Abstract
We examine how three types of marital commitment—personal, structural, and moral—are associated with perceptions of fairness in chores (N = 1,839 married couples or 3,678 spouses). Using the Actor–Partner Interdependence Model, we find distinct relationships by type of marital commitment and gender. For personal commitment, both actor and partner effects were positively associated with wives’ perceived fairness of chores, with actor effects more influential to perceptions of fairness for wives versus husbands. For structural commitment, actor and partner effects were negatively associated with husbands’ perceived fairness of chores, with actor effects more influential to perceptions of fairness for husbands versus wives. For moral commitment, actor effects were positively associated with husbands’ perceived fairness, without any gender differences found. Given that employed wives continue to do the disproportionate amount of housework, we discuss how differences by gender in marital commitment explain perceptions of fairness in household chores.
Changes in women’s labor force participation invoke an evaluation of equity in marriage and stimulate a debate about traditional roles of husbands and wives in terms of housework. Studies show that the amount of housework alone does not explain fully perceptions of fairness in chores (Braun, Lewin-Epstein, Stier, & Baumgartner, 2008; Lee & Waite, 2010). Quite often, perceptions of fairness in chores are studied using equity or social exchange theory with constructs including gender ideology, alternatives or options, power, and time availability (Braun et al., 2008; Lennon & Rosenfield, 1994; Sanchez & Kane, 1996). For example, gender ideology, or “a set of beliefs about men and women and marital roles” (Hochschild & Machung, 2003, p. 200), could constrain how spouses negotiate a fair division of housework (DeMaris & Longmore, 1996). Yet, relationship factors, such as beliefs about marital relationships and dyadic interdependency that influence perceptions of fairness in marriage, are understudied.
Here, we incorporate tenets from equity and social exchange theories and relevant findings as appropriate. Our guiding theory for the article, however, is commitment theory (Adams & Jones, 1999; Johnson, 1999) given that commitment motivates partners to act in the best interest of the relationship or of their partners (Murray & Holmes, 2011). Commitment is a cognitive variable that dictates individuals’ beliefs and has an ultimate effect of binding an individual to a course of action (Johnson, 1973; Kiesler & Sakumura, 1966). Although research finds associations between commitment and equity in marriage (Crawford, Feng, Fischer, & Diana, 2003; Weigel, Bennett, & Ballard-Reisch, 2006), earlier thinking about commitment often treated commitment as unidimensional (e.g., Rusbult, Coolsen, Kirchner, & Clarke, 2006). Commitment theory and a growing body of research, however, suggest that a multidimensional approach to commitment is needed (Adams & Jones, 1999; Johnson, 1999; Rhoades, Stanley, & Markman, 2010). Adapted from Johnson (1999), marital commitment includes three types: personal (want to commit), structural (have to commit), and moral (ought to commit). Furthermore, because spouses’ perceptions of fairness change according to the actions taken, or not taken, by both members of a couple, we use the actor–partner interdependence model (APIM; Kashy & Kenny, 2000) to guide our analysis. The purpose of our study is to test how personal, structural, and moral commitment from both members of a dyad are associated with perceptions of fairness in chores in a sample of dual-income married couples, and how the relationships vary between genders.
Perceptions of Fairness in Chores
Even though gender-associated responsibilities are more negotiable between men and women in recent decades, women still do more housework than men. In 2009, men spent on average 2 hours per day in housework, whereas women spent on average 2.6 hours per day (U.S. Bureau of Labor Statistics, 2010). Although increases in workforce participation have reduced time spent in housework by women, and men have also increased housework participation, increases in time spent by men on chores only partially offsets the reduction in women’s household labor (Gershuny & Sullivan, 2003). Nevertheless, the relative hours spent in housework between men and women in a dual-income household does not explain fully the perceptions of fairness in chores (Coltrane, 2000; Lee & Waite, 2010).
Framed from equity and social exchange theories, certain constructs are more associated with perceptions of fairness in household labor than others, although findings across studies lack consistency. Alternatives versus dependence on the relationship and economic resources are important in understanding perceptions of fairness in some studies (Ferree, 1987; Greenstein, 2009; Sanchez & Kane, 1996); for example, women with fewer alternatives to marriage and economic resources perceived the division of housework as fair, whereas women with more alternatives perceived the same division as unfair (Lennon & Rosenfield, 1994). In other studies, however, little support is found for alternatives, with greater emphasis instead on constructs such as equity, gender ideology, or both (DeMaris & Longmore, 1996; Greenstein, 1996). Equity is defined as when one’s output and input ratio equals the partner’s output and input ratio (Adams, 1965). For instance, DeMaris and Longmore (1996) found that the greater the husband’s relative contribution to chores, the less likely either partner was to say housework was unfair to the wife. For gender ideology, these researchers reported that more egalitarian husbands saw housework as more unfair to their wives, and more egalitarian wives saw housework as unfair to them.
