Introduction

The past several decades have witnessed major social changes related to gender roles in the United States. First, the rising labor force participation rates of women in the United States have led to increasing numbers of dual-earner couples who struggle to balance paid work and family life (Moen and Yu 2000). Their struggles are complicated by a stalled revolution characterized by workplaces failing to attend to the changing needs of the workforce and women remaining largely responsible for household labor (Hochschild 1989). A second major social change is the increasing liberalization of gender ideologies since the 1960s in the United States (for a review see Brewster and Padavic 2000). Despite this trend, contemporary dual-earner couples in the United States continue to operate according to the dictates of traditional gender roles in many respects (Risman 1998). The present study uses a dataset of couples from the United States, but we are hopeful that the study has broad importance as many, though not all, countries have experienced rising labor force participation rates of women, and many dual-earner couples across the globe struggle with balancing work and family life (Hill et al. 2004; Padavic and Reskin 2002).

We note that all literature reviewed for the study is based on U.S. samples unless otherwise noted. Scholars have sought to address the ramifications of these social changes for contemporary marital relationships in the United States. Some scholars, emphasizing the permeable boundaries between work and family life, have examined how work experiences shape marital relationships (Hughes et al. 1992; Kanter 1977; MacEwen and Barling 1988; MacEwen and Barling 1994; Matthews et al. 1996; Repetti 1989; Roberts and Levenson 2001). Other scholars have called attention to how gender ideology continues to influence marital dynamics, especially the division of household labor (Greenstein 1995; 1996a, 1996b; Hochschild 1989). Little research, however, has integrated these two areas of study to explore the influence of gender ideologies in how the navigation of work and family life impacts marital relationships, although rare exceptions exist (e.g. Zvonkovic et al. 1994). Further, we know of no studies that examine whether gender ideologies shape how work-to-family conflict relates to marital satisfaction. The present study provides a nuanced exploration of how gender ideology interacts with work-to-family conflict in predicting marital satisfaction. Our study’s main contribution lies in an investigation of whether gender ideology moderates the relationship between work-to-family conflict and marital satisfaction. The analysis is guided by the perspective that gender ideologies provide a lens through which work-to-family conflicts are viewed (Greenstein 1995; 1996a, 1996b). Further, we propose that dual-earner couples who have similar gender ideologies will have higher levels of marital satisfaction than couples with differing gender ideologies (Hochschild 1989). The research questions are examined using data from a random sample of dual-earner couples from a western state of the United States (N = 156 dual-earner couples).

Literature Review and Theory

Work-to-Family Conflict and Marital Outcomes

In the context of the stalled revolution, many dual-earner couples struggle to find balance between work and family life (Hochschild 1989; 1997; Moen and Roehling 2005; Moen and Yu 2000). Work-to-family conflict, defined as inter-role conflict that occurs when the demands of work make it difficult to attend to family needs, has been linked to a number of negative marital outcomes (Burley 1995; Greenhaus and Beutell 1985; Voydanoff 1988, 2002). For instance, a meta-analysis of the outcomes associated with work-to-family conflict demonstrated a moderate negative correlation between work-to-family conflict and marital satisfaction (Allen et al. 2000). Further, the findings from other studies (including some studies of Canadian populations) suggest that work-to-family conflict, work-related psychological distress, high workload, and work stress lead to marital withdrawal, marital anger, negative marital interactions, and decreased marital adjustment and marital quality (Hughes et al. 1992; MacEwen and Barling 1988; 1994; Matthews et al. 1996; Repetti 1989; Roberts and Levenson 2001; Rogers and Amato 2000).

Work-family scholarship has also emphasized how the workplace experiences of an individual can influence his or her spouse (Crossfield et al. 2005; Gareis et al. 2003; MacDermid and Harvey 2006; Perry-Jenkins et al. 2000; Westman 2001). Several studies (including one conducted in Israel), have shown that the work-related stress of one partner can cross over into the family domain resulting in marital conflict, negative marital interactions, stress, and decreased marital warmth of the other partner (Barnett 1998; Barnett and Brennan 1997; Billings and Moos 1982; Crouter et al. 1989; Doumas et al. 2003; Bolger et al. 1989; Westman and Etzion 1995). Altogether, the findings from previous studies suggest the work-to-family conflict of both partners is likely to impact marital relationships, including marital satisfaction. Here we consider the role of each spouse’s work-to-family conflict in shaping marital satisfaction.

Gender Ideology and Marital Outcomes

Following other scholars we define gender ideology as “how a person identifies herself or himself with regard to marital and family roles that are traditionally linked to gender” (Greenstein 1996a, p. 586). Scholars have proposed that gender ideology provides a lens through which marital dynamics, such as the allocation of housework (Greenstein 1995; 1996a; 1996b), wives’ economic independence (Sayer and Bianchi 2000), spousal support (Mickelson et al. 2006) and power tactics in Israeli couples (Schwarzwald et al. 2008), are viewed. These studies have focused on examining the potential moderating role of gender ideology in predicting marital outcomes. For instance, Greenstein (1995; 1996b) argued that women’s gender ideologies help to determine how they view inequalities in housework and whether such inequalities impact women’s marital quality. As such, women with egalitarian gender ideologies are expected to react more negatively to inequalities, including experiencing reduced marital happiness, compared to women with traditional gender ideologies; previous research on women in the United States and Israel supports these claims (Greenstein 1996b; Lavee and Katz 2002). More recent research by Greenstein (2009), which included data from 30 nations, found that perceived fairness in the division of household labor was more strongly related to the family life satisfaction of women with egalitarian gender ideologies than women with traditional ideologies.

