Maureen Perry-Jenkins & Elizabeth Turner. Handbook of Contemporary Families. Editor: Marilyn Coleman & Lawrence H Ganong. Sage Publication. 2004.

“Each marriage bears the footprints E of economic and cultural trends which originate far outside the marriage” (Hochschild, 1989, p. 11). Dual-earner marriages, in particular, have emerged from, and borne the costs and benefits of, the dramatic social and economic changes that have occurred over the past half-century. In particular, one of the most striking and consistent demographic changes over the past 30 years has been the steady and linear upward trend in women’s labor force participation (Cohen & Bianchi, 1999). From a historical perspective, the structures and functions of families in the United States are far different today than they were at the beginning of the 20th century. In 1920, 75% of working households were composed of single-earner, married couples, and dual-earner, married households made up only approximately 9% of working families. In less than a century, those statistics have shifted dramatically. The latest data from the Bureau of Labor Statistics (2001) indicate that 63.2% of all married couples with children under 18 are dual earners, and 55.9% of all married couples with children under 6 are dual earners, whereas families in which only the husband works constitute only 19.4% of all married-couple families.

As dual-earner families became the normative household type over the 20th century, the question of how families cope with and negotiate new work and family role responsibilities became the topic of much research and scholarly debate. Although some family scholars suggest that we are currently in a stalled revolution where we lack new models and social arrangements to support the dual-earner lifestyle (Hochschild, 1989), others point to families who are successfully creating and managing their dual-earner arrangements (Deutsch, 1999). Scholars from a variety of disciplines, incorporating multiple theoretical perspectives and using a wide range of methodological approaches, have focused much attention on dual-earner families. Perhaps the most important revelation to emerge from this effort is that there appears to be no one, uniform experience for couples attempting to “have it all,” namely, jobs, a relationship, and children. The first goal of this chapter is to review the theoretical approaches that have been used to study dual-earner families and to provide a summary of the key findings linking work and family life in dual-earner households. Building on this knowledge base, the second goal of this chapter is to point to new directions where research on dual-earner families could go in our efforts to better understand how social contexts differentially shape the work and family processes that occur within dual-earner families.

Theoretical Approaches to the Study of Dual-Earner Families

The study of dual-earner families arose in direct response to women’s dramatic increase in labor force participation. In a review of the work-family literature in the 1990s, Perry-Jenkins, Repetti, and Crouter (2000) highlighted four broad conceptual approaches that have guided this field of study. The first approach, one that informed much of the early research on dual-earner families, comes primarily from developmental psychology and is most accurately referred to as the maternal employment literature. In over six decades of research, scholars have examined the effects of maternal employment on children’s development. Although this approach is rich in its attention to developmental outcomes for adults and children, its conceptualization of work is limited. In this early work, maternal and child outcomes were often compared in single-earner versus dual-earner households, with the overwhelming conclusion being that women’s employment alone is unrelated to either positive or negative developmental outcomes for children (Harvey, 1999). From an ecological perspective, this practice of comparing different groups of individuals, such as dual- and single-earner couples, on some psychological or relational outcome is referred to as the social address model (Bronfenbrenner & Crouter, 1983). As Bronfenbrenner and Crouter noted, social address approaches, which focus only on group comparisons, provide no insight into the processes that link work and family. Thus, the conclusion that mothers’ work hours are unrelated to children’s developmental outcomes does not mean that employment is unimportant or unrelated to family life; rather, it means that conditions of work, such as schedules, satisfaction, autonomy, support, and complexity, are more significant indicators of individual and family outcomes than the simple fact of work.

It should be noted that, historically, the maternal employment literature primarily examined the effects of mothers’ work hours on children’s social, emotional, and cognitive development, although the majority of working mothers being studied at the time lived in dual-earner households with employed spouses. This early work rested on the assumption that it was women’s employment and men’s unemployment that were the causes of most concern for children’s development. As the field developed, fathers’ employment began to receive greater attention in this literature (Marsiglio, Amato, & Day, 2000), and more attention was paid to issues of the temporal patterns of parents’ employment, such as shift work and seasonal work, that could, in turn, affect the temporal rhythms of family life (Perry-Jenkins et al., 2000).

The second theoretical approach that has been used to explore the effects of employment on family life stems from sociology and focuses much greater attention on characteristics of employment; it is referred to as the work socialization perspective. Specifically, scholars in this tradition examine how conditions of employment, such as autonomy and job complexity, socialize workers’ values, which, in turn, spill over into family life (Kohn, 1995; Kohn & Schooler, 1982). Findings have revealed, for example, that mothers who take on jobs low in complexity show decrements over time in the quality of home environments they provide their children (Menaghan & Parcel, 1995), whereas increases in mothers’ job complexity have been linked to enhanced reading scores for children (Parcel & Menaghan, 1994b).

A third approach to the study of work-family connections is the occupational stress literature, with roots in both clinical and health psychology, which explores how short-and long-term stresses at work are related to employee well-being and family relationships. One of the most important findings to emerge from this line of research points to the role of individuals’ assessments and internal responses to workplace stressors as an important mediator linking workplace conditions to mental health (Perry-Jenkins et al., 2000). Especially relevant for dual-earner families are studies that explore crossover effects, whereby the work experiences of one spouse affect the well-being of his or her partner.

Finally, the multiple-roles literature, which stems from both social psychology and sociology, focuses on how individuals manage the roles of parent, spouse, and worker and the consequences of this balance for health and family relationships. Although debate has arisen as to whether multiple roles compromise or enhance mental health and family relationships, it appears that an important third factor to consider is workers’ perceptions of the quality of each role. Moreover, the multiple-roles literature has addressed the interactive nature of roles whereby a positive marital relationship may serve as a buffer against stressful and negative work experiences (Barnett, Marshall, & Pleck, 1992; Repetti, 1992).

All of these traditions have examined the effects of work on family life, and in a few cases family effects on work, and each approach has its own strengths and weaknesses. When findings from these different research traditions are examined in their totality, it is clear that much has been learned about how work shapes the lives of workers and their families. What has been missing, however, from much of the theoretical literature on the work-family interface is attention to the role of sociocultural contexts as they shape work-family processes. From an ecological perspective, Bronfenbrenner (1986) proposed that we must carefully attend not only to aspects of social context, such as race, class, gender, and family structure, but to the idea that family processes may differ across contexts and ultimately hold different consequences for human development. In the dual-earner literature, there is still a paucity of studies that explore how aspects of race and ethnicity, social class, and gender (e.g., same-sex households) may affect the type of work-family processes that occur within families. Thus, in the following review of research on dual-earner families, it will be important to keep a critical eye on what we have learned thus far and the extent to which our research has been sensitive to the ways in which race/ethnicity, social class, and gender may moderate relationships between work experiences, workers’ well-being, and their family relationships.

Review of Key Research Findings

The question of how the dual-earner lifestyle shapes the lives of individual family members and their relationships has been the focus of much empirical research. Scholars have explored the relationships among spouses’ employment and individuals’ mental health, their marital relationships, parent-child relationships, and children’s development. The following review will highlight the key findings in each of these areas.

Women’s and Men’s Mental Health in Dual-Earner Families

An implicit assumption guiding the early research on women’s employment was that managing two jobs while maintaining a marriage and family life takes both a psychological and a physical toll on the adults involved in this endeavor. Contrary to expectation, however, there is little empirical evidence of any direct negative effects of paid employment on women’s or men’s mental health (Steil, 1997). Married men, whether their wife is employed or unemployed, report significantly lower levels of psychopathology than unmarried men (Steil, 1997). For wives, in general, employment appears to enhance their psychological well-being, even across race (Guarnaccia, Angel, & Worobey, 1991) and social class (Ferree, 1976; Scarr, Phillips, & McCartney, 1989). Two competing hypotheses have been advanced linking multiple roles (e.g., worker, spouse, parent) to psychological well-being (Hyde, DeLamater, & Hewitt, 1998). The scarcity hypothesis holds that there is only limited time to fulfill the responsibilities of multiple roles and that as the demands for time become too high in dual-earner households, inter-role conflict will occur. This conflict will, in turn, lead to negative mental health outcomes for both spouses (Baruch, Biener, & Barnett, 1987; Voydanoff, 1987), especially for women working full time (Rogers, 1996). In contrast, the enhancement hypothesis posits that holding multiple roles enhances psychological functioning (Baruch et al., 1987).

Theorists holding this view posit that success in one role can buffer disappointment or failure in another, that the diversity of experiences across roles enhances one’s perspective on life, and, finally, that the economic contribution of two workers decreases financial strain. To date, the empirical findings, at least in terms of mental health outcomes, support the enhancement hypothesis. Specifically, a number of studies have found that women who are employed or who increase their labor force participation report lower distress than their unemployed counterparts (Glass & Fujimoto, 1994; Wethington & Kessler, 1989).

Research indicates, however, that the positive effects of employment accrue differentially on the basis of a number of occupational conditions. Specifically, time demands, flexibility, pressure, lack of control, and unsupportive supervisors and/or co-workers can all shape employees’ experiences of their jobs and their well-being. Not surprisingly, these conditions vary by social class, suggesting that those employed in hourly, unskilled, or service occupations are most at risk for bearing the negative effects of employment. Research indicates, however, that how job characteristics are interpreted and experienced by the worker is a key mediator variable linking work conditions to mental health.

For example, the chronic stress transfer literature indicates that it is not objective job characteristics that are related to worker well-being but rather feelings of job stress that have been related to self-reports of depression, which, in turn, have been linked to poorer marital relations (Barling & MacEwen, 1992; Sears & Galambos, 1992). Moreover, a number of studies have found that employees who experience more work-family overload also report greater emotional distress (Gerstel & Gallagher, 1993; Guelzow, Bird, & Koball, 1991; Paden & Buehler, 1995). Some research with dual-earner couples has attempted to examine crossover effects where one partner’s job experiences are related to the other partner’s mental health. For example, some studies have linked overload and pressure at work reported by one spouse to feelings of overload and depression reported by his or her partner (Crouter, Bumpus, Maguire, & McHale, 1999; Wortman, Biernat, & Lang, 1991), whereas others have not detected any crossover effects (Sears & Galambos, 1992). Future research is needed to understand for whom and under what conditions one spouse’s job has the greatest impact on his or her partner’s mental health.

An implicit assumption in much of the dual-earner literature is that the cumulative, long-term effects of the dual-earner lifestyle will eventually compromise the quality of spouses’ mental health and family relationships. Few studies, however, have explicitly examined these relationships longitudinally. In an exception, Barnett, Raudenbush, Brennan, Pleck, and Marshall (1995) examined how changes in job quality and marital experiences were related to change in psychological distress in a sample of white, middle-class dual-earner couples. Findings revealed that for both husbands and wives, when job quality decreased, distress increased. The relation between marital change and distress differed by gender, revealing that if the marital relationship deteriorated over time, women were likely to experience greater psychological distress than men. These findings are important for two reasons. First, the researchers move past static, cross-sectional models to explore the inherently dynamic nature of work and family life. Second, the use of hierarchical linear modeling, with its capacity to explore the dependency between spouses’ reports, reflects the fact that when individuals live together one spouse’s sense of distress or marital harmony influences the other spouse’s. Future research must continue to explore this important and inherent dependency in couple’s lives.

Dual-Earner Lifestyle and the Marital Relationship

The implications of the dual-earner lifestyle for couples’ marital relations have received the most attention in the work-family literature. Rogers (1996) tested the scarcity and enhancement hypotheses described above in explaining marital happiness and conflict. Findings supporting the scarcity hypothesis revealed that full-time employed mothers with high family demands (i.e., more children) reported less marital happiness and more marital conflict. For mother-stepfather families, however, as family demands increased, mothers’ full-time employment was associated with greater marital happiness and lower marital conflict. In a more recent study, the reciprocal effects of income and marital quality were examined (Rogers, 1999). It was found that increases in marital discord were associated with increases in wives’ income, specifically by increasing the likelihood that nonemployed wives would enter the labor force.

In a rare look at the sex lives of dual-earner couples, Hyde et al. (1998) examined sexual satisfaction, frequency of intercourse, and decreased sexual desire for a sample of dual- and single-earner couples across the transition to parenthood. The authors reported no support for the scarcity hypothesis because sexual satisfaction and activity did not suffer in dual-earner families. Their findings did reveal that role quality, especially job role quality, was related to sexual satisfaction. Though some support has emerged for a direct link between work and marital quality, a more fruitful line of analysis has examined processes whereby work conditions are related to perceptions of job quality or mental health that, in turn, affect the marriage.

In fact, the literature that has explored linkages between work and marriage has uncovered few strong and consistent direct links between work status and marital outcomes (Hyde et al., 1998); rather, it appears that the variable of spouses’ perceptions of the quality of their roles is the important mediator linking work roles to marital quality. Thus, the division of unpaid work in the home, or the home-caretaker role, becomes a major point of contention among many dual-earner couples. Hochschild (1989) highlighted how women’s movement into the workforce has challenged couples to examine assumptions regarding the division of unpaid work in the home, specifically household chores and child care. Although the majority of dual-earner couples believe in the ideology of fairness in sharing paid and unpaid work, in the reality of day-to-day life, equality in family work is not the norm (Coltrane, 2000.

Although questions concerning how women’s employment in dual-earner families is related to the division of household and child care tasks and, in turn, shapes perceptions of marriage have been addressed by many researchers (Coltrane, 2000; Spitze, 1988), the answers remain ambiguous. Women continue to perform the lion’s share of family work, though men’s participation has increased slightly over the past few decades (Coltrane, 2000). The division of unpaid labor in dual-earner families has been linked to partners’ mental health and marital relationships. Specifically, the more housework is shared, the less depression wives report (Ross, Mirowsky, & Huber, 1983) and the higher marital quality and well-being wives report (Barnett & Baruch, 1987). These same relationships examined for husbands reveal less clear results. Researchers have found that shared housework is unrelated to men’s depression (Ross et al., 1983) but is positively related to enhanced well-being (Pleck, 1985).

Although the aforementioned relationships between the division of unpaid labor and spouses’ mental health and marital relationships have been found, the findings, though significant, are often not strong or consistent. One of the most surprising findings in the division-of-labor literature is that although women, on average, consistently perform more household chores than men, only a minority of women view this unequal division of labor as unfair (Thompson & Walker, 1989). This finding even holds up cross-culturally, where Zuo and Bian (2001) found that for a sample of dual-earner Chinese couples, the actual division of labor was largely unrelated to perceptions of fairness in the relationship. These studies have led family scholars to propose that beliefs about the fairness of household labor may be as, or even more, critical in relation to marital quality than the relative or absolute amounts of household chores completed (Blair & Johnson, 1992; Thompson, 1991; Wilkie, Ferree, & Ratcliff, 1998). In testing this hypothesis, Wilkie et al. (1998) found that the division of labor affects marital satisfaction mainly through perceptions of fairness. Similarly, Blair (1993) reported that the strongest predictor of both husbands’ and wives’ perceptions of marital conflict was wives’ assessment of unfairness in the marriage. Perry-Jenkins and Folk (1994) found that for middle-class dual-earner couples, the variable of perceptions of fairness regarding the division of labor was a key mediator linking reported division of labor to marital conflict, with higher perceptions of fairness being linked to lower reports of marital conflict. In contrast, this relationship did not emerge for working-class dual-earner couples. Thus, the research to date points to the importance of not only the division of labor but also perceptions of its fairness as important indicators of marital quality in dual-earner families. However, Perry-Jenkins and Folk’s (1994) findings raise the question of how social contexts, such as race, ethnicity, and social class, moderate these relationships.

With regard to gender role ideology, research has found that African Americans and Mexican Americans have more positive attitudes toward working wives than European Americans but are also more likely to endorse the idea that the man should be the main breadwinner (Blee & Tickamyer, 1995; Kamo & Cohen, 1998). When it comes to gendered behaviors, a number of studies have found that, compared to European American men, African American men perform more child care and household tasks; however, African American women still perform the majority of family work (McLoyd, 1993). In addition, African American husbands or partners are more likely than their white counterparts to view the division of labor as unfair to their wives or partners (John, Shelton, & Luschen, 1995). In a related study, Shelton and John (1993) found that Hispanic men perform more female-typed tasks than European American men.

The question of how these different values and belief systems within different racial groups may lead to unique patterns of relationships among work hours, work conditions, and marital relations is an important new direction for research. Because research indicates that the meaning and interpretation that spouses give to their paid and unpaid work ultimately hold consequences for mental health and marital relations and because this meaning appears to vary by race, class, and gender, it is unlikely that our current findings based primarily on white, middle-class families will extrapolate to families of different races and social classes.

Dual-Earner Lifestyle, Parent-Child Relationships, and Child Development

The approach one uses to understand work-family connections affects what facets of this relationship are focused on (Crouter & McHale, 1993). Researchers focusing primarily on the workplace have explored (a) work as a socializer of values that are transmitted to children; (b) work as a setting that provides opportunities (e.g., learning new skills) or constraints (e.g. time inflexibility) that can positively or negatively affect one’s ability to parent; and (c) the workplace as a place that shapes the emotional states of workers, moods that may then be brought home to affect the quantity and quality of parent-child interactions.

The work socialization perspective, grounded in the sociology of work and occupations, rests on the basic assumption that occupational conditions, such as autonomy and complexity, shape the beliefs and values of workers, which in turn influence parenting styles and child outcomes. For example, Parcel and Menaghan (1993) found that higher levels of occupational complexity for fathers served as a protective factor against later child behavior problems. In a related study, Cooksey, Menaghan, and Jekielek (1997) found that when they controlled for family structure, maternal employment characteristics such as more autonomy, working with people, and problem solving predicted decreases in child behavior problems. The actual processes whereby work conditions shape parental values and ensuing parental behavior style, though they can be inferred from Parcel and Menaghan’s (1994b) work, need more direct empirical investigation. Thus, although a few studies have linked work conditions to parental behaviors and, in turn, to child outcomes, a process that supports the work socialization perspective (Greenberger, O’Neil, & Nagel, 1994; Grimm-Thomas & Perry-Jenkins, 1994; Whitbeck et al., 1997), we need more longitudinal and quasi-experimental studies that can more effectively tease apart causality issues. In addition, greater attention must be paid to selection effects that lead certain types of parents to choose particular occupations and the extent to which child characteristics affect the types of jobs parents choose.

Research on work as a setting that provides opportunities and/or constraints has shown that skills acquired at work, such as problem solving and group decision making, can be translated into more democratic or authoritative parenting styles at home (Crouter, 1984). Finally, at an even more proximal level, researchers have begun to elucidate processes whereby day-to-day variation in work-induced moods affects subsequent parent-child interactions at home. Repetti and Wood (1997), for example, found that employed parents tend to withdraw from family interactions following a high-stress workday. Specifically, mothers were more withdrawn from their preschoolers on days when they experienced greater workloads or interpersonal stress at work. All these findings suggest that, ranging from macro levels (i.e., worldviews and values) to more micro, proximal levels (i.e., work-induced moods), experiences of work can affect everything from parental values, to parenting styles, to patterns of parent-child interactions.

Crouter and McHale (1993) pointed out that research on the work-family relationship that focuses primarily on the family as its point of entry has attended to quite different processes linking parental work to child outcomes. This approach has paid less attention to conditions of work and focused more on the fact that employed parents may face unique challenges such as finding time to share activities with children, to monitor children’s activities, and to maintain close, satisfying parent-child relations. Thus, the issue of how much and when parents work has been the focus of much research over the past decade (Hochschild, 1997; Perry-Jenkins et al., 2000; Schor, 1991) because the amount of time parents work and when they work affect parents’ ability to be involved in child-oriented activities and/or their ability to monitor children’s daily activities (Crouter, MacDermid, McHale, & Perry-Jenkins, 1990; Macoby & Martin, 1983). At least one pair of researchers suggest that paternal monitoring of children differs across cultural groups: Toth and Xu (1999) found that African American and Latino fathers reported supervising their children’s activities more than European American fathers and that Latino fathers spent more time in shared activities with their children than European American fathers. These findings highlight the importance of culture as a critical variable that may moderate work-parenting relationships.

The question of whether the sheer amount of time parents work is related to child outcomes has been raised as an important issue, especially given the increased time that employed men and women are spending on the job (Crouter, Bumpus, Head, & McHale, 2001; Schor, 1991). Parcel and Menaghan (1994a) found that working more than 40 hours per week is related to more negative child outcomes; however, occupational complexity interacts with maternal work hours by decreasing the negative effects of long work hours. In one of the rare studies to look beyond outcomes for young children, Crouter and colleagues found that the combination of fathers’ long work hours and high overload was consistently associated with less positive father-adolescent relations.

Beyond the number of work hours, Presser’s (1994, 1999) research on shift work highlights how temporal patterning of work hours in dual-earner families has important implications for family life. Her research suggests that the less husbands’ and wives’ work hours overlap, the more family work and child care men perform. Shift work may affect parents’ ability to interact with their children and supervise their activities (Presser, 2000), although Presser found this pattern of relationships to be complex and a function of the shift the parent worked, the gender of the parent, the type of parent-child activity assessed, and the developmental stage of the child. For instance, parents of a preschooler working evening or night shifts could have more time with their child than those working day shifts. In contrast, working an evening shift would provide parents of school-aged children with few or no overlapping hours at home with their children during their workweek. If we identify parent-child interaction as an important mediator variable, our next step is to explore how parent-child interactions may mediate the link between nonstandard work schedules and children’s developmental outcomes in multiple areas, such as social, emotional, and cognitive dimensions of development.

Thus far, our discussion has focused on the direct and indirect influences of parental work on children’s development. But just as important to children’s development are the issues of where and how children are spending their time when parents are at work. In a recent Urban Institute report, 41% of children under 5 spend 35 or more hours in nonparental care, and an additional 25% spend between 15 and 34 hours in non-parental care (Capizzano & Adams, 2000). Thus, one of the most important factors affecting children’s development is the quality of nonparental care that children receive. Findings from the Early Child Care Research Network of the National Institute of Child Health and Human Development [NICHD] revealed few simple effects of nonmaternal care in the first year of life. However, results did indicate more complex relationships: Poor-quality care, unstable care, and more than minimal hours of care coupled with insensitive mothering were related to poorer child outcomes and less secure mother-infant attachments (NICHD, 1997a, 1997b).

Care and supervision issues remain important issues for school-aged children and adolescents. In a recent study using nationally representative data from the Panel Study of Income Dynamics, it was found that 73% of children ages 5 to 12 go straight home from school (with only 14% of those children being alone), 11% go to child care, 8% stay at school, and 8% go somewhere else (Hofferth & Sandberg, 2001). The question of how after-school time is spent and its relation to children’s development has not received a great deal of attention. For lower-income, but not middle-income, children, unsupervised after-school time was related to more externalizing behaviors (e.g., yelling, hitting), whereas attending an after-school program was related to fewer internalizing behaviors (e.g., introversion) (Marshall et al., 1997). Adolescents performed better academically when engaged in supervised after-school activities than when unsupervised (Muller, 1995). Clearly, these studies point to the importance of including child care factors into the study of work and family issues for dual-earner families.

Contemplating the Future of Dual-Earner Families

How dual-earner families will change and cope with work-family challenges over the next few decades is an intriguing question that scholars and researchers must address with strong theoretical models and new methodologies. In this last section, we propose new directions for research that will more fully capture the complexity and realities of life in dual-earner households.

The Sociocultural Context of Dual-Earner Families

An ecological perspective emphasizes the importance of extrafamilial contexts as they influence family and individual functioning (Bronfenbrenner, 1986). To date, the predominant focus in both the broad work-family literature and dual-earner research in particular has been on white, middle-class, two-parent families, a trend that has sharply restricted an understanding of how work-family relationships differ across social class, racial, and ethnic lines. A more ecologically valid approach to studying dual-earner families would place greater emphasis on the diversity of contexts that can shape parents’ experiences.

Family scholars have emphasized that experiences of dual-career families are not generalizable to dual-earner families due to differences in the nature of the problems confronted by these two types of families, their resources, and the solutions they devise to deal with their problems (Mortimer & London, 1984; Rubin, 1976, 1994). Nevertheless, only a handful of studies have examined the implications of women’s employment in working-class families. In her classic study, Komarovsky (1962) demonstrated how values about work and family lifeproviding and mothering were quite clearly delineated in the working-class families of the 1960s. More recently, a number of studies have refuted the idea that working-class families divide household tasks strictly along gender lines and that employed women in the working class are employed only for financial reasons (Ferree, 1987; Rubin, 1994; Ybarra, 1982). Ybarra (1982) found that in working-class, Latino families, wives’ employment was correlated with more egalitarian family roles. In addition, it was found that the majority of these working-class women preferred to be employed. Thompson and Walker (1989) suggested that the realities of class lead working-class families to more equitably divide up paid and unpaid work. Perry-Jenkins and Folk (1994) found different processes linking division of labor, perceptions of its fairness, and marital quality for working-class versus middle-class couples. Thus, we must move beyond our tendency to control for variability in social class indicators, such as education, income, and occupational prestige, and explore how these aspects of families’ lives moderate the ways in which work and family are connected.

Research on gender and race effects on occupational characteristics suggests that both factors differentially influence occupational attainment and earnings. With respect to earnings, gender effects are more discernible than race effects: Women earn less than men, although black men earn less than white men. In contrast, occupational prestige is influenced by both race and gender, with 13% of white women in high-prestige jobs and only 10% of black women and 9% of black men in high-prestige jobs (Xu & Leffler, 1996). These differences in income and job prestige, which arise as a function of race and gender, are likely to influence how well dual-earner families are able to cope with work-family stressors. Moreover, the literature on working-class occupations suggests that workers in lower-level jobs have less autonomy and control on the job, as well as fewer supportive workplace policies again, conditions that hold consequences for workers’ ability to manage work and family life. In short, gender, race, and class matter, and “they matter because they structure interactions, opportunities, consciousness, ideology and the forms of resistance that characterize American life” (Anderson, 1996, p. ix). It is expected that as we look more carefully within the contexts of race and class, as well as family structure, we will be able to outline common and unique factors both at work and at home that affect how parents and children in these different contexts fare in terms of their mental health and family relationships.

Research on work and family issues in African American families is rare. In a study that explored work and family roles in black families, Broman (1991) found that involvement in multiple roles had differential effects on black women and men. Specifically, employed wives reported the lowest levels of family life satisfaction, whereas married, employed men reported the highest family satisfaction. In addition, couples’ organization of work and family role responsibilities was related to assessments of family life satisfaction but not to psychological well-being outcomes. Broman suggested that the usefulness of social role theory may be domain specific for blacks: useful in understanding family life satisfaction but not mental health. He and McLoyd (1993) argued further that the different historical experiences of blacks and whites in the United States have implications not only for racial differences in work and family role configurations but for different associations between role patterns and family and individual outcomes for blacks and whites.

The few studies of Latino families reveal work-family relationships similar to those often described in European American dual-earner families. Specifically, Chicano men endorse the idea that wives’ income is needed to support families’ standard of living, and Chicano women report less depression when husbands are involved in family work (Saenz, Goudy, & Lorenz, 1989; Williams, 1988).

Some of the most important advances over the past decade are the new conceptual frameworks that urge us to place families in context and to take into account that ethnicity and race do not exist in isolation from class and gender hierarchies. A major challenge for the new century is translation of these sophisticated and nuanced models into sound empirical research (McLoyd, Cauce, Takeuchi, & Wilson, 2000, p. 1087).

Future work should take on the challenge presented by McLoyd and her colleagues by examining how race, gender, and family structure intersect within dual-earner families to influence the well-being and family relationships of women and men as they negotiate work and family responsibilities.

Work-Family Processes in Context

Although an ecological perspective suggests that broader social systems directly and indirectly influence adult and child development, there is a basic need to determine how distal factors (i.e., class, race, family structure) influence development through proximal processes (Bronfenbrenner, 1986; O’Connor & Rutter, 1996). As O’Connor and Rutter (1996) pointed out, “[T]he link between culture and proximal processes must go beyond demonstrating group mean differences to include information on how key features are perceived differently in different groups … It is equally necessary to consider variability within cultural groups” (p. 787). For example, the occupational stress literature documents ways in which pressure and stress at work can affect a worker’s mood, which in turn can lead to social withdrawal and/or conflict at home. Future research that is attentive to process issues needs to integrate the short- and long-term stress transmission processes with attention to ways in which job pressures may be reduced in situations of job control and autonomy. Finally, some of the most important findings over the past decade point to the importance of interaction effects in work-family models. For example, Parcel and Menaghan’s (1994a, 1994b) work highlights ways in which occupational complexity combined with specific family circumstances (e.g., structure, class) affects home environments and parenting styles.

Methodological Issues

Although the dual-earner literature has clearly moved past the social address approach to explore within-family processes that link work and family life, a few scholars have begun to explicitly and systematically explore the heterogeneity of dual-earner families (Crouter & Manke, 1997; Kinnunen & Mauno, 2001). Crouter and Manke (1997) proposed that categorizing dual-earner couples on the basis of mothers’ work hours (e.g., part time, full time, homemaker) or men’s occupational prestige (e.g., dual earner vs. dual career) oversimplifies the dual-earner experience and ignores covariations that occur around multiple work characteristics. To address this shortcoming, they used cluster analysis to typologize dual-earner families on four dimensions: hours, prestige, role overload, and job involvement. Three groups emerged: (a) high-status dual earners, (b) low-stress dual earners, and (c) main-secondary provider families. These types, based on characteristics of both parents’ jobs, were then related to other aspects of family life such as parental monitoring and marital quality. Parents in the low-stress group were better monitors than parents in the other two groups, and these parents also reported more marital satisfaction than the high-status group. In a partial replication of this work, Kinnunen and Mauno also explored dual-earner typologies and obtained some similar but unique findings related to job exhaustion, which was most significantly linked to negative family outcomes. We highlight these studies because they look beyond the social address of dual-earner couples to examine within-group variability and pay attention to the dyadic level of analysis by combining data from both parents’ work situations. Spouses and children in dual-earner families must deal with the effects of two jobs and their intersection, yet our analyses often ignore this interdependence.

Once we begin to examine how dual-earner families differ as a function of race, ethnicity, social class, and life stage, it becomes apparent that exactly what we mean by the very terms work and family should be examined. As Thorne (2001) pointed out, “[F]ruitful topics will illuminate social processes that don’t necessarily stop at the pre-specified boundaries of work and family” (p. 374). Work often extends into family life as computers, beepers, and cell phones come into our homes, and our jobs can include multiple family members. Family boundaries usually extend far beyond the oft-used definition of family found in our research that is limited to “mother, father, and target child.” We would greatly benefit from research that starts to give the terms work and family social and cultural relevance for the individuals that live in dual-earner families.

Although conceptual models often accentuate the dynamic nature of both work and family experiences, empirical research is much less sensitive to within- and between-individual change over time. Distress is relational for dual-earner couples (Barnett, Marshall, Raudenbush, & Brennan, 1993), in that change over time in one partner’s psychological distress is related to change in the other partner’s distress. Models must take this level of dependency into account.

Hierarchical linear modeling (HLM) provides a unique conceptual orientation and analytical technique for the study of psychological and relationship change over time. HLM is uniquely suited to explore change over time in individuals’ and couples’ work and family experiences and makes it possible to identify significant predictors of differences in individuals’ variation in job stress or marital love over time. Predictors of change in mental health or family relationships may be due to more structural and time-invariant variables, such as family size, parental age, and child temperament, or predictors that may change concurrently with the dependent variables, such as job autonomy, role overload, and workplace support (time-variant variables). For example, Barnett et al. (1995) explored how changes in job experiences are related to change in psychological distress for both partners in dual-earner marriages. To address the dependence of data for married couples, the longitudinal model for individuals was combined with a cross-sectional model for matched pairs (Barnett et al., 1995; Maguire, 1999). Future research on dual-earner couples will benefit greatly from this analytic strategy, which makes it possible to look past averages and independent reports and explore issues of change and dependence among family members.

Another way to conceptualize change over time from a more macro perspective is to examine challenges faced by dual-earner families from a life course perspective. The predominant focus in the dual-earner literature has been on families with young children. We know little about how work-family issues may differ for families with adolescents, for empty nesters, or for families with young adults who return home. Thus, although our research has focused on those families whose resources are most tightly stretched, dual-earners with young children, we neglect to emphasize that this is a relatively short period in a family’s life course (although for those living through it, it may feel like an eternity). A life course perspective challenges us to examine how work-family strategies and relationships change over time and, in part, as a function of historical, social, and family factors.

Work-Family Policy

A sentiment that underlies much of the dual-earner literature, as well as societal reactions to this lifestyle, is sadness over the loss of an unhurried, simpler, happier time for families, even though our historical data suggest that those simpler, happier times exist only as myths (Coontz, 1992). In fact, empirical data actually suggest that dual-earner families fare quite well on mental health, marital, and child indicators, yet societal ambivalence regarding maternal employment, and consequently dual-earner families, still exists (Pleck, 1992). As Pleck suggests, this ambivalence is reflected in our lack of social policies to support working families.

In fact, to date, research on work and family has rarely been translated into policies at work, state, or federal levels, with the possible exception of the Family and Medical Leave Act (FMLA). FMLA legislation, which provides for 12 weeks of unpaid leave for parental responsibilities, was passed in 1993; however, it is sorely inadequate. It excludes 95% of employers and 50% of employees (because it applies only to workplaces with 50 or more employees), and as unpaid leave it offers little support to low-income families who cannot afford unpaid time off (Gerstel & McGonagle, 1999). On a brighter note, California has become the first state to pass legislation to provide partial pay for parental leave for working parents.

Looking beyond social policy, another potential site for change is the workplace (MacDermid & Targ, 1995). The research evidence is clear: Jobs that have some degree of autonomy, control, flexibility, and support are good for workers, their mental health, and their family relationships. Some companies have been extremely creative in making what appears to be monotonous, highly controlled work into participative and engaging work (Crouter, 1984; Lambert, 1999). When this happens, benefits to the worker accrue as well as benefits to the employer in terms of less turnover, fewer sick days, and greater productivity.

It is our belief that, given current knowledge about workplace policies and conditions that have been associated with the well-being of workers and their families, a fruitful direction for the next decade would be to develop experimental or quasi-experimental studies that begin to pinpoint how changes in workplace polices (i.e., flexible schedules) or changes in work conditions (i.e., increases in autonomy, supervisor supports) affect workers’ mental health and family relationships. A confound in our non-experimental designs is selection effects, non-random effects that cloud our understanding of work-family processes because certain types of individuals choose particular types of employment. Carefully designed studies with control and experimental groups will not only highlight those factors that enhance family life but allow us to get a handle on selection effects.

Finally, at the level of the family, some research points to the effective strategies that couples have developed to better manage their work and family lives. In their qualitative study of 100 individuals in dual-earner families, Becker and Moen (1999) identified three separate strategies that couples used to cope with work and family demands. The first strategy was placing limits on the number of hours they worked and reducing expectations for career advancement. There were gender differences in this strategic approach, with two thirds of women and only one third of men using this technique. The second approach, trading off, reflected a value couples shared about egalitarianism. This was a life course strategy where partners would take turns, over their lives, giving priority first to one partner’s career and then the other’s. Finally, many couples coped by having one-job, one-career families. In 40% of the families interviewed, one person had what both spouses perceived as a job while the other had a career. This strategy was gendered, with over two thirds of the couples falling into the wife-job/husband-career category.

Deutsch (1999) documented similar strategies used by equal-sharing dual-earner couples. In her study, 150 couples were identified as equal sharers, couples in which both partners reported that family work was divided equally between parents. Deutsch’s data highlight the importance of a life course perspective in examining work-family issues because approximately half of her equal-sharing couples reported a period of inequality when their children were very young (under 3 years of age) (F. Deutsch, personal communication, November 7, 2002). Although this time was often characterized by high levels of paternal involvement, mothers performed significantly more of the family work. However, once children entered more formal care arrangements (e.g., preschool or kindergarten), couples quickly transitioned into an equal-sharing mode. Deutsch also noted that another common approach taken by equal sharers was a cutting-back strategy, which looked different in her middle-class versus working-class dual-earner couples. Middle-class career couples limited both work hours and opportunities for career advancement as a way to successfully combine work and family life, thus at times limiting their potential for career success. For working-class couples, equal shares were more likely to have two spouses who worked full time, so that husbands did not need to work two or three jobs to make ends meet. This allowed both parents an equal opportunity to share in family life. Often these couples relied on shift work and alternating work schedules not only to avoid the costs of child care but to give both parents equal opportunity to raise their children.

Both of these qualitative studies provide important insights into the strategies couples use to reduce work and family conflict. They are important because they move us beyond the question of “How does work affect family?” to “How do individuals and couples shape their work and family lives successfully?” The next steps should include attention to how these strategies might work for families of different races and ethnicities, for workers with different job arrangements such as shift workers, and for seasonal and contingent workers. In addition, we would benefit from more information about family relations and supports that extend beyond the nuclear parent-child family to examine broader family networks that include the role of kin and fictive kin in supporting working parents.

Imagining what the future holds for dual-earner families is an interesting exercise. The growing use of technology means work can happen at any place and at any time. These technological changes create greater flexibility for workers along with greater demands to work around the clock. The projected shortage of workers over the next three decades is expected to lead to increases in benefits that attract skilled workers (Judy & D’Amico, 1997), but the fate of low-income, unskilled workers may be at risk. Future research on dual-earner families must continue to focus on processes that link employment conditions to family life, with greater attention to the bidirectional nature of these relationships and greater sensitivity to the ways in which race, class, and gender moderate these relationships. Finally, as we began this chapter, we noted that all marriages bear the imprints of economic and cultural trends. To understand the dynamic nature of dual-earner marriages, our lens must continually change focus to include broader macro-level social and economic events that ultimately affect our vision of micro-level relationships and processes in families.