Which of the following is inconsistent with typical divisions of labor in relationships in the west?

Am Sociol Rev. Author manuscript; available in PMC 2014 Dec 22.

Published in final edited form as:

PMCID: PMC4273893

NIHMSID: NIHMS648399

Abstract

Changes in the nature of marriage have spurred a debate about the consequences of shifts to more egalitarian relationships, and media interest in the debate has crystallized around claims that men who participate in housework get more sex. However, little systematic or representative research supports the claim that women, in essence, exchange sex for men’s participation in housework. Although research and theory support the expectation that egalitarian marriages are higher quality, other studies underscore the ongoing importance of traditional gender behavior and gender display in marriage. Using data from Wave II of the National Survey of Families and Households, this study investigates the links between men’s participation in core (traditionally female) and non-core (traditionally male) household tasks and sexual frequency. Results show that both husbands and wives in couples with more traditional housework arrangements report higher sexual frequency, suggesting the importance of gender display rather than marital exchange for sex between heterosexual married partners.

Keywords: gender, household labor, marriage, sexual frequency

In the United States, a new narrative is emerging to describe contemporary marriage. Challenging the notion of marriage as an institution ensnared in a stalled gender revolution, this new perspective asserts that today’s marriages are more egalitarian, flexible, and fair than those of the past (Sullivan 2006; Sullivan and Coltrane 2008). The theme of convergence between wives’ and husbands’ roles has taken center stage at high-profile conferences on the family, such as those of the Council on Contemporary Families, in policy pieces on marriage and feminism (Gornick 2002; Marshall and Sawhill 2004), and in academic work predicated on the demise of the male breadwinner model in the industrialized West (Crompton 1999).

The debate about how much heterosexual marriage has changed from traditional models often boils down to changes in the division of labor (cf. Bianchi et al. 2000). More specifically, it hinges on whether married men’s participation in household work has increased meaningfully. Advocates of the gender-role convergence perspective argue that recognition of change has been lost, because scholars highlight women’s larger share of household work but fail to recognize married men’s greater participation in housework and childcare as a response to the dramatic rise in wives’ employment and paid labor (Sullivan 2006). Debates about the importance of housework—and under what conditions men and women do more housework—have recently come to the fore again (England 2011; Risman 2011; Schneider 2012; Sullivan 2011).

Although this debate can resemble a struggle over whether the glass is half-empty or half-full, evidence is accumulating that U.S. husbands are, in fact, doing more unpaid family work, particularly in the realm of child-care, than did their counterparts of yesteryear. From the 1960s to the beginning of the twenty-first century, men’s contribution to housework doubled, increasing from about 15 to over 30 percent of the total (Bianchi et al. 2000; Fisher et al. 2006; Robinson and Godbey 1997). Similar trends are evident for industrialized countries throughout the world: men’s proportional contribution to family work (including housework, childcare, and shopping) increased, on average, from less than one-fifth in 1965 to more than one-third by 2003 (Hook 2006).

Accompanying the effort to track secular change in wives’ and husbands’ work patterns are efforts to document how egalitarian work arrangements affect other components of marriage. Interest in the connections among role similarity or complementarity, a couple’s cohesiveness, and marital well-being is longstanding in social science (Becker 1981; Parsons and Bales 1955), but it seems to have intensified in tandem with the recent claims of work-role convergence (Amato et al. 2003, 2007; Brines and Joyner 1999). Here again, special attention is devoted to the household division of unpaid family work. For example, research shows that when men do more housework, wives’ perceptions of fairness and marital satisfaction tend to rise (Amato et al. 2003; Stevens, Kiger, and Mannon 2005) and couples experience less marital conflict (Coltrane 2000).1 Other research shows that U.S. couples who have more equal divisions of labor are less likely to divorce than are couples where one partner specializes in breadwinning and the other partner specializes in family work (Cooke 2006).

The claim that couples who share housework have more sex has captured substantial public attention. In the popular imagination, husbands’ contributions to housework seem decisive, the implications of which were recently spun in a headline: “Men: Want More Sex? Do the Laundry!” This claim appears to have originated in an unpublished survey conducted by Chethik (2006). It so captured the popular imagination (or at least that of reporters) that it led to an Associated Press story subsequently featured online by media giants ABC, NBC, CBS, CNN, FOX, and smaller sites like the Huffington Post and China Daily.

Sex is an understudied but important component of marriage, continuing to be a central area of spousal concern and conflict (Elliott and Umberson 2008). Sexual activity is linked to marital satisfaction, but there have been few recent attempts to understand the organization of sexual frequency in marriage (Call, Sprecher, and Schwartz 1995; although see Gager and Yabiku 2010; Yabiku and Gager 2009). Romantic and sexual scripts are often highly gendered outside marriage (Udry and Chantala 2004), and we suspect they remain so within marriage. Sexual activity, in addition to being important in its own right, also offers a view about the functioning of gender relations in marriage at the close of the twentieth century.

Although the notion that egalitarian marriages are sexier was widely broadcast in the media, there is little empirical support for this view. The claim rests on results of a small-scale (N = 300) survey and reports of couples in therapy conducted by Chethik, which, while intriguing, are difficult to evaluate (Chethik 2006; cf. North 2007). Moreover, other research suggests that for all the benefits of peer marriage, more egalitarian couples are more likely to have unsatisfactory sex lives and experience a lack of passion due to habituation, and these differences are not explained by a shortage of time (Schwartz 1995). While couples in more traditional marriages may experience a range of marital difficulties, lower sexual interest is especially a problem among egalitarian couples (Schwartz 1995). More recent research finds that husbands’ housework is positively linked to sexual frequency, but women’s own housework hours are even more strongly associated with sexual frequency, suggesting that greater egalitarianism may not be associated with higher sexual frequency (Gager and Yabiku 2010).

In this article, we begin by outlining two bodies of theory that offer competing predictions about the relationship between sexual frequency and the household division of labor among heterosexual married couples. We first discuss predictions derived from exchange theory, then predictions from an approach that stresses the gendered nature of sexual scripts, and finally turn to a range of important control variables derived from the existing literature that emphasizes constraints and opportunities for sex. One key innovation is that rather than consider all housework as identical, we separately examine men’s and women’s time spent in traditionally masculine and traditionally feminine tasks. We use nationally representative data to test whether and how sexual frequency is linked to the household division of labor. Our results do not support the notion that more egalitarian divisions of labor are associated with higher sexual frequency. Instead, we find that households in which men do more traditionally male labor and women do more traditionally female labor report higher sexual frequency. This suggests that among heterosexual couples, the relationship between housework and a couple’s sex life is governed by a gendered set of sexual scripts.

SEX IN MARRIAGE: EGALITARIANISM AND EXCHANGE

Sex in marriage, and what leads to more or less of it, reliably excites the popular imagination, but interest in these questions has a more uneven history in the social sciences. Kinsey’s early attempts to develop a science of the terra incognita of human sexual behavior found that marital intercourse was, as described by Blumer (1948:522), “the chief medium of sex outlet” for the adults in his samples (Kinsey, Pomeroy, and Martin 1948). However, changing family demographics and related policy preoccupations have arguably steered attention away from research on sexual frequency in marriage over the last quarter-century. Over the past few decades, scholars have noted the scarcity of research on sexual activity among married and committed couples (Call et al. 1995; Christopher and Sprecher 2000; Greenblat 1983), despite the emergence of several nationally representative surveys that gathered data on respondents’ sexual behavior in the late 1980s and 1990s.

Nonetheless, sex is an important component of marriage. Blumstein and Schwartz’s classic, American Couples (1983), identified sex (in addition to money and power) as a key good around which marriages—indeed all intimate partnerships—are organized. Sexual frequency is of interest for researchers because it is positively linked to emotional satisfaction and physical pleasure, and couples with greater sexual frequency are less likely to divorce or break up (Waite and Joyner 2001; Yabiku and Gager 2009; Yeh et al. 2006). Throughout this article, we assume that greater sexual frequency is generally a desired good: conflict may exist over the timing and frequency of sex (Elliott and Umberson 2008), but more frequent sex is linked to higher sexual and marital satisfaction for both men and women.2 Couples believe sex is an important part of marriage, but there is also substantial marital conflict over sex, largely because men and women differ in their desire for sex. This suggests caution in a straightforward interpretation of sexual frequency as purely unproblematic or reflecting desire (Elliott and Umberson 2008).

The difference in men’s and women’s desire for sex underpins a key perspective on sex: sex can be used as a resource for exchange. Predictions of social exchange theory are of particular interest (Homans 1961; Sprecher 1998).3 Because spouses (the parties to the exchange) possess different resources, they benefit from exchanging a resource one possesses for another scarce resource the other possesses. Sex, in this view, is a resource that partners might use for exchange. A self-interested view of social exchange suggests that individuals exchange when each party benefits. Partners thus trade sex for other scarce resources such as time, money, commitment, or other goods when they both benefit (Baumeister and Vohs 2004).

Although the condition of mutual benefit suggests a gender-free venue for exchange, both popular and scholarly understandings see sex as a female, rather than male, resource. Baumeister and Vohs (2004) argue compellingly that sex should be seen as a female resource due to the principle of least interest— if men want sex more than women, they must induce women to engage in sex by offering other benefits. A review of a wide variety of measures of sex drive suggests that men want sex more than women (Baumeister, Catanese, and Vohs 2001).4 Whether men’s greater sexual desire results from biological or cultural factors is immaterial; either condition results in women’s possession of a scarce resource.

An exchange perspective, combined with the assumption that men desire sex more than women, suggests that women could trade sex for resources men control. This could apply to any set of bargaining goals (e.g., decision-making, monetary or gift exchange, or time spent on any task), but we focus here on the application to household labor, because labor has been at the center of a discussion about how much marriages have changed. In addition, a long research tradition investigates whether and how women exchange another resource they control—their earnings—for men’s participation in housework (Brines 1994; Greenstein 2000; Gupta 2007; Lundberg and Pollak 1993).5 A central assumption of this line of research, which we follow, is that both men and women prefer to avoid housework, but housework is more likely to fall into women’s sphere of responsibilities by default (cf. Lundberg and Pollak 1993). The implication is that women are likely to use their resources—in this case, sex—to bargain their way out of performing housework.

Qualitative evidence supports the view that some women exchange sex for men’s participation in household labor, or, alternatively, withhold sex when men do not participate in household labor. In Hochschild and Machung’s (1989:45) The Second Shift, one respondent (Nancy) notes, “When Evan refused to carry his load at home … I used sex. I said, ‘Look, Evan, I would not be this exhausted and asexual every night if I didn’t have so much to face every morning.’” Similarly, one of Elliott and Umberson’s (2008:401) respondents (Chantelle) said, “[I tell Anthony,] ‘If I have had a really good day, and you have been helpful, I would say you took out the trash and you brought the trashcans in and you mowed the lawn and everything. Those are the things that work for me to kind of get me going.’”6

Rather than direct exchange, it is possible that sexual frequency and an egalitarian division of household labor are linked via marital satisfaction. Recent studies show that husbands’ participation in household labor is often associated with wives’ reports of higher marital quality (Amato et al. 2003; Stevens et al. 2005). Other work (Chethik 2006) appears to draw from this result to explain why husbands’ sharing of housework might lead to greater frequency of sex in marriage: wives feel more supported and happier in their marriages when their husbands do more chores, and these positive feelings promote more sex as a side benefit. More generally, theoretical work ranging from the stipulation that a sense of distributive justice in marriage promotes coital frequency (Jasso 1987) to economic models that locate today’s marital gains in partner similarities that maximize joint consumption rather than joint production (Lam 1988; Lundberg and Pollak 1996) also lend credibility to the idea that an egalitarian division of labor results in a happier marriage and is more conducive to sexual activity.

An exchange perspective would predict a positive relationship between men’s household labor and sexual frequency: sexual frequency should be high when husbands do more housework and low when husbands do less. This prediction reflects an understanding of marriage as a site characterized by the exchange of scarce resources between partners, and is concordant with popular and scholarly understandings of sex in marriage. Nevertheless, given research linking marital satisfaction to husbands’ participation in household labor and some research that suggests the importance of marital satisfaction for sexual frequency (Rao and Demaris 1995), we are open to the possibility that egalitarian arrangements increase satisfaction in relationships and thus lead to greater sexual frequency. We include controls for marital satisfaction to test this possibility.

SEXUAL SCRIPTS: GENDER, DIFFERENCE, AND DESIRE

There are reasons to predict a very different relationship between the division of household labor and sexual frequency. First, gender continues to play a central role in organizing the division of household labor. Women continue to do more housework than men, and differences are not explicable by a range of economic factors. The importance of gender in organizing labor and marriage suggests that housework itself may lie outside the realm of conventional possibilities for exchange. Second, heterosexual attraction and intimacy seem to be organized around the enactment of difference or complementarity between the sexes (Goffman 1977; Rich 1980). Among heterosexual couples in their teens, pairs with a self-rated very masculine boy and self-rated very feminine girl are most likely to have sex, and to have sex sooner, than are other romantic pairs (Udry and Chantala 2004). Gender’s role in marital sex is less well documented, but Schwartz (1995, 2007:2) reports that egalitarianism in committed heterosexual adult relationships is associated with occasional boredom and a “sibling-like” tonality to the relationship that undermines sexual desire. Schwartz (2007:2) avers that “introducing more distance or difference, rather than connection and similarity, helps to resurrect passion in long-term, stable relationships.”

These observations suggest a conceptualization of heterosexual marriage as an institution in which gender still plays a central role (Berk 1985; Coltrane 1998). Drawing on this central insight and on a sexual scripts approach (Gagnon and Simon 1973), we argue that sexual activity is more likely in households with more gender-traditional divisions of household labor. A sexual script approach suggests that for intercourse to occur, a script must exist that defines a situation as sexual (Gagnon and Simon 1973). Sexual scripts specify when, why, and how individuals should act sexually (Laumann et al. 1994). As a simple example of a script, intercourse typically takes place in a series of relatively tightly delineated stages, moving from kissing to fondling and then to coitus (Gagnon and Simon 1973). The approach suggests that scripts exist at three levels: the cultural or collective, which broadly defines available sets of scripts; the interpersonal, used when individuals improvise or adapt cultural scripts for particular scenarios; and the intrapsychic, which helps individuals script their own behaviors and align their own desires (Simon and Gagnon 1986). In this article, we assume that internalized dominant cultural scripts govern sexual behavior, although interpersonal and intrapsychic scripts may also structure sexual behavior in marriage.

How might sexual scripts work in marriage? There is relatively little work on this topic, but the logic of a sexual scripting argument generally suggests that women’s and men’s sexual activity is governed by internalized cultural scripts.7 Among teens, sexual scripts are highly gendered and link sexual activity to masculinity and femininity (Storms et al. 1981; Udry and Chantala 2004). Other recent research finds that men experience greater sexual dysfunction when their partners spend more time with the men’s friends than men do themselves, suggesting that behaviors that threaten men’s independence and masculinity lead to greater sexual dysfunction (Cornwell and Laumann 2011). Given the general importance of gender, we suspect that scripts continue to link sexual behavior to masculinity or femininity among heterosexual married couples. If so, expressions of gender difference should help to create sexual desire. Household labor and its performance—or lack thereof—is centrally tied to notions of what constitutes appropriate behavior for men and women and thus masculinity and femininity (Berk 1985; Bittman et al. 2003; Brines 1994; Greenstein 2000; South and Spitze 1994). If appropriate performances of masculinity and femininity are prerequisites for sexual behavior or sexual desire in marriage, and housework is a key way of engaging in these performances, then the extent to which husbands and wives do housework in ways that signify masculinity or femininity should be linked to sexual frequency.

We note three plausible mechanisms that might link sexual frequency to gender-traditional divisions of housework: (1) gender-traditional divisions of labor increase sexual desire and thus sexual frequency, (2) both result from gender traditional beliefs or are ways of doing gender, and (3) gender-traditional arrangements may increase relationship satisfaction, which in itself leads to greater sexual frequency.

The first mechanism—that sexual scripts activate desire and sexual behavior in the presence of gendered activity—would operate in the following fashion. Traditional gender performances serve as cues of masculine and feminine behavior; these cues activate individuals’ internalized cultural sex scripts, creating sexual desire and activity. In essence, traditionally masculine and feminine behaviors consciously or unconsciously serve as turn-ons for individuals. We do not argue that this takes place instantly, but rather over time, individuals perceive their spouse as more masculine or feminine as they engage in gender-traditional behaviors, and this increases sexual attraction. To the extent that masculinity and femininity are central parts of both the household division of labor and sexual attraction and activity, we expect that households with more traditionally gendered divisions of labor will experience greater sexual frequency. We note that this argument—that sexual behavior is linked to gender identity and expression—is entirely consistent with a mechanism proposed by Cornwell and Laumann (2011:177–78): “in the context of sexual relationships, masculinity is expressed through ‘erection, penetration, and climax,’ so it is possible that threats to gender identity … manifest as sexual problems.”

A second possibility is that couples with more gender-traditional divisions of housework hold more traditional beliefs and act in more gender-typical ways, which leads to more frequent sex. More masculine-identified men may value more frequent sex, and more highly feminine-identified wives may refuse sex less often because they view providing sex as part of being a good wife. Thus, men may initiate sex more frequently, and wives refuse less, with no link to desire. In essence, this mechanism suggests that both housework and sexual behavior are ways that couples do gender, and any observed relationship between the two would reflect couples’ underlying orientations toward gender rather than causal influence. An alternative possibility reflecting similar intuitions is that there is greater coercion among households with traditional divisions of labor, leading to greater sexual frequency. However, as we show in the Appendix, wives’ reported satisfaction with their sex life has the same relationship to men’s participation in housework as sexual frequency. This suggests coercion is not an important mechanism, because coercion should lead to higher sexual frequency but lower sexual satisfaction among women.

A third possibility is simply that gender-traditional arrangements are linked to sexual activity because couples perceive greater affection and love when partners do more (albeit in traditionally gendered ways) in the household. Rather than couples engaging in more sexual activity because traditional divisions of housework act as signals of masculinity and femininity, couples may instead feel more affection and satisfaction within their relationships under traditional gender divisions of labor, and this leads to more frequent sex. Doing housework can convey affection, although often in traditionally gendered ways. As DeVault (1991:324) notes, “the gender relations of feeding and eating seem to convey the message that giving service is part of being a woman, and receiving it fundamentally part of being a man.” Gender-traditional beliefs and practices are often associated with greater marital happiness and men’s emotion work in the family (Wilcox and Nock 2006).

How does a sexual scripts approach translate into testable hypotheses about the link between beliefs about gender, the division of housework, and sexual frequency in marriage? In short, a sexual scripts perspective (and the associated alternative mechanisms we noted) suggests that couples with more egalitarian divisions of household labor will have less active sex lives. Because these couples engage in less traditionally feminine and masculine behaviors, they are less likely to activate scripts linking displays of difference to desire. In contrast, couples in which husbands and wives engage in more gender-traditional behaviors should report more frequent sexual activity.

We attempt to determine whether the alternative mechanisms we noted could explain any association we find. We thus test for two additional effects. First, to check whether gender ideology is responsible for any association, we include measures of gender ideology and religious affiliation, because religion is often correlated with gender ideology and traditional behavior. Second, we include measures of satisfaction with marriage and with a spouse’s contribution to housework to check whether the division of household labor is associated with sexual frequency only because it increases satisfaction in marriage. We are thus able to offer tests for alternative mechanisms.

SEXUAL OPPORTUNITIES AND CONSTRAINTS

The perspectives outlined above have much to say about theoretical relationships between sexual activity and marital characteristics, but little existing research focuses on these theories. Instead, research typically looks at the role of opportunities and constraints for sex in marriage, focusing on a variety of demographic correlates. What we do know about sexual frequency in marriage is that older couples report lower sexual frequencies than younger couples (Blumstein and Schwartz 1983; Brewis and Meyer 2005; Call et al. 1995; Greeley 1991; Greenblat 1983; Rao and Demaris 1995). Biological aging is the most common explanation for this decline. The negative correlation between age and sexual frequency has also been attributed to marital duration and habituation (Blumstein and Schwartz 1983; James 1981). However, marital duration has no significant effect after the first year of marriage in models that control for other time-related variables (Call et al. 1995). Marital satisfaction, in contrast, is the strongest correlate of sexual frequency, after age (Call et al. 1995).

Constraints and opportunity also play a central role in understanding sexual frequency, although results are inconsistent (Christopher and Sprecher 2000). Certainly, some constraints matter, such as the presence of young children or pregnancy (Call et al. 1995; Gager and Yabiku 2010; Greeley 1991). Time constraints appear to be less important. Sexual frequency does not decline when both partners are employed full-time or with the number of hours husbands and wives spend in paid work (Call et al. 1995; Gager and Yabiku 2010; Greeley 1991; Hyde, DeLamater, and Hewitt 1998). Non-standard work, however, is associated with more sexual problems and dissatisfaction (White and Keith 1990), suggesting that some, but not all, opportunity constraints impose costs on couples’ sex lives.

In addition, a recent article by Gager and Yabiku (2010) explicitly takes up the relationship between time spent in housework and sexual frequency, asking whether time spent in housework serves as a constraint preventing couples from engaging in sex. Instead, they find that both men’s and women’s time in housework is related to greater sexual frequency. They conclude that this relationship is due to unmeasured tendencies toward greater activity in both areas: individuals who work hard also “play” hard. As we discuss below, our theoretical approach leads us to focus on different measures of the extent to which particular types of housework are gendered. However, to account for the theoretical relationship Gager and Yabiku suggest, we also include measures of the total amount of time spent in housework. Our discussion of results further compares our model with theirs.

DATA

To investigate the relationship between sexual frequency and division of household labor among married couples, we use data from Wave II of the National Survey of Families and Households (NSFH) (Sweet and Bumpass 1996).8 The age of the data may limit generalizability to the present day (interviews occurred from 1992 to 1994), but to our knowledge, it is the only dataset with detailed measures of both sexual frequency and actual participation in household labor.9 Despite the age of the data, we consider these results relevant for contemporary discussions of marriage and the family. We return to the topic of generalizability to the present day in the conclusion.

Given the sensitive nature of our dependent variable—self-reports of frequency of sex—one problem we encounter is missing data. Roughly 10 percent of respondents have missing values for sexual frequency, including those who report “don’t know,” and nearly 25 percent of respondents have missing data on this or another variable in our analysis. Rather than lose these cases, we used two procedures for missing data: for housework variables, we used procedures developed by South and Spitze (1994); for other missing data, we relied on multiple imputation.

We dealt with missing and extreme values on the housework variables using a slight modification of procedures described by South and Spitze (1994). First, we excluded respondents in which both members of a couple had missing values on more than eight items of housework. This eliminated 444 respondents, many of whom had missing values on other key variables.10 We then recoded reports of hours spent beyond the 95th percentile of the distribution for each housework item to the 95th percentile of the distribution for each gender. For respondents with missing values, we replaced missing values with the mean for each item for other respondents. For respondents who gave a value of zero to all core or non-core items, we placed men’s share at zero.11 We also experimented with multiple imputation for missing housework items. Results were nearly identical, so we chose the simpler method.

We used multiple imputation for other missing values because missingness on sexual frequency is likely correlated with one’s actual sexual frequency. Multiple imputation uses correlations between variables in an analysis to generate replacement values for missing values, adding in an error term and generating multiple estimates to capture the variability. Estimates from each imputation are then generated and combined. We used all variables in our analysis for multiple imputation, using the ICE program in Stata. ICE is a regression-based program for imputation, meaning that variables are imputed using all other variables as regressors for each of the other variables. We included a partner’s reports of sexual frequency as an auxiliary variable to improve imputation, but no other variables because inclusion of auxiliary variables does little to reduce bias unless the correlations between auxiliary variables and variables with missing data are high (.9) and the proportion of missing data is high (e.g., 50 percent missing) (Collins, Schafer, and Kam 2001). We generated 20 imputations because the rule of thumb of three to five imputations is often insufficient (Graham, Olchowski, and Gilreath 2007). We used logistic regression and ordered logit models to impute non-continuous variables because using linear methods and rounding to maintain categorical or binary variables creates biased estimates (Horton, Lipsitz, and Parzen 2003).

Another potential source of missing data in Wave II of the NSFH is attrition from the original sample interviewed at Wave I, roughly five to seven years earlier. Roughly 18 percent of Wave I respondents were lost by Wave II because they could not be found, were too ill to be interviewed, or did not participate for another reason. Attrition could lead to bias if these couples had lower sexual frequency or less egalitarian divisions of labor. Additionally, some respondents divorced between Waves I and II. As with those lost from the sample, if these marriages were different than other couples on measures of sexual frequency or the division of household labor, our results could be biased. To test for this possibility, we performed t-tests for differences in sexual frequency and the amount of housework done by men and women in NSFH Wave I. T-tests (not shown) indicate no significant differences between respondents who remained married, remained in the sample but divorced, and were missing at Wave II but had been married at Wave I.

MEASURES

We measured the share of household labor performed by men across two types of household labor forming a rough approximation of male and female typed labor. Following existing literature, we separated tasks into core and non-core categories (Bianchi et al. 2000). Core tasks include preparing meals, washing dishes, cleaning house, shopping, and washing and ironing; non-core tasks include outdoor work, paying bills, auto maintenance, and driving. In some sense, core tasks are female simply because women do more of them, and non-core tasks are male for the same reason. Yet beyond this, traditional notions of masculinity and femininity are attached to these tasks. Core and non-core tasks are roughly divided along indoor/ outdoor, nurturing/worldly, and private/public dimensions that reflect gendered expectations in place since development of the separate spheres ideology (Padavic and Reskin 2002).

We calculated share measures using respondents’ reports of their own and their spouses’ time spent on these activities. We relied only on self-reports of individuals’ and their spouses’ labor because husbands and wives did not complete the survey at the same time, so their reports may diverge because of differences in the labor performed during the reported week. Table 1 shows women’s and men’s hours, and the ratio of women’s to men’s time, for core and non-core household labor using women’s and men’s reports. As other research has demonstrated (Bianchi et al. 2000), women do the majority of core tasks that need to be done on a daily basis, and men do relatively more non-core tasks. Similarly, we find that men’s and women’s reports diverge slightly: men’s reports indicate more time on both core and non-core activities than their wives’ reports of the same work. Interestingly, men report that women spend more time on non-core housework but less time on core housework than women report for themselves.