We also propose that dual-earner couples who have similar gender ideologies, regardless of whether they are both more traditional or more egalitarian, will have higher levels of marital satisfaction. Hochschild’s classic book The Second Shift (1989) described the marital stresses that sometimes resulted from conflicting gender ideologies. In contrast, similar gender ideologies may result in dual-earner couples more readily agreeing on solutions to work-family issues, which can reduce marital conflict (Blumstein and Schwartz 1983). For example, a dual-earner couple with both partners holding egalitarian gender ideologies might not think twice about the husband increasing his housework performance to reduce stress experienced by his wife (Amato and Booth 1995). In contrast, a dual-earner couple with differing gender ideologies might encounter difficulties agreeing on such a solution. Additionally, the findings from one study indicate that both partners holding traditional gender ideologies may be related to higher levels of marital satisfaction for men and women (Zvonkovic et al. 1994), and a study of Chinese couples found that men with less egalitarian values than their wives reported higher marital disharmony (Pimental 2006).

Gender Ideology and Work-to-Family Conflict

We argue that gender ideology will interact with each spouse’s work-to-family conflict in predicting marital satisfaction, because gender ideology will provide a lens through which these conflicts are viewed (Greenstein 1995; 1996b). For those subscribing to a traditional gender ideology, men’s paid work is seen as natural and normalized; whereas women’s paid labor force participation is frowned upon (Davis and Greenstein 2009; Deutsch and Saxon 1998). Hence, men and women subscribing to a traditional gender ideology will see men’s work-to-family conflict as a normal part of men’s appropriate participation in paid work, and it will be unlikely to negatively impact marital satisfaction. In contrast, for men with traditional gender ideologies, women’s work-to-family conflict will be viewed as interfering with their primary responsibility of caretaking, and hence may negatively impact marital satisfaction (Vannoy and Philliber 1992). Women’s work-to-family conflict may even serve as an unwelcome reminder to men holding traditional gender ideologies that their lives are, in some fashion, different than what they believe is most correct, which can lead to marital conflict (Hochschild 1989; Deutsch and Saxon 1998). For instance, studies have generally shown that wives’ incomes and economic independence bear stronger relationships to reduced marital quality and marital dissolution when men hold traditional gender ideologies (Brennan et al. 2001; Sayer & Bianchi, 2000). Along similar lines, Atkinson et al. (2005) found that wives’ relative income was related to the abuse of wives, but only among men with traditional gender ideologies.

Men who hold egalitarian gender ideologies view the paid labor force participation of both men and women as being worthwhile (Voydanoff 2007). Hence, we think that the experience of work-to-family conflict by men will be normalized in such contexts, and will be seen as a natural and inevitable part of the dual-earner lifestyle. Further, men who hold egalitarian gender ideologies will be less negatively impacted by their wives’ work-to-family conflict than men holding more traditional gender ideologies. Studies have shown that men holding egalitarian gender ideologies tend to be more supportive of their wives’ paid work than men with traditional gender ideologies. For instance, a study by Amato and Booth (1995) found that husbands with egalitarian gender ideologies were more accepting and accommodating of their wives’ paid work commitments than men with traditional gender ideologies, and the findings from a study of Spanish couples indicated that women married to egalitarian men reported greater career salience than women married to traditional men (Moya et al. 2000).

Women holding traditional gender ideologies who experience high levels of work-to-family conflict will likely view work-to-family conflict as an inevitable part of juggling work and family responsibilities. Accordingly, such women will interpret work-to-family conflict as an expected cost associated with their paid labor force participation, and it will be unlikely to negatively shape their marital satisfaction. A different picture emerges when we consider egalitarian women, as such women consider their labor force participation to be valuable. A number of studies, with samples from a variety of industrialized nations, have shown that women with egalitarian gender ideologies have higher earnings, greater career salience, and work more hours than other women—suggesting that women holding egalitarian gender attitudes are likely to place a higher premium on paid work (Christie-Mizell 2006; Christie-Mizell et al. 2007; Moya et al. 2000; Nordenmark 2004; Stickney and Konrad 2007). Such women may take issue with work-to-family conflict, because they believe that women should be able to work without negative consequences (such as work-to-family conflict). Further, egalitarian women are likely aware that they navigate paid work and family life in a social context of gender inequality. In particular, “women shoulder more responsibility for domestic work, and they also face larger obstacles in the workplace, including less autonomy and flexibility on the job and more pressure to make career sacrifices by cutting back time at work…” (Jacobs and Gerson 2004, p. 115). Such inequalities can impede the abilities of women to climb corporate ladders and to successfully integrate paid work and family life, and we propose that egalitarian women are more likely to be cognizant of these gender inequalities, and thereby more likely to take issue with the experience of work-to-family conflict. The experience of work-to-family conflict by egalitarian women, therefore, may be more likely to result in decreased marital satisfaction, because it is an unwelcome reminder of barriers that impede women’s abilities to juggle paid work and family. Traditional women, on the other hand, are less likely to be aware of such issues and are unlikely to be upset by them, as these inequalities are largely consistent with their gender ideologies.

Demographic Control Variables

The present study includes a number of demographic control variables (average labor force hours, presence of children, length of marriage, level of education, and total household income) that can be conceptualized as resources and demands, and that have been shown to affect marital relationships (Voydanoff 2005). Labor force hours, conceptualized as a time demand, limit the time and energy available to attend to marital needs, thus potentially negatively affecting marital relationships (Hughes et al. 1992; Kingston and Nock 1987; Voydanoff 1988, 2005; Zvonkovic et al. 2006). Indeed, previous studies have linked women’s labor force hours to marital disruption and decreased marital quality (Amato et al. 2003; Greenstein 1990; Perry-Jenkins et al. 2000; Rogers 1996).

We also conceptualize children as creating demands, as children increase time spent providing child care and housework (Voydanoff 2007). Previous scholarship indicates that children, in general, tend to adversely impact spousal companionship, marital quality, and marital satisfaction (Glenn and McLanahan 1982; Ross et al. 1990; Teachman et al. 1999; Twenge et al. 2003). Further, when considering the role of work-to-family conflict in shaping marital satisfaction, it is important to take into account total household income, which we conceptualize as an important resource that can be utilized to mitigate work-family issues. This is because dual-earner couples with high incomes are able to use their resources to obtain a smoother balance between paid work and family life by purchasing goods and services. Previous research has found that higher incomes buffer stresses encountered in married life and lower incomes heighten stresses (Belsky and Rovine 1990; Ross et al. 1990). Education can also be considered a resource that individuals can draw upon as they navigate paid work and family life. For instance, those with higher levels of education may be more likely to work in occupations that provide greater access to family-friendly benefits and other resources. Length of marriage is also a key variable to take into account, as research suggests that marital quality peaks in the first years of marriage and then steadily declines until midlife (Orbuch et al. 1996).

The Present Research and Hypotheses

As outlined earlier, we argue that gender ideology will shape the marital satisfaction of dual-earner men and women by providing a lens through which conflicts between work and family are viewed. As such, how a person’s marital satisfaction is impacted by work-to-family conflict and their spouse’s work-to-family conflict will depend on their gender ideology. For instance, women with egalitarian gender ideologies may be negatively impacted by their own work-to-family conflict, whereas women with traditional gender ideologies will view work-to-family conflict as an inevitable feature of their workforce participation. Both egalitarian and traditional women are unlikely to be negatively impacted by their spouse’s work-to-family conflict, as men’s work-to-family conflict is viewed as normal and appropriate. Egalitarian men are unlikely to be negatively impacted by their spouse’s work-to-family conflict, whereas traditional men’s marital satisfaction may be lower when women’s work-to-family conflict is higher. Further, the interaction between spouses’ gender ideologies will also be related to marital satisfaction. Specifically, it is expected that when the ideologies of spouses are similar marital satisfaction will be buttressed. Based on this theoretical perspective, we put forth the following hypotheses concerning the role of gender ideology in predicting marital satisfaction:

  1. Hypothesis 1.

    Gender ideology will moderate the relationship between work-to-family conflict and marital satisfaction for women.

    1. Hypothesis 1a:

      Work-to-family conflict among women with more egalitarian gender ideologies will be negatively related to such women’s marital satisfaction.

    2. Hypothesis 1b:

      Work-to-family conflict among women with more traditional gender ideologies will be unrelated to such women’s marital satisfaction.

  2. Hypothesis 2.

    Gender ideology will moderate the relationship between spouse’s work-to-family conflict and marital satisfaction for men.

    1. Hypothesis 2a:

      Spouse’s work-to-family conflict among men with more traditional gender ideologies will be negatively related to such men’s marital satisfaction.

    2. Hypothesis 2b:

      Spouse’s work-to-family conflict among men with more egalitarian gender ideologies will be unrelated to such men’s marital satisfaction.

  3. Hypothesis 3.

    The interaction of spouses’ gender ideologies will be significantly and positively related to marital satisfaction, such that more similar gender ideologies will be related to higher levels of marital satisfaction.

We address these hypotheses using data from a random sample of U.S. dual-earner couples in a western state (N = 156 couples). First, we present descriptive statistics and paired t tests. A bivariate correlation matrix of the variables is then provided. Finally, the multivariate analysis is performed using separate ordinary least squares (OLS) regression analyses for men and women. We use two models to predict marital satisfaction. Model 1, the base model, includes the following predictor variables: total household income, education, presence of children, length of marriage, each spouse’s paid work hours, each spouse’s gender ideology, and each spouse’s work-to-family conflict. This model is testing if there are direct relationships between work-to-family conflict, spouse’s work-to-family conflict, and gender ideology on the marital satisfaction of men and women. Model 2 includes the same predictor variables with the addition of the following interaction terms: the interaction between gender ideology and work-to-family conflict (to examine Hypothesis 1, 1a and 1b), the interaction between gender ideology and partner’s work-to-family conflict (to examine Hypothesis 2, 2a, and 2b), and the interaction between spouses’ gender ideologies (to examine Hypothesis 3).

Method

Sample

Data for this study originate from a sample of co-residential dual-earner couples in the northern part of a western state. City officials from the selected cities provided a list of residential addresses, which served as the sampling frame for the study. Using a computer program we then sampled residential addresses using simple random sampling without replacement. We then used a drop-off-pick-up data collection technique in which a member of the research team approached each randomly selected household in order to determine if they were eligible for the study (Riley and Kiger 2002). To qualify for inclusion in the study, households had to be headed by a couple in which both were employed in the paid labor force. There was no predetermined number of hours that employed individuals had to work to quality for inclusion in the study. If dual-earner couples agreed to participate, two self-administered questionnaires were dropped off, and respondents were instructed not to discuss the questions with their partner. A research assistant then returned within 48 hr to pick up the surveys. Only dual-earner couples for whom both questionnaires were completed were included in the final sample. This resulted in a total of 156 dual-earner couples, or 312 individuals, an 83% response rate. The majority of the respondents were White (94%), which is reflective of the area from which the sample was drawn. The median family income was $47,500. The average age of men was 39.7 years old (SD = 11.53) and the average age of women was 37.2 (SD = 12.69). Roughly 66 percent of the sample had at least one child under the age of 18 living in the home, and about a third of the dual-earner couples had pre-school aged children in the home.

Measures

Dependent Variable

Marital satisfaction was measured using a subscale derived from Spanier’s (1976) index of marital adjustment. Respondents were asked to rate their satisfaction with these seven aspects of their relationship: “(a) The way money is handled in your relationship; (b) The things you and your partner do together when you go out visiting or for entertainment; (c) The amount of affection in your relationship; (d) The way you and your partner deal with in-laws; (e) Sexual relations in your relationship; (f) Religious beliefs in your relationship; and (g) The way chores around the house are performed in your relationship.” The response categories were “always dissatisfied,” “often dissatisfied,” “sometimes satisfied,” “often satisfied,” and “always satisfied.” Non-missing responses to the items were averaged to create an index with higher scores indicating higher levels of satisfaction. The alpha reliability coefficient was .78 for women and .81 for men.

Independent Variables

Household income was measured by asking respondents to indicate their total household income. Response categories began with 1 = “less than $10,000” and increased in 5,000 increments to 20 = “100,000 and over.” The variable was somewhat skewed, so it was symmetrized with a square root transformation. Length of marriage was measured by asking the respondents to indicate how long they had been married. Skewness on this variable was also corrected with a square root transformation. Presence of children was measured by asking the respondents to indicate how many children currently live in their household in each of the following categories: 0–5 years old, 6–11 years old, and 12–18 years old. Respondents were asked to write in a 0 if they have no children in a given category. A dummy variable was constructed where a value of 1 indicates at least one child in any category, and 0 otherwise. Years of education was measured by asking respondents to circle the highest grade they completed in school. Possible choices ranged from 1 to 17 or more years of education. Average hours worked was measured by asking respondents to indicate the average hours they worked per week in the paid labor force.

The work-to-family conflict scale was adapted from Kirchmeyer (1992, 1993) and was the average of non-missing responses to six items: “(a) My job keeps me from spending time with my partner; (b) Our relationship suffers because of my work; (c) My job makes it difficult for me to enjoy my free time outside of work; (d) The amount of time I spend working interferes with how much free time I have; (e) My job makes it difficult for me to get household chores done; and (f) I spend so much time working that I am unable to get much done at home.” Respondents were asked to indicate the extent to which they agreed with each statement as it applied to them most of the time. The response categories were “strongly disagree,” “disagree,” “agree,” and “strongly agree.” The alpha reliability coefficient was .89 for women and .88 for men.

Gender ideology was measured using a ten-item attitudinal scale adapted from Spence and Helmreich’s Attitudes toward Women Scale (1978). According to previous scholars, the Attitudes toward Women Scale “has been the most widely used measure of gender role attitudes” (Olson et al. 2007, p. 301). Due to space constraints, only 10 items from the original scale were included on the questionnaire, and items were selected on the basis of relevance to the lives of couples juggling work and family responsibilities, with items that seemed antiquated excluded. This version of the scale has been used in other published studies (Minnotte et al. 2007; Stevens et al. 2001). Respondents were asked to indicate whether they agreed or disagreed with the following items: “(a) Swearing and obscenity are more repulsive in the speech of a woman than a man; (b) When the wife works outside the home, the husband should share in household tasks such as washing dishes and doing the laundry; (c) Women should worry less about their rights and more about becoming good wives and mothers; (d) Women should assume positions in business and all professions, along with men; (e) A woman should not expect to have quite the same freedom of action as a man; (f) The leadership of a community should be largely in the hands of men; (g) Women should be given equal opportunity with men for apprenticeship in the various trades; (h) Sons in a family should be given more encouragement to go to college than daughters; (i) In general, the father should have greater authority than the mother in rearing of children; and (j) There are many jobs in which men should be given preference over women in being hired.” Response categories were “disagree strongly,” “disagree somewhat,” “agree somewhat,” and “agree strongly.” The appropriate items were reverse coded so that the higher the scale score, the more traditional the respondent’s gender ideology, with a final score of the average of all non-missing responses. The alpha reliability coefficients were .82 for men and .73 for women.

For ease of interpretation and discussion, all variables in the models but one were standardized for the regression analyses. Thus reported coefficients will represent the mean change in standardized marital satisfaction found from increasing the independent variable in question by one standard deviation while holding the remainder constant. The sole exception was the dummy variable for presence of children, whose coefficient may be viewed simply as the mean change in standardized marital satisfaction due to the presence of children in the household.

Results

To investigate differences by gender in the study variables, Repeated Measures MANOVA was employed to account for the paired nature of the couple data. Gender was found to be highly significant in distinguishing these sets of variables, F (1, 129) = 45.80, p < .001. With clear differences between the sexes, we employed a paired t-test on each variable with a Bonferonni correction to adjust for multiple tests (Cohen 2001, pp. 376–377). Table 1 presents descriptive statistics and paired t-tests for the study variables.

Table 1 Descriptive Statistics (N = 156 dual-earner couples).
Full size table

Unsurprisingly, the household variables of income, marriage length, and presence of children were not significantly different between genders. The transformed household income variable for the dual-earner couples in our sample had an average of approximately 2.94, equivalent to about $48,000, with standard deviations of about 0.76. The square root of years married had a mean of about 3.22, equivalent to about 10 years, with standard deviations of about 1.65. About 66% of couples had children living in the household (SD = 0.48).

Women had an average of 13.84 years of education (SD = 2.10) compared to men with 14.57 years (SD = 2.09). Men worked an average of 45.75 hr (SD = 12.17), whereas women worked an average of 36.20 hr per week (SD = 13.35). Mean scores for traditional gender ideology indicated that men (M = 1.79, SD = 0.52) in our sample were somewhat more traditional than women (M = 1.64, SD = 0.44). Two-tailed paired t-tests, using a Bonferroni correction for the eight comparisons conducted, indicated that men and women differed significantly at the 0.05 level on these three variables, with men reporting higher mean levels of education, greater mean hours worked, and more traditional mean gender ideologies than women. Work-to-family conflict scores were fairly similar for men and women, with men reporting an average of 2.31 (SD = 0.63) on the scale compared to women’s average of 2.30 (SD = 0.68). Finally, mean scores for marital satisfaction were similar for men (M = 3.80, SD = 0.66) and women (M = 3.73, SD = 0.68). Table 2 presents the bivariate correlations for the variables.

Table 2 Zero-Order Correlation Matrix of Variables Used in the Analysis for Men and Women (N = 156 dual-earner couples).
Full size table

In order to test for potential issues of multicollinearity, variance inflation factors were determined for the independent variables for both models. The models showed no symptoms of multicollinearity, with all variance inflation factors less than 2. Upon running the models, diagnostic plots were examined, with an emphasis on added-variable plots (also known as partial-regression plots; Fox 2002, p. 200) and partial residual plots (Faraway 2005, p. 72). All plots suggested that the models and, in particular, the linear terms for the control variables were appropriate.

The results from the two OLS regression models are reported in Table 3. The first model examined the relationships between marital satisfaction and all the predictor variables (household income, years of education, length of marriage, number of children, average work hours, spouse’s average work hours, work-to-family conflict, spouse’s work-to-family conflict, gender ideology, and spouse’s gender ideology), while excluding interaction terms. Model 2, to test the proposed hypotheses, added the following interactions terms: the interaction between respondent’s work-to-family conflict and respondent’s gender ideology, the interaction between spouse’s work-to-family conflict and respondent’s gender ideology, and the interaction between respondent’s and spouse’s gender ideologies. We can see that for both men and women the F test for the change in R 2 shows that the improvement in prediction of marital satisfaction due to the interaction terms is significant.

Table 3 Regression Coefficients for Models Examining Marital Satisfaction for Women and Men (N = 156 dual-earner couples).
Full size table

Hypothesis 1 predicts that gender ideology will moderate the relationship between work-to-family conflict and marital satisfaction for women. This basic hypothesis was supported, as the significant positive interaction term is equivalent to a test on the difference of simple slopes of marital satisfaction on work-to-family conflict for different levels of gender ideology (Aiken and West 1991, pp. 19–21). Figure 1 demonstrates this interaction by plotting the effect of the women’s work-to-family conflict on women’s marital satisfaction, while holding all other variables constant at their respective means, for women at five points along the gender ideology continuum, from two standard deviations below the mean (much more egalitarian) to two standard deviations above the mean (much more traditional). Figure 1 indicates that, although high levels of work-to-family conflict are generally detrimental to women’s marital satisfaction, this effect is much stronger for more egalitarian women and milder-to-nonexistent, or even mildly reversed, for traditional women. These findings support the specific predictions found in Hypotheses 1a and 1b; work-to-family conflict appears to be more detrimental to marital satisfaction for strongly egalitarian women than for women toward the traditional end of the scale. Simple slope tests (Aiken and West 1991, p.16) verify these results. The simple slope of marital satisfaction on work-to-family conflict conditioned on gender ideology is negative and significantly different from 0 at the 0.05 level for women with any gender ideology less than 0.58 standard deviations above the mean, including the entire egalitarian side of the distribution. In contrast, the simple slope of marital satisfaction on work-to-family conflict is not significantly different from 0 for more traditional women, with gender ideology scores higher than this value.

Fig. 1
figure 1

Examples of interaction effects between women’s gender ideology and women’s work- to-family conflict on women’s marital satisfaction. Numeric values are standard deviations below (negative) and above (positive) the means on gender ideology, work-to-family conflict, and marital satisfaction ratings.

Full size image

Hypothesis 2 concerning the moderating role of gender ideology on the effect of women’s work-to-family conflict on their husband’s marital satisfaction was not supported. Hypothesis 2a was supported, in that our findings suggest that predicted marital satisfaction is significantly and negatively impacted by their spouses’ work-to-family conflict for all men, including those who are more traditional. However, the lack of a significant interaction term suggests that this effect is not tied to gender ideology. Egalitarian men appear to be as affected as traditional husbands by their wives’ work-to-family conflict, which fails to support Hypothesis 2b.

Hypothesis 3 predicts that the interaction of spouses’ gender ideologies will be significantly and positively related to marital satisfaction, such that similar gender ideologies will be related to higher levels of marital satisfaction. For men, Model 2 provides support for this hypothesis, as the significant interaction term of spouses’ gender ideologies for men’s marital satisfaction is positive. Figure 2 plots the effect of the wife’s gender ideology on the husband’s marital satisfaction, while holding all other variables constant at their respective means, for men at five distinct points along the gender ideology spectrum, from two standard deviations below the mean (much more egalitarian) to two standard deviations above the mean (much more traditional). Men’s marital satisfaction tends to be higher when men’s gender ideologies are similar to their spouse’s. Tests of simple slopes particularly confirm this effect for more traditional men. Men with gender ideologies more than 0.14 standard deviations above the mean have a significantly (α = 0.05) positive simple slope of marital satisfaction on wife’s gender ideology, suggesting higher marital satisfaction when such men are married to women with more traditional gender ideologies. On the other hand, only the most egalitarian men, with gender ideology more than 2.05 standard deviations below the mean, show a corresponding significantly negative simple slope. The remaining egalitarian-to-moderate men have simple slopes on wife’s gender ideology that are not significantly different from 0. Therefore, we find that traditional men and extremely egalitarian men who are married to women whose ideologies match their own experience higher marital satisfaction. Although women also demonstrated a positive coefficient on the gender ideology interaction term, it was far from significant, and we must conclude that Hypothesis 3 was not supported for women.

Fig. 2
figure 2

Examples of interaction effects between spouses’ gender ideologies on men’s marital satisfaction. Numeric values are standard deviations below (negative) and above (positive) the means on gender ideology and marital satisfaction ratings.

Full size image

Discussion

A voluminous literature has developed examining the experiences of dual-earner couples and how they negotiate the work-family interface (e.g. Hochschild 1989; Moen and Roehling 2005; Moen and Yu 2000). Another related literature explores the division of household labor and how gender ideologies of individuals shape how such inequalities affect their marital quality (Greenstein 1995, 1996a, 1996b). As yet, far fewer studies have examined how gender ideologies shape the navigation of the work-family terrain, although important exceptions exist (e.g. Zvonkovic et al. 1994), and we know of no studies until this one that investigate how gender ideology interacts with work-to-family conflict in predicting marital satisfaction.

The present study utilizes a quantitative approach to study the relationships between gender ideology, work-to-family conflict, and marital satisfaction using data from 156 U.S. dual-earner couples from a western state. Using a theoretical perspective that highlights how gender ideology provides a lens through which the negotiation of paid work and family is viewed, we hypothesized that gender ideology would interact with both work-to-family conflict and spouse’s work-to-family conflict in predicting marital satisfaction. We also hypothesized that the gender ideologies of each spouse would interact such that spouses with similar gender ideologies would report higher levels of marital satisfaction (Zvonkovic et al. 1994). By addressing these research questions we contribute to the literature by providing a nuanced examination of the gendered nature of the work-family interface.

Women’s marital satisfaction, in Model 2, was largely explained by the interaction between work-to-family conflict and women’s gender ideology. Yet in the absence of the interaction term (Model 1), neither main effect of gender ideology is significant. These results suggest that gender ideologies matter in predicting marital satisfaction for women only in terms of how they interact with other factors such as work-to-family conflict. This finding is consistent with previous studies that emphasize the importance of considering women’s gender ideologies when determining how women’s work characteristics impact marital outcomes (Vannoy and Philliber 1992). The interaction of gender ideology with work-to-family conflict in predicting women’s marital satisfaction, however, has yet to be demonstrated by previous research.

We find that the interaction term of gender ideology and work-to-family conflict is positively related to women’s marital satisfaction. More traditional women with high levels of work-to-family conflict demonstrate higher levels of marital satisfaction than the main effects alone would suggest, as do more egalitarian women with low levels of work-to-family conflict. Conversely, more traditional women with lower levels of work-to-family conflict and more egalitarian women with higher levels of work-to-family conflict experience lower marital satisfaction than would be predicted without the interaction term. When combined with the strong negative main-effect coefficient on work-to-family conflict and a strong positive one on gender ideology, we arrive at the predictions depicted in Fig. 1. Although most women’s marital satisfaction is negatively impacted by work-to-family conflict, this effect is much stronger for those toward the egalitarian end of the gender ideology scale than for their more traditional counterparts, whose marital satisfaction does not significantly depend on work-to-family conflict. The women with the lowest levels of marital satisfaction are strongly egalitarian women with high levels of work-to-family conflict. These results were in line with our hypotheses. We think the primary explanation for the finding lies in what likely occurs when women seek to reduce their experiences of work-to-family conflict. Egalitarian women, as a way to deal with work-to-family conflict, are more likely to seek increased housework participation from their husbands which may lead to marital disagreements. As Greenstein (1995) has suggested, women with traditional gender ideologies likely view inequalities in the division of household labor as fair, hence such women are unlikely to press their husbands for increased housework participation. Further, when egalitarian women with high work-to-family conflict experience unmet expectations concerning the housework participation of their husbands this may negatively impact their marital satisfaction. Such an interpretation is consistent with Greenstein’s finding that egalitarian women’s marital quality is more likely to be negatively affected by perceived inequities in the division of household labor than traditional women’s marital quality (Greenstein 1996b). It should be noted, however, that Greenstein did not examine work-to-family conflict in his studies. We call for future research to explore how work-to-family conflict is related to conflict over the division of household labor dependent on spousal gender ideologies.

We found that men’s marital satisfaction, in Model 2, appeared to be heavily shaped by gender ideology. For men, we found a significant and positive relationship between the interaction of gender ideologies and marital satisfaction. As shown in Fig. 2, this result suggests that men’s marital satisfaction tends to be higher when the gender ideologies of both spouses are similar, relative to what the main effects alone would suggest, at least for men in the extremes of the gender ideology spectrum. Traditional men demonstrate significantly higher marital satisfaction when married to more traditional women. Despite the fact that such dual-earner couples are leading lives that differ from their professed gender ideologies, these men’s marital satisfaction remains high. The changing economic context in the United States has led dual-earner couples to become the modal type of family. Indeed, among married couples 51.4% had both partners in the paid labor force in 2008, and if you consider married couples with children the percentage increases to 62.1% (U.S. Department of Labor, 2009). Men in such couples gain the economic benefit of a second income contributing to the total household income, which enhances the standard of living of the couple. Yet among dual-earner couples, traditional men with traditional wives likely face the least pressure to increase their housework participation (Amato et al. 2003; Hochschild 1989). In combining the financial benefits of two incomes with hypothetically low housework expectations, it easy to see how these men might have few complaints.

Our interaction plots also indicate that egalitarian men married to egalitarian women report above-average levels of marital satisfaction, although the simple slope tests note that the negative slope is only significant for especially egalitarian men. Given that such men appear to be leading lives closest to their professed gender ideologies (i.e. both partners in the paid labor force), we were not surprised by their high levels of marital satisfaction.

The men with the lowest levels of marital satisfaction appear to be highly traditional men married to highly egalitarian women. We believe this is likely connected to the fact that such men will face greater pressures to increase their child care and housework participation than other men (Amato et al. 2003; Hochschild 1989). Further, traditional men married to egalitarian women are likely to encounter conflict when the stresses of negotiating paid work and family are evident, as they are unlikely to readily agree on potential solutions and strategies (Hochschild 1989). This interpretation is consistent with Blumstein and Schwartz’s (1983) suggestion that similar gender ideologies provide couples with almost pre-determined solutions to stresses that they may encounter; whereas couples with differing ideologies have no such cut and dried solutions to rely on. The relationships between gender ideologies and men’s marital satisfaction did not emerge until the interactions of spouses’ gender ideologies were added to the model, suggesting that without taking both spouses into account we gain an incomplete understanding of marital relationships. This finding is consistent with Greenstein’s (1996a) assertion that we must consider the interaction of both spouses’ gender ideologies to fully understand marital dynamics. When taking into account only one spouse’s gender ideology, we fail to gain a complete picture of any given marriage.

