributions people make for the outcome of the event (i.e., the reason for their success or failure). These specific emotions are attribution-dependent, and include, for example, serenity and surprise for positive outcomes and shame and anger for negative outcomes (Weiner, 1985).
The influence individuals’ work life has on their family life can be positive, when work experiences improve family experiences (i.e, work–family enrichment; Greenhaus & Powell, 2006) or negative, when in some respect, “the role pressures from work and family domains are mutually incompatible” (i.e., work–life conflict; Greenhaus & Beutell, 1985, p. 77). Work–family conflict can thus be conceptualized on the “failure side” on a continuum reflecting success versus failure in balancing work and family roles. This implies that work–family conflict not only elicits broad negative affective states that are outcome dependent (and attribution independent), but also results in specific attribution-dependent negative emotions: hopelessness, shame, guilt, frustration, and anger. It is our contention that individuals’ specific attributions regarding the causes of work–family conflict and the specific emotions that these attributions elicit have important implications for future actions aimed at managing the work–family interface.
Martinko et al. (2011) recently argued that, in general, organizational sciences largely underutilize attributional mechanisms as explanations for psychological processes and organizational behavior. Indeed, there is virtually no research on the role of employees’ causal beliefs about the occurrence of work family–conflict in influencing its outcomes. Nevertheless, a large body of research on causal attributions deals with the differentiation of emotions following an event or experience according to the specific attribution that individuals make for the cause of the event or experience (e.g., Covington & Omelich, 1984; Weiner, 1985; Weiner, Russell, & Lerman, 1979). These studies have indicated that following negative events and outcomes, such as work–family conflict, shame, guilt, anger, frustration, and hopelessness prevail. There is also theory and research suggesting that different emotions, including moral emotions such as guilt and shame have different action tendency and behavioral consequences (e.g., Roseman et al., 1994; Tangney, 1991; Tangney et al., 2007). Research hardly addressed the decision-making or behavioral action consequences of work–family conflict. In this paper, we argue that in order to understand what actions individuals take when they realize that their work prevents them from functioning effectively in their family role, we must understand the differentiated emotional reactions and associated adaptive functions of emotions that are elicited in response to the occurrence of work–family conflict.
In reaction to work–family conflict, employees can engage in both adaptive and maladaptive behaviors to alleviate the negative emotions related to work–family conflict. Although there has been little research addressing the behavioral outcomes of work–family conflict, studies on work–family conflict prevention (e.g., Baltes & Heydens-Gahir, 2003; Kossek, Lautsch, & Eaton, 2006) and choice making between work and family demands (Greenhaus & Powell, 2003) suggested several possible actions employees could undertake as a reaction to work–family conflict. Based on these studies and on research on discrete emotions we decided to focus on maladaptive behaviors such as withdrawal from the family, work withdrawal, turnover, and aggressive work behaviors. These behaviors may reduce work–family conflict and the induced negative emotions on the short term, but are likely to also have negative consequences, such as estranged family and work relations, loss of family, or job loss. We also focus on adaptive behaviors that are aimed at solving the problem: self-development and time management, to be better and more timely able to deal with the work demands, and job crafting, that is, changing the boundaries of the job to better fit the needs of the job incumbent. These latter behaviors are likely to effectively reduce work family conflict without concurring negative side effects.
In short, our model proposes that depending on the attribution for the occurrence of work–family conflict, the experience of work-family conflict influences different specific emotions (guilt, shame, anger, frustration, and hopelessness) that have different consequences for actions and behaviors relevant to balancing employees’ work and family roles. The general model that we propose is presented in Figure 1.
General representation of the attributional model of work–family conflict.
Consequences of work–family conflict
As the general model (Figure 1) shows, work–family conflict arises when work demands and family demands are incompatible, and employees are prevented from effectively fulfilling their family roles, resulting in what we call poor family performance. Following attribution theory (Kelley, 1973; Weiner, 1985), work–family conflict will generate attribution-independent emotions. Because poor family performance constitutes, in essence, a failure in managing the work–family interface, we expect that when employees realize that their work demands interfere with their family life in such a way that their performance of their family role suffers, they will experience general negative affective reactions, such as sadness, dissatisfaction, and unhappiness. This expectation is in line with both qualitative and quantitative reviews on the emotional consequences of work–family conflict that indicate that work-family conflict is associated with broadly defined negative emotional-based outcomes such as decreased job and life satisfaction (e.g., Allen et al., 2000; Kossek & Ozeki, 1998), higher emotional exhaustion and burnout (e.g., Eby et al., 2005; Kossek & Ozeki, 1999), and less positive and more negative affect (e.g., Eby et al., 2010).
Individuals’ affective reactions (and affective feelings in general) can be conceptualized and operationalized either as broad (or global) affective dimensions or as discrete emotions (Watson, 2000). Examining the affective outcomes of work–family conflict and associated behavioral consequences, we agree with the view of Lazarus and Cohen-Charash (2001) that “The differences among discrete emotions may be more important (or at least as important) for understanding problems of adaptation than the similarities within each grouping” (2001, p. 52). Investigating discrete emotions as specific outcomes of work–family conflict instead of focusing on the broad dimension of negative affect is a necessary expansion to understand how individuals react to the experience of work–family conflict. Therefore, having specified the link between work–family conflict and attribution-independent broad affective reactions, we now turn to the discrete emotions resulting from it.
Research on the relationships between work–family conflict and specific emotions is sparse. One diary study (Judge et al., 2006), showed that work–family conflict is related to feelings of guilt and hostility. A diary study by Livingston and Judge (2008), however, found no support for a main effect of work–family conflict on feelings of guilt. Their study revealed that only employees who do not conform to traditionally accepted gender norms feel guilty when experiencing high levels of work–family conflict. Taken together, there is some evidence that work–family conflict results into specific emotional reactions, such as guilt and hostility. However, results are not entirely consistent across studies, which emphasizes the need for more elaborated theoretical models to make more focused predictions about the specific emotional outcomes of work–family conflict.
Attributions for and discrete emotional reactions to work–family conflict
In order to regain some sense of control over their environments and to increase their ability to predict future outcomes, individuals often, if not always, engage in causal searches in order to determine the causes of events or outcomes (Heider, 1958; Weiner, 1985). With respect to work–family conflict, we believe employees will attribute causes for the occurrence of it when they realize that conflict has occurred, as a means to gain control over and improve the balance between their work and family roles.
As Weiner noted in most general terms, people want to understand the causes of events and are in a constant pursuit of “why” (1985, p. 548). Ascribing causes to effects concerning success and failure is especially functional because the ascribed causes have direct implications for future action. Attribution theory (Heider, 1958; Kelley, 1973; Weiner, 1985), distinguishes individuals’ explanations (i.e., ascribed causes) for success and failure along three dimensions: causal locus (whether the cause is internal to the person or external), stability (whether the cause is constant over time), and controllability (whether the cause is under the person’s volitional control). The type of causal attribution that one makes for an event has important influences on future expectancies, emotions, and actions. For example, when one attributes failure at a task to internal and controllable causes, one would be more likely to decide to put forth more effort to improve performance, compared to when the failure is attributed to external or uncontrollable causes. In Figure 2 we illustrate how causal attributions for the occurrence of work–family conflict can be distinguished along these three dimensions and also show the different emotional reactions and resulting behaviors associated with different causal attributions.
Causal attributions, emotional reactions, and behaviors.
First, the causes of work–family conflict can be either stable or unstable. When causes are perceived as stable (i.e., not likely to change in the future), they are also uncontrollable, and whether their locus is internal or external has little influence on the emotional consequences of the attribution. The stability of an outcome relates to “the magnitude of expectancy change following success or failure” (Weiner, 1979, p. 8). When individuals perceive the causes of negative outcomes to be stable, their expectations for future positive outcomes will be low. Therefore, they are likely to experience hopelessness after an attribution for a negative outcome to a stable cause (e.g., Weiner et al. 1979). As Weiner (1985) described the emotional consequences of stable causal attributions, “if the future is anticipated to remain as bad as the past, then hopelessness is experienced” (p. 563). Thus, we propose:
When the attribution for work–family conflict involves unstable causes, the ascribed causes can be internal or external, and also controllable or uncontrollable (see Figure 2). When employees attribute the occurrence of work–family conflict to internal causes, they are likely to experience either shame or guilt (Tracy & Robins, 2006), self-conscious emotions “that are evoked by self-reflection and self-evaluation” (Tangney et al., 2007, p. 347). Shame and guilt are also considered moral emotions, because their experience follows past behavior or action that was inconsistent with some set of standards that is often moral of nature (Tangney, 1990). Guilt is considered “more moral” than shame because guilt is an adaptive emotion associated with reparative action tendencies that can benefit individuals and their relationships (Tangney, 1990; Tangney et al., 2007), a point to which we return in a subsequent section.
According to the appraisal model of self-conscious emotions (e.g., Tracy & Robins, 2004), internal attributions underlie both shame and guilt. However, shame is elicited when individuals consider the causes of the outcome to be uncontrollable. When individuals consider the causes of outcomes to be controllable, they experience guilt (Tracy & Robins, 2006). Hence, an important distinction between guilt and shame concerns the focus on behavior (guilt) versus self (shame). Based on the above, work–family conflict should elicit guilt when it is attributed to a controllable internal cause (e.g., lack of effort or planning in dealing with work demands), whereas work–family conflict should elicit shame when it is attributed to an uncontrollable internal cause (e.g., difficulty in dealing with work demands because of low ability). Indeed, guilt and shame have been found to be related to internal-controllable and internal-uncontrollable attributions, respectively, in previous research (e.g., Covington & Omelich, 1984). Based on the theoretical arguments and empirical evidence reviewed above, we propose:
Whereas internal attributions produce the self-conscious emotions shame or guilt (depending on the extent to which the cause is seen as controllable or not), external attributions induce non-self-conscious emotions such as frustration and anger (e.g., Poposki, 2011; Weiner, 1985). External attribution elicits frustration when the external cause is seen as uncontrollable and anger when the external cause is conceived of as controllable (e.g., Averill, 1983; Weiner, 1985). Frustration is typically associated with the experience of having one’s goals (balancing work and family roles, in our case) blocked by an external cause that cannot be controlled. Anger concerns, more than anything else, the attribution of blame (Averill, 1983) and is associated with an event blocking one’s goals that appears to be caused intentionally by a responsible other (e.g., Roseman et al., 1994; Tangney et al., 2007). If work–family conflict is induced by unreasonable (objective) work demands—an external cause because it is managed by others (e.g., one’s supervisors)—that prevent the employees from fulfilling their family roles effectively, they will experience anger or frustration, depending on whether they perceive that the external party (e.g., their supervisors) can change the cause (i.e., job demands) as to diminish work–family conflict.
In the first six propositions we specified our expectations with respect to the differentiated emotional reactions elicited by work–family conflict either directly (attribution-independent emotions) or indirectly, through causal attributions. Earlier, in the first part of this article, we contended that it is important to understand such differential emotional reactions to the experience of work–family conflict (and the attribution for its occurrence) because specific emotions are associated with different action tendencies and thus elicit different behaviors. In the following section we present propositions specifying the differentiated emotion–behavior relationships induced by work–family conflict and the attributions about its causes.
Behavioral actions induced by emotional reactions to work–family conflict
We distinguished five discrete emotional reactions that, depending on the causal attributions individuals make, may follow the experience of work–family conflict: (a) hopelessness, when the causes of work–family conflict are seen as stable, (b) guilt, when work–family conflict is attributed to unstable, internal, but controllable causes, (c) shame, when work–family conflict is attributed to unstable, internal, and uncontrollable causes, (d) anger, when work–family conflict is attributed to unstable, external, but controllable causes, and (e) frustration, when work–family conflict is attributed to unstable, external, uncontrollable causes. In what follows, we address the different behavioral reactions to work–family conflict employees may engage in, depending on their causal attributions for it and their experienced, discrete emotions.
Hopelessness
Stable causal attributions for work–family conflict imply acknowledging that one cannot influence its causes. As we have delineated in Proposition 2, following Weiner (1985), we believe that when employees ascribe stable causes for the occurrence of work–family conflict, they will experience the attribution-dependent emotion of hopelessness. Hopelessness can be defined as “an expectation for failure or lack of positive beliefs about future orientations” (Gonzalez, Jones, Kincaid, & Cuellar, 2012, p. 110).
Although research on organizational behavior hardly addressed hopelessness, research in other areas has indicated that hopelessness increases the risk for maladaptive behaviors. For instance, research in the field of psychiatry has consistently shown that hopelessness relates to suicide among individuals diagnosed as having mood disorders, schizophrenia, and bipolar disorder (e.g., Acosta et al., in press; Gray & Otto, 2001; Hor & Taylor, 2010). Research on cultural diversity has shown that hopelessness negatively relates to psychological adjustment of youths, as reflected in higher levels of anxiety, depression, and withdrawal and positively relates to aggressive and delinquent behaviors (Gonzalez et al., 2012).
Hopelessness has also been associated with a present-fatalistic time perspective (e.g, Boniwell & Zimbardo, 2003). Time perspective as a general construct is described as the relative temporal focus that (often nonconsciously) guides and influences individuals’ decisions, actions, and goals (e.g., Zimbardo & Boyd, 1999) and comprises five perspectives: a past-positive perspective, a past-negative perspective, a future perspective, a present-hedonistic perspective, and a present-fatalistic perspective. The present fatalistic time perspective has been defined as “a fatalistic, helpless, and hopeless attitude toward the future and life” (Zimbardo & Boyd, 1999, p. 1275). Individuals with a present-fatalistic time perspective are characterized by general pessimism and self-destructive behaviors (Henson, Carey, Carey, & Maisto, 2006) and the perception of having no influence over their own lives (Zimbardo & Boyd, 1999). Most research on time perspective combined the present-fatalistic with the present-hedonistic perspective (or only studied effects of the present-hedonistic perspective), and showed that a present time perspective relates to maladaptive behaviors such as the use of alcohol and illegal drugs, smoking (Keough, Zimbardo, & Boyd, 1999), gambling (Hodgins & Engel, 2002), and risky driving (Zimbardo, Keough, & Boyd, 1997). Research that specifically focused on a present-fatalistic perspective indicated, however, that individuals holding this perspective lack goals, have low levels of energy, lack the need for excitement, and do not experience bitterness (Zimbardo & Boyd, 1999), suggesting that they might not engage in risky behaviors, but rather may do nothing but resign oneself to the situation. Indeed, Henson et al. (2006) found no significant relationships between a present-fatalistic perspective and proactive risky health behaviors (i.e., substance use, number of sexual partners), but significant negative relationships between a present-fatalistic perspective and protective health behaviors such as seat belt use and the use of birth control.
The above suggests that feelings of hopelessness are likely to reduce active coping behaviors and instead result in resignation and withdrawal (Weiner, 1985). We therefore propose that when employees feel hopeless and are convinced that they cannot change the demands that produce work–family conflict, they will either further withdraw from the family (especially when hopelessness is accompanied by shame) or withdraw from the job (i.e., absenteeism, turnover) that imposes the unreasonable and uncontrollable demands, in an attempt to regain some control over their life.
Shame and guilt
As we proposed earlier on, when employees attribute the occurrence of work–family conflict to internal causes, they are likely to experience either shame or guilt. Shame and guilt are both moral, self-conscious emotions (Tangney et al., 2007) that have a strong impact on individuals’ behavior regulation (Scarnier, Schmader, & Lickel, 2009). Of these two emotions, guilt, which is induced by attribution to internal causes that are perceived as being controllable, is more adaptive than shame, which is induced by attribution to internal causes that are perceived as uncontrollable (Tracy & Robins, 2006).
Shame
When individuals experience shame, they focus on the self; their deep-seated flaws (Scarnier et al., 2009) and the unfavorable characteristics of themselves (Tracy & Robins, 2006). Because “one’s core self—not simply one’s behavior is at stake” (Tangney et al., 2007, p. 349) shame is considered a painful and maladaptive emotion that is accompanied by feelings of worthlessness, powerlessness (Tagney et al., 2007), and the feeling of being exposed to and scrutinized by others (Smith, Webster, Parrott, & Eyre, 2002). Shame induces the desire to defend the self (Tangney, Stuewig, & Hafez, 2011) and to detach oneself from the negative outcome or event (e.g., Schmader & Lickel, 2006; Tangney, Miller, Flicker, & Barlow, 1996). Shame also promotes defensiveness, separation, and distance (e.g., Lindsay-Hartz, 1984; Morris & Keltner, 2000; Tangney, 1991; Tangney et al., 2007), and stimulates maladaptive behaviors such as denying the shame-inducing event or hiding or escaping from the unfavorable situation (Tangney et al., 2011).
Extending these findings to work–family conflict, when employees feel shame for not being able to fulfill their family roles, they will likely further withdraw from family activities in an attempt to hide their inability of balancing their work and family roles and separate themselves, psychologically, from their family role. Such family withdrawal can include behaviors such as avoiding social interactions at home, not paying attention to one’s family members, or appearing bored and uninterested in family activities (e.g., Repetti, 1989). Hence:
Guilt
Similar to experiencing shame, experiencing guilt is painful. However, because individuals who experience guilt focus on the things they did (or did not do), that is, on the “negative aspects of their behavior” (Tracy & Robins, 2006, p. 1340) rather than on the negative aspects of their core self (Tangney et al., 2007), guilt is considered to be less disruptive and leads to more adaptive behaviors than shame (Tangney et al., 2011). Individuals who feel guilty are likely to ruminate over the negative outcome, wishing they had acted in a different way. Because they reconsider the behaviors that caused the negative outcome, they do not feel the need to defend themselves as those individuals experiencing shame. Instead, they are motivated towards prosocial and reparative behaviors, such as apologizing, undoing, and repairing the damage (Tangney et al., 2011). Hence, from a functionalist perspective, guilt can be seen as a constructive emotion that promotes good relationships with others at work (i.e., one’s employer and colleagues) and in the family by spurring efforts to repair the harm that one has done, once one realizes that such harm occurred (Morris & Keltner, 2000; Tangney et al., 2007).
Indeed, there is convincing empirical evidence supporting the proposition that experiencing guilt motivates reparative-oriented behaviors (e.g., Roseman et al., 1994; see Tangney et al., 2007, for a recent review). Extending the above to work–family conflict, guilt is likely to promote behaviors aimed at reducing the negative impact of high job demands on one’s performance in one’s family role. This can be accomplished by better managing the demands at work through time management or self-development efforts (Baltes & Heydens-Gahir, 2003).
Frustration and anger
As we argued earlier on, when employees attribute the occurrence of work–family conflict to external causes, they are likely to experience either frustration or anger. Both anger and frustration are outward-focused emotions that are elicited when individuals consider other people and/or other people’s behaviors as the cause of a negative outcome (Tangney & Dearing, 2002).
Frustration
We proposed that employees are likely to experience frustration following an external and uncontrollable attribution for the occurrence of work–family conflict (Proposition 5). According to Roseman et al. (1994), when feeling frustrated individuals sense that they are blocked and think about what blocks them. They want to get past the obstacles and will take vigorous action to overcome them. Two possible actions employees could take to decrease work–family conflict and frustration would be to craft their job or to withdraw from their job.
Wrzesniewski and Dutton (2001) defined job crafting as “the actions employees take to shape, mold, and redefine their jobs” (2001, p. 180) and distinguished three forms of job crafting. Job crafting involves the active redesigning of one’s job by (a) changing the physical boundaries of the job (e.g., taking on fewer or different tasks, changing work procedures), (b) changing the relational boundaries of the job (e.g., changing the quality or quantity of interactions with others), and (c) changing how the tasks are cognitively perceived (e.g., changing the meaningfulness of the work). Engaging in job crafting can be considered an adaptive response to work–family conflict. By actively crafting their job, employees change the locus of the cause from others to self, thereby increasing their control over the processes that influence work–family balance, which should alleviate frustration.
Rather than effectively coping with work–family conflict and frustration, frustrated employees could also withdraw from their work activities. Work withdrawal refers to behaviors that “restrict the amount of time working to less than is required by the organization” (Spector et al., 2006, p. 450) and includes behaviors such as missing meetings, arriving too late or leaving work early, and being absent from work. Withdrawal is a maladaptive reaction to work–family conflict and engaging in withdrawal behaviors is potentially costly for employees themselves (Hanisch & Hulin, 1991). However, withdrawal might be an instrumental means of coping with work–family conflict (Koslowsky, 2000), as withdrawal will also result in alleviated frustration because job demands are less likely to interfere with employee’s family activities when the employee does not attend to the job demands.
Anger
When employees attribute work–family conflict to external, controllable causes, they are likely to experience anger (Proposition 6). Roseman et al. (1994) summarized how participants in their study recalled anger experiences: they “felt blood rushing through the body and felt as if they would explode, thought about how unfair something was and of violence toward other people, felt like yelling and like hitting someone, said something nasty, and wanted to hurt and get back at someone” (1994, pp. 212–213).
Anger is a negative emotion that often concurs with aggressive behavior (e.g., Glomb, 2002; Roseman et al., 1994; Rubin, 1986). The relationship among and the distinction between anger and aggression are complex (Glomb, 2002). However, the scope of this article prevents us from a full consideration of this issue. For our purposes we adopt the view that workplace aggression, in its various forms, is an expression of the feeling of anger (e.g., Glomb, 2002; Roseman et al., 1994). This view is consistent with our main thesis that specific attribution-dependent emotions have differential behavioral consequences. Given the range of possible work behaviors associated with anger, we propose that anger may result in both intentional aggressive work behaviors (Glomb, 2002) and in workplace incivility, that is, in low-intensity interpersonal mistreatment enacted with ambiguous intent to harm the target (e.g., Andersson & Pearson, 1999; Cortina, Magley, Williams, & Langhout, 2001). Both intentional aggressive work behaviors and workplace incivility are maladaptive behaviors. They do not decrease the experience of work–family conflict and they both have important negative consequences for organizations and individuals alike (e.g, Glomb, 2002; Lim, Cortina, & Magley, 2009).
In summary, our attributional model of work–family conflict proposes that work-family conflict is likely to trigger attribution-independent negative emotions (sadness, dissatisfaction, and unhappiness) directly (see Figure 1). Our model furthermore suggests that individuals’ specific emotional and behavioral reactions to work–family conflict will vary as a function of the attributed causes of the work-family conflict. As depicted in Figure 2, when employees attribute the occurrence of work–family conflict to stable causes that are inherently uncontrollable, they are likely to experience hopelessness and withdraw from their work and/or their family in order to regain some control over their lives.
When employees attribute the occurrence of work–family conflict to unstable causes, such causes can be internal (self) or external (others), as well as controllable or uncontrollable. An internal uncontrollable attribution will induce shame, which can trigger defensive behaviors such as family withdrawal, while an internal controllable attribution will produce feelings of guilt, which can motivate adaptive behaviors such as time management and self-development. An external uncontrollable attribution is likely to create frustration, which can be alleviated by adaptive behaviors such as job crafting or maladaptive behaviors related to work withdrawal. Finally, an external controllable attribution is likely to produce anger, which can lead to maladaptive, uncivil, and aggressive behaviors.
Discussion and directions for research
Our proposed model addresses two important gaps in theory and research on work–family conflict. First, our model focuses on specific emotions rather than broad affective reactions. This is especially important given the differential impact specific emotions have on individuals’ cognition, choices, actions, and behaviors (Frijda, 1986; Gooty et al., 2009; Izard, 2009; Roseman et al., 1994; Tangney, 1991; Tangney, et al., 2007). Second, our model proposes the ultimate effects of work–family conflict on decisions, actions, and behaviors, through these emotions, which have, so far, largely been ignored. Now that we have presented both the model and the rationale behind it, there are several issues that need to be addressed: the role of individual-difference variables and of the job context, the implications of causal attributions for work–family conflict for outcomes related to employees’ well-being, and the outcomes of attribution-independent emotions (sadness, dissatisfaction, unhappiness).
Individual dispositions and job context
With regard to the attribution of external causes for work–family conflict, our propositions have focused on the controllability of such causes by external parties (e.g., a supervisor). However, it is possible that some people perceive external causes as self-controllable, for instance, individuals who are high on proactive personality and internal locus of control. Arguably, the extent to which individuals perceive external causes as self-controllable might potentially affect their subsequent behaviors. For example, work–family conflict could be caused by the inherently demanding nature of a specific project at work (e.g., tight deadlines, limited resources). However, if the employee perceives that he or she can impact the content of his or her work, the employee will be more likely to engage in adaptive behaviors such as job crafting. While this is a possibility that can be explored in future studies, previous research (De Pater, van Vianen, Bechtoldt, & Klehe, 2009) has indicated that most work tasks or projects are assigned by superiors, and employees are unlikely to possess full autonomy in choosing the types of assignment that they prefer. Thus, the probability of employees making such external self-controllable attributions might be relatively low in many job contexts.
Furthermore, while our proposed model suggests specific links between attributed causes for work–family conflict and emotions and behaviors, we do not claim that such links are invariant across individuals (and contexts). For example, individuals may attribute work–family conflict to a lack of ability to complete certain tasks (an unstable, internal, and uncontrollable cause), but not feel shame, depending on, for instance, their self-esteem. Indeed, past research has revealed that individuals differ in their proneness to experience emotions such as guilt and shame (e.g., Tangney 1990, 1991), and we believe that such individual differences in the propensity to experience certain emotions are likely to moderate the proposed relationships between attributed causes and the emotions associated with different causal attributions. In a similar vein, individuals may attribute work–family conflict to a personal lack of effort at work (an unstable, internal, and controllable cause), and not feel guilt, due to the norms and values of the people around them.
Also, individual characteristics and contextual variables could affect the links between the emotions and behaviors associated with each causal attribution. For example, individuals may differ in their preferred coping styles in response to stressful situations (e.g., Carver, Scheier, & Weintraub, 1989), and such coping styles can impact their behavioral responses to work–family conflict. Individuals who feel frustrated due to work overload might be more likely to engage in job crafting if they are task-oriented in their coping style, but the same frustration could lead to work withdrawal behaviors if they are avoidance-oriented (Endler & Parker, 1990). Situational constraints such as job characteristics and cultural norms can also play a role in moderating the relationships between the emotions and behaviors associated with work–family conflict. For example, individuals who are angry with coworkers who are loafing on the job are likely to engage in aggressive or uncivil behaviors at work only if such behaviors are tolerated in the workplace.
Causal attributions for work–family conflict and well-being outcomes
Due to the focus of this paper, we did not examine links between the causal attribution process for work–family conflict and employee well-being outcomes. We do, however, believe that the differential emotions and behaviors elicited by specific causal attributions have important implications for employee well-being. As we noted before, meta-analytic research has shown that increased work–family conflict is associated with stress-related outcomes such as depression, burnout, physical health problems, and alcohol abuse (Allen et al., 2000). We suspect that such negative outcomes are likely to be exacerbated by maladaptive actions such as family or work withdrawal. On the other hand, adaptive actions such as job crafting, self-development, and family engagement are likely to reduce the negative effects of work–family conflict on employee well-being. In other words, we propose that certain causal attributions are useful in alleviating the negative outcomes of work–family conflict on employees’ well-being. In particular, employees who perceive work–family conflict as arising from unstable, internal, and controllable causes (e.g., lack of effort) are most likely to cope by engaging in adaptive behaviors while those who perceive work–family conflict as arising from internal, stable, uncontrollable causes (e.g., lack of ability) are most likely to adopt maladaptive behaviors. This also has important practical implications for employees who experience work–family conflict. As the causal attribution for work–family conflict is largely a subjective process, interventions aiming to help employees cope with it can focus on encouraging employees to engage in a more effortful search for possible internal and controllable causes. This would make it more likely for employees to perceive that they can do something to improve the situation and thus engage in adaptive behaviors such as those specified in our model.
Outcomes of attribution-independent emotions
Because our main focus is on identifying the outcomes of attributions for work–family conflict, we did not offer formal propositions for the outcomes of attribution-independent emotions (sadness, dissatisfaction, unhappiness). Nonetheless, these emotions can have important implications for how employees respond to work–family conflict. For example, individuals who experience sadness in response to work–family conflict might feel like doing nothing and desire to be comforted (Roseman et al., 1994), while those who experience dissatisfaction might seek to proactively change the situation. Such attribution-independent emotions can also occur in tandem with attribution-dependent emotions to result in certain behavioral tendencies (e.g., a wider range of withdrawal behaviors as a result of sadness and shame). Indeed, it is also possible for employees to attribute the experience of work–family conflict to more than one cause (e.g., internal and external), which can then trigger multiple emotions. Thus in addition to considering potential moderators of the relationships proposed in our model, future research could also explore the behavioral implications of experiencing a combination of different causal attributions and emotions.
Future research
We believe it is important to rigorously test our model, in order to advance knowledge on behavioral reactions to work–family conflict and the underlying mechanisms that explain these behavioral reactions. Not only may this knowledge better predict employees’ work behavior, from a practical perspective, but this knowledge may be applied to prevent work–family conflict (for instance by encouraging job crafting) or to develop interventions that alleviate its negative consequences.
To test our model, there is a need to measure the experience of work–family conflict, the attributed causes, as well as the resulting emotions and behaviors. While behaviors can be assessed by both self-report and peer or spouse ratings, the other constructs might be best assessed using a self-report questionnaire due to their inherently subjective nature.
As highlighted by Allen et al. (2000), most of the current research on work–family conflict has been cross-sectional, and thus not able to track changes in the effects of work-family conflict over time. We therefore underline the need for longitudinal research that allows researchers to measure changes in the experiences of work–family conflict and causal attributions, as well as to monitor the corresponding changes in the emotional and behavioral outcomes over time. Changes over time can also be measured over a short time period using the experience-sampling or diary approach (e.g., Ilies, Schwind, Wagner, et al., 2007), which allows one to capture the employees’ experience of work–family conflict and its associated outcomes on a daily basis. Finally, as mentioned, we do believe that individual characteristics and contextual variables may influence how individuals feel and react when they experience work–family conflict. Future research should also address these issues.
Conclusion
In sum, we hope that this essay will stimulate research on attributions for work–family conflict and help advance the literature on this topic. As previous research on work–family conflict has tended to focus on negative outcomes, our proposed model provides a new perspective by suggesting that the experience of work-family conflict can potentially result in positive adaptive behaviors as a function of the emotions associated with the causal attribution for work–family conflict. We also hope that understanding the link between the attributed causes of work–family conflict and their emotional and behavioral outcomes will provide some practical insights for helping employees cope effectively with work demands and situations that interfere with their family lives.
Footnotes
Funding
This research was funded by a start-up grant, R317-000-098-133, provided by the National University of Singapore to Remus Ilies.