Table 1

Wives’ and Husbands’ Time in Household Labor

Our measure of sexual frequency is a self-reported response to the question, “About how often did you and your husband/wife have sex during the past month?” We recoded values of sexual frequency past the 95th percentile to values at the 95th percentile and imputed values for cases with missing data, including cases where respondents did not know or refused to answer. Table 2 shows descriptive statistics for sexual frequency and other variables. As Table 2 shows, women reported having sex with their spouses slightly more than five and a half times in the past month, and men reported lower frequencies, about .4 times fewer over the past month. Although it may appear surprising that husbands’ reports are lower than their wives’, existing research comparing husbands’ and wives’ reports has found similar results (Clark and Wallin 1964; Kinsey et al. 1948).

Table 2

Means and Standard Deviations of Sexual Frequency and Other Couple Characteristics

As we noted earlier, other mechanisms could explain a relationship between the division of household labor and sexual frequency. We thus included variables to test for the presence of some of these mechanisms. As a measure of the extent to which men and women engage in gender-traditional behaviors, we included measures of husbands’ and wives’ participation in paid labor. In the models we present, we relied simply on the number of hours spent by husbands and wives in paid work, because these are also measures of potential constraints on time availability. In other models, we tested whether male-breadwinner/ female-homemaker households were significantly different and found no significant results.

We also tested to see whether gender ideology and gender beliefs dictate housework arrangements and sexual frequency by including two sets of variables. First, we included a measure of gender ideology, consisting of the sum of a respondent’s agreement or disagreement with the following four statements: “It is much better for everyone if the man earns the main living and the woman takes care of the home and family”; “A husband whose wife is working full-time should spend just as many hours doing housework as his wife”; “Both the husband and wife should contribute to family income”; and “It is all right for mothers to work full-time when their youngest child is under age 5.” In addition, we controlled for religious affiliation and church attendance as another way to tap into gender traditionalism and distinctive patterns of sex and housework. Earlier research indicates that Catholics report lower sexual frequency, and conservative Protestants have more traditionally gendered divisions of labor and distinct sexual patterns (Call et al. 1995; Wilcox 2004). We thus included dichotomous variables for respondents’ religious affiliation, following the coding scheme suggested by Steensland and colleagues (2000) as closely as possible using the NSFH data. We ended up with black Protestant, evangelical Protestant, mainline Protestant, Jewish, Catholic, conservative Christians, and an “other” category combining the remaining smaller categories from the coding scheme (nonreligious is the reference category).

Finally, to control for the possibility that any relationship between wives’ and husbands’ share of housework functions through its effects on marital quality, we included controls for happiness in marriage. We measured this with responses on a seven-point scale to the question, “Taking things all together, how would you describe your marriage?” and with spouses’ housework contributions, measured with responses to the question, “How happy are you with the work your spouse does around the house?” Additionally, because joint religious attendance is a good predictor of relationship quality (Ellison, Burdette, and Wilcox 2010), we included a dichotomous variable measuring whether both spouses attend church weekly or more frequently.

We also included measures of family structure and stage in the life cycle, because these may be important for sexual frequency and the division of household labor. First, we included a measure of marriage within the past year to control for the possibility of a honeymoon effect in recent marriages leading to greater sexual frequency; 3 percent of couples in our sample were married within the previous year. We also included measures of the wife’s and husband’s age because age is often an important predictor of sexual frequency and is related to the division of household labor. We included measures of the number of children living in the household below age 2 years, between 2 and 6 years, and between 6 and 13 years. We controlled for husbands’ and wives’ economic contributions, using the share of the income provided by the wife and total household income.12 For these measures, values in Table 2 are the same for men and women. This is because they are based on either primary respondents’ responses, as in the case of the number of children, or self-reports from each spouse, as with wife’s age and husband’s age.

As an additional control for opportunity, we included a measure of time spent alone with the spouse over the past month. Respondents answered the question, “During the past month, about how often did you and your husband/wife spend time alone with each other, talking, or sharing an activity?” and were given six response options: never, about once a month, two or three times a month, about once a week, two or three times a week, and almost every day. We collapsed these six categories to four. Interestingly, nearly 40 percent of respondents said they spent time alone with their spouse once a week or less during the previous month.

In addition, we included controls for education. We measured education using a series of dichotomous variables for completion of high school, attendance of some college, and completion of a college degree using husbands’ and wives’ self-reports. Finally, we included measures of wife’s and husband’s self-rated health. Health was self-reported using a five-point scale, with higher values representing greater perceived health.

METHODS

Because sexual frequency is distributed as a count variable but is overdispersed—that is, the distribution of sexual frequency violates the assumption in Poisson regression that the mean and variance are equal—the use of ordinary least squares or Poisson regression is inappropriate. We used negative binomial regression to assess links between men’s and women’s participation in different types of household labor and sexual frequency. Negative binomial regression models the count-generating process but relaxes the assumption that variance of the distribution is equal to the mean.

The model takes the following form:

log(λi) = xiβ, var(Y) = λτ

where λi= E(Y), Y is sexual frequency, β is a set of coefficients produced by the model including an intercept, xi is the set of independent variables, and τ is a shape parameter modifying the variance of Y.

The NSFH has two types of respondents: primary respondents for a household and their spouses. To take full advantage of the available data, we relied on both respondents. We present four regression models: men separately, women separately, one pooling men and women into a single analysis, and one using pooled data but relying on the opposite spouse’s report of sexual frequency. Our analysis using women’s reports alone thus relies on female primary respondents and female spouses of male primary respondents, and uses women’s reports of most individual and household characteristics, including sexual frequency and their and their spouse’s hours spent in housework. Similarly, our analysis for men relies on male primary respondents and male spouses and uses men’s reports of key variables.

We present two additional results that leverage the fact that we have reports from both husbands and wives. To account for the fact that husbands and wives are located within the same households and standard errors may be biased by unobserved shared characteristics, we present results from a regression in which we pool male and female respondents into a single analysis and use cluster-robust standard errors. To deal with the possibility of same-source bias—that our independent and dependent variables may be correlated because they are reported by the same individual—we conducted a pooled analysis where the dependent variable is not a self-report of sexual frequency but spouses’ reports of sexual frequency (cf. Amato and Rivera 1999). Because husbands and wives do not necessarily complete the questionnaire on sexual frequency and hours spent in housework at the same time, we restricted the sample to partners who completed the survey in the same month. This reduces the sample size to 7,002 for this analysis.

We first present results for the overall models. We then present results showing whether variables representing alternative mechanisms mediate the relationship between sexual frequency and the household division of labor, as well as models investigating whether this link varies with respondents’ gender ideologies.

DETERMINANTS OF SEXUAL FREQUENCY

Table 3 shows results from the regression models described earlier. Column 1 shows coefficients and p-values generated using cluster-robust standard errors from the pooled analysis of husbands and wives, using self-reports of their own and their spouses’ hours spent in housework and sexual frequency. For purposes of brevity, we limit discussion of results that do not bear on our main theoretical question of interest. These findings are similar to much previous research on sexual frequency: self-rated health, wife’s and husband’s age, young children in the home, and the amount of time respondents reported spending alone with their spouse are all significant predictors of sexual frequency. In addition, we find a positive and significant effect of household income in pooled results, although the effect does not reach significance using only women’s reports and is significant only at the .05 level using men’s reports.

Table 3

Effects of Husbands’ Share of Core and Non-core Housework on Sexual Frequency

Our main question of interest, however, is whether and how men’s participation in household labor is linked to sexual frequency. Our results suggest that sexual frequency is highest in households with traditionally gendered divisions of labor. As Table 3 shows, the coefficient for men’s share of core house-hold labor is negative: households in which men do more female-typed (core) tasks report lower sexual frequency. The coefficient for men’s share of non-core household labor, on the other hand, is positive: households in which men do more male-typed (non-core) tasks report more sex. These effects are statistically significant and substantively large. Overall, these results suggest that sexuality is governed by enactments of femininity and masculinity through appropriately gendered performances of household labor that coincide with sexual scripts organizing heterosexual desire.

To illustrate the substantive size of these effects, Figure 1 shows predicted values for sexual frequency, varying the share of household labor performed by men while setting all other variables to their means. As the figure shows, shifting from a household in which women perform all of the core household tasks to one where women perform none of the core household tasks is associated with a decline in sexual frequency of nearly 1.6 times per month. Given a mean sexual frequency in this sample of slightly over five, this is a large difference. The figure represents two extreme values, but even households in which men do 40 percent of core household task hours report substantially lower sexual frequency than households in which women perform all core housework. The effect for men’s share of non-core housework is similar although somewhat smaller.

Predicted Sexual Frequency by Men’s Share of Core and Non-core Housework, Results from Pooled Self-report Model (column 1 of Table 3)

These models include variables that represent possible common causes of both a traditional gender division of labor and higher sexual frequency: men’s and women’s work hours, which may serve as a measure of broader masculinity or femininity; religious affiliation; and gender ideology. Not only do these variables fail to reduce the relationship between men’s share of both types of housework and sexual frequency to zero, most are not statistically significant in their own right, with the exception of two religious categories: black Protestants and conservative Christians report higher sexual frequency. Still, most important is not these specific differences, but that their existence does not eliminate the relationship of theoretical interest.

Other models largely confirm findings from the analysis pooling men’s and women’s self-reports. Column 2 of Table 3 relies on reports of independent variables from one partner and a report of sexual frequency from the opposite partner. This eliminates the possibility of same-source bias, that correlations between independent and dependent variables exist solely because both are reported by the same individual. Coefficients and levels of significance are nearly identical, with the exception of estimates for how often individuals spent time alone with their spouse in the past month. This may be because spouses who filled out the survey in the same month are more likely to share activities even if they do not share time, reducing the size of this effect. Finally, we present models using men’s and women’s self-reports of all items except household-level measures. These demonstrate whether there are differences between effects reported by women and men. Column 3 presents results using women’s self-reports and column 4 presents results using men’s self-reports. There are few differences between these and earlier estimates, although results using only men’s reports show no significant effects of non-core housework. Still, the coefficient remains positive and husbands’ share of core housework is still negative and significant.

These results—whether using both men’s and women’s reports in a pooled analysis, relying on opposite spouses for reports of our key variables, or relying on men’s or women’s results alone—show that households with a more gender-traditional division of labor report having more sex. The pattern of results suggests the existence of a gendered set of sexual scripts, in which the traditional performance and display of gender is important for creation of sexual desire and performance of sexual activity. Because we lack data on sexual desire or related variables, it is difficult for us to untangle mechanisms linking sex to a traditional division of labor.

Existing data do, however, allow additional tests for the possibilities that the relationship between sexual frequency and a traditional division of labor is mediated by marital satisfaction or linked by common causes. If spouses are happier with each other’s contributions in more gender-traditional divisions of household labor, and happiness leads to greater sexual frequency, then any relationship between sexual frequency and a traditional division of labor should disappear with inclusion of measures of happiness, and the link would have little to do with sexual scripts. We test for this possibility and variation by gender ideology. To the extent that individuals hold different beliefs about gender, their reaction to gender-traditional behaviors could vary. Table 4 shows tests for these two possibilities and includes coefficients for our main variables of interest: men’s share of both types of housework, happiness with marriage and with a spouse’s contribution to housework, and gender ideology interactions with men’s share of housework. We show results from the pooled analysis for purposes of brevity.13 Finally, we show results using men’s and women’s total hours in core and non-core work rather than shares to show that results are robust to alternative specifications of contributions to housework.

Table 4

Selected Coefficients: Tests of Alternative Specifications, Happiness as a Mediator and Interaction with Gender Ideology Using Pooled Self-reports

These results do not show support for the possibilities of mediation or an interaction. The first model shows the effect of including happiness in marriage and happiness with a spouse’s contribution to housework. The effect of happiness in marriage is positive—indicating that individuals who are happier with their marriage report higher sexual frequency—but it does not reduce the effect of men’s share of these two types of housework to nonsignificance. In contrast, the effect of happiness with spouse’s contribution to housework is not significantly different from zero. Although happiness in marriage has an important link to sexual frequency, we conclude that it does not account for the association observed. The second model shows the effect of including interactions between our measure of gender ideology and men’s share of core and non-core housework. Neither of these interactions reaches statistical significance. Their inclusion does lead the primary share of housework variables to be nonsignificant, but this lack of significance appears to reflect collinearity; when we subtract means of variables before generating interaction terms, the share of housework variables remain significant and in the expected direction.

Finally, the alternative specification of the model using men’s and women’s hours in housework is consistent with our earlier findings. Men’s hours in core—female-typed—house-work are negatively associated with sexual frequency, and women’s hours in core housework are positively associated. For non-core housework, only men’s hours are significantly associated with sexual frequency, and the coefficient is positive. Checking effects of the total number of hours is important, because we could see a negative effect of a share if coefficients for men’s and women’s hours are similarly signed but one is simply larger than the other. In this case, however, we find different effects of men’s and women’s work, and these effects differ by task. Thus, when men do more core work, reported sexual frequency is lower; when men do more non-core work, reported sexual frequency is higher, consistent with the notion of sexual scripts. To compare our model to another recent paper measuring the effects of housework hours (Gager and Yabiku 2010), we considered regression models in which we used measures of men’s and women’s total housework hours, combining core and non-core hours into a single measure. Our results are very similar to those previously reported: we find significant and positive relationships between total hours and reports of sexual frequency, likely because most of women’s hours are in core labor, which is positively signed, and most of men’s hours are in non-core labor, which is also positively signed. We note, however, that measures of model fit are better using our measures than using total number of hours, and we suggest these measures better capture the relationship between sexual frequency and household labor.

The lack of significance for tests of marital happiness and gender ideology leads us to conclude that the arrayed evidence—that households with more traditional arrangements report more frequent sexual activity, and that this relationship is not mediated by happiness, religion, gender ideology, or a range of other variables—is concordant with a gendered sexual scripts perspective. The lack of interactions or mediation lends support, we argue, to the notion that the operating mechanism is one that links within-couple displays of masculinity and femininity to sexual scripts leading to sexual frequency. Still, our understanding of the exact dynamics is limited due to the use of quantitative data. Men or women may, in essence, be turned on (however indirectly) when partners in a marriage do more gender-traditional work. Of course, men and women could also be turned off by doing work that is not traditional for their gender. Similarly, it is unclear whether women’s or men’s reactions to these performances are more important. These questions cannot be untangled with existing quantitative data.

CONCLUSIONS

This article began by noting that American marriages are more egalitarian today than they were in the past, but scholars have found it difficult to offer a clear interpretation of how egalitarianism has changed the nature of marriage itself. One broad interpretation of egalitarianism is that couples exchange resources across various domains. Moves toward more equality in one area, such as earnings, might thus induce more equal distributions in other areas, like housework, a suggestion that has certainly received extensive investigation. In this article, we asked whether men and women use housework and sex as resources for exchange, or whether other logics govern sexual frequency within marriage.

Following up on the widely publicized claim that by doing more housework, husbands in more egalitarian marriages got more sex, we sought to investigate the links between men’s participation in housework and sexual frequency using nationally representative data. Our findings suggest the importance of gender display for sexual frequency in heterosexual marriage: couples where men participate more in core tasks—work typically done by women—report lower sexual frequency. Similarly, couples where men participate more in non-core, traditionally masculine tasks report higher sexual frequency, suggesting the importance of gender-typed participation in household labor. Additionally, although our main results examined core and non-core labor separately, we note that regressions using the share of total housework (core and non-core combined) also show a negative and significant coefficient for men’s share of housework.

Because the bulk of housework done in U.S. households involves the traditionally female or core tasks of cooking, cleaning, and laundry (Bianchi et al. 2000), our findings stand in marked contrast to the published claim motivating this study: that husbands who do more housework get more sex. At the same time, one can understand how this claim might have gained currency. First, men’s contributions are important for wives’ satisfaction in marriage. Marital satisfaction is associated with sexual frequency, and it may be the case that husbands in more satisfied relationships qualitatively perceive that they have more frequent sex even though they quantitatively do not. Second, to the casual observer, husbands who do more of the traditionally masculine tasks in a marriage may in fact populate the mental category of husbands who do more to help around the house. Although men who do more yard work, car maintenance, household repairs, and the like might make sizable contributions to the division of labor at home, to characterize these efforts as emblematic of egalitarianism is misleading. At the very least, our results are difficult to reconcile with the idea that women trade sex to men for doing what is traditionally viewed as women’s work. Based on our findings, sex seems to lie outside the realm of conventional exchange.

The data we selected—Wave II of the National Survey of Families and Households—are the most recent data we are aware of that include objective measures of both sexual frequency and the division of household labor. These data are dated, though, as they were collected roughly 20 years prior to the time of writing. Because these data are older, there is a possibility that the relationships we document have changed. In particular, the script that men exchange housework for sex is a relatively recent one and, as such, may not have been evident at the time of this research. However, given the durability of some features of marriage, including the gendered division of labor, we suspect our results would still hold despite the time that has passed since the data were collected. Additionally, conclusions about the shift to egalitarianism and effects of this shift are often based on similarly aged data from the late 1980s and early 1990s. Although we may be unable to comment specifically on patterns of marriage in the present day, our results are easily applicable to claims about shifts in marriage.

Sexual frequency appears to lie in the realm of sexual scripts, but couples are not purely interested in the amount of sex they have—they undoubtedly also care about the quality of sex. Although sexual frequency is correlated with sexual satisfaction, the correlation is far from perfect. We focus on sexual frequency in this article in part as a response to existing media claims about the topic, but also because sex and housework are enduring components of marriage, historically predating romantic love and sexual satisfaction (Coontz 2005). The importance of sexual frequency for sexual satisfaction, marital stability, and marital satisfaction for egalitarian versus traditional marriages are testable questions, but not the ones this article asks. If scripts define a moment as sexual, and govern sexual initiation, then the sexual scripts theory explains sexual frequency, not sexual satisfaction. Even if egalitarian couples have the least but most satisfying sex, the scripts perspective would not be invalidated. Still, the question of satisfaction is undoubtedly important and should be pursued in future research.

In addition to encouraging further research on the relevancy of sexual scripts for other components of couples’ sexual relationships, our research also brings up questions about the relationships among sexual scripts, sexual frequency, and housework for other types of couples, including same-sex spouses and cohabiting partners. The past several decades have seen shifts in whether couples marry or cohabitate, and there are now more same-sex partners in the pool of married spouses and long-term committed partnerships. We suspect the saliency of sexual scripts and housework for sexual frequency within cohabiting and same-sex couples hinges on whether sexual activity and housework have a similar meaning for them as they do for married heterosexual couples. Research suggests that the division of household labor among gay, lesbian, and cohabiting couples is influenced by earnings and gender, but differences remain in how these couples divide household labor compared to married heterosexual couples (Carrington 1999; South and Spitze1994). We thus caution against assuming that our findings apply to other types of couples, but we encourage further investigation into the role of housework and sexual scripts in shaping sexual behavior across different types of couples.

One contribution of this study is to offer and test models of the role of sexual activity within marriage. Existing research acknowledges that sex lives and the frequency of sex are important concerns for couples, even if they remain contested terrain (Elliott and Umberson 2008), but little research suggests how sex is organized. This article offers a systematic test of exchange perspectives on sex in marriage and provides a new perspective to explain sexual behavior in marriage: namely, one that emphasizes the continued importance of gendered sexual scripts.

Our research indicates that changes in sexual scripts have not kept pace with changes in the division of household labor. In some ways, this finding should not be surprising. Scholars continue to assert that shifts toward gender equality across multiple arenas occur at uneven paces, with the organization of romantic relationships being particularly stagnant (England 2010). The association we observed between sex and traditional gender behavior corresponds with the persistence of other traditional gender mores within heterosexual romance, including the double standard that penalizes young women and rewards young men for sexual agency (England, Shafer, and Fogarty 2008; Hamilton and Armstrong 2009). One area we did not investigate that could offer promise for the future is men’s and women’s work behaviors outside the home, for example, whether the gender-type of one’s occupation also influences sexual frequency (Schneider 2012).

The notion that sex within marriage is bound to traditional sexual scripts does not necessarily put egalitarianism at odds with sexual frequency. Rather, the saliency of traditional sexual scripts suggests that if maintaining certain features of marriage, such as sexual frequency, is desired, increased egalitarianism in one area of marriage must be paired with comparable shifts away from traditional gender behaviors, attitudes, and scripts in others. One potential change may be women’s sexual agency. As we noted earlier, Baumeister and colleagues (2001) document substantial differences in sexual interest and activity between men and women, reflecting double standards that penalize girls and young women for sexual activity while often rewarding sexually active young men. To the extent these double standards become internalized, heterosexual women may subjugate their own desires and may not feel as free to initiate sex. One potential interpretation of our results is that husbands’ participation in core housework increases their stress levels and makes them less likely to initiate sex. If wives do not feel empowered to initiate sex, then husbands’ housework and ensuing fatigue would reduce the frequency of intercourse. In this interpretation, it is not necessarily the case that egalitarianism in household labor is incompatible with sexual activity itself, but rather that egalitarianism is incompatible with current sexual scripts. Gendered sexual scripts punish women for being sexually agentic and encourage men to be sexual initiators. If these scripts were to change and both men and women initiated intercourse, then the division of household labor would presumably be less consequential.14

In conclusion, these results shed new light on an area of marriage—sex—that has received relatively little recent attention. More broadly, they expand our understanding of how couples make bargains in households, suggesting that straightforward exchange relations do not govern sexual behavior in marriage. Instead, a more complex, socialized set of beliefs and scripts related to gender link wives’ and husbands’ performances of household labor and sexual frequency, much as gendered scripts govern a wide range of behavior. The importance of gender has declined over time, but it continues to exert a strong influence over individual behaviors, including sexual frequency within marriage.

Acknowledgments

We presented a previous version of this work at the 2009 American Sociological Association conference, and we would like to thank the audience and the presider from that session for their valuable comments. We would also like to thank Pepper Schwartz, Rebecca Sheehan, Jon Agnone, Tyler Corwin, and anonymous reviewers for their helpful comments, although we note that mistakes and conclusions are solely our own.

Biographies

• 

Sabino Kornrich is a Junior Researcher at the Center for Advanced Studies at the Juan March Institute in Madrid. He will be Assistant Professor of Sociology at Emory University beginning 2013. Much of his current research focuses on parents’ monetary investments in children and how these have changed over time. He is also interested in processes that structure relations between income, expenditures, time in housework, and the role of gender for these processes. His earlier research examined discrimination charges in organizations and sources of black-white labor market inequality.

• 

Julie Brines is Associate Professor of Sociology at the University of Washington. Her research examines how the principles that organize work and allocation decisions within families are tied to durable patterns of inequality. Current projects analyze the effects of changes in local labor and housing markets immediately before and during the Great Recession on county-level rates of filing for divorce. She is also studying how trends in women’s and men’s employment have altered the terms of social comparison, the use of power, and perceptions of justice in marriage.

• 

Katrina Leupp is a PhD Candidate in Sociology and an NICHD Trainee at the Center for Studies in Demography and Ecology at the University of Washington. Her interests include family demography, work, and gender. Her dissertation examines the consequences of employment trajectories and family care demands on health and well-being. Other projects analyze patterns of nonstandard employment hours across the life course, and the effects of gender attitudes on mothers’ employment outcomes.

APPENDIX

Sexual Satisfaction and Housework

This appendix examines the relationship between housework and satisfaction with one’s sex life. We present these additional results as a potential way to gain leverage on two questions. First, as we mentioned in the main text, one possible concern is that households with more traditional gender divisions of labor may have higher sexual frequency due to coercive sexual behavior. To the extent this is the case, wives in more traditional households should have lower satisfaction with their sex lives, and men in these households should have greater satisfaction. As Tables A1 and A2 show, however, this is not what we find. Instead, wives are more likely to report greater sexual satisfaction when their husbands report higher shares of housework, and husbands’ sexual satisfaction is unrelated to their wives’ reports of men’s share of housework.

Another possibility is simply that more egalitarian households are likely to engage in a range of behaviors that would not count as sex but that might lead to greater sexual satisfaction. If couples with more egalitarian divisions of household labor are more likely to engage in (unreported) sexual activities that prioritize women’s sexual satisfaction, we would expect to find the opposite relationship between women’s sexual satisfaction and men’s participation in household labor. Instead, we still find the same relationship as for sexual frequency, suggesting this possibility is unlikely.

These multinomial logistic regression analyses separately examine the relationship between housework and sexual satisfaction for husbands and wives using opposite spouse reports of sexual satisfaction and housework. For wives’ results, the dependent variable of satisfaction is reported by wives and housework measures are reported by husbands; for husbands’ results, the dependent variable is reported by husbands and housework measures are reported by wives.

Table A1

Wives’ Sexual Satisfaction and Housework

Table A2

Husbands’ Sexual Satisfaction and Housework

Footnotes

1.Amato and colleagues (2003) also show, however, that men’s increasing share of housework seems to depress their own marital satisfaction.

2.Authors’ calculations from National Survey of Families and Households data are available on request.

3.Although we rely on social exchange theory, similar predictions could be generated using economic or quasi-economic theories of household bargaining, such as separate-spheres bargaining models or a dependency model (Brines 1994; Lundberg and Pollak 1993). Sprecher (1998) also proposes a model based on equity, rather than exchange, that would lead to similar predictions.

4.To list only a few differences, Baumeister and colleagues (2001) found that men desired sex more often, were more frequently aroused, initiated sex more frequently, refused sex less frequently, and had more permissive attitudes. Other evidence on initiation and refusal can be found in Byers and Heinlein (1989).

5.There is certainly debate over these findings. Gupta (2007) argues that a better model is one of women’s autonomy. Recent research continues to investigate when different models work (Killewald and Gough 2010).

6.Note that the gender-typical work described is consistent with the sexual scripts approach we will develop.

7.To the extent that gender works through sexual scripts, it suggests that individuals have internalized gendered scripts. This is in contrast to some theoretical perspectives on gender that suggest gender is performed to meet others’ expectations (e.g., West and Zimmerman 1987).

8.We exclude cohabiting couples because some evidence suggests relationships among cohabitors are different from those among married couples.

9.Other datasets typically contain measures of satisfaction with the division of household labor and one’s sex life, but not measures of sexual frequency or the actual amount of time spent on household labor.

10.South and Spitze excluded respondents with more than four missing items. Because we consider individuals’ reports of their own and their spouse’s activities, we double the number of potential missing items to eight. Including respondents with different numbers of items has little substantive impact on results.

11.Few couples fell into this category. Nine men reported all zeroes for items of core housework for themselves and their wives, one man reported all zeroes for non-core housework, and two women reported all zeroes for their own and their husbands’ non-core housework.

12.We calculated the share using wives’ and husbands’ reports of their own income, because nearly half of respondents had missing data for the question about their spouses’ income.

13.In results from other models, men’s share of core housework always remains significant, and men’s share of non-core housework is significant except in the pooled model and using only men’s reports.

14.We would like to thank an anonymous reviewer for suggesting this possibility.

References

• Amato Paul R, Booth Alan, Johnson David R, Rogers Stacy J. Alone Together: How Marriage in America is Changing. Boston, MA: Harvard University Press; 2007. [Google Scholar]
• Amato Paul R, Johnson David R, Booth Alan, Rogers Stacy J. Continuity and Change in Marital Quality between 1980 and 2000. Journal of Marriage and the Family. 2003;65:1–22. [Google Scholar]
• Amato Paul R, Rivera Fernando. Paternal Involvement and Children’s Behavior Problems. Journal of Marriage and the Family. 1999;61:375–384. [Google Scholar]
• Baumeister Roy F, Catanese Kathleen R, Vohs Kathleen D. Is There a Gender Difference in Strength of Sex Drive? Theoretical Views, Conceptual Distinctions, and a Review of Relevant Evidence. Personality and Psychology Review. 2001;5:242–273. [Google Scholar]
• Baumeister Roy F, Vohs Kathleen D. Sexual Economics: Sex as Female Resource for Social Exchange in Heterosexual Interactions. Personality and Social Psychology Review. 2004;8:339–363. [PubMed] [Google Scholar]
• Becker Gary S. A Treatise on the Family. Cambridge, MA: Harvard University Press; 1981. [Google Scholar]
• Berk Sarah F. The Gender Factory: The Apportionment of Work in American Households. New York: Plenum Press; 1985. [Google Scholar]
• Bianchi Suzanne M, Milkie Melissa A, Sayer Liana C, Robinson John P. Is Anyone Doing the Housework? Trends in the Gender Division of Household Labor. Social Forces. 2000;79:191–228. [Google Scholar]
• Bittman Michael, England Paula, Folbre Nancy, Sayer Liana, Matheson George. When Does Gender Trump Money? Bargaining and Time in Household Work. American Journal of Sociology. 2003;109:186–214. [Google Scholar]
• Blumer Herbert. A Sociologist Looks at the ‘Kinsey Report.’ Ecology. 1948;29:522–524. [Google Scholar]
• Blumstein Philip, Schwartz Pepper. American Couples: Money, Work, Sex. New York: William Morrow Publishers; 1983. [Google Scholar]
• Brewis Alexandra, Meyer Mary. Marital Coitus across the Life Course. Journal of Biosocial Science. 2005;37:499–518. [PubMed] [Google Scholar]
• Brines Julie. Economic Dependency, Gender, and the Division of Labor at Home. American Journal of Sociology. 1994;100:652–688. [Google Scholar]
• Brines Julie, Joyner Kara. The Ties that Bind: Principles of Cohesion in Cohabitation and Marriage. American Sociological Review. 1999;64:333–365. [Google Scholar]
• Byers E. Sandra, Heinlein Larry. Predicting Initiations and Refusals of Sexual Activities in Married and Cohabiting Heterosexual Couples. Journal of Sex Research. 1989;26:210–231. [Google Scholar]
• Call Vaughn, Sprecher Susan, Schwartz Pepper. The Incidence and Frequency of Marital Sex in a National Sample. Journal of Marriage and the Family. 1995;57:639–652. [Google Scholar]
• Carrington Christopher. No Place Like Home: Relationships and Family Life among Lesbians and Gay Men. Chicago: University of Chicago Press; 1999. [Google Scholar]
• Chethik Neil. VoiceMale: What Husbands Really Think about Their Marriages, Their Wives, Sex, Housework, and Commitment. New York: Simon and Schuster; 2006. [Google Scholar]
• Christopher F. Scott, Sprecher Susan. Sexuality in Marriage, Dating, and Other Relationships: A Decade Review. Journal of Marriage and the Family. 2000;62:999–1017. [Google Scholar]
• Clark Alexander L, Wallin Paul. The Accuracy of Husbands’ and Wives’ Reports on the Frequency of Marital Coitus. Population Studies. 1964;18:165–173. [Google Scholar]
• Collins Linda M, Schafer Joseph L, Kam Chi-Ming. A Comparison of Inclusive and Restrictive Strategies in Modern Missing Data Procedures. Psychological Methods. 2001;6:330–351. [PubMed] [Google Scholar]
• Coltrane Scott. Gender and Families. Newbury Park, CA: Pine Forge Press; 1998. [Google Scholar]
• Coltrane Scott. Research on Household Labor: Modeling and Measuring the Social Embeddedness of Routine Family Work. Journal of Marriage and Family. 2000;62:1208–1233. [Google Scholar]
• Cooke Lynn Prince. ‘Doing Gender’ in Context: Household Bargaining and the Risk of Divorce in Germany and the United States. American Journal of Sociology. 2006;112:442–472. [Google Scholar]
• Coontz Stephanie. Marriage, A History: How Love Conquered Marriage. New York: Viking Penguin; 2005. [Google Scholar]
• Cornwell Benjamin, Laumann Edward O. Network Position and Sexual Dysfunction: Implications of Partner Betweenness for Men. American Journal of Sociology. 2011;117:172–208. [PMC free article] [PubMed] [Google Scholar]
• Rosemary Crompton., editor. Restructuring Gender Relations and Employment: The Decline of the Male Breadwinner. Oxford, UK: Oxford University Press; 1999. [Google Scholar]
• DeVault Marjorie L. Feeding the Family: The Social Organization of Caring as Gendered Work. Chicago: University of Chicago Press; 1991. [Google Scholar]
• Elliott Sinikka, Umberson Debra. The Performance of Desire: Gender and Sexual Negotiation in Long-Term Marriages. Journal of Marriage and Family. 2008;70:391–406. [PMC free article] [PubMed] [Google Scholar]
• Ellison Christopher G, Burdette Amy M, Wilcox W. Bradford. The Couple That Prays Together:Race and Ethnicity, Religion, and Relationship Quality among Working-Age Adults. Journal of Marriage and Family. 2010;72:963–975. [Google Scholar]
• England Paula. The Gender Revolution: Uneven and Stalled. Gender & Society. 2010;24:149–166. [Google Scholar]
• England Paula. Missing the Big Picture and Making Much Ado about Almost Nothing: Recent Scholarship on Gender and Household Work. Journal of Family Theory and Review. 2011;3:23–26. [Google Scholar]
• England Paula, Shafer Emily Fitzgibbons, Fogarty Alison CK. Hooking Up and Forming Romantic Relationships on Today’s College Campuses. In: Kimmel M, Aronson A, editors. The Gendered Society Reader. 3rd ed. New York: Oxford University Press; 2008. pp. 531–546. [Google Scholar]
• Fisher Kimberly, Egerton Muriel, Gershuny Jonathan I, Robinson John P. Gender Convergence in the American Heritage Time Use Study (AHTUS) Social Indicators Research. 2006;82:1–33. [PMC free article] [PubMed] [Google Scholar]
• Gager Constance T, Yabiku Scott T. Who Has the Time? The Relationship between Household Labor Time and Sexual Frequency. Journal of Family Issues. 2010;31:135–163. [Google Scholar]
• Gagnon John H, Simon William. Sexual Conduct: The Social Sources of Human Sexuality. Chicago: Aldine Publishing Company; 1973. [Google Scholar]
• Goffman Erving. The Arrangement between the Sexes. Theory and Society. 1977;4:301–331. [Google Scholar]
• Gornick Janet C. Is the Marriage Debate Over? James Q. Wilson and Janet C. Gornick Discuss Whether Liberal and Conservative Takes on Matrimony Can Get Hitched. The American Prospect Online. 2002 Retrieved November 29, 2012 ( http://prospect.org/article/marriage-debate-over)
• Graham John W, Olchowski Allison E, Gilreath Tamika D. How Many Imputations Are Really Needed? Some Practical Clarifications of Multiple Imputation Theory. Prevention Science. 2007;8:206–213. [PubMed] [Google Scholar]
• Greeley Andrew M. Faithful Attraction: Discovering Intimacy, Love and Fidelity in American Marriage. New York: Doherty; 1991. [Google Scholar]
• Greenblat Cathy S. The Salience of Sexuality in the Early Years of Marriage. Journal of Marriage and the Family. 1983;45:289–299. [Google Scholar]
• Greenstein Theodore N. Economic Dependence, Gender, and the Division of Labor in the Home: A Replication and Extension. Journal of Marriage and the Family. 2000;62:322–335. [Google Scholar]
• Gupta Sanjiv. Autonomy, Dependence, or Display? The Relationship between Married Women’s Earnings and Housework. Journal of Marriage and Family. 2007;69:399–417. [Google Scholar]
• Hamilton Laura, Armstrong Elizabeth A. Gendered Sexuality in Young Adulthood: Double Binds and Flawed Options. Gender & Society. 2009;23:589–616. [Google Scholar]
• Hochschild Arlie, Machung Anne. The Second Shift: Working Parents and the Revolution at Home. New York: Viking; 1989. [Google Scholar]
• Homans George C. Social Behavior: Its Elementary Forms. New York: Harcout, Brace, and World; 1961. [Google Scholar]
• Hook Jennifer. Care in Context: Men’s Unpaid Work in 20 Countries, 1965–2003. American Sociological Review. 2006;71:639–660. [Google Scholar]
• Horton Nicholas J, Lipsitz Stuart R, Parzen Michael. A Potential for Bias When Rounding in Multiple Imputation. American Statistician. 2003;57:229–232. [Google Scholar]
• Hyde Janet Shibley, DeLamater John D, Hewitt Erri C. Sexuality and the Dual-Earner Couple: Multiple Roles and Sexual Functioning. Journal of Family Psychology. 1998;12:354–368. [Google Scholar]
• James William H. The Honeymoon Effect on Marital Coitus. Journal of Sex Research. 1981;17:114–123. [Google Scholar]
• Jasso Guillermina. Choosing a Good: Models Based on the Theory of the Distributive-Justice Force. Advances in Group Processes: Theory and Research. 1987;4:67–108. [Google Scholar]
• Killewald Alexandra, Gough Margaret. Money Isn’t Everything: Wives’ Earnings and Housework Time. Social Science Research. 2010;39:987–1003. [PMC free article] [PubMed] [Google Scholar]
• Kinsey Alfred C, Pomeroy Wardell B, Martin Clyde E. Sexual Behavior in the Human Male. Philadelphia: Saunders; 1948. [Google Scholar]
• Lam David. Marriage Markets and Assortative Mating with Household Public Goods. Journal of Human Resources. 1988;22:462–487. [Google Scholar]
• Laumann Edward O, Gagnon John H, Michael Robert T, Michaels Stuart. The Social Organization of Sexuality. Chicago: University of Chicago Press; 1994. [Google Scholar]
• Lundberg Shelly, Pollak Robert A. Separate Spheres Bargaining and the Marriage Market. Journal of Political Economy. 1993;101:988–1010. [Google Scholar]
• Lundberg Shelly, Pollak Robert A. Bargaining and Distribution in Marriage. Journal of Economic Perspectives. 1996;10:139–158. [Google Scholar]
• Marshall Will, Sawhill Isabel V. Moynihan DP, Smeeding TM, Rainwater L. The Future of the Family. New York: Russell Sage Foundation; 2004. Progressive Family Policy in the 21st Century; pp. 198–230. [Google Scholar]
• North Scott. Book Review: Chethik, Neil. (2006). VoiceMale: What Husbands Really Think about Their Marriages, Their Wives, Sex, Housework, and Commitment. New York: Simon and Schuster. Men and Masculinities. 2007;10:121–123. [Google Scholar]
• Padavic Irene, Reskin Barbara. Women and Men at Work. 2nd ed. Thousand Oaks, CA: Sage Publications; 2002. [Google Scholar]
• Parsons Talcott, Bales Robert F. Family Socialization and Interaction Process. Glencoe, IL: Free Press; 1955. [Google Scholar]
• Rao KV, Demaris Alfred. Coital Frequency among Married and Cohabiting Couples in the United States. Journal of Biosocial Science. 1995;27:135–150. [PubMed] [Google Scholar]
• Rich Adrienne. Compulsory Heterosexuality and Lesbian Existence. Signs. 1980;5:631–660. [Google Scholar]
• Risman Barbara. Gender as Structure or Trump Card? Journal of Family Theory and Review. 2011;3:18–22. [Google Scholar]
• Robinson John, Godbey Geoffrey. Time for Life: The Surprising Ways Americans Use Their Time. University Park: Pennsylvania State University Press; 1997. [Google Scholar]
• Schneider Daniel. Gender Deviance and Household Work: The Role of Occupation. American Journal of Sociology. 2012;117:1029–1072. [PubMed] [Google Scholar]
• Schwartz Pepper. Love Between Equals: How Peer Marriage Really Works. New York: The Free Press; 1995. [Google Scholar]
• Schwartz Pepper. What Sexual Scientists Know About … Sexual Satisfaction in Committed Relationships. Allentown, PA: The Society for the Scientific Study of Sexuality; 2007. Retrieved January 1, 2009 ( http://www.sexscience.org/dashboard/articleImages/SSSS-SexualSatisfactionInCommittedRelationships.pdf). [Google Scholar]
• Simon William, Gagnon John H. Sexual Scripts: Permanence and Change. Archives of Sexual Behavior. 1986;15:97–120. [PubMed] [Google Scholar]
• South Scott J, Spitze Glenna. Housework in Marital and Nonmarital Households. American Sociological Review. 1994;59:327–347. [Google Scholar]
• Sprecher Susan. Social Exchange Theories and Sexuality. Journal of Sex Research. 1998;35:32–43. [Google Scholar]
• Steensland Brian, Park Jerry Z, Regnerus Mark D, Robinson Lynn D, Wilcox W. Bradford, Woodberry Robert D. The Measure of American Religion: Toward Improving the State of the Art. Social Forces. 2000;79:291–318. [Google Scholar]
• Stevens Daphne Pedersen, Kiger Gary, Mannon Susan E. Domestic Labor and Marital Satisfaction: How Much or How Satisfied? Marriage and Family Review. 2005;37:49–67. [Google Scholar]
• Storms Michael D, Stivers Margaret L, Lambers Scott W, Hill Craig A. Sexual Scripts for Women. Sex Roles. 1981;7:699–707. [Google Scholar]
• Sullivan Oriel. Changing Gender Relations, Changing Families: Tracing the Pace of Change over Time. Lanham, MD: Rowman and Littlefield; 2006. [Google Scholar]
• Sullivan Oriel. 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 and Review. 2011;3:1–13. [Google Scholar]
• Sullivan Oriel, Coltrane Scott. Men’s Changing Contribution to Housework and Child Care: A Discussion Paper on Changing Family Roles. Chicago: Council on Contemporary Families; 2008. Retrieved November 1, 2008 ( http://www.contemporaryfamilies.org/marriage-partnership-divorce/menchange.html). [Google Scholar]
• Sweet James A, Bumpass Larry L. The National Survey of Families and Households - Waves 1 and 2: Data Description and Documentation. Madison, WI: Center for Demography and Ecology, University of Wisconsin-Madison; 1996. ( http://www.ssc.wisc.edu/nsfh/home.htm). [Google Scholar]
• Udry J. Richard, Chantala Kim. Masculinity-Femininity Guides Sexual Union Formation in Adolescents. Personality and Social Psychology Bulletin. 2004;30:44–55. [PubMed] [Google Scholar]
• Waite Linda J, Joyner Kara. Emotional Satisfaction and Physical Pleasure in Sexual Unions: Time Horizon, Sexual Behavior, and Sexual Exclusivity. Journal of Marriage and Family. 2001;63:247–264. [Google Scholar]
• West Candace, Zimmerman Don. Doing Gender. Gender and Society. 1987;1:125–151. [Google Scholar]
• White Lynn, Keith Bruce. The Effect of Shift Work on the Quality and Stability of Marital Relations. Journal of Marriage and the Family. 1990;52:453–462. [Google Scholar]
• Wilcox W. Bradford. Soft Patriarchs, New Men: How Christianity Shapes Fathers and Husbands. Chicago: University of Chicago Press; 2004. [Google Scholar]
• Wilcox W. Bradford, Nock Steven L. What’s Love Got to Do with It? Equality, Commitment and Women’s Marital Quality. Social Forces. 2006;84:1321–1345. [Google Scholar]
• Yabiku Scott, Gager Constance T. Sexual Frequency and the Stability of Marital and Cohabiting Unions. Journal of Marriage and Family. 2009;7:983–1000. [Google Scholar]
• Yeh Hsiu-Chen, Lorenz Frederick O, Wickrama KAS, Conger Rand D, Elder Glen H., Jr Relationships among Sexual Satisfaction, Marital Quality, and Marital Instability at Midlife. Journal of Family Psychology. 2006;20:339–343. [PubMed] [Google Scholar]