Not all of the proposed hypotheses were supported in the analysis. Hypothesis 3 concerning the potential positive impact of similar gender ideologies on marital satisfaction was unsupported for women. We think this is because gender ideology likely matters only in conjunction with other variables, such as work-to-family conflict, in shaping dual-earner women’s marital satisfaction. It may be the case that women’s marital satisfaction is shaped by actual behaviors of their husbands, rather than their husbands’ gender ideologies. This interpretation is consistent with the findings from a previous study which indicated that gender attitudes and women’s employment characteristics jointly shape marital quality (Vannoy and Philliber 1992).

Our findings do not support Hypothesis 2. This hypothesis proposed that gender ideology would moderate the relationship between spouse’s work-to-family conflict and marital satisfaction for men. The direct effects show that the work-to-family conflict experienced by men’s partners is negatively related to men’s marital satisfaction, but gender ideology does not appear to play a role in this effect. In particular, Hypothesis 2b, which proposed that spouse’s work-to-family conflict among men with more egalitarian gender ideologies will be unrelated to such men’s marital satisfaction, was not supported. Hypothesis 2a, which suggested that the marital satisfaction of more traditional men would be negatively impacted by their wives’ work-to-family conflict, was supported, inasmuch as this appears to be true for all men, more egalitarian as well as more traditional.

Although the interaction results for Hypothesis 2 are nonsignificant, we note that the strong significance of the women’s gender ideology and work-to-family conflict interaction term shows that this study had the power to detect such effects. The lack of a significant interaction term between gender ideology and (own or partner’s) work-to-family conflict for men shows that if there is such interaction for men, it is likely much weaker than that exhibited by women. Such a result strikes us as interesting and potentially important, as it suggests that men’s gender ideologies do not matter in shaping how work-to-family conflict (that of spouse or partner) relates to marital satisfaction. In particular, women’s work-to-family conflict, regardless of men’s gender ideologies, is associated with men reporting lower levels of marital satisfaction. Egalitarian gender ideologies do not appear to protect men from whatever stresses are associated with women’s work-to-family conflict. Egalitarian men are responding to women’s work-to-family conflict in a way that appears to be the same as more traditional men. In other words, it may be the case that egalitarian men’s reactions to their wives’ work-to-family conflict are not aligned with their professed gender ideologies.

The findings from the present study should be interpreted in light of its limitations. First, the study uses a regional U.S. sample, which limits the external validity of the study and our ability to generalize beyond the sampling frame. Related to the use of the regional sample are the relatively high levels of marital satisfaction and fairly traditional gender ideologies that characterize the sample, as this may restrict the applicability of our findings to samples containing dual-earner couples that are largely dissatisfied and/or dual-earner couples who are more egalitarian in their beliefs. Nonetheless, we are hopeful that there is enough variability in the current dataset to help negate some of these concerns. Moreover, the unique strengths of this particular dataset in terms of the availability of couple data and the high response rate partially mitigate this concern. Second, while our study does shed light on gender ideologies as a key mechanism that shapes how the negotiation of paid work and family roles affect marital satisfaction, we must recognize that gender ideologies are considerably more nuanced and complex than how they are measured in our study. Qualitative work highlights that people’s gender ideologies are often somewhat fractured with discrepancies between how people represent themselves and how people actually feel (Hochschild 1989). Nonetheless, the present study does highlight how gender ideologies (measured quantitatively) do seem to play a role in how work-to-family conflict is related to marital satisfaction.

In conclusion, the present study suggests that considering the gender ideologies of each spouse enhances our understanding of how conflicts between work and family impact marital satisfaction. The findings also have highlighted the importance of contextualizing the experiences of individuals navigating the work-family terrain by taking into account gender and situating individuals in dyads where appropriate. It is critical to take into account both spouse’s perspectives when examining the experiences of dual-earner couples, as each spouse often has their own experience of the marriage that may differ from their partner’s (Bernard 1971). We suggest that future research consider not only the behaviors and experiences of individuals and their spouses, but also the gendered beliefs, attitudes, and perceptions that shape how people interpret their lived experiences. In particular, research using larger datasets could more carefully examine the interplay of both spouses’ work-family variables and both spouses’ gender ideologies in predicting marital outcomes. Such research could integrate the use of more complex interactions that consider simultaneously the role of each spouse’s work-to-family conflict and each spouse’s gender ideologies. We also argue that future research should move further towards the integration of the work-family literature and the division of household labor literature. Given that many of the interpretations of the present study’s findings center on the division of household labor, future studies could more fully consider how housework interacts with work-to-family conflict and gender ideologies in shaping marital outcomes. We think future work could start by considering more fully whether gender ideologies moderate the amount of housework men and women perform in response to their partner’s work-to-family conflict. Do egalitarian men really perform more housework in response to women’s work-to-family conflict than traditional men do? If so, what are the potential impacts of such relationships on the marital quality reported by both spouses? The possibilities for future researchers are rich.