One conclusion from this literature is that the division of housework is a dyadic phenomenon. Although individuals’ own attitudes could explain perceptions of fairness in chores and gender differences (Sanchez & Kane, 1996), the interactions during the process of sharing household labor also explain perceptions of fairness, especially for women (Mannino & Deutsch, 2007; Thompson, 1991). For example, husbands’ appreciation of their wives’ housework explains the relationship between relative resources and wives’ perceptions of fairness (Blair & Johnson, 1992; Hochschild & Machung, 2003).
This emphasis on the division of housework as dyadic gives rise to the inclusion of both spouses in the study and emphasizes the importance of commitment theory for studying perceptions of fairness in chores. The more committed a person is to the marriage, the more the spouse depends on the partner (Murray & Holmes, 2011). Furthermore, an individual’s behavior reflects the person, the partners, and the couple as a group (Kenny, 1996). That is, observations from a pair of spouses are correlated and should be examined simultaneously. Thus, we use the APIM to capitalize on the dyadic nature of the division of housework, and to understand how perceptions of fairness of chores are associated with individuals’ commitment (actor effects) and their partners’ commitment (partner effects).
Marital Commitment and Its Relationship With Perceptions of Fairness
We study dyadic interdependence using commitment theory because committed individuals are motivated “to justify the sunk costs that come from dependence on the relationship” (Murray & Holmes, 2011, p.104). Although perceived inequity can be a cost of staying in a relationship and then evokes relational doubt, commitment has the effect of counterweighting the cost because commitment is driven by personal dedication and consistency in action (Johnson, 1973). For example, a husband may perform more than his fair share of housework and perceive it is fair because of his dedication to his wife’s welfare or to the stability of the marriage. Specifically, commitment can be conceptualized as three types: personal, structural, and moral commitment.
Personal commitment, defined as the “want to” aspect of commitment, arises because of attraction to the relational partner, the relationship itself, and/or the couple identity (Johnson, 1999). Studies examining personal commitment (typically assessed as global commitment that taps personal commitment; Johnson, 1999) conclude that greater relationship commitment is positively associated with perceptions of equity and negatively related to perceptions of inequity (Crawford et al., 2003; Weigel et al., 2006). One explanation for this positive relationship between relationship commitment and equity is that couples with higher personal commitment have less difficulty in problem solving (Swensen & Trahaug, 1985) and are active in maintaining the quality of the relationship (Ogolsky, 2009). As such, individuals who have greater personal commitment or being with a partner whose personal commitment is high should be associated with a higher likelihood of perceiving the division of chores as fair.
Structural commitment, defined as the “have to” aspect of commitment, is conceptualized such that one feels that he or she must stay in the relationship because of external constraints. These include unfavorable alternatives (e.g., disadvantageous consequences after divorce), irretrievable investments in the marriage (e.g., time and resources), and social pressure (e.g., from in-laws; Adams & Jones, 1999). For example, wives with more resources and alternatives outside of marriage (i.e., lower structural commitment) may feel the division as unfair to them because they feel less appreciated for the housework that they do (Lee & Waite, 2010). For individuals with greater structural commitment, maintaining a relationship makes sense as long as there are no attractive alternatives (Michaels, Acock, & Edwards, 1986). Thus, individuals who stay married because of high structural constraints or are married to a partner whose structural commitment is high should be more likely to report the division as fair because there is no better alternative (Lennon & Rosenfield, 1994).
Moral commitment, defined as the “ought to” aspect of commitment, comprises obligations and social rules, such as commitments to institutional systems including marriage. One may stay married because of internal constraints, such as concerns about divorce, children, and morals. Moral commitment is related to religious attitudes and church participation (Adams & Jones, 1999; Johnson, Caughlin, & Huston, 1999). For moral commitment, spouses who perceived marriage as spiritual, and had greater joint religious activities, have less marital conflict and greater collaboration (Mahoney et al., 1999). Greater perceived constraint because of concerns about consequences after breakup or reactions from one’s social network is associated with higher relationship adjustment (Rhoades et al., 2010). These findings imply that individuals with greater moral commitment should be more likely to accept the current division as it is and to consider it fair. Furthermore, marrying a partner whose moral commitment is high should also be associated with a higher likelihood of responding to the division as fair given that the partner may share a similar idea of the division, or emphasize the benefits for the marital union more than for the individual.
Associations among personal, structural, and moral commitment show minimal correlations with one another (Johnson et al., 1999), supporting the distinctiveness of each type. Studies showed variability in the direction and the strength of the relationships between each of the three types commitment and spouses’ contributions to housework (Tang, 2012) or relationship maintenance behaviors (Ramirez, 2008; Rhoades et al., 2010). Given that the literature reviewed above all suggests a positive relationship between each type of marital commitment and equity, the distinctiveness of the three types of commitment should be reflected in the strength of the relationship. That is, all three types of commitment should be related to equity in the same direction while the strength of these relationships is expected to vary. Yet, information concerning how each type of commitment is related to marital processes in the presence of other types of commitment is rare, which gives rise to the current study.
Specific to the dyadic nature of housework and partner effects, this is an area of sparse research. Husbands’ gratitude and gender ideology account for both the division of housework and perceptions of fairness (Coltrane, 2000; Hochschild & Machung, 2003; Sanchez & Kane, 1996), illustrating the importance of examining both actor and partner effects. Just as Thompson (1991) discussed, some wives experience husbands as responsive in family work even if the husbands did not do housework. Instead, these husbands were responsive in other ways, including showing appreciation for the family work done by their wives, or supporting the wives’ wage work. In a similar fashion, the examination of wives’ commitment to the relationship, as well as the husbands’ commitment to the relationship, could help researchers understand perceptions of fairness in chores from an interdependent perspective.
Among the few studies of commitment in which the authors investigated spousal interdependence, results of actor and partner effects of marital commitment differ. Ramirez (2008) found similar patterns of actor and partner personal or moral commitment in associations with maintenance behavior, whereas Tang (2012) found differences in actor and partner personal or moral commitment as associated with time spent in routine housework. Differences between these two studies also existed in the patterns between actor and partner structural commitment in relation to maintenance behavior. Other studies examining the three types of commitment and their association with perceptions of fairness in chores have not been done, to our knowledge, emphasizing the need to study such associations.
Here, we focus on examining the relationship between perceptions of fairness in chores and the three types of marital commitment for both actor and partner effects. For both actor and partner effects of each type of commitment (personal, structural, and moral), we hypothesize positive associations with perceiving the division in household labor as fair to both.
Another limitation of the aforementioned studies is that the findings may not be applicable to both genders because they typically include only one gender. That is, in studies of different types of marital commitment (e.g., Ramirez, 2008; Rhoades et al., 2010), the authors did not investigate gender effects. In studies in which authors have included both husbands and wives, gender differences did not emerge (Tang, 2012; Johnson et al., 1999). That being said, gender differences remain understudied, especially as specific to our study variables. Thus, by including both genders, we can make comparisons regarding the relationship between the three types of marital commitment and perceptions of fairness between genders with the following research question:
Research Question: Is there a significant gender difference in the relationship between each type of marital commitment and perceptions of fairness in chores?
This Research Question is aimed at examining significant group difference between men and women on the commitment variables under study in relation to perceptions of chores. In contrast, for the hypotheses, we test significant patterns in the relationship between the commitment variables and perceptions of fairness in chores within each gender group.
Given influences from both equity and social exchange theory, and along with previous literature, we controlled for variables that influence marital commitment and/or the perceptions of fairness in chores, including gender ideology, education, age, tendency to keep opinions to self, and paid work hours reported by both spouses (Coltrane, 2000; Kluwer, Heesink, & Van de Vliert, 2002). Because housework is frequently perceived as tedious and time-consuming, stress generated from performing housework may evoke a sense of unfairness, especially when spouses are dual earners. Thus, we control for the ratio of hours spent performing low-schedule-control (LSC) housework, which is associated with psychological distress for both men and women (Barnett & Shen, 1997). We also control for the overall housework ratio because the relative amount of housework between spouses partly accounts for perceptions of fairness in chores (Coltrane, 2000; Lee & Waite, 2010). Furthermore, we control for couple characteristics including relative economic independence, race, marital length, and number of children younger than the age of 13 years (e.g., Barnett & Shen, 1997; Braun et al., 2008; Coltrane, 2000). Finally, we included six gender and commitment interaction terms (the three types of commitment from both members of a dyad) to identify significant gender differences in the relationship between perceptions of fairness and marital commitment.
Method
We use data from the National Survey of Families and Households (NSFH; Sweet & Bumpass, 1996). The NSFH is one of the most commonly used data sets to study the division of housework. Participants were interviewed and asked to recall the amount of time they spent on housework in a general week. This type of method (i.e., retrospective survey) is highly correlated with division of household data collected through diary methods (Press & Townsley, 1998). When questions concerning household tasks are more narrowly defined, as they are in the NSFH survey, estimates are more accurate (Shelton & John, 1996). Thus, the hours of housework from the NSFH are considered informative and valid.
Data for this study are from NSFH2 and 10,005 primary respondents (response rate 77%) from NSFH1 were reinterviewed successfully in 1992 to 1994 through both personal interviews and self-administered questionnaires (Sweet & Bumpass, 1996). The NSFH1 includes 9,637 households plus an oversampling of Blacks, Puerto Ricans, Mexican Americans, single-parent families, families with step-children, cohabiting couples, and recently married individuals. Only 5,624 spouses from NSFH1 were successfully interviewed again as current spouses in NSFH2. The design and content of the NSFH are described in greater detail by Sweet, Bumpass, and Call (1988). We did not include data from Wave 1 because of a mismatch between Waves 1 and 2 concerning items that were used to construct marital commitment. Furthermore, we did not include data from Wave 3 because NSFH3 excluded respondents younger than 45 years old with no children. Here, we focus on married couples. Next, we focus on dual-earner couples because contributions to housework can differ if one spouse, typically the wife, is not employed, versus if both spouses are employed outside the home. Finally, we exclude responses specific to division as unfair to husbands, as these responses were very limited in number (6% husbands; 5% wives).
The final sample size was 1,839 paired spouses who (a) had been married since NSFH1 (excluding 1,573couples; 28% of 5,624 couples), (b) both worked for pay (excluding 623 couples with no income and 1,389 single-income couples; about 50% of 4,051 couples), and (c) reported the division as fair to both or unfair to the wife (excluding 200 couples in which either spouse responded the division as unfair to the husband; about 10% of 2,039 couples). The majority of respondents reported ethnicity as non-Hispanic White (78.2%) and 818 couples (44.5%) had no children younger than 13 years of age.
Measures
Perceptions of fairness in chores, the dependent variable, is a dichotomous variable. We coded both husbands’ and wives’ responses so that 0 = unfair to the wife and 1 = fair to both. In accordance with studies reviewed previously, most respondents in the current study reported the division as fair to both. See Table 1 for the descriptive statistics of perceptions of fairness, as well as the other variables in the current study.
Descriptive Statistics of Studied Variables (N = 1,839 Couples).
Note. LSC = low-schedule-control. Reliability is reported based on Cronbach’s α.
0 = Unfair to wife; 1 = Fair to both.
0 = Less than college education; 1 = At least some college education.
For commitment, we performed a principal component factor analysis using Promax rotation with Kaiser Normalization on the 14 items. Results showed three factors: personal, structural, and moral commitment. Items on each commitment factor had factor loadings greater than .60. Together, the three factors accounted for 56% of the total variance in commitment.
Personal commitment included four items. One question was, “What do you think the chances are that you and your husband/wife will eventually separate or divorce” (1 = very low to 5 = very high; reverse coded). For the other three questions, respondents were asked to rate their perceptions of happiness in the marriage, including how happy or unhappy respondents were with “the understanding you receive from your spouse,” “the love and affection you get from your spouse,” and “your sexual relationship” (1 = very unhappy to 7 = very happy).
Structural commitment consisted of six items regarding quality of life in the standard of living, job opportunities, economic security, friendships, sex life, and leisure time. Respondents were asked, if they and their spouse were to separate, what changes in quality of life would they expect (1 = much worse to 5 = much better). Responses for all six items were reverse coded.
Moral commitment included four questions. The three questions were “Marriage is a lifetime relationship and should never be ended except under extreme circumstances,” “The bible is God’s word and everything happened or will happen exactly as it says,” and “It is all right for a couple with an unhappy marriage to get a divorce when their youngest child is under age 5” (1 = strongly agree to 5 = strongly disagree). We reverse coded the first two items to match the direction of the third item and the next item. The final item for this scale consisted of the frequency of participating in church-affiliated groups (responses were 0 = never, 1 = once a year or less, 2 = about once a month, 3 = about once a week, and 4 = several times a week).
Reliabilities of the commitment measures are in Table 1. Although a measure with an alpha greater than .70 is preferred (Nunnally & Bernstein, 1994), a reliability value lower than .70 is common in survey studies and when measuring a multidimensional psychological construct (Kline, 1999). Also, the lower internal consistency of moral commitment may result from different aspects of moral attitudes (i.e., moral obligation about being a spouse or parent vs. religiosity). Responses of the 14 variables were standardized (M = 0 and SD = 1) before being used to compute the average score for personal, structural, and moral commitment. Each type of commitment was centered using the group mean of the entire sample. For each type of commitment, a positive score indicates greater commitment than the average score whereas a negative score indicates lower commitment than the average score.
Control variables
For gender ideology, individuals were asked four questions about individuals’ attitudes toward men’s and women’s family roles and responsibilities. Questions included “It is much better for everyone if the man earns the main living and the woman takes care of the home and family,” “It is all right for children under three years old to be cared for all day in a day care center (reversed),” “It is all right for mothers who work full-time when their youngest child is under age 5 (reversed),” and “Preschool children are likely to suffer if their mother is employed.” Responses ranged from 1 (strongly agree) to 5 (strongly disagree). We computed a mean across the four items, with lower scores indicating more traditional beliefs about gender behaviors and higher score indicating more egalitarian beliefs.
Paid work hours were defined as total work hours for all jobs in the last week, or during a typical workweek if the respondent was absent during the previous workweek. The tendency to keep opinions to self during conflict (keeping opinions) was measured based on one item that asked individuals’ inclination to keep their opinions to themselves during conflict. Responses to this question were measured on a 5-point scale (1 = never, 5 = always). Economic independence was assessed based on a formula frequently used by researchers (e.g., Greenstein, 2000): Economic independency = (self’s earnings − partner’s earnings)/(self’s earnings + partner’s earnings). The score for this measure ranged from −1 (i.e., one is completely dependent on spouse’s economic support) to +1 (i.e., one is completely independent).
LSC housework ratio is the ratio of one’s time spent on LSC housework versus perceived spouse’s time on LSC housework. LSC housework includes preparing meals, washing dishes, cleaning house, washing/ironing, and shopping (α = .72 for husbands and .80 for wives for own perceptions, and α = .89 for husbands and .73 for wives for perceptions of spouses’ time). Higher LSC housework ratio means that individuals consider themselves to have spent more time on LSC housework than the spouse. Similarly, overall housework ratio is the ratio of one’s time spent on all housework versus perceived spouse’s time spent on all housework. Higher overall housework ratio indicates perceptions of more time spent on housework compared to the spouse.
Binary codes included gender (0 = woman; 1 = man), educational status (1 = at least some college education; 0 = less than college education), and ethnicity (1 = both members reported as non-Hispanic White; 0 = all others). Age and marriage length were coded on number of years. Children younger than13 years and living in the household were recoded as number of children. Because children 13 years or older do more housework than children younger than 13 years of age (Gager, Sanchez, & Demaris, 2009), we used 13 years old as a cutoff age for this variable.
Analytical Approach
We used pooled data in which each married couple is a cluster, and in which each cluster contains two genders. Because the dependent variable was dichotomized, we used logistic regression. Such an analysis has been used in other studies of perceptions of fairness in chores (e.g., Braun et al., 2008; DeMaris & Longmore, 1996). We also accounted for dyadic interdependence using clustered robust standard error method, which we explain below.
We note here that two problems arise when examining binary outcomes using dyadic data. First, the data violate the independence assumption in that significant correlations emerged between spouses. We found a few significant between-gender correlations (all ps < .001), including fairness in chores (r = .26), personal, structural, and moral commitment (r = .48, .18, and .58, respectively), and gender ideology (r = .43). Also, we calculated the pairwise intraclass correlation, a Pearson-type of correlation to estimate the intraclass correlation for binary outcomes (Gonzalez & Griffin, 2000; Ridout, Demetrio, & Firth, 1999). Results indicated significant interdependence in perceptions of fairness in chores within couples (pairwise intraclass correlation = 0.256, z = 10.965, p < .001). Second, results of the dyadic data analysis can be problematic if using hierarchical linear models. According to Raudenbush (2008) problems of statistical precision occur when analyzing binary responses from “many small groups.”
We implemented two strategies to take into account spousal interdependence and to analyze binary responses from many small groups. First, in line with the APIM (Kashy & Kenny, 2000), we included measures from both actors and partners to estimate perceptions of fairness of a spouse. For example, a wife’s perception of fairness is not only estimated by her own three types of commitment and personal characteristics (e.g., gender ideology, age, and education) but also by her husbands’ three types of commitment and personal characteristics.
The second strategy was to apply a logistic regression model and treat each couple as a cluster unit (see UCLA: Academic Technology Services, Statistical Consulting Group, n.d.). By treating a couple as a cluster, we allowed the standard errors to be correlated within a couple but independent across couples (see vce_option in StataCorp, 2007). By including both husbands and wives in the same model, we were able to compare gender differences regarding the relationship between marital commitment and perceptions of fairness. Furthermore, the robustness of standard errors was preserved when we considered each couple as an interdependent cluster.
Within each gender, explanatory variables present low correlations with each other except for the pair of age and marital length (r = .76 for husbands; .81 for wives, ps < .001). Other moderate within-gender correlations (all ps < .001) include the pair of age and number of children (r = −.59 for husbands; r = −.60 for wives), the pair of LSC ratio and overall housework ratio (r = .55 for husbands; r = .58 for wives), and the pair of marital length and number of children (r = −.51 for husbands and wives). Minor within-gender correlations were also observed between personal and structural commitment (r = .33 for husbands and wives) and between moral commitment and gender ideology (r = −.38 for husbands and wives).
Explanatory variables excluding race, gender, age, and number of children contained missing data. Most variables were missing in less than 2% of the sample, with the largest percentage missing for LSC housework ratio (10%). For missing values in each factor, a multiple regression imputation procedure was used to create 20 imputed data sets. Statistical analyses were conducted separately on each of the 20 data sets, and results were combined to produce a single set of pooled estimates (Johnson, & Young, 2011; StataCorp, 2011). Rubin’s combination rules (Rubin, 1987) were used to form the pooled estimates and to adjust the estimated standard errors.
Gender and commitment variables were entered in the model as a first step. Control variables including individual and marital characteristics were entered in the model along with the interaction terms in the last step. Besides commitment variables, other variables obtained from both spouses include the individual characteristics of gender ideology, work hours, the tendency to keep opinions to self, education, and age.
Results
Personal Commitment
As expected, higher actor or partner personal commitment was associated with higher odds of responding to the division of chores as fair both. However, the association was significant for wives but not for husbands. For actor personal commitment, wives with one point above the average score were 91% more likely to respond to the division as fair to both (see odds ratio in Table 2). For partner personal commitment, living with a husband whose personal commitment was one point above the average, the odds that the wife responded the division being fair to both were 18% higher.
An Actor–Partner Interaction Model of the Relationship Between Perceived Fairness and Personal, Structural, and Moral Commitment for Each Gender and Difference Between Genders.
Note. OR = odds ratio; LSC = low-schedule-control.
p < .001. **p < .01. *p < .05.
Structural Commitment
Contrary to our hypothesis, we found negative associations between structural commitment and perceptions of fairness. For husbands only, higher actor or partner structural commitment was negatively associated with perceiving the division in household labor being fair to both. For actor structural commitment, husbands with one point higher than the average score were 22% less likely to respond the division being fair to both. For partner structural commitment, husbands were 16% less likely to respond the division as fair to both if marrying to a wife whose structural commitment was one point higher than the average.
Moral Commitment
As expected, perceptions of fairness are positively associated with higher score of moral commitment reported by actors and partners. However, only the relationship between actor moral commitment and husbands’ perceptions of fairness reached statistical significance. Husbands with one point more than the average score of moral commitment were 21% more likely than husbands with average moral commitment to respond the division as fair to both.
Gender Difference
Overall, results from the logistic regressions indicated that gender was a significant moderator mostly for the relationship between one’s own perceptions of fairness and actor personal or structural commitment (Research Question). First, we found a significant gender difference in the relationship between actor personal commitment and perceptions of fairness in chores (see Table 2). For both men and women, actors’ personal commitment was positively associated with the odds of responding to the division as fair to both. However, this positive relationship is only significant for men but not for women. The figure for this significant gender difference shows that the increasing rate was sharper for wives than for husbands (see Figure 1A; the effects of covariates on logistic regression estimates were considered using guidelines from Mitchell & Chen, 2005).
(A) Logits of responding to the division as unfair to the wife versus fair to both by actor personal commitment for each gender. (B) Logits of responding to the division as unfair to the wife versus fair to both by actor structural commitment for each gender. (C)* Logits of responding to the division as unfair to the wife versus fair to both by actor moral commitment for each gender
Second, gender moderated the relationship between actor structural commitment and perceptions of fairness in chores (see Table 2). As discussed earlier, structural commitment as an actor variable was associated with husbands’ perception of fairness but not for wives, and the declining rate was sharp for husbands’ (see Figure 1B).
We found no significant gender difference between perceptions of fairness and actor moral commitment (see Figure 1C). Nor was there a significant gender difference in the relationship between perceptions of fairness and partners’ personal, structural, or moral commitment.
Control Variables
We found actors’ variables, including gender ideology, education, and age, to be significantly related to actor’s perceptions of fairness in chores. In contrast, none of the control variable reported by the spouse was significantly associated with actor’s perceptions of fairness in chores. Individuals who indicated higher values of the LSC housework ratio, or who reported as non-Hispanic White couple, were less likely to report the division as fair to both.
Discussion
Our goal was to identify how marital commitment of wives and husbands could be used to estimate perceptions of fairness in chores. We examine how personal, structural, and moral commitment was associated with perceiving the division of housework as either fair to both spouses or unfair to the wife. Given that marital commitment could promote actions toward relationship maintenance, such as performing housework (Tang, 2012), this study fills an important gap by adding to the literature as to how marital commitment by both members of a dyad is associated with perceptions of fairness for both actors and partners, and for each gender.
Greater Actor and Partner Personal Commitment, Greater Fairness in Chores for Wives
For wives, we find that they are more likely to report the division of chores as fair to both if their own or their husbands’ personal commitment are higher. That is, the quality of the marital relationship and the affection experienced in the relationship (personal commitment) matter more than moral guidelines or structural constraints for women’s perceptions of fairness in chores, the latter two of which are not significant. Perhaps wives understand that by committing to the marriage, they also are committing to the embedded inequity in the division of household labor. Their willingness to take on extra work and to respond to the needs of the family represent symbolic meanings of love and rewards in the relationship (Hochschild & Machung, 2003). Because of the interdependent nature of the marriage, positive action taken by the husbands, such as expressing affection to their wives, could balance out the injustice in the division of household labor for women (Bolak, 1997; Hochschild & Machung, 2003; Thompson, 1991). In contrast, husbands’ lack of expressing understanding and affection to their wives are associated with greater perceived injustice in the division of household labor for women (Wilcox & Nock, 2006). By simultaneously examining all three types of commitment from both spouses, our study demonstrates the importance of personal commitment from both actors and partners in relation to women’s sense of fairness in chores.
Greater Actor and Partner Structural Commitment, Lower Fairness in Chores for Husbands
For husbands, they are less likely to consider the current division of housework as fair to both if they or their wives are concerned with a decline in various aspects of quality of life due to divorce (structural commitment). This result contradicts the tenets of equity and social exchange theories, with these theories suggesting that the perceptions of fairness increase along with a decline of alternatives outside of the relationship. This finding was also opposite to what we had hypothesized using commitment theory.
Instead, this finding may be better explained using role and interdependence perspectives. Husbands traditionally are expected to contribute more hours to paid work and to earn more than their wives do (Oakley, 1974). Men’s higher economic power compared with women’s also gives men the “entitlement” of doing less housework than women (Blair & Lichter, 1991; Greenstein, 1996). Because of better social and material resources, husbands on average have lower structural commitment than wives do, as reflected in the current study. Consequently, for men the leverage within the marriage is associated with downplaying, or ignoring, the significance of inequality in relationship (Sanchez & Kane, 1996). In contrast to the “average husbands,” husbands with higher structural commitment may consider themselves less fulfilling or more dependent on their marriage. Receiving more marital benefits than their wives might make husbands feel overbenefitted from the marriage, especially when perceiving their wives contribute to paid and unpaid work equally or more than they do (Dempsey, 2002). Kluwer et al. (2002) showed that feeling relationally overbenefitted can decrease perceptions of fairness, explaining the negative relationship between husbands’ structural commitment and their perceived fairness in chