Linking Work Control to Depressive Symptoms Through Intrapersonal and Marital Processes

Abstract

Using latent growth curve modeling and data from a sample of 308 middle-aged husbands and wives over a 3-year period, this study examines the influence of work control experiences on marital processes, specifically focusing on changes in conflict management behaviors, perceived spousal support, and depressive symptoms, and the connection between these attributes over time. Consistent with human ecological and work socialization perspectives, our findings suggest that adults’ positive experiences at work serve as a resource for developing personal and interpersonal skills that transfer to behaviors exhibited in their marriages and this process consequently contributes to individual depressive symptoms. Husbands’ and wives’ positive work experience such as control over work increases their use of positive conflict management behaviors and spousal support through personal control, which in turn decreases their depressive symptoms over time.

Keywords

work and family family health spousal support marital relationships family processes

An individual’s work experience has implications for both quality of individual and family life (Grzywacz & Marks, 2000; Menaghan, 1991). In particular, positive work experience has been linked to improved quality of family life and increases in individuals’ general life satisfaction and less depression (Grzywacz & Bass, 2003). Furthermore, higher work control is associated with effective parental practices (Whitbeck et al., 1997) and marital interactions (Grzywacz & Marks, 2000). These findings suggest that values, behaviors, or skills acquired at work affect the way individuals interact and behave in the family (Crouter, 1984). In addition, psychological control and positive marital attributes link aspects of a person’s occupation to their health (Wickrama, Lorenz, Conger, Matthews, & Elder, 1997).

However, little research has investigated specific pathways linking experiences in both the work and family domains, or the combined influence of these experiences on individuals’ health over time. In this study, we aim to enhance the understanding of the linkages between work, marriage, and health by investigating the pathways between what occurs in the work and family domains and the ultimate implication for individuals’ depressive symptoms. We elaborate how the effects of individuals’ work experiences (i.e., work control) influence intrapersonal process (i.e., personal control) and interpersonal process (i.e., conflict management behaviors and spousal support) and, in turn, affect depressive symptoms. The associations between the levels and changes in the time-varying attributes such as conflict management behaviors, spousal support, and depressive symptoms may be parallel. That is, the levels and changes in these attributes are correlated among each other resulting in parallel trajectories over time. These parallel trajectories are known as “interlocking trajectories” (Wickrama, Beiser, & Kaspar, 2002).

Evidence suggests that positive synergy from work to family increases across the adult life course because an individual’s accumulated work experience and expertise in jobs can maximize this synergy (Grzywacz, Almeida, & McDonald, 2002). As such, we expect marital relationships to change over time as a consequence of the change of work contexts (Umberson, Williams, Powers, Chen, & Campbell, 2005). In turn, changes in marital relationships will contribute to change in health over time (Umberson, Williams, Powers, Hui, & Needham, 2006). Traditional analytical approaches that often examine average individuals’ changes over time and the covariations or correlations between them generate limited information and unlikely reveal complexity of individual changes (Karney & Bradbury, 1995). Thus, some researchers suggest that trajectories are the proper avenues to analyze change because trajectories explicitly focus on individuals’ growth/decline over time and examine a correspondence between individual trajectories within the same person and across respondents (Rogosa & Willett, 1985). Growth curve analysis allows us to examine the interlocking individual trajectories over time providing strong evidence for the associations between those time-varying attributes (Duncan, Duncan, & Strycker, 1999).

Using prospective longitudinal data and growth curve analysis, the current study examines (a) individual trajectories of conflict management behaviors, spousal support, and depressive symptoms over time; (b) the associations reflecting “interlocking” trajectories of those attributes over time; and (c) the influence of antecedents such as work control and personal control on this parallel processes (Wickrama et al., 2002). The present study responds to a call to shift from the examination of only self-report measures and global evaluations of one’s marriage to multiple dimensions of marital quality (Leach & Butterworth, 2012). We investigate husbands’ and wives’ reports of their spouses’ conflict management behaviors and their self-reported perceived spousal support in the marital relationship to examine the positive spillover process of work–family influencing depressive symptoms.

Theoretical Background

This study is rooted in the blending of several theoretical perspectives: human ecology, positive spillover framework, and work socialization. A human ecological perspective, which reinforces the importance of examining how individuals interact within and across multiple systems (Bronfenbrenner & Ceci, 1994), offers important insights to understanding how socioeconomic contexts shape individuals’ and families’ behaviors, beliefs, and lifestyles. The linkages and cross-domain process between multiple systems can occur through resource generation and spillover process (Voydanoff, 2004). Drawing from a work socialization perspective, it was posited that psychological resources at work, which are beneficial for personal growth can enhance worker’s cognitive skills (Kohn & Slomczynski, 1990; Schooler, Mulatu, & Oates, 2004). Specifically, self-directive work experience associated with work control spills over to greater sense of personal control (Wickrama, Surjadi, Lorenz, & Elder, 2008) as well as to a greater interpersonal problem solving in family life (Menaghan, 1991). Thus, work contexts may shape individual marital attributes through the generation of individual resources such as personal control. This may help husbands and wives manage marital conflicts better, feel supported by each other, and ultimately lead to less depression. The conceptual model of this study is shown in Figure 1, and further described below.

Figure 1.

Conceptual model.

Work Control, Personal Control, and Marital Interactions: Intrapersonal Processes

Work socialization research has focused on the psychosocial work environment and investigated the positive associations between the psychosocial characteristics of work and family life (Grzywacz & Marks, 2000). One such psychosocial work characteristics, work control, has been shown to be a fairly strong and consistent predictor variable affecting individuals’ psychosocial resources (Grzywacz & Bass, 2003; Karasek & Theorell, 1990) and their family life. Work control (Karasek & Theorell, 1990) is composed of two dimensions, including skill discretion, which are the various opportunities to use one’s skills adequately, and decision authority, which is the amount of control one has over how work tasks are addressed and accomplished. Consistent with a work socialization perspective, a high-control work environment fosters individuals’ involvement in joint decision-making processes which promotes a flexible orientation toward society, self, and circumstances (Schooler et al., 2004). In addition, if individuals have opportunities and successful experiences in using their abilities or skills and self-directiveness at their work, the accumulation of these experiences contributes to their perceptions of mastery and an enhanced sense of empowerment (Kohn & Slomczynski, 1990; Wickrama et al., 2008). These cognitive and behavioral processes are likely to shape individuals’ beliefs related to self-worth and personal control, which are key elements of productive and successful adult development (Kohn & Slomczynski, 1990). Wickrama et al. (2008), using longitudinal data over 10 years, showed that changes in work control influences changes in sense of control, which in turn influence changes in individual physical health of middle-aged men. Consistent with prior research, we posit that if individuals accomplish their work tasks successfully using their authority and adequate skills, this would lead to a greater sense of personal control.

Work-influenced changes in cognitive abilities can be displayed in other life settings (Menaghan, 1991). Crouter (1984) argued that positive attitudes, beliefs, and behavioral orientation acquired from work experience carry over to the family environment. In other words, individuals’ gained work experience contributes to their capacity to navigate different situations and engage in intellectually flexible (e.g., joint decision making with spouse) and socially flexible (e.g., positive interactions with spouse) behaviors to solve challenges in the family environment. We expect that experience of high work control enhances one’s personal control and in turn may alter the ways couples interact with each other and develop one’s ability to manage conflict constructively at home. Ultimately, this process enhances their interactions over time which, in turn, likely improves their perceptions of spousal support.

Marital Interactions and Depressive Symptoms: Marital Processes

Husbands and wives who engage in more constructive problem-solving behaviors, which are, listen to their partner’s ideas, try to see their partner’s point of view, and compromise to help solve the problem, are better able to handle disagreements and marital difficulties (Rueter & Conger, 1995). Specific negative conflict management behaviors, such as criticism, defensiveness, and withdrawal, result in a rapid rate of deterioration in couples’ marital satisfaction (Johnson et al., 2005). Moreover, constant negative marital interactions can serve as a chronic life stressor putting husbands and wives at greater risk for developing depression (Kiecolt-Glaser & Newton, 2001) and accelerate individuals’ negative health outcomes over time (Umberson et al., 2006).

Interestingly, some research has documented that subjective reports on spousal interactions or support have a strong influence on depressive symptoms (Schafer, Wickrama, & Keith, 1996). How husbands and wives perceive and interpret their spouses’ disconfirming behaviors has a greater impact on marital happiness and depressive symptoms. Negative and disconfirming behaviors may affect one’s perceptions of spousal support, which in turn may lead to greater risk for depressive symptoms. Additionally, individuals with high job demands and low job control reported less positive support from their partners in marriage-like relationships (Leach & Butterworth, 2012).

Work Control, Personal Control, and Depressive Symptoms

Extant research supports additive effects of work control on psychological distress (Dollard et al., 2012). Individuals who have lower levels of skill discretion are more likely to experience high emotional exhaustion (Rafferty, Friend, & Landsbergis, 2001), and lower levels of decision latitude at work is associated with poor mental health such as fatigue, depression, and anxiety (Griffin, Fuhrer, Stansfeld, & Marmot, 2002). In contrast, individuals’ work control is positively associated with their personal resources, which in turn decreases the risk of depression (O’Neal, Wickrama, & Bryant, 2014). Therefore, we posit that individuals’ higher work control experience is associated with their depressive symptoms through their personal control.

Gender Differences

Research suggests that marriage and marital processes may have unique influences on husband and wives. For example, men receive more psychosocial benefits from marriage than women (McRae & Brody, 1989). And, whereas marriage itself seems beneficial for men, marital quality is more crucial for women’s psychosocial well-being (Williams, 1988). In fact, a stronger association between marital satisfaction and depressive symptoms has been found for wives as compared with husbands (Bradbury, Beach, Fincham, & Nelson, 1996). In general, women have higher expectations for intimacy within marriage (Williams, 1988) and a strong desire to be connected to others when they are under conditions of stress (Taylor et al., 2000). This tendency may explain why women can be more vulnerable to their stressful marital interactions especially when they most need to be supported.

Similarly, work-related stress seems to have a more substantial and negative influence on marital satisfaction for wives than for husbands (Repetti, Wang, & Saxbe, 2009). In addition, wives who report experiencing less control either at work or in their family relationships are at highest risk for depression and anxiety (Griffin et al., 2002). As gender likely enhances the associations between work, marriage, and depressive symptoms, the present study further explores variations or similarities between husbands and wives.

The Current Study

The present study elucidates hypothesized spillover processes by examining the interlocking or parallel trajectories among conflict management behaviors (influenced by work control through personal control), spousal support, and depressive symptoms. We contend that such an analysis of change over time will provide compelling evidence for the dynamic associations between these attributes. We hypothesized that husbands’ and wives’ work control through increasing personal control be (a) positively associated with their engagement in conflict management behaviors, (b) positively associated with perceived spousal support, and (c) negatively associated with depressive symptoms. Using longitudinal data and latent growth curve models to estimate individual trajectories of change in these variables, we hypothesized that the level and rate of change in conflict management behaviors will increase the level and rate of change in marital quality, which in turn will decrease the level and rate of change in depressive symptoms. Considering gender moderation in the hypothesized relationships, we expect to find a stronger association among conflict management behaviors, spousal support, and depressive symptoms for wives when compared with husbands. In addition, stressful work experience related to less work control will have stronger consequences for wives’ as opposed to husbands’ marital experiences and depressive symptoms.

Method

Sample

The data used to examine our theoretical model (Figure 1) are from the Iowa Youth and Family Project, a longitudinal study examining family adaptation to the Farm Crisis of the 1980s in Iowa. The study seeks to identify family interactional dynamics that influence how families and children adjust to rural economic stress.

From the larger sample of 424 couples, we only selected employed husbands and wives in 1990 (n = 368; 87%) who also both provided complete data on the variables of interest in 1990 (W1), 1991 (W2), and 1992 (W3). This reduced the sample examined in the present study to 308 (76%) husbands and wives. Comparisons between our final sample (n = 308) and those excluded (n = 60; i.e., unavailable, discontinued participation in this study, or experienced unemployment between 1990 and 1992) revealed no significant differences in respondent characteristics documented at W1 (e.g., years of education and annual family income). In 1989, at the time of the first wave of data collection, the median ages for the husbands, the wives, and their adolescent children were 39, 37, and 12 years, respectively. Median years of education for both the husbands and wives was 13 years, which was comparable to the median years of education for Caucasians between 35 and 44 years of age in the United States in 1989 (U.S. Census Bureau, 1990). Couples in the sample had been married for at least 14 years. In 1989, 97% of the husbands and 78% of the wives were employed and most (97%) of husbands and 50% of the wives were full-time workers. Regarding their occupational status, husbands were employed as professionals, managers, and officials (33.6%), technicians or skilled workers like craftsmen or operatives (29.8%), farmers (22%), labors (5.5%), and clerical/sales workers and others (9.1%). Wives reported their occupational status as professionals, managers, or owners (38.5%), clerical or sales workers (33.2%), household workers (18.9%), and other (9.4%). Because specific occupational status such as irregular/regular work hours, the number of work hours, or seasonal/temporary employment may be associated with the study variables, they are considered as control variables in this study. The mean annual family income was about $40,000 and 11% of the families had an annual income below the federal poverty line in 1989. In terms of residency, 34% of the families lived on a farm, 12% resided in the countryside but not on a farm, and 54% lived in small towns (5,000 people or less).

Procedures

The aim of the Iowa Youth Family Project was to examine the effects of economic stress on rural families’ experiences and well-being over time (Conger & Conger, 2002). The sample selected from a rural and predominantly agricultural eight-county area in North Central Iowa included families with at least two children, one in seventh grade and a near sibling, and two parents living together at the beginning of the study in 1989. The qualified families were recruited from the 34 public and private schools across in these areas. Families received a letter explaining the project and were asked to participate over the phone. Approximately 78% of eligible families who met the strict criteria for inclusion in the study agreed to participate in the first wave of the study. Since racial/ethnic minorities rarely reside in these rural areas, all participating families were Caucasian. Each family was visited each year by research staff. During the visit, trained field interviewers asked each family member to complete a questionnaire package independently.

Although the data used in this study reflect the experiences of rural-dwelling families during a specific economic era (Conger & Conger, 2002), they share pivotal characteristics with what generally happens to individuals, families, and communities in economic hard-times. This includes uncertainty about the future, strains on family relationships, drastic or permanent lifestyle changes, challenges to individual emotional health, as well as physical health. At the community level, effects are evident in strains on small businesses, public services, and overall quality of life. So, while we do not promulgate the idea that our Iowa data have a one-to-one correspondence with any other era, on the other hand, how families are affected is remarkably similar. There are other important constants in people’s lives that are reflected in this current analysis, including how conflict is managed, where work fits into the quality of family life, and how individual mental well-being is influenced; each of these has relevance across economic eras.

Measures

Work Control

In 1990, six items (Conger, 1988) were used to assess participants’ feelings about their work conditions ranging from skill utilization (e.g., “My job allows me to use my skills and abilities”) to authority over work and decision latitude (e.g., “My job matches what I like to do”). Responses ranged from 1 (strongly agree) to 5 (strongly disagree), and a mean score was computed such that higher scores reflect greater work control (α = .76 and .75, for husbands and wives, respectively).

Personal Control

In 1990, participants responded to four items from Pearlin’s Mastery Scale (Pearlin, Menaghan, Lieberman, & Mullan, 1981). Examples of items include “Sometimes I feel that I am being pushed around in life” and “I have little control over the things that happen to me.” Responses ranged from 1 (strongly agree) to 5 (strongly disagree) and were coded and averaged such that higher scores indicated higher levels of mastery (α = .62 and .69, for husbands and wives, respectively).

Conflict Management Behavior

In 1990, 1991, and 1992, six items from the family problem-solving questionnaire (Conger, 1988) were used to ask each spouse about their spouse’s behaviors during conflict (e.g., How often your partner “listen[s] to your ideas how to solve the problem?” “Show[s] a real interest in helping solve the problem?”). Responses ranged from 1 (always) to 7 (never) and were coded and averaged so that higher scores reflected more frequent use of constructive conflict management behaviors (α = .82 to .92 across years).

Spousal Support

In 1990, 1991, and 1992, spouses were asked four items about marital interpersonal sensitivity (Conger, 1988). Sample items included “How much does she or he show concern for your feelings and problem?” and “How much would you say she or he understands the way you feel about things?” Responses ranged from 1 (a lot) to 4 (not at all) and were coded and averaged such that higher scores reflect greater spousal support (α = .76 to .86 across years).

Depressive Symptoms

In 1990, 1991, and 1992, using the Symptoms Checklist–90–Revised (SCL-90-R) depression scale (Derogatis, 1983), participants were asked to assess their level of distress (1 = not at all to 5 = extremely) during the previous week. Responses to the 13 symptoms (e.g., feeling down, loss of sexual interest or pleasure, crying easily, and feeling no interest in things) were averaged, with higher scores representing a higher level of depressive symptoms (α = .87 to .92 across years).

Analysis

Latent growth curve models were used to estimate individual trajectories of change in couples’ conflict management behaviors, spousal support, and depressive symptoms. This analysis tool is particularly suitable for this study as it provides valuable information not only for estimating the individual changes over time but also for investigating associations between the changes in the variables (Duncan et al., 1999). We tested the conceptual model (Figure 1) through two stages. First, six univariate growth curve parameters were computed for the three attributes (i.e., conflict management behaviors, spousal support, and depressive symptoms) over time (1990, 1991, and 1992) to estimate average levels (intercepts) and rates of linear change (slopes) for husbands and wives separately. Second, the associated growth curve for the three attributes were estimated simultaneously including work control and personal control as predictors in a single modeling framework. To examine gender differences, we tested the models for husbands and wives separately and then performed a multiple group comparison analysis. To test fit of the conceptual model, we used the maximum likelihood estimation from Mplus version 6.0. Model fit indices used to evaluate the conceptual model include chi-square statistics, comparative fit index (CFI), and root mean square error of approximation (RMSEA). If RMSEA value is less than 0.06 and CFI value is close to 1.0 (favorable 0.90), it is believed to indicate a good model fit.

Results

Descriptive Statistics

Table 1 shows means, standard deviations, and for all of the study variables. Paired-sample t tests were conducted to examine whether mean-level differences existed for husbands and wives. On average, no significant differences were found between husbands and wives on work control and personal control. However, compared with husbands, on average, wives’ levels of conflict management behaviors (reported by their husbands) were higher than those of husbands, in 1991 (5.25 and 5.13 for wives and husbands, respectively) and in 1992 (5.27 and 5.14 for wives and husbands, respectively). As well, compared with husbands, on average, wives reported lower spousal support and a higher level of depressive symptoms over time.

Table 1.

Descriptive Statistics for Husbands and Wives (N = 308 Couples).

Variables	Husbands (n = 308)		Wives(n = 308)		Husbands–wives (paired-samples t test)
Variables	M (SD)	α	M (SD)	α	t	df
Conflict management behaviors
1990	5.13 (0.83)	.84	5.20 (0.75)	.84	−1.31	307
1991	5.13 (0.93)	.82	5.25 (0.85)	.92	−2.17*	307
1992	5.14 (0.95)	.89	5.27 (0.84)	.87	−2.08*	307
Spousal support
1990	3.63 (0.43)	.76	3.52 (0.51)	.76	3.71**	307
1991	3.62 (0.45)	.76	3.47 (0.58)	.80	4.44**	307
1992	3.60 (0.49)	.82	3.44 (0.60)	.86	4.86**	306
Depressive symptoms
1990	1.31 (0.36)	.87	1.50 (0.48)	.88	−5.71**	307
1991	1.29 (0.41)	.92	1.49 (0.57)	.92	−5.60**	305
1992	1.29 (0.43)	.90	1.45 (0.50)	.91	−4.38**	307
Work control 1990	3.66 (0.48)	.76	3.63 (0.46)	.75	0.84	307
Personal control 1990	3.85 (0.53)	.62	3.85 (0.58)	.69	0.02	307

Note. df = degrees of freedom.

p < .05. **p < .01.

Univariate Growth Curves

The estimates of growth curve parameters are presented in Table 2. The growth curves for conflict management behaviors, spousal support, and depressive symptoms for both husbands and wives had significant variances for intercept parameters (τ₀ = .60, .16, and .09 for husbands; τ₀ = .54, .28, and .23 for wives, respectively). This indicated that there was significant variability around the individuals’ means of these three attributes (γ₀ = 5.11, 3.62, and 1.30 for husbands; γ₀ = 5.18, 3.51, and 1.49 for wives, respectively) at the initial time point. The variances of change for these three attributes were significant only for wives (τ₁ = .09, .04, and .05 for conflict management behaviors, spousal support, and depressive symptoms, respectively) demonstrating that there are significant individuals’ variations in wives’ rates of changes for all three attributes over time. While there was a significant decreasing trend in wives’ spousal support (γ₁ = −0.04, p < .05) over time (1990-1992), there were substantial individual differences in changes for wives (τ₁ = .04, p < .05). This suggests that while some wives experienced a rapid decrease in spousal support over this period, other wives’ spousal support remained mostly unchanged.

Table 2.

Estimate for Univariate Growth Curves’ of Conflict Management Behaviors, Spousal Support, Depressive Symptoms for Husbands and Wives (N = 308 Couples).

	Initial level		Rate of change		χ²(df)	RMSEA/CFI
	Mean (γ₀)	Variance (τ₀)	Mean (γ₁)	Variance (τ₁)	χ²(df)	RMSEA/CFI
Husbands
Conflict management behaviors	5.11*	.60*	.01	.05	.08*(1)	.00/1
Spousal support	3.62*	.16*	−.01	.01	.04*(1)	.00/1
Depressive symptoms	1.30*	.09*	−.01	.00	.60*(1)	.00/1
Wives
Conflict management behaviors	5.18*	.54*	.04	.09*	.46*(1)	.00/1
Spousal support	3.51*	.28*	−.04*	.04*	.16*(1)	.00/1
Depressive symptoms	1.49*	.23*	−.02	.05*	.55*(1)	.00/1

Note. df = degrees of freedom; RMSEA = root mean square error of approximation; CFI = comparative fit index. Factor loadings for intercepts λ₁₁ = λ₂₁ = λ₃₁; for slopes λ₁₂ = 0, λ₂₂ = 1, λ₃₂ = 2 for all models.

p < .05.

To explore what characteristics may differentiate these two groups of wives, we conducted post hoc analyses whereby wives were mean-split divided into either a decreasing spousal support group (n = 122) or no change/increasing spousal support group (n = 186). Using independent-sample t tests, between-group differences on marital commitment, marital instability, and conflict over financial issues were examined. Significant between-group differences were only found on conflict over financial issues, t(306) = 2.09, p < .01, such that wives who experienced decreasing spousal support over time reported more marital conflict over family finances (M = 2.52, SD = 1.17) when compared with those who experienced no change or increasing spousal support (M = 2.25, SD = 1.06).

Associated Latent Growth Curves

To test the hypothesized model, we estimated associated growth curves for the three attributes simultaneously, including personal control and work control in a comprehensive model. These models are shown in Figure 2 for wives and Figure 3 for husbands. For both the wives’ and husbands’ models, the goodness-of-fit indices show a generally good fit with the data for the models (RMSEA = .07, CFI = .97 for wives; RMSEA = .04, CFI = .99 for husbands).

Figure 2.

Linking work, marriage, and depressive symptoms through 1990 to 1992 for wives.

Figure 3.

Linking work, marriage, and depressive symptoms through 1990 to 1992 for husbands.

Wives

Consistent with the hypothesis, wives’ work control contributed to the level of conflict management behaviors through personal control. There is no direct correlation between wives’ work control and their conflict management behaviors. The unstandardized path coefficient between work control and personal control was β = .22 (p < .01) and the coefficient between personal control and conflict management behaviors was β = .28 (p < .01). This implies that wives’ higher work control is positively associated with their sense of personal control; consequently, this attribute spills over from work to the family setting, contributing to wives’ use of more positive practices to manage marital conflicts at home. In addition, wives’ personal control was significantly and positively related to the level of spousal support (β = .11, p < .05) and negatively related to the level of depressive symptoms (β = −.25, p < .01). The initial level of spousal support was significantly and positively related to the initial level of conflict management behaviors (β = .34, p < .01) and negatively associated with the initial level of depressive symptoms (β = −.16, p < .01). This indicated that wives who had higher starting points for conflict management behaviors trajectories also had higher starting points for their spousal support and lower starting points for their depressive symptoms.

The findings also showed an “interlocking trajectory” among these three variables, which means the change in conflict management behaviors from 1990 to 1992 was parallel and positively associated with the change in spousal support from 1990 to 1992 (β = .16, p < .05). The changes in spousal support in turn, significantly and negatively related to the changes in their level of depressive symptoms from 1990 to 1992 (β = −.27, p < .05), linking changes in conflict management behaviors with changes in depressive symptoms. This association between individual changes gives strong evidence that the increase in constructive conflict management behaviors positively predicted the increase in spousal support and the decrease in the risk for wives’ depressive symptoms over time.

Husbands

For husbands, the results for the associated growth curve analysis were quite similar to those for wives (Figure 3). Husbands’ work control contributed to their level of conflict management behaviors only through personal control. There is no direct correlation between husbands’ work control and their conflict management behaviors. The unstandardized path coefficient between work control and personal control was .23 (p < .01), and the coefficient between personal control and conflict management behaviors was .38 (p < .01). In addition, husbands’ personal control was significantly and negatively related to their level of depressive symptoms only (β = −.22, p < .01). This implies that husbands’ higher work control positively influenced their sense of personal control, contributing to lower depressive symptoms from 1990 to 1992. Also, the initial level of spousal support was significantly and positively related to the initial level of conflict management behaviors (β = .22, p < .01), but not significantly related to the level of depressive symptoms. This indicated that husbands who had higher starting points for conflict management behaviors trajectories also had higher starting points for their spousal support and but not for their depressive symptoms. The husbands’ findings showed an “interlocking trajectory” among these three variables. Increasing use of constructive conflict management behaviors was positively related to the increases in spousal support across time (β = .34, p < .01), which in turn was associated with reduced risk for depressive symptoms from 1990 to 1992 (β = −.32, p < .01).

Interestingly, husbands’ personal control was significantly and negatively related to change in their conflict management behaviors across time (1990-1992; β = −.07, p < .05). In order to examine this more closely, husbands were divided into higher personal control and lower personal control groups and two univariate growth curves for husbands’ conflict management behaviors were estimated to compare each group’s trajectories over time (see Figure 4). The means of conflict management behaviors were 4.96 and 5.25 for the lower and higher personal control groups, respectively. An average slope for the univariate growth curve for the lower personal control group was .14 (p < .05), indicating an increased trajectory of constructive conflict management behaviors across the 3 years. An average slope for the univariate growth curve for the higher personal control group was only −.07 (p = .10), indicating their constructive conflict management behaviors remained fairly constant across the 3 years. These findings suggest a ceiling effect for husbands who have higher personal control. Once individuals developed high-level conflict management behaviors influenced by high personal control, they may not experience further significant change in conflict management behaviors over time.

Figure 4.

High personal control group versus low personal control group for husbands (N = 308).

Gender Differences

To inspect gender differences in the model, a multiple group comparison was performed. With the fully unconstrained model as a starting point, we allowed parameters one by one to be constrained. The outcomes showed that only two associations were significantly different between husbands and wives. First, the associations between the initial level of conflict management behaviors and the initial level of spousal support was significant, Δ × 2 (1, N = 308) = 5.63, p < .05. Wives had a stronger association between these two attributes (β = .34, p < .01) compared with husbands (β = .22, p < .01). Second, the association between the initial level of spousal support and the initial level of depressive symptoms was significantly different between husbands and wives, Δ × 2 (1, N = 308) = 6.21, p < .05. The initial levels of spousal support significantly and negatively contributed to the initial level of depressive symptoms for wives (β = −.16, p < .01), but not for husbands.

Discussion

Latent growth curve methodology was used to examine hypothesized pathways whereby dimension of work control affects husbands’ and wives’ personal and marital life. The current study suggested that husbands’ and wives’ work control experiences are associated with their use of more positive practices to manage conflict, albeit through sense of personal control. This consequently emphasizes the importance of psychosocial work environment influences on enriching personal resources and fostering positive interpersonal interactions in other areas of employees’ personal lives. Consistent with other studies (Griffin et al., 2002), the present study reinforces that work control is salient to husbands’ and wives’ personal and marital well-being.

The present study extends the existing research by demonstrating the additive influence of work control on marriage and health over three waves by estimating individual trajectories of the three attributes simultaneously. Unlike other studies using cross-sectional data which only detect associations based on the rank order in terms of average levels of individuals at one point in time, the use of longitudinal data with the growth curve approach enabled the investigation of individual trajectories of time-varying study variables. This approach allowed us to consider within-individual changes and interindividual variability in those changes that yielded stronger conclusions about the associations between these changes. This methodological approach has two important features. First, this provides important insight into how the level and the rate of changes in one attribute influences the level and the rate of changes in other attributes. The interlocking trajectories among growth parameters of these key variables offer compelling evidence for their systematic and dynamic association between changes in the variables (Duncan et al., 1999). The interlocking trajectories and the association between these variables reveal that the interactions among them were always synchronized over time. Our results showed that higher levels of constructive conflict management contributed to a higher level of spousal support while also operating as a significant predictor for changes in these attributes over time for both husbands and wives. Whereas previous research has only focused on the independent influence of conflict management behaviors on marital satisfaction and stability (Johnson et al., 2005), our results highlight the significance of taking a nuanced approach to studying marital relationships by examining the interdependence between various variables.

The second important feature involves findings that husbands’ and wives’ work control experiences, through personal control, uniquely contributed to their level of conflict management behaviors but did not result in significant changes in conflict management behaviors over time. This result provides evidence for the contemporaneous effects of work control and personal control on marital interactions. Individuals’ work control experience initiated trajectories up to a certain level but did not shape the trajectories of change. This ceiling effect was especially true for husbands, such that husbands with a higher level of personal control also tended to engage more frequently in behaviors that facilitated more positive conflict management but did not experience dramatic changes in these behaviors over time.

Previous research has demonstrated how various workforce productivity outcomes are influenced by marital quality (Bardasi & Taylor, 2008). Our findings suggest that employers may benefit from supporting their employees’ marriages through enhancing the quality of work control experiences. For example, employers could provide employees more opportunities to practice their skills and knowledge so as to reinforce their sense of personal control. This cognitive process is associated with their engagement in behaviors at home that facilitate their ability to more effectively manage marital conflicts, which consequently propels the associated processes of increasing spousal support and reducing the risk of mental illnesses such as depression (Umberson et al., 2006). This pattern has important implications for practitioners who guide couples in developing skills to effectively manage conflicts in their relationship. For example, research has demonstrated the efficacy of relationship education programs in promoting positive changes in interpersonal behaviors and overall marital quality (Halford & Snyder, 2012). Our findings suggest that such programs which focus on strengthening individuals’ ability to more effectively manage conflicts may yield positive long-term changes in their perception of spousal support as well as improvements over time in mental health.

Another important consideration in interpreting the current study’s results is the role of gender on the spillover effects of work on marital and individual well-being. Our findings revealed that more frequent attempts by both men and women to manage marital conflicts using constructive conflict management behaviors similarly affected their increased perceptions of spousal support and subsequent experiences of less depressive symptoms. Interestingly, a slightly stronger association between marital conflict behaviors and perceived spousal support was demonstrated in wives. Compared with husbands, wives’ actual behaviors of constructive conflict management strongly affected their own perceptions of husbands’ support. Consistent with past research (Bradbury et al., 1996), our results demonstrated that wives’ depressive symptoms were significantly and negatively affected by their own perceptions of spousal support, an effect not seen in husbands. Although the current study demonstrated that husbands and wives are similarly susceptible to the negative effects of marital discord, wives’ mental health status may be more vulnerable to marital stresses and emotional support from their spouses or marital quality is more meaningful for wives compared with husbands. Thus, while there are clear gender differences between men’s and women’s health and experiences at home, gender may play a subtle role in shaping the process linking work, family, and health outcomes. This has important practical implications such that services aimed at supporting individuals in mastering skills that reinforce positive experiences at work and in their marriages should adjust to the needs of men and women alike.

In discussing the meaning of the study’s findings, there are several possible limitations. First, lack of spousal support and frequency of negative conflict management behaviors can be affected by individuals’ depressive symptoms. Sustained depressive symptoms in one spouse can stain his or her marital relationships and this mental illness might prevent a depressed person from receiving adequate help and support from their spouse. Also, chronic mental health problems can inhibit work productivity. Thus, although the present findings provide strong evidence of the spillover process from work to marriage and depressive symptoms, we cannot discount potential reverse causation and/or reciprocal influences. Future research should take into account this perspective in investigating these dynamic associations. Second, to increase general applicability, replication of these analyses with urban populations, individuals with different occupational characteristics, and individuals with different social backgrounds would enable the examination of these linkages to a broader cross-section of the population. In addition, since the racial makeup of the study sample was 100% Caucasian, the characteristics of the study sample limit the generalizability of the study results. Prior research suggests that unique social contexts for different racial groups, such as racial discrimination, minority status, and cultural background, may shape their lives differently (Bryant et al., 2010). Further investigation of the models tested in the current study that examine the moderating effects of race is warranted. Third, missing data are mostly due to attrition and nonparticipation in some of the waves, rather than missing at random. This is largely because missing cases are related to respondents’ employment status (underemployment and unemployment) that may be associated with our study variables particularly, work control. So, we believe that the imputation methods and full information maximum likelihood approach may not be appropriate for handling missing data and the use of respondents who have complete data are more conservative approach. Finally, a few of the study’s measurement issues need to be addressed. Modest reliability of the personal control measure is a limitation of the study. Lack of reliable data can limit the scope of analysis and can deter from finding a meaningful pattern. In this study, we used both self-reported and spousal-reported information. Self-reported information was subject to potential biases that could introduce inaccuracies in reporting. Thus, using the measurements of multiple reporters in one model could reduce the possibility of reporters’ potential bias and strengthen the results of future studies.

The current study provides valuable insight into the effects of married individuals’ work control experience on their marital and personal well-being. The study confirms that adults’ work experience must be considered as a resource for growing psychological resource and interpersonal skills. This process consequently affects their later depressive symptoms. Future research should focus on these dynamic processes and specific intervention/prevention programs need to focus on vulnerable population groups who have lower levels of control at either work or home.

Footnotes

Acknowledgements

This research uses data from Iowa Youth Family Project, a project directed by Rand D. Conger, at the University of California, Davis. Detailed information is available on the Family Transitions Project website ().

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Bardasi

Taylor

(2008). Marriage and wages: A test of the specialization hypothesis. Economica, 75, 569-591. doi:10.1111/j.1468-0335.2007.00630.x

Bradbury

T. N.

Beach

S. R. H.

Fincham

F. D.

Nelson

(1996). Attributions and behavior in functional and dysfunctional marriages. Journal of Consulting and Clinical Psychology, 64, 569-576. doi:10.1037/0022-006X.64.3.569

Bronfenbrenner

Ceci

S. J.

(1994). Nature-nuture reconceptualized in developmental perspective: A bioecological model. Psychological Review, 101, 568-586.

Bryant

C. M.

Wickrama

K. A. S.

Bolland

Bryant

B. M.

Cutrona

C. E.

Stanik

C. E.

(2010). Race matters, even in marriage: Identifying factors linked to marital outcomes for African Americans. Journal of Family Theory & Review, 2, 157-174.

Conger

R. D.

(1988). Iowa Youth and Families Project (IYFP). Washington, DC: National Institute of Mental Health.

Conger

R. D.

Conger

K. J.

(2002). Resilience in Midwestern families: Selected findings from the first decade of a prospective, longitudinal study. Journal of Marriage and Family, 64, 361-373. doi:10.1111/j.1741-3737.2002.00361.x

Crouter

A. C.

(1984). Participative work as an influence on human development. Journal of Applied Developmental Psychology, 5, 71-90.

Derogatis

L. R.

(1983). SCL-90-R: Administration, scoring and procedures manual (2nd ed.). Towson, MD: Clinical Psychometric Research.

Dollard

M. F.

Opie

Lenthall

Wakerman

Knight

Dunn

MacLeod

(2012). Psychosocial safety climate as an antecedent of work characteristics and psychological strain: A multilevel model. Work & Stress, 26, 385-404.

10.

Duncan

T. E.

Duncan

S. C.

Strycker

L. A.

(1999). An introduction to latent variable growth curve modeling concepts, issues, and applications. Mahwah, NJ: Lawrence Erlbaum.

11.

Griffin

J. M.

Fuhrer

Stansfeld

S. A.

Marmot

(2002). The importance of low control at work and home on depression and anxiety: Do these effects vary by gender and social class? Social Science & Medicine, 54, 783-798.

12.

Grzywacz

J. G.

Almeida

D. M.

McDonald

D. A.

(2002). Work–family spillover and daily reports of work and family stress in the adult labor force. Family Relations, 51, 28-36. doi:10.1111/j.1741-3729.2002.00028.x

13.

Grzywacz

J. G.

Bass

B. L.

(2003). Work, family, and mental health: Testing different models of work-family fit. Journal of Marriage and Family, 65, 248-261.

14.

Grzywacz

J. G.

Marks

N. F.

(2000). Reconceptualizing the work–family interface: An ecological perspective on the correlates of positive and negative spillover between work and family. Journal of Occupational Health Psychology, 5, 111-126.

15.

Halford

W. K.

Snyder

D. K.

(2012). Universal processes and common factors in couple therapy and relationship education. Behavior Therapy, 43, 1-12.

16.

Johnson

M. D.

Cohan

C. L.

Davila

Lawrence

Rogge

R. D.

Karney

B. R.

Bradbury

T. N.

(2005). Problem-solving skills and affective expressions as predictors of change in marital satisfaction. Journal of Consulting and Clinical Psychology, 73, 15-27.

17.

Karasek

R. A.

Theorell

(1990). Healthy work. New York, NY: Basic Books.

18.

Karney

B. R.

Bradbury

T. N.

(1995). Assessing longitudinal change in marriage: An introduction to the analysis of growth curves. Journal of Marriage and the Family, 57, 1091-1108.

19.

Kiecolt-Glaser

J. K.

Newton

T. L.

(2001). Marriage and health: His and hers. Psychological Bulletin, 127, 472-503. doi:10.1037/0033-2909.127.4.472

20.

Kohn

M. L.

Slomczynski

K. M.

(1990). Social structure and self-direction: A comparative analysis of the United States and Poland. Oxford, England: Basil Blackwell.

21.

Leach

Butterworth

(2012). Psychosocial adversities at work are associated with poorer quality marriage-like relationships. Journal of Population Research, 29, 351-372.

22.

McRae

J. A.

Jr. Brody

C. J.

(1989). The differential importance of marital experiences for the well-being of women and men: A research note. Social Science Research, 18, 237-248.

23.

Menaghan

E. G.

(1991). Work experiences and family interaction processes: The long reach of the job? Annual Review of Sociology, 17, 419-444.

24.

O’Neal

C. W.

Wickrama

K. S.

Bryant

C. M.

(2014). Control over work, positive self, and health among African American husbands and wives. Journal of Family Issues, 35, 384-410. doi:10.1177/0192513X12468435

25.

Pearlin

L. I.

Menaghan

E. G.

Lieberman

M. A.

Mullan

J. T.

(1981). The stress process. Journal of Health and Social Behavior, 22, 337-356.

26.

Rafferty

Friend

Landsbergis

P. A.

(2001). The association between job skill discretion, decision authority and burnout. Work & Stress, 15, 73-85.

27.

Repetti

Wang

Saxbe

(2009). Bringing it all back home: How outside stressors shape families’ everyday lives. Current Directions in Psychological Science, 18, 106-111. doi:10.1111/j.1467-8721.2009.01618.x

28.

Rogosa

D. R.

Willett

J. B.

(1985). Understanding correlates of change by modeling individual differences in growth. Psychometrika, 50, 203-228.

29.

Rueter

M. A.

Conger

R. D.

(1995). Interaction style, problem-solving behavior, and family problem-solving effectiveness. Child Development, 66, 98-115.

30.

Schafer

R. B.

Wickrama

K. S.

Keith

P. M.

(1996). Self-concept disconfirmation, psychological distress, and marital happiness. Journal of Marriage and the Family, 58, 167-177. doi:10.2307/353385

31.

Schooler

Mulatu

M. S.

Oates

(2004). Occupational self-direction, intellectual functioning, and self-directed orientation in older workers: Findings and implications for individuals and societies. American Journal of Sociology, 110(1), 161-197.

32.

Taylor

S. E.

Klein

L. C.

Lewis

B. P.

Gruenewald

T. L.

Gurung

R. A. R.

Updegraff

J. A.

(2000). Biobehavioral responses to stress in females: Tend-and-befriend, not fight-or-flight. Psychological Review, 107, 411-429.

33.

Umberson

Williams

Powers

D. A.

Chen

M. D.

Campbell

A. M.

(2005). As good as it gets? A life course perspective on marital quality. Social Forces, 84, 487-505.

34.

Umberson

Williams

Powers

D. A.

Hui

Needham

(2006). You make me sick: Marital quality and health over the life course. Journal of Health and Social Behavior, 47, 1-16.

35.

U.S. Census Bureau. (1990). Statistical abstract of the United States. Washington, DC: U.S. Department of Commerce.

36.

Voydanoff

(2004). The effects of work demands and resources on work-to-family conflict and facilitation. Journal of Marriage and Family, 66, 398-412. doi:10.1111/j.1741-3737.2004.00028.x

37.

Whitbeck

L. B.

Simons

R. L.

Conger

R. D.

Wickrama

K. S.

Ackley

K. A.

Elder

G. J.

(1997). The effects of parents’ working conditions and family economic hardship on parenting behaviors and children’s self-efficacy. Social Psychology Quarterly, 60, 291-303. doi:10.2307/2787091

38.

Wickrama

K. S.

Beiser

Kaspar

(2002). Assessing the longitudinal course of depression and economic integration of south-east Asian refugees: An application of latent growth curve analysis. International Journal of Methods in Psychiatric Research, 11, 154-168. doi:10.1002/mpr.133

39.

Wickrama

K. S.

Lorenz

F. O.

Conger

R. D.

Matthews

Elder

G. J.

(1997). Linking occupational conditions to physical health through marital, social, and intrapersonal processes. Journal of Health and Social Behavior, 38, 363-375. doi:10.2307/2955431

40.

Wickrama

K. S.

Surjadi

F. F.

Lorenz

F. O.

Elder

G. J.

(2008). Influence of work control trajectories on men’s mental and physical health during the middle years: Mediational role of personal control. Journals of Gerontology: Series B: Psychological Sciences and Social Sciences, 63B, S135-S145.

41.

Williams

D. G.

(1988). Gender, marriage, and psychosocial well-being. Journal of Family Issues, 9, 452-468.