Sex Roles DOI 10.1007/s11199-017-0848-6

ORIGINAL ARTICLE

Relative Earnings and Depressive Symptoms among Working Parents: Gender Differences in the Effect of Relative Income on Depressive Symptoms Karen Z. Kramer 1 & Sunjin Pak 2

# Springer Science+Business Media, LLC 2017

Abstract The relationship between income and psychological well-being is well established. Yet, most of this research is conducted at the individual level without taking into account the role played by relative earnings at the couple level. In the present study we estimate the effect of share of family income on depressive symptoms of individuals. Specifically, we examine whether within-person change in the share of family income has differential effects on the level of depressive symptoms of mothers and fathers. Using data from the U.S. National Longitudinal Surveys of Youth (NLSY79), we follow the same individuals over 4 years and analyze their data using a cross-lagged structural equation model. Controlling for net income, we find that an increase in one’s share of family income is related to an increased level of depressive symptoms among mothers and a decreased level of depressive symptoms among fathers. When looking at a subsample of stay-at-home parents, we find that a change from providing some share of the family income to stay-at-home parent status over time is related to higher level of depressive symptoms among fathers but not mothers. Furthermore, we find that egalitarian gender ideology moderates this relationship for Electronic supplementary material The online version of this article (https://doi.org/10.1007/s11199-017-0848-6) contains supplementary material, which is available to authorized users. * Karen Z. Kramer [email protected] Sunjin Pak [email protected] 1

Department of Human Development and Family Studies, University of Illinois, 905 South Goodwin Ave, Urbana, IL 61801, USA

2

School of Labor and Employment Relations, University of Illinois, Champaign, IL, USA

mothers but not for fathers. We discuss potential implications of our findings to the work-family and gender literature and to counselors and therapists who specialize in treating depression. Keywords Depression . Depressive symptoms . Gender ideology . Parents . Relative earnings . Share of family income Recent decades have witnessed growing variations in the division of paid and unpaid labor in two-parent U.S. families. The increase in women’s labor force participation and education relative to that of men has resulted in a significant decrease in the number of families in which the traditional division of labor, a primary-earner father and a primary-caregiving mother, is observed (Chesley and Moen 2006; Coltrane 2000; Gerson 2010). The division of work and care roles has become increasingly diverse in modern families, reflected in the fact that families in which the mother is earning a significant portion of the family income are growing in number (Kramer and Kramer 2016; Sussman and Bonnell 2006). A growing number of studies examine the relationship between the way in which families divide work, the share of each spouse’s earning, and psychological well-being. Past research has consistently shown that lower income and social class are related to lower reported health and higher rates of depression, either directly (Martikainen et al. 2003) or through employment status and financial strain (Lee et al. 2011; Zimmerman and Katon 2005). Previous studies also have shown that unemployment has a negative effect on psychological well-being, especially on men’s (see meta-analysis by McKee-Ryan et al. 2005). Fewer studies have examined the effects of the share of each spouse’s earnings within families on family outcomes. The share of one’s earnings within

Sex Roles

families is important to consider because in many families, the traditional division of labor between a breadwinner father with a significant share of family income and a caregiving mother with a small share of family income is being replaced by a more egalitarian division of paid labor. The scant research on share of family income has focused on its relationship with marital satisfaction (Brennan et al. 2001; Furdyna et al. 2008) and the division of care and household roles (Bittman et al. 2003; Gupta 2007; Raley et al. 2012; Winslow-Bowe 2006), but has not considered the psychological well-being of spouses. In the present paper we study the effect of spousal share of family income on depressive symptoms of individuals who were born between 1957 and 1965. We consider depressive symptoms as an indicator of psychological well-being (Ryff 1989; Ryff and Keyes 1995). Depression is often referred to as an epidemic and is the leading cause of disability worldwide (World Health Organization 2017). Our main research question is whether share of family income is related to depressive symptoms and whether this relationship is different for men and women. We hypothesize that an increase in share of family income will be related to an increase in depressive symptoms for women and a decrease in depressive symptoms for men. We focus on married other-sex parents in which at least one spouse is working and exclude single-parent families for which share of family income is fixed. We do not examine non-parent families because there is a large difference in paid and unpaid work arrangements between parents and nonparents (Wang et al. 2013), and given our interest in gender roles, we do not include same-sex parents. In addition, we study the role that gender ideology plays in the relationship between share of family income and depressive symptoms. Although gender ideology has, at least up until the 1990s, slowly shifted toward egalitarianism, research has shown that this shift has been much slower for men than for women (Brewster and Padavic 2000) and has mostly stalled since the mid-1990s (Cotter et al. 2011). We expect individuals with a more egalitarian gender ideology to experience a weaker relationship between their share of family income and the level of depressive symptoms when their share of family income violates traditional gender roles—that is, women with more egalitarian gender ideology will have a lower level of depressive symptoms when they increase their share of family income whereas men with less egalitarian gender ideology will have a higher level of depressive symptoms when they decrease their share of family income. The contributions of our paper to the current literature are threefold. First, we expand theoretically and empirically on our understanding of the effect of share of family income on individual depressive symptoms. This is an important topic given the growing prevalence of two-parent families in which the mothers are the primary-earners, equal-earners, or provide a significant share of family income. This demographic shift

affects not only individuals and families, but also organizations, labor markets, and policy. The severity of the depression epidemic in the United States has taken a huge toll on the nation’s mental health (Levine 2007; Rosenzweig and Amsterdam 1992) and demands more investigation at all levels, including at the nexus of family life, income, and work arrangements. Second, we estimate this relationship using a robust cross-lagged structural equation model (SEM) that follows the same individuals over time while controlling for a base depressive symptoms level. As such, we are able to test the relationship between relative earnings and depressive symptoms with a robust design. Finally, we test the roles of gender and gender ideology in individual and family outcomes, providing further evidence for the prevalence of gender disparities in the effect of paid work on health and depressive symptoms.

Share of Family Income and Depression It has long been recognized that both paid and domestic work have gendered meanings. Both are imbued with beliefs and assumptions about gendered roles, as well as how gender is constructed in our society (Chesley 2016; De Henau and Himmelweit 2013; Doucet 2015; Legerski and Cornwall 2010; Thompson and Walker 1989; Wilkie et al. 1998). The gendered division of paid and unpaid work is not a result of a rational reflection of resources and demands that are part of employment and domestic work but rather of socially constructed gender boundaries that structure caregiving work roles as better aligned with the traits frequently assigned to women (and especially mothers) and paid work roles as better aligned with traits traditionally assigned to men (Davis and Greenstein 2013; Thompson and Walker 1989). There, are, however, examples of how these gendered meanings change when income distribution changes within a household. Employed wives, for example, are more likely to expect their husbands to share domestic work than non-employed wives, thus challenging the long-held breadwinner husband norm (Bianchi et al. 2012; Potuchek 1992; Wilkie et al. 1998). Previous studies that have examined the role played by one’s share of income in families have shown its importance to different individual, couple, and family outcomes. For example, an increase in a woman’s share of the family income is positively related to a woman’s marital satisfaction and happiness (Furdyna et al. 2008; Rogers and DeBoer 2001) and negatively related to a husband’s perception of marital quality (Brennan et al. 2001). A husband’s share of family income is negatively related to the probability of a wife being physically or verbally abused (Atkinson et al. 2005; Kaukinen 2004), whereas a wife’s economic independence and higher share of family income are related to higher divorce rates (Cooke 2006; Teachman

Sex Roles

2010), especially when a wife perceives the quality of marriage as low (Nock 2001; Sayer and Bianchi 2000). Mixed results have been found regarding the effect of share of family income on housework. Some studies find that a wife’s higher share of income is related to lower levels of housework participation for her and higher housework participation for her husband (Killewald and Gough 2010; Lam et al. 2012). Other studies suggest that wives who take a more significant role in providing income to the family are engaged in “doing gender” and “gender display” and perform more housework compared to their husbands (Bittman et al. 2003; Brines 1994; Chesley 2011; Greenstein 2000) in order to compensate for their gender deviance in terms of their income role reversal in what has come to be known as “gender deviance neutralization” (Bittman et al. 2003; Schneider 2012; Thébaud 2010). More specifically, earlier studies have shown that greater earning by women is associated with lower investment at home by women and an increased investment by men, as would be predicted by exchange-bargaining theory. But, at a certain level of earnings, couples start to compensate for this violation of gendered expectations. In families in which couples deviate substantially from the normative standard of earnings, men and women may increase behaviors that meet gendered expectation. For example, women might do even more housework and will not push their husband to perform more housework because it would further feminize the men (Bittman et al. 2003; Brines 1994). Other studies support the existence of gender deviance neutralization, especially for men in contexts where work identity and income are highly values (Thébaud 2010) and across different cultures (Simister 2013), although there has been a considerable debate in the literature whether gender deviance neutralization is dependent on culture, actual earnings, and the value individuals assign to paid-work and pay (Gupta 2007; Kluwer 2011; Sullivan 2011a, 2011b). Specifically, Gupta (2006, 2007) criticized the gender deviance neutralization argument by showing that women’s housework is dependent on their absolute income and not on their relative earnings and suggests that high-earning women may, for example, outsource some housework chores if they earn more money, regardless of their partner’s income. An abundance of research that has examined how employment affects psychological well-being has generally shown strong evidence for the positive relationship between employment and psychological well-being for men (Simon 2014) and a more complex relationship between employment and psychological well-being for women. For example, Warr and Parry (1982) summarized results from 38 studies and argued that employment positively affects women’s psychological well-being when they are single, but it has no relationship with psychological well-being when they are married. More recent studies found that women with heavier workloads have lower levels of psychological well-being than do women with lighter

workloads because women still bear the primary burden of taking care of children and home (Gjerdingen et al. 2001; Kleiner and Pavalko 2010). Other studies have shown that women’s employment, and especially full-time employment, is positively related to women’s health (Damaske et al. 2014; Frech and Damaske 2012; Schnittker 2007), but long working hours (over 40 h) is related to poorer health (Kleiner and Pavalko 2010). In addition, longitudinal evidence supports the positive relationship between employment and psychological well-being for both men and women when desired working time matches actual working time but a negative relationship for both underemployment, especially for men, and over-employment, especially for women (Angrave and Charlwood 2015; Kleiner and Pavalko 2010). Some research has shown that the ambiguous findings about the relationship between employment and psychological well-being for women are a result of not taking into account the gender ideology of women (Dereuddre et al. 2014; Perry-Jenkins et al. 1992). Women who work but possess traditional views about gender ideology suffer from lower psychological well-being than do working women who hold beliefs about gender ideology that are more egalitarian (Goldberg and Perry-Jenkins 2004). In general, it seems that the lack of congruence among gender ideology, work preferences (i.e., not working, working part-time, or working fulltime), and actual work is related to higher level of depression among women than when congruence between these factors exists (Klein et al. 1998; Perry-Jenkins et al. 1992). Although there is strong evidence of the relationship among employment, earnings, and psychological well-being at the individual level, there are very few studies that have examined the effects of this relationship at the family level. Specifically, studies that have analyzed how employment of both spouses and share of family income affect the psychological well-being of both husbands and wives are scarce (Dereuddre et al. 2014). There is some evidence that an increase in women’s share of the family income is positively related to their psychological well-being but negatively related to their spouses’ psychological well-being (Rogers and DeBoer 2001). Similarly, gender egalitarianism in employment is associated with lower depression for women, but not for men (Bourne 2006). The relationship between share of family income and psychological well-being can be different for men and women. A larger share of the family income provides greater bargaining power to its holder (Greenstein 2000) and, as a result, may positively affect the psychological well-being of the higher earner who, for example, can devote more time to healthpromoting behavior such as exercise, relaxation, and leisure. However, this may be true for fathers more than for mothers because of social expectations about gendered division of labor, especially where children are concerned (Acker 1990; Sanchez and Thomson 1997; Yavorsky et al. 2015).

Sex Roles

Specifically, whereas being a father and a primary-earner are congruent with the prevailing gender ideology and gendered social expectations, being a mother and a primary-earner are incongruent with them (Chesley 2016; Damaske 2011; Livingston 2014b). And, although mothers today are expected to work outside the home, the predominant expectation is that fathers will not have primary responsibility for household and caregiving roles whereas the expectations held for mothers are that they will be secondary earners with greater responsibility for childcare and household, especially when they have young children at home (Kramer and Kramer 2016; Poortman and Van der Lippe 2009). In sum, we suggest that an increase in the share of family income will have different effect on mothers’ and fathers’ depressive symptoms. The gender deviance neutralization perspective suggests that women and men respond to their share of income in the household in terms of the amount of time they spend on unpaid household work (Simister 2013). Such change in behaviors entails more effort (in the household) from mothers (and not from fathers) that increases when they increase their share of family income. Further, increase in share of family income will positively affect mothers depressive symptoms even if they do not exert more effort at home (or even if they exert less effort at home) because of the deviance from expected gendered behaviors and roles that will create more stress for mothers. For fathers, the opposite is true. An increase in the share of family income is likely to decrease their participation in unpaid-work at home and allow them to have more time for health-promoting behavior, relaxation, and leisure which will translate to lower levels of depressive symptoms. It also reaffirms fathers’ expected gendered behavior of being a primary provider. In our study, we do not compare men and women directly because traditional depression measures show much higher depression levels for women than for men, a topic that is highly debated in the depression literature (Van de Velde et al. 2010). Comparing women to women and men to men accounts for any gender biases in the depression measure used in our study. In addition, we note that although we analyze relative share of family income, our analysis is done at the individual level and not at the couple level because depressive symptoms are measured only at the individual level.

Depressive Symptoms of Stay-at-Home Parents Stay-at-home parents are qualitatively different from secondary-earner parents not only in earnings but also in terms of their responsibilities at home. Even more importantly, stayat-home mothers and fathers are different from each other in their characteristics (Kramer et al. 2015). Married stay-athome mothers were the majority among working-age U.S. mothers until the 1970s, but their population has shrunk

consistently since then, becoming stable at around 25% in the early 1990s to present times (Cohn et al. 2014). Later cohorts of married stay-at-home mothers tend to hold more traditional gender ideology than working mothers do, are less educated and poorer, and stay at home to take care of children (Cinamon and Rich 2002; Damaske and Frech 2016; GarcíaManglano 2015). Societal perceptions of stay-at-home mothers tend to be more positive than those of career women and primary-earner mothers, who are often depicted as being more concerned with their personal success and career than the well-being of their children and family (Dillaway and Paré 2008; Johnston and Swanson 2003, 2004). Past research has found that primary-earner mothers tend to report that they experience the stress of being responsible for the family’s financial health as well as the guilt associated with not having enough time to be more involved with their children (Chesley 2016). However, most married mothers are not primaryearners in their household, especially in older generations. As such, although it might be expected that a decision by a mother to stay at home to take care of children will alleviate some of the stress that is associated with needing to meet both work and family roles (Chesley 2016), for most women, a decision to switch from being a secondary-earner to a fulltime stay-at-home mother will not represent a significant change in responsibilities and demands. Specifically, for mothers who change from having some share of the family income to no share at all, we expect that depressive symptoms will not change. In contrast to stay-at-home mothers, the majority of stay-athome fathers are not working because of illness, disability, or inability to find work, and the minority among the (about 20% of all stay-at-home U.S. fathers in the 2000s) are fathers who tend to hold a more egalitarian gender ideology and choose to stay at home (Kramer et al. 2015). Stay-at-home fathers are estimated to be only 3–7% of the U.S. parent population, depending on whether the definition is more restrictive to include only fathers who do not work at all and have a fulltime working spouse (Kramer et al. 2015) or fathers who may be working but are the main providers of care for children (Livingston 2014a). Societal perceptions of stay-at-home fathers are still very negative; they are often perceived as lazy and incompetent men who are lacking in traditionally-respected masculine traits, and in general they are viewed by both men and women as nonnormative (Doucet 2004; Fischer and Anderson 2012; Rochlen et al. 2010). Furthermore, because work identity is critical for men’s self-esteem, not working is likely to have an unfavorable effect on men’s depressive symptoms. Therefore, it is likely that the effect of shifting from working for pay and providing a share of the family income to being a stay-at-home father who does not provide any income will be associated with increased depressive symptoms among fathers.

Sex Roles

The Moderating Role of Gender Ideology

Method

From the 1960s until the mid-1990s in the United States, gender ideology had progressed toward greater egalitarianism, but has changed little since then in what has been described as the uneven and stalled revolution (Cotter et al. 2011; England 2010, 2011). Specifically, women’s perceptions of gendered roles and femininity have changed significantly more than men’s perceptions of gendered roles and normative masculinity. Part of this uneven change in gendered perceptions has been attributed to the low rewards women receive for performing unpaid work (e.g., childcare) which incentivize women to leave these roles and join the labor market. Men on the other hand do not have any substantial incentive to perform roles that are associated with women’s roles and hence their change in gender ideology is uneven (England 2010, 2011). Furthermore, social class may play a role here, with women mostly entering male-dominated professional and managerial jobs and not blue-collar male-dominated jobs. However, individuals and families vary greatly in their gender ideology (Bianchi et al. 2000; Raley et al. 2012). More egalitarian gender ideology has been shown to lead to more egalitarian division of labor at home (Bianchi et al. 2000; Raley et al. 2012) and to greater participation of women in the labor market, women’s accelerated career growth, and increase in work hours (Corrigall and Konrad 2007; Hakim 2002; Nordenmark 2004). Men with more egalitarian gender ideology are more likely to care for children and do more housework (Chesley and Flood 2017; Fischer and Anderson 2012; Lam et al. 2012; Sanderson and Thompson 2002). Greater acceptance of egalitarian gender ideology may moderate the relationship between share of family income and depressive symptoms for both fathers and mothers. For fathers, egalitarian gender ideology may not only negate the effect of earning less than a spouse or not having any income from work, but may also increase fathers’ satisfaction with their functioning in the family domain if they perceive it as central to their identity (Shows and Gerstel 2009). As such, the level of depressive symptoms of fathers is likely to be affected less by their share of family income when they possess more egalitarian gender ideology. For mothers, a more egalitarian gender ideology may increase an acceptance of their own role as a primary-earner and reduce the pressure of doing gender. As a result, mothers who hold a more egalitarian gender ideology will experience less pressure in respect to the expectation that they will be the primary caregiver and that their higher share of family income represents a deviation from expected gendered roles (Chesley 2016; Raley et al. 2012). Hence, more egalitarian gender ideology will reduce the unfavorable effect of increasing one’s share of family income on mothers’ level of depressive symptoms.

Data for our study comes from the National Longitudinal Surveys of Youth (NLSY79), a nationally representative panel study administered by the United States Department of Labor and the Bureau of Labor Statistics. Starting with 12,686 individuals in 1979, participants were interviewed annually through 1994 and on a biennial basis since then. Participants were 14–22 years -old when first surveyed in 1979. Funding constraints led to the omission of 1079 individuals in 1984 and of 1643 individuals in 1990. All respondents to the NLSY79 were born between 1957 and 1965 and are part of the late Baby-Boom generation. Although this sampling may restrict the generalizability of our findings to later birth cohorts, it also provides more robust estimates of the proposed hypotheses because many exogenous variables are fixed at the generation level. Retention rate of participants in the NLSY79 is extremely high, with about 89% retention rate for the study’s period (NLSY79 2016). An analysis conducted of item nonresponse of NLSY79 data concluded that only 1.5% of non-responses in the data could be classified as missing data, and about half of those were “don’t know” responses (NLSY79 2016). Overall, the NLSY79 is a high-quality sample and has been used in more than 5000 peer-reviewed publications (NLS 2016). Sample Our study is focused on share of family income as an antecedent of depressive symptoms among mothers and fathers. Hence, only two-parent other-sex households were included in the sample. Specifically, the sample is restricted to participants who met the following criteria: Individuals had to be married or in a committed relationship and at least one of the spouses had to be working full time so that share of family income could be calculated. Because the study focuses on parents, only families with at least one child under age 13 at home (adopted or biological) were included. This age cutoff was selected based on previous studies which used a similar cutoff and because children 13 and older receive lower levels of involvement by parents (e.g., Brayfield 1995; Higgins et al. 1994; Hofferth and Sandberg 2001; Morrissey and Warner 2009). Because low income and class are strongly associated with depressive symptoms and may trump any other effect (Lee et al. 2011; Martikainen et al. 2003), we have only included families with incomes that were above the poverty threshold (United States Census Bureau 2016). Our final sample includes 1463 fathers and 1769 mothers between 1991 (Time 1) and 1994 (Time 4). It is important to note that although share of family income is calculated at the couple-level, all analysis is done at the individual level (i.e., no couples are included in our study) because depressive symptoms were only measured for individual respondents.

.19 .21 30.06

Black a Hispanic b Age

Education (T1) Unemployment rate (T1) Family income (T1) Family income (T3) Egalitarian gender ideology Share of family income (T1)

Share of family income (T3) Stay at home status (T1) Stay at home status (T3) Depression (T2) Depression (T4)

1. 2. 3.

4. 5. 6. 7. 8. 9.

10. 11. 12. 13. 14.

.28 .20 .21 3.34 3.12

2.40 32.54 160,690 37,910 .95 .28

.39 .41 2.25

SD

.38 .21 .18 4.33 4.03

13.13 74.44 63,182 48,231 .23 .37

.21 .20 30.08

M

Female

.32 .41 .38 4.04 4.14

2.11 33.58 137,813 35,352 .99 .32

.41 .40 2.24

SD

2 −.25* – −.06 −.18* .17* −.07 −.07 −.11* .10* .02 −.02 .01 .04 .08

1 – −.24* −.03 −.05 −.13* −.03 −.09* .09* −.04 −.05 .00 −.01 .10* .02

Correlations

.00 .02 .02 −.04 −.03

.03 −.05 .09* .14* −.07 .05

−.02 −.01 –

3

−.03 −.04 −.06 −.11* −.16*

– −.12* .26* .38* .24* −.10

.01 −.16* .04

4

.02 −.01 −.01 −.01 .02

−.07* – −.09* −.15* −.09* .06

−.09* .14* −.01

5

−.08* −.01 −.01 −.03 .01

.20* −.04 – .51* .07 −.26*

−.05 −.03 .06

6

−.24* .01 −.08* −.08* −.11*

.32* −.11* .47* – .10* −.21*

−.15* −.05 .12*

7

−.06 −.04 −.06 −.11* −.04

.20* −.06 .07* .07* – −.11*

.08* −.06 −.06

8

.50* −.47* −.20* .00 −.01

.06 −.05 −.20* −.16* .16* –

.19* .04 −.03

9

– −.21* −.53* −.02 −.05

.05 −.04 −.08* −.24* .15* .67*

.21* .04 −.04

10

−.41* – .35* .10* .07

−.10* .05 −.01 −.05 −.18* −.58*

−.08* .00 −.01

11

−.55* .51* – .03 .06

−.06 .01 .06 −.02 −.11* −.39*

−.06 −.06 .00

12

.03 .04 .02 – .29*

−.09* −.02 −.07* −.13* −.04 .05

.09* .01 −.03

13

.06 .03 .00 .40* –

−.18* −.01 −.07* −.14* −.05 .01

.07* .01 −.05

14

*

b

a

p < .01

Denotes dummy variable codes as 1 = yes, 0 = White

Denotes dummy variable coded as 1 = yes, 0 = White

n = 1463 for men n = 1769 for women. Correlations for men are reported below the diagonal; for women, above. For the correlations with egalitarian gender ideology, n = 1,340 for men; n = 1,799 for women. For the share of family income and depressive symptoms, the raw scores of means and standard deviations are reported. Sample is restricted to married individuals with at least one child and earnings above poverty line, which vary according to the number of children

.69 .04 .05 3.30 2.64

13.02 75.12 73,906 51,633 −.18 .68

M

Male

Descriptive statistics of study variables by gender

Variables

Table 1

Sex Roles

Sex Roles

In Table 1 we present descriptive statistics and correaltions between the study’s variables for men and women. As can be seen, men have a significantly lower level of depressive symptoms (p < .001) and a higher share of family income (p < .05) than do women. There is no statistically significant difference between men and women in terms of age, educational level, and local unemployment rate. Respondents at Time 4 were, on average, 33.20 years old. In our sample, men’s average share of family income is 69% and women’s average share is 38%, and men have a lower level of depressive symptoms than women do (2.97 and 4.18, respectively). Measures Depressive Symptoms Depressive symptoms were measured using the seven-item scale of the Center for Epidemiological Studies Depression Scale (CES-D) at Time 2 (1992) and Time 4 (1994). They were not measured in any other waves of data until individuals reached age 40 and 50 (starting in 1998 and 2008). Two sample items are: “I felt that everything I did was an effort” and “I felt depressed.” Higher averaged scores indicate a higher level of depressive symptoms and range from 0 (rarely/none of the time) to 3 (most/all of the time/5–7 days last week). The sevenitem scale was aggregated and standardized by predicting a regression factor score using a principal components method (Distefano et al. 2009). Cronbach’s alphas for the depressive symptoms measure are .75 at Time 2 and .80 at Time 4. The depressive symptoms score was normalized by logarithmic transformation because the distribution of the raw score shows a highly skewed distribution (Keene 1995). Share of Family Income Each year, participants were asked about their income and about the total net family income in past calendar year. The share of family income at Times 1 and 3 were computed by dividing the individual’s income by the total family income at Times 1 and 3, respectively. The time lag between the measurement of share of family income and depressive symptoms is less than one year (i.e., 5 to 12 months) due to the time of data collection (between May and December each year). For example, at Time 4 between May to December, respondents reported their depressive symptoms during the past week and also reported their income as well as their family income in the previous year. Because the inference is based on the ratio of income, we used the natural log of the individual’s income divided by family income. We do that because sole-earners have small baseline values (total net family income) and this could cause biased estimates of the effects of share of family income if not using log transformation (Keene 1995).

Egalitarian Gender Ideology In 1987, participants were asked about their opinion regarding women’s roles in work and family domains using eight items rated on a 4-point scale from (strongly agree) to 4 (strongly disagree). Two sample (reverse-scored) items are: “A women’s place is in the home, not in the office or shop” and “It is much better for everyone concerned if the man is the achiever outside the home and the women takes care of the home and family.” We constructed the variable such that a higher averaged score represents a higher level of egalitarian gender ideology. The eight-item scale was aggregated and standardized by predicting a regression factor score using a principal components method. Cronbach’s alpha for the egalitarian gender ideology measure is .76. Covariates Individual characteristics including race/ethnicity (White, Black, and Latino), age, number of years of education (Time 1), unemployment rate (Time 1), and log of total net family income (Time 1 and 3) were used as control variables for predicting the initial share of family income (Time 1) and depressive symptoms (Time 2) to control potential extraneous variance in our analysis. Analysis We test the cross-lagged share of family income → depressive symptoms relationship using structural equation modeling (SEM) with STATA 14.0. Cross-lagged analysis using SEM has several advantages compared to other cross-sectional design (Lang et al. 2011). Specifically, our analysis allows taking into account initial level of depressive symptoms at Time 2 and observing the effect of share of family income on changes in depressive symptoms from Time 2 to Time 4. However, cross-lagged relations have a possibility of endogeneity in that the reciprocal relationship may be influenced by shared sources (Antonakis et al. 2010). Therefore, following prior recommendations (Antonakis et al. 2010), we included exogenous variables at Time 1 (i.e., race/ethnicity, age, years of education, unemployment rate, and family income) that do not depend on share of family income at Time 1 and 3 and depressive symptoms at Time 2 and 4 (Maynard et al. 2014). This modeling technique provides unbiased estimates of the relationship between share of family income and depressive symptoms. To evaluate the fit of the models, we report Chisquare, root mean square error of approximation (RMSEA) and its 90% confidence interval, comparative fit index (CFI), and standardized root mean square residual (SRMR). The indicators of a good fit to the data are CFI ≥ .90, RMSEA ≤ .08 or .06, and SRMR ≤ .08 (Hu and Bentler 1998). As a robustness check we have also conducted within-person (fixed-

Sex Roles

effects) analyses which are available as an online suppelement. The results of these analyses were consistent with the results presented in this paper.

Results Table 1 presents descriptive statistics and correlations of the study variables by gender. Correlations for women are above diagonal and for men are below diagonal. The correlation matrix provides support for the established negative relationship between depressive symptoms on the one hand and income and education on the other hand. Other variables are not consistently correlated with depressive symptoms as measured at Time 2 and Time 4.

reported by mothers and fathers in our sample. The overall fit of our structural model was at an acceptable level, χ2 (40) = 316.30, p < .001 (RMSEA = .07, 90% CI [.06, .07], CFI = .90, SRMR = .03). Path coefficients for mothers and fathers are shown to the right and left of forward slashes, respectively, in Fig. 2. We find that for mothers, a stay-athome status was not associated with depressive symptoms (β = −.01, p = .66). In contrast, for fathers, a stay-at-home status was positively associated with depressive symptoms (β = .07, p = .01). In other words, fathers who stopped working and have a wife who is the sole-earner experience an increase in their depressive symptoms over time. In sum, we find that becoming a non-working parent is unrelated to changes in mothers’ depressive symptoms but is unfavorably related to an increase in fathers’ depressive symptoms over time.

Share of Family Income and Depressive Symptoms We first test the effect of share of family income on depressive symptoms of mothers and fathers in our sample. The overall fit of our structural model was at acceptable level, χ2(40) = 363.08, p < .001 (RMSEA = .07, 90% CI [.06, .08], CFI = .92, SRMR = .03). Path coefficients for mothers and fathers are shown to the right and left of forward slashes, respectively, in Fig. 1. We found a significant positive relationship between share of family income and depressive symptoms of mothers. Specifically, controlling for prior level of depressive symptoms, a mother’s share of family income at Time 3 is positively associated with her depressive symptoms at Time 4 (β = .07, p = .02). This result can be interpreted as an increase of .07 in a mother’s level of depressive symptoms when her share of income increases by one standard deviation over time. The results from the fathers’ sample show an opposite relationship. Specifically, we found that fathers in our sample show a significant negative relationship between their share of family income at Time 3 and their depressive symptoms at Time 4 (β = −.10, p < .001). This can be interpreted as an average decrease of .10 in a father’s depressive symptoms level when his share of family income increases by one standard deviation over time. In sum, one’s share of family income has opposite effects on mothers and fathers; mothers’ share of family income is related unfavorably to changes in level of depressive symptoms over time, whereas fathers’ share of family income is related favorably to changes in the level of depressive symptoms over time. Depressive Symptoms among Parents Who Stopped Working We continue to test the effect of stay-at-home status (i.e., parents who stopped working and no longer have any share of family income) on the level of depressive symptoms as

The Moderating Role of Egalitarian Gender Ideology Finally, we tested the moderating effect of egalitarian gender ideology on the relationship between share of family income and depressive symptoms using moderated regression analysis (see Table 2). Egalitarian gender role and share of family income were mean-centered before the interaction term was created (Aiken and West 1991). We find that egalitarian gender ideology did not moderate the relationship between share of family income and depressive symptoms among fathers, but did moderate it for mothers. Fathers who experience a decrease in their share of family income experienced an increase in their depressive symptoms (β = −.06, p = .04) regardless of their egalitarian gender ideology. For mothers, however, the effect of share of family income on depressive symptoms was moderated by egalitarian gender ideology in the expected direction. As can be seen in Fig. 3, simple slope tests confirmed that mothers who hold more egalitarian gender ideology (one standard deviation above the mean) experience a decrease in their depressive symptoms (β = −.06, p = .04) as their share of family income increases. In contrast, mothers who hold less egalitarian gender ideology (one standard deviation below the mean) experience an increase in their depressive symptoms (β = .06, p = .04) as their share of family income increases. For mothers who hold an average level of egalitarian gender ideology, we find no relationship between share of family income and depressive symptoms. We also test the hypotheses using fixed-effects (withinperson) analyses. The results from the fixed-effects estimation were consistent with the results from cross-lagged path analysis and provide further support for our findings. Results of these analyses are available as an online supplement (see Tables 1s and 2s).

Sex Roles

Fig. 1 Path coefficients of the hypothesized model for the effect of share of family income by gender. For clarity, correlations among exogenous variables are omitted (see Table 1), and factor loadings that are not significant for both genders are omitted. n = 1463 for fathers and n = 1769 for mothers. Values shown are standardized parameter.

Coeffcients on the left of forward slashes are for fathers whereas coefficients on the right are for mothers. Solid black lines indicate paths that are signfiicant for both genders, dashed black lines are signficant for only gender, and dotted grey lines are nonsignificant for both genders. *p < .05. **p < .01. ***p < .001

Fig. 2 Path coefficients of the hypothesized model for the effect of stayat-home status (by gender). For clarity, correlations among exogenous variables are omitted (see Table 1), and factor loadings that are not significant for both genders are omitted. n = 1463 for fathers and n = 1769 for mothers. Values shown are standardized parameter.

Coeffcients on the left of forward slashes are for fathers whereas coefficients on the right are for mothers. Solid black lines indicate paths that are signfiicant for both genders, dashed black lines are signficant for only gender, and dotted grey lines are nonsignificant for both genders. *p < .05. **p < .01. ***p < .001

Sex Roles Table 2 Summary of regression analyses predicting depressive symptoms (Time 4)

Men Independent variables

Women β

B

SE B

β

2.12

***

.50

.03 .03

.02 .05

.11 −.05

***

.00 .00

−.02 −.11

−.01 −.07

−.03 −.09

.00 −.11

B

Constant

1.24

Black a

.02 .04

SE B ***

.20

Hispanic b Age (Time 1) Education (Time 1)

.00 −.02

Unemployment (Time 1) Family income (Time 3)

.00 −.04

***

.00 .01

Share of family income (Time 3)

−.12

*

.06

−.06

.22

Egalitarian gender role Share of family income × Egalitarian gender role R2

.00 −.01

.01 .06

−.01 −.01

−.08 −.24

.03***

.07 .07

.04 −.02

***

.01 .01

−.02 −.13

***

.00 .04

−.04 −.08

.13

.05

.03 .12

−.07 −.06

* *

.04***

n = 1,422 for men; n = 1,751 for women. Share of family income and egalitarian gender role were centered at their means a

Denotes dummy variable coded as 1 = yes, 0 = White

b

Denotes dummy variable codes as 1 = yes, 0 = White

*p < .05. **p < .01. ***p < .001

Discussion The present study suggested that the relationship between share of family earnings and depressive symptoms is different for men and women. Supporting our general hypothesis, we find that an increase in the share of family income over time is related to a decrease in the level of depressive symptoms among men and an increase in the level of depressive symptoms among women. However, we also show that for women, Fig. 3 Moderating effect of egalitarian gender ideology on the relationship of share of family income (T3) and depressive symptoms (T4) among women. Depressive symptoms variable (Time 4) was standardized

but not for men, this relationship is moderated by gender ideology: women with more egalitarian gender ideology who increase their share of family income experience a decreasing level of depressive symptoms over time. Finally, we find that the level of depressive symptoms is increasing among men, but does not change among women, when they cease working and have become financially dependent on their spouse. The division of paid and unpaid labor in two-parent U.S. families has been going through substantial changes in recent

Sex Roles

decades. In the 1950s, at least among Whites (Boustan and Collins 2014), an overwhelming majority of fathers had primary responsibility for paid work and very little responsibility for unpaid work. In contrast, an overwhelming majority of mothers had primary responsibility for unpaid work and very little responsibility for paid work. Over time, the division of labor has gradually evolved to embody a more egalitarian division of paid and unpaid work. And, although the gendered division of paid and unpaid labor is far from parity, U.S. women’s share of family income has been constantly increasing. For example, the percent of wives who share equally in providing income increased from 9% in 1970 to 24% in 2001 (Raley et al. 2006), and more mothers today out-earn their husbands than ever before (Wang et al. 2013), with the number of wives who are primary or sole earners increasing from 4% in 1970 to 12% in 2001 (Raley et al. 2006). Yet, this rapid change in women’s share of family income has not been followed by a similar change toward a more egalitarian gender ideology, which has changed slowly, especially in the last 20 years (England 2010). As a result, a violation of gendered roles may result in negative outcomes for individuals. We find that mothers who earn a greater share of family income over time, as well as fathers who earn a lower share of family income over time, experience an increase in their depressive symptoms, perhaps as a result of the nonconformity of their pattern of relative earnings. In our study, we have focused on how change in relative earnings affects the level of depressive symptoms of mothers and fathers who are part of the Baby-Boomers cohort. This focus has afforded us the opportunity to observe what the consequences of this “stalled revolution” look like in finer detail, as well as arguably what its psychological toll has been on families pushing at the margins beyond traditional gendered meanings of family and professional life. In the general population, women suffer from major depression at almost twice the rate of men (Kendler and Prescott 1999; Martin et al. 2013). Indeed, we observe a statistically significant and substantial difference in levels of depressive symptoms between men and women in our study. Yet, once mothers and fathers are categorized into groups based on their relative earnings, the gender differences in depressive symptoms show a pattern that is based either on conformity, or lack of conformity, to the demands of aligning oneself with gendered expectations. Mothers who increase their share of family income report higher level of depressive symptoms whereas fathers who increase their share of family income over time report a lower level of depressive symptoms. Furthermore, depressive symptoms of mothers who stop working do not change, but fathers who stop working report a significant increase in depressive symptoms. Although these findings may be unique to the more traditional late BabyBoomers cohort, they support the overarching hypothesis of our study: More depressive symptoms are reported by mothers

and fathers who violate gendered expectations about the division of paid labor and are lower for mothers and fathers who conform to the gendered expectations of the division of paid labor. The results of our study suggest that the change toward more egalitarian gender ideology is lagging behind more egalitarian, gender-based relative earnings. It is possible that while human capital (e.g., the growing gap in education in favor of women; DiPrete and Buchmann 2006) and changes in the labor market (e.g., the decline of male-dominated occupations and the rise of female-dominated occupations) are pushing toward an increase in women’s relative share of family income, perceptions of what the relative earnings “should be like” is still more traditional and less egalitarian. As Ridgeway (1997, p. 218) succinctly puts it: “…changes in the status dimension of gender stereotypes lag behind changes in resource inequalities.” It seems that social norms regarding gendered roles are much slower to change than are labor market opportunities for women and that this lag may lead mothers and fathers who do not conform to gendered expectations about earnings to suffer from a higher level of depressive symptoms. A possible interpretation of our findings, although we are unable to test it, is that there may be a conflict between work and family identities and conventionally-accepted responsibilities for mothers and fathers, especially among individuals belonging to older cohorts. Mothers who increase their share of family income but still have a working spouse are likely to still bear the primary responsibility at home. As such, mothers may feel pressure from both life domains (Hochschild 1989) and perceive their effort and success at both work and family as insufficient (Jones 2012). This perceived deficiency will result in increases in depressive symptoms of mothers compared to fathers. Our study also indirectly supports the gender deviance neutralization argument (Bittman et al. 2003; Schneider 2012; Thébaud 2010). Mothers who increase their share of family income and, presumably, their bargaining power, report an increase in their depressive symptoms. This may happen because, as proposed by the gender deviance neutralization framework, they also increase their housework to compensate for their presumed deviance. Their depressive symptoms may therefore increase directly or indirectly. They may increase directly as a result of the stress created by violating norms (i.e., by increasing their share of family income) or indirectly as a result of growing demands on their resources at both work and home. Fathers may perceive their decreased share of family income as a threat to their identity and masculinity or their increased share of family income as a confirmation of their identity (Chesley 2011, 2016; Doucet 2004). Fathers who decrease their share of family income may also perceive their performance in domestic work and as caregivers as poor (Hudson et al. 2001). As a result, they may also suffer from

Sex Roles

a higher level of depressive symptoms when they decrease their share of family income (or cease to work) or benefit from a lower level of depressive symptoms when increasing their share of family income. Finally, we find that egalitarian gender ideology moderates the share of family income and depressive symptoms relationship of mothers but not of fathers. This pattern supports past findings that gender ideology became especially more egalitarian among women but not among men (Brewster and Padavic 2000) and thus provides further support for the uneven revolution (England 2010). Specifically, mothers who had a more egalitarian gender ideology experienced a decrease in their reported level of depressive symptoms when their share of family income increased, suggesting that the affirmation of their and work identity along with increased bargaining power at home (Greenstein 2000) favorably impacted their level of depressive symptoms. Fathers, however, did not experience this moderating effect, suggesting that perhaps work identity and the traditional role of primary earner are still critical for men, even when they hold more egalitarian gender ideology. It also is possible that this finding is a result of the much slower pace toward more egalitarian perceptions of gendered roles and the persistence of normative masculinity among men. Limitations and Future Research Directions No study is without limitations. Although using the NLSY79 offers many advantages in terms of a longitudinal representative sample, it also includes some inherent restrictions. Our study is limited to a cohort of mothers and fathers who were born in the late 50s and early 60s (late Baby-Boomers) and as such, they are likely to have similar gender ideology, which might limit the ability to test its moderating effect. It is also possible that for later cohorts, the relationships reported in our study are different, assuming they hold, on average, more egalitarian perceptions of gendered roles and expectations. In addition, depressive symptoms are measured for all respondents at the same time only in 1992 and 1994. It would have been more beneficial to examine the relationship between earner status and depressive symptoms over a longer, but also more contemporaneous, period. Still, given the minor changes in both women’s labor participation rate and gender ideology over the last 20 years, the results of our study are relevant to researchers and the literature. Future studies may want to explore if the relationship between relative earning and depressive symptoms varies across different social classes. Although we do control for income and include only families with income above the poverty line in our study, past research has shown that income is strongly related to depression and depressive symptoms (Martikainen et al. 2003). It is possible that, for example, at lower levels of income, total income trumps share of family income because

the financial stress and difficult economic realities are detrimental to depressive symptoms and other well-being factors (Lee et al. 2011; Martikainen et al. 2003). It might be beneficial to take a more comprehensive view of social class in general when studying depression. Future studies may also want to examine the relationship between relative earnings and depressive symptoms over an extended period and with different cohorts. It is also important to examine the mechanisms that underlie the relationship between relative earnings and depression. For example, it is possible that congruence between the desired amount of weekly working hours (which are related to income) and the actual amount of weekly working hours mitigates the relationship between relative earnings and depression. Such a future study will be able to identify, for example, if incongruence in the form of over-employment is different in its relationship with depression than incongruence in the form of under-employment, and whether such differences are also dependent on gender. Practice Implications Depression is a debilitating illness that is widespread in the U.S. population. Approximately 16% of Americans experience major depression any given year (Kessler et al. 2003). Although the antecedents of depressive symptoms vary greatly at the individual level, our study suggests that practitioners may also consider not just net income and financial stress as antecedents of depressive symptoms but also relative income. For example, a man who experienced a sudden increase in his wife’s share of family income may experience an increased level of depressive symptoms. Similarly, women who substantially increase their share of family income may also experience an increased level of depressive symptoms because they may compensate for their “deviance” from prescribed gendered expectation by doing gender. As a result, a therapist may, for example, encounter a couple which will present a paradox: a substantial increase in family income (and presumably, financial stability) may increase depressive symptoms of both spouses if the wife is the one who increased her income, especially if the woman holds a more traditional gender ideology. Conclusion Gender inequality is persistent and difficult to change. Even as labor market outcomes become more egalitarian, other gender inequalities persist and may be much harder to change. The opposite effect for men and women when their share of family income increases on their depressive symptoms suggests that gender inequalities should not be addressed just at the labor market level but also at the perceptual and attitudinal individual level if families and society are to truly achieve gender equality.

Sex Roles Acknowledgements This work was supported by the USDA National Institute of Food and Agriculture, Hatch project 232659. We would like to thank Deborah Ostrovsky for her helpful review and comments. Compliance with Ethical Standards This paper and its authors fully comply with the ethical standards set forth by the University of Illinois and Sex Roles. Conflict of Interest There is no conflict of interest and the research utilizes publicly available data.

References Acker, J. (1990). Hierarchies, jobs, bodies: A theory of gendered organizations. Gender and Society, 4, 139–158. https://doi.org/10.1177/ 089124390004002002. Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting interactions. Thousand Oaks: Sage Publications. Angrave, D., & Charlwood, A. (2015). What is the relationship between long working hours, over-employment, under-employment and the subjective well-being of workers? Longitudinal evidence from the UK. Human Relations; Studies Towards the Integration of the Social S c i e n c e s , 6 8 , 1 4 9 1 – 1 5 1 5 . h t t p s : / / d o i . o r g / 1 0 . 11 7 7 / 0018726714559752. Antonakis, J., Bendahan, S., Jacquart, P., & Lalive, R. (2010). On making causal claims: A review and recommendations. The Leadership Quarterly, 21, 1086–1120. https://doi.org/10.1016/j.leaqua.2010. 10.010. Atkinson, M. P., Greenstein, T. N., & Lang, M. M. (2005). For women, breadwinning can be dangerous: Gendered resource theory and wife abuse. Journal of Marriage and Family, 67, 1137–1148. https://doi. org/10.1111/j.1741-3737.2005.00206.x. Bianchi, S. M., Milkie, M. A., Sayer, L. C., & Robinson, J. P. (2000). Is anyone doing the housework? Trends in the gender division of household labor. Social Forces, 79, 191–228. https://doi.org/10. 2307/2675569. Bianchi, S. M., Sayer, L. C., Milkie, M. A., & Robinson, J. P. (2012). Housework: Who did, does or will do it, and how much does it matter? Social Forces, 91, 55–63. https://doi.org/10.1093/sf/sos120. Bittman, M., England, P., Sayer, L., Folbre, N., & Matheson, G. (2003). When does gender trump money? Bargaining and time in household work. American Journal of Sociology, 109, 186–214. https://doi.org/ 10.1086/378341. Bourne, H. (2006). Gender ideology, depression, and marital quality in working-class, dual-earner couples across the transition to parenthood (Unpublished doctoral dissertation). Amherst, Massachusetts: University of Massachusetts Amherst. Boustan, L. P., & Collins, W. J. (2014). The origin and persistence of Black-White differences in women’s labor force participation. In L. Platt Boustan, C. Frydman, & R. A. Margo (Eds.), Human capital in history: The American record (pp. 205–240). Chicago: University of Chicago Press. Brayfield, A. (1995). Juggling jobs and kids: The impact of employment schedules on fathers' caring for children. Journal of Marriage and the Family, 57, 321–332. https://doi.org/10.2307/353686. Brennan, R. T., Barnett, R. C., & Gareis, K. C. (2001). When she earns more than he does: A longitudinal study of dual-earner couples. Journal of Marriage and Family, 63, 168–182. https://doi.org/10. 1111/j.1741-3737.2001.00168.x. Brewster, K. L., & Padavic, I. (2000). Change in gender-ideology, 1977– 1996: The contributions of intracohort change and population turnover. Journal of Marriage and Family, 62, 477–487. https://doi.org/ 10.1111/j.1741-3737.2000.00477.x.

Brines, J. (1994). Economic dependency, gender, and the division of labor at home. American Journal of Sociology, 100, 652–688. https://doi. org/10.1086/230577. Chesley, N. (2011). Stay-at-home fathers and breadwinning mothers: Gender, couple dynamics, and social change. Gender and Society, 25, 642–664. https://doi.org/10.1177/0891243211417433. Chesley, N. (2016). What does it mean to be a “breadwinner” mother? Journal of Family Issues. Advance online publication. https://doi. org/10.1177/0192513X16676857. Chesley, N., & Flood, S. (2017). Signs of change? At-home and breadwinner parents' housework and child-care time. Journal of Marriage and Family, 79, 511–534. https://doi.org/10.1111/jomf.12376. Chesley, N., & Moen, P. (2006). When workers care: Dual-earner couples' caregiving strategies, benefit use, and psychological well-being. American Behavioral Scientist, 49, 1248–1269. https://doi.org/ 10.1177/1744629511410922. Cinamon, R. G., & Rich, Y. (2002). Gender differences in the importance of work and family roles: Implications for work–family conflict. Sex Roles, 47, 531–541. https://doi.org/10.1023/A:1022021804846. Cohn, D., Livingston, G., & Wang, W. (2014). After decades of decline, a rise in stay-at-home mothers. Retrieved from http://www. pewsocialtrends.org/2014/04/08/after-decades-of-decline-a-rise-instay-at-home-mothers/. Coltrane, S. (2000). Research on household labor: Modeling and measuring the social embeddedness of routine family work. Journal of Marriage and Family, 62, 1208–1233. https://doi.org/10.1111/j. 1741-3737.2000.01208.x. Cooke, L. P. (2006). “Doing” gender in context: Household bargaining and risk of divorce in Germany and the United States. American Journal of Sociology, 112, 442–472. https://doi.org/10.1086/506417. Corrigall, E. A., & Konrad, A. M. (2007). Gender role attitudes and careers: A longitudinal study. Sex Roles, 56, 847–855. https://doi. org/10.1007/s11199-007-9242-0. Cotter, D., Hermsen, J. M., & Vanneman, R. (2011). The end of the gender revolution? Gender role attitudes from 1977 to 2008. American Journal of Sociology, 117, 259–289. https://doi.org/10. 1086/658853. Damaske, S. (2011). For the family? How class and gender shape women's work. New York: Oxford University Press. Damaske, S., & Frech, A. (2016). Women’s work pathways across the life course. Demography, 53, 365–391. https://doi.org/10.1007/s13524016-0464-z. Damaske, S., Smyth, J. M., & Zawadzki, M. J. (2014). Has work replaced home as a haven? Re-examining Arlie Hochschild’s time bind proposition with objective stress data. Social Science & Medicine, 115, 130–138. https://doi.org/10.1016/j.socscimed.2014.04.047. Davis, S. N., & Greenstein, T. N. (2013). Why study housework? Cleaning as a window into power in couples. Journal of Family Theory & Review, 5, 63–71. https://doi.org/10.1111/jftr.12004. De Henau, J., & Himmelweit, S. (2013). Unpacking within-household gender differences in partners' subjective benefits from household income. Journal of Marriage and Family, 75, 611–624. https://doi. org/10.1111/jomf.12027. Dereuddre, R., Missinne, S., Buffel, V., & Bracke, P. (2014). Gender specific effects of financial and housework contributions on depression: A multi-actor study among three household types in Belgium. Health Sociology Review, 23, 78–90. https://doi.org/10.1080/ 14461242.2014.11081963. Dillaway, H., & Paré, E. (2008). Locating mothers how cultural debates about stay-at-home versus working mothers define women and home. Journal of Family Issues, 29, 437–464. DiPrete, T. A., & Buchmann, C. (2006). Gender-specific trends in the value of education and the emerging gender gap in college completion. Demography, 43, 1–24. https://doi.org/10.1177/ 0192513X07310309.

Sex Roles Distefano, C., Zhu, M., & Mindrila, D. (2009). Understanding and using factor scores: Considerations for the applied researcher. Practical Assessment, Research & Evaluation, 14(20), 1–11. Doucet, A. (2004). "It's almost like I have a job, but I don't get paid": Fathers at home reconfiguring work, care, and masculinity. Fathering: A Journal of Theory, Research, and Practice about Men as Fathers, 2, 277–303. https://doi.org/10.3149/fth.0203.277. Doucet, A. (2015). Parental responsibilities: Dilemmas of measurement and gender equality. Journal of Marriage and Family, 77, 224–242. https://doi.org/10.1111/jomf.12148. England, P. (2010). The gender revolution uneven and stalled. Gender a n d S o c i e t y, 2 4 , 1 4 9 – 1 6 6 . h t t p s : / / d o i . o r g / 1 0 . 11 7 7 / 0891243210361475. England, P. (2011). Reassessing the uneven gender revolution and its slowdown. Gender and Society, 25, 113–123. https://doi.org/10. 1177/0891243210391461. Fischer, J., & Anderson, V. N. (2012). Gender role attitudes and characteristics of stay-at-home and employed fathers. Psychology of Men & Masculinity, 13, 16–31. https://doi.org/10.1037/a0024359. Frech, A., & Damaske, S. (2012). The relationships between mothers’ work pathways and physical and mental health. Journal of Health and Social Behavior, 53, 396–412. https://doi.org/10.1177/ 0022146512453929. Furdyna, H. E., Tucker, M. B., & James, A. D. (2008). Relative spousal earnings and marital happiness among African American and White women. Journal of Marriage and Family, 70, 332–344. https://doi. org/10.1111/j.1741-3737.2008.00485.x. García-Manglano, J. (2015). Opting out and leaning in: The life course employment profiles of early baby boom women in the United States. Demography, 52, 1961–1993. https://doi.org/10.1007/ s13524-015-0438-6. Gerson, K. (2010). The unfinished revolution: How a new generation is reshaping family, work, and gender in America. New York: Oxford University Press. Gjerdingen, D., McGovern, P., Bekker, M., Lundberg, U., & Willemsen, T. (2001). Women's work roles and their impact on health, wellbeing, and career: Comparisons between the United States, Sweden, and The Netherlands. Women & Health, 31, 1–20. https:// doi.org/10.1300/J013v31n04_01. Goldberg, A. E., & Perry-Jenkins, M. (2004). Division of labor and working-class women's well-being across the transition to parenthood. Journal of Family Psychology, 18, 225–236. https://doi.org/ 10.1037/0893-3200.18.1.225. Greenstein, T. N. (2000). Economic dependence, gender, and the division of labor in the home: A replication and extension. Journal of Marriage and Family, 62, 322–335. https://doi.org/10.1111/j.17413737.2000.00322.x. Gupta, S. (2006). Her money, her time: Women’s earnings and their housework hours. Social Science Research, 35, 975–999. https:// doi.org/10.1016/j.ssresearch.2005.07.003. Gupta, S. (2007). Autonomy, dependence, or display? The relationship between married women’s earnings and housework. Journal of Marriage and Family, 69, 399–417. https://doi.org/10.1111/j.17413737.2007.00373.x. Hakim, C. (2002). Lifestyle preferences as determinants of women's differentiated labor market careers. Work and Occupations, 29, 428– 459. https://doi.org/10.1177/0730888402029004003. Higgins, C., Duxbury, L., & Lee, C. (1994). Impact of life-cycle stage and gender on the ability to balance work and family responsibilities. Family Relations, 43, 144–150. https://doi. org/10.2307/585316. Hochschild, A. (1989). The second shift: Working parent and the revolution at home. New York: Viking. Hofferth, S. L., & Sandberg, J. F. (2001). How American children spend their time. Journal of Marriage and Family, 63, 295–308. https:// doi.org/10.1111/j.1741-3737.2001.00295.x.

Hu, L., & Bentler, P. M. (1998). Fit indices in covariance structure modeling: Sensitivity to underparameterized model misspecification. Psychological Methods, 3(4), 424–453. https://doi.org/10.1037// 1082-989x.3.4.424. Hudson, D. B., Elek, S. M., & Fleck, M. O. (2001). First-time mothers’ and fathers’ transition to parenthood: Infant care self-efficacy, parenting satisfaction, and infant sex. Issues in Comprehensive Pediatric Nursing, 24, 31–43. https://doi.org/10.1080/ 014608601300035580. Johnston, D., & Swanson, D. (2003). Invisible mothers: A content analysis of motherhood ideologies and myths in magazines. Sex Roles, 49, 21–33. https://doi.org/10.1023/A:1023905518500. Johnston, D., & Swanson, D. (2004). Moms hating moms: The internalization of mother war rhetoric. Sex Roles, 51, 497–510. https://doi. org/10.1007/s11199-004-5460-x. Jones, B. (2012). Women who opt out: The debate over working mothers and work-family balance. New York: New York University Press. Kaukinen, C. (2004). Status compatibility, physical violence, and emotional abuse in intimate relationships. Journal of Marriage and Family, 66, 452–471. https://doi.org/10.1111/j.1741-3737.2004. 00031.x. Keene, O. N. (1995). The log transformation is special. Statistics in Medicine, 14, 811–819. https://doi.org/10.1002/sim.4780140810. Kendler, K. S., & Prescott, C. A. (1999). A population-based twin study of lifetime major depression in men and women. Archives of General Psychiatry, 56, 39–44. https://doi.org/10.1001/archpsyc. 56.1.39. Kessler, R. C., Berglund, P., Demler, O., Jin, R., Koretz, D., Merikangas, K. R., ... Wang, P.S. (2003). The epidemiology of major depressive disorder: Results from the National Comorbidity Survey Replication (NCS-R). JAMA, 289, 3095–3105. https://doi.org/10.1001/jama. 289.23.3095. Killewald, A., & Gough, M. (2010). Money isn’t everything: Wives’ earnings and housework time. Social Science Research, 39, 987– 1003. https://doi.org/10.1016/j.ssresearch.2010.08.005. Klein, M. H., Hyde, J. S., Essex, M. J., & Clark, R. (1998). Maternity leave, role quality, work involvement, and mental health one year after delivery. Psychology of Women Quarterly, 22, 239–266. https://doi.org/10.1111/j.1471-6402.1998.tb00153.x. Kleiner, S., & Pavalko, E. K. (2010). Clocking in: The organization of work time and health in the United States. Social Forces, 88, 1463– 1486. https://doi.org/10.1353/sof.0.0301. Kluwer, E. S. (2011). Psychological perspectives on gender deviance neutralization. Journal of Family Theory & Review, 3, 14–17. https://doi.org/10.1111/j.1756-2589.2010.00075.x. Kramer, K. Z., & Kramer, A. (2016). At-home father families in the United States: Gender ideology, human capital, and unemployment. Journal of Marriage and Family, 78, 1315–1331. https://doi.org/10. 1111/jomf.12327. Kramer, K. Z., Kelly, E. L., & McCulloch, J. B. (2015). Stay-at-home fathers: Definition and characteristics based on 34 years of CPS data. Journal of Family Issues, 36, 1651–1673. https://doi.org/10.1177/ 0192513X13502479. Lam, C. B., McHale, S. M., & Crouter, A. C. (2012). The division of household labor: Longitudinal changes and within-couple variation. Journal of Marriage and Family, 74, 944–952. https://doi.org/10. 1111/j.1741-3737.2012.01007.x. Lang, J., Bliese, P. D., Lang, J. W., & Adler, A. B. (2011). Work gets unfair for the depressed: Cross-lagged relations between organizational justice perceptions and depressive symptoms. Journal of Applied Psychology, 96, 602–618. https://doi.org/10.1037/ a0022463. Lee, C. Y. S., Lee, J., & August, G. J. (2011). Financial stress, parental depressive symptoms, parenting practices, and children's externalizing problem behaviors: Underlying processes. Family Relations, 60, 476–490. https://doi.org/10.1111/j.1741-3729.2011.00656.x.

Sex Roles Legerski, E. M., & Cornwall, M. (2010). Working-class job loss, gender, and the negotiation of household labor. Gender and Society, 24, 447–474. https://doi.org/10.1177/0891243210374600. Levine, B. E. (2007). Surviving America's depression epidemic: How to find morale, energy, and community in a world gone crazy. White River Junction: Chelsea Green Publishing. Livingston, G. (2014a). Growing number of dads home with kids. Pew Research Center. Retrieved from http://www.pewsocialtrends.org/ 2014/06/05/growing-number-of-dads-home-with-the-kids/#fn19605-3. Livingston, B. A. (2014b). Bargaining behind the scenes spousal negotiation, labor, and work–family burnout. Journal of Management, 40, 949–977. https://doi.org/10.1177/0149206311428355. Martikainen, P., Adda, J., Ferrie, J. E., Smith, G. D., & Marmot, M. (2003). Effects of income and wealth on GHQ depression and poor self-rated health in white collar women and men in the Whitehall II study. Journal of Epidemiology & Community Health, 57, 718–723. https://doi.org/10.1136/JECH.49.1.48. Martin, L. A., Neighbors, H. W., & Griffith, D. M. (2013). The experience of symptoms of depression in men vs women: Analysis of the National Comorbidity Survey Replication. JAMA Psychiatry, 70, 1100–1106. https://doi.org/10.1001/jamapsychiatry.2013.1985. Maynard, M. T., Luciano, M. M., D’Innocenzo, L., Mathieu, J. E., & Dean, M. D. (2014). Modeling time-lagged reciprocal psychological empowerment–performance relationships. Journal of Applied Psychology, 99, 1244–1253. https://doi.org/10.1037/a0037623. McKee-Ryan, F., Song, Z., Wanberg, C. R., & Kinicki, A. J. (2005). Psychological and physical well-being during unemployment: A meta-analytic study. Journal of Applied Psychology, 90, 53–76. https://doi.org/10.1037/0021-9010.90.1.53. Morrissey, T. W., & Warner, M. E. (2009). Employer-supported child care: Who participates? Journal of Marriage and Family, 71, 1340–1348. https://doi.org/10.1111/j.1741-3737.2009.00672.x. NLS. (2016). NLS annotated bibliography. Retrieved from https://www. nlsinfo.org/bibliography/search/cohort=NLSY79. NLSY79. (2016). Retention & reasons for noninterview. Retrieved from https://www.nlsinfo.org/content/cohorts/nlsy79/intro-to-the-sample/ retention-reasons-noninterview . Nock, S. L. (2001). The marriages of equally dependent spouses. Journal of Family Issues, 22, 755–775. Nordenmark, M. (2004). Does gender ideology explain differences between countries regarding the involvement of women and of men in paid and unpaid work? International Journal of Social Welfare, 13, 233–243. https://doi.org/10.1177/019251301022006005. Perry-Jenkins, M., Seery, B., & Crouter, A. C. (1992). Linkages between women’s provider-role attitudes, psychological well-being, and family relationships. Psychology of Women Quarterly, 16, 311–329. https://doi.org/10.1111/j.1471-6402.1992.tb00257.x. Poortman, A. R., & Van der Lippe, T. (2009). Attitudes toward housework and child care and the gendered division of labor. Journal of Marriage and Family, 71, 526–541. https://doi.org/10.1111/j.17413737.2009.00617.x. Potuchek, J. L. (1992). Employed wives’ orientations to breadwinning: A gender theory analysis. Journal of Marriage and the Family, 54, 548–558. https://doi.org/10.2307/353241. Raley, S. B., Mattingly, M. J., & Bianchi, S. M. (2006). How dual are dual-income couples? Documenting change from 1970 to 2001. Journal of Marriage and Family, 68, 11–28. https://doi.org/10. 1111/j.1741-3737.2006.00230.x. Raley, S., Bianchi, S. M., & Wang, W. (2012). When do fathers care? Mothers’ economic contribution and fathers’ involvement in child care. American Journal of Sociology, 117, 1422–1459. https://doi. org/10.1086/663354. Ridgeway, C. L. (1997). Interaction and the conservation of gender inequality: Considering employment. American Sociological Review, 62, 218–235. https://doi.org/10.2307/2657301.

Rochlen, A. B., McKelley, R. A., & Whittaker, T. A. (2010). Stay-athome fathers' reasons for entering the role and stigma experiences: A preliminary report. Psychology of Men & Masculinity, 11, 279– 285. https://doi.org/10.1037/a0017774. Rogers, S. J., & DeBoer, D. D. (2001). Changes in wives' income: Effects on marital happiness, psychological well-being, and the risk of divorce. Journal of Marriage and Family, 63, 458–472. https://doi. org/10.1111/j.1741-3737.2001.00458.x. Rosenzweig, M. H., & Amsterdam, J. D. (1992). Serious depression. JAMA, 267, 2960–2960. Ryff, C. D. (1989). Happiness is everything, or is it? Explorations on the meaning of psychological well-being. Journal of Personality and Social Psychology, 57, 1069–1081. https://doi.org/10.1037//00223514.57.6.1069. Ryff, C. D., & Keyes, C. L. M. (1995). The structure of psychological well-being revisited. Journal of Personality and Social Psychology, 69, 719–727. https://doi.org/10.1037/0022-3514.69.4.719. Sanchez, L., & Thomson, E. (1997). Becoming mothers and fathers parenthood, gender, and the division of labor. Gender and Society, 11, 747–772. https://doi.org/10.1177/089124397011006003. Sanderson, S., & Thompson, V. L. S. (2002). Factors associated with perceived paternal involvement in childrearing. Sex Roles, 46, 99– 111. https://doi.org/10.1023/A:1016569526920. Sayer, L. C., & Bianchi, S. M. (2000). Women's economic independence and the probability of divorce: A review and reexamination. Journal of Family Issues, 21, 906–943. https://doi.org/10.1177/ 019251300021007005. Schneider, D. (2012). Gender deviance and household work: The role of occupation. American Journal of Sociology, 117, 1029–1072. https://doi.org/10.1086/662649. Schnittker, J. (2007). Working more and feeling better: Women’s health, employment, and family life, 1974-2004. American Sociological R e v i e w, 7 2 , 2 2 1 – 2 3 8 . h t t p s : / / d o i . o r g / 1 0 . 1 1 7 7 / 000312240707200205. Shows, C., & Gerstel, N. (2009). Fathering, class, and gender: A comparison on physicians and emergency medical technicians. Gender a n d S o c i e t y, 2 3 , 1 6 1 – 1 8 7 . h t t p s : / / d o i . o r g / 1 0 . 11 7 7 / 0891243209333872. Simister, J. (2013). Is men's share of housework reduced by 'gender deviance neutralization'? Evidence from seven countries. Journal of Comparative Family Studies, 44, 311–325. https://doi.org/10.2307/ 23644604. Simon, R. W. (2014). Twenty years of the sociology of mental health: The continued significance of gender and marital status for emotional well-being. In R. J. Johnson, J. R. Turner, & B. G. Link (Eds.), Sociology of Mental Health (pp. 21–51). New York: Springer. Sullivan, O. (2011a). An end to gender display through the performance of housework? A review and reassessment of the quantitative literature using insights from the qualitative literature. Journal of Family Theory & Review, 3(1), 1–13. https://doi.org/10.1111/j.1756-2589. 2010.00074.x. Sullivan, O. (2011b). Gender deviance neutralization through housework: Where does it fit in the bigger picture? Response to England, Kluwer, and Risman. Journal of Family Theory & Review, 3(1), 27–31. https://doi.org/10.1111/j.1756-2589.2010.00078.x. Sussman, D., & Bonnell, S. (2006). Wives as primary breadwinners. Perspectives on Labour and Income, 18, 20–27 Retrieved from http://www.statcan.gc.ca/pub/75-001-x/index-eng.htm. Teachman, J. (2010). Wives’ economic resources and risk of divorce. Journal of Family Issues, 31, 1305–1323. https://doi.org/10.1177/ 0192513X10370108. Thébaud, S. (2010). Masculinity, bargaining, and breadwinning understanding men’s housework in the cultural context of paid work. Gender and Society, 24, 330–354. https://doi.org/10.1177/ 0891243210369105.

Sex Roles Thompson, L., & Walker, A. J. (1989). Gender in families: Women and men in marriage, work, and parenthood. Journal of Marriage and the Family, 51, 845–871. https://doi.org/10.2307/353201. United States Census Bureau. (2016). Poverty thresholds. Retrieved from https://www.census.gov/data/tables/time-series/demo/incomepoverty/historical-poverty-thresholds.html. Van de Velde, S., Bracke, P., & Levecque, K. (2010). Gender differences in depression in 23 European countries: Cross-national variation in the gender gap in depression. Social Science & Medicine, 71, 305– 313. https://doi.org/10.1016/j.socscimed.2010.03.035. Wang, W., Parker, K., & Taylor, P. (2013). Breadwinner moms. Retrieved from http://www.pewsocialtrends.org/files/2013/05/Breadwinner_ moms_final.pdf. Warr, P., & Parry, G. (1982). Paid employment and women's psychological well-being. Psychological Bulletin, 91, 498–516. https://doi. org/10.1037/0033-2909.91.3.498.

Wilkie, J. R., Ferree, M. M., & Ratcliff, K. S. (1998). Gender and fairness: Marital satisfaction in two-earner couples. Journal of Marriage and the Family, 60, 577–594. https://doi.org/10.2307/353530. Winslow-Bowe, S. (2006). The persistence of wives’ income advantage. Journal of Marriage and Family, 68, 824–842. https://doi.org/10. 1111/j.1741-3737.2006.00298.x. World Health Organization. (2017). Depression. Retrieved from http:// www.who.int/mediacentre/factsheets/fs369/en/. Yavorsky, J. E., Kamp Dush, C. M., & Schoppe-Sullivan, S. J. (2015). The production of inequality: The gender division of labor across the transition to parenthood. Journal of Marriage and Family, 77, 662– 679. https://doi.org/10.1111/jomf.12189. Zimmerman, F. J., & Katon, A. (2005). Socioeconomic status, depression disparities, and financial strain: what lies behind the incomedepression relationship? Health Economics, 14, 1197–1215. https://doi.org/10.1002/hec.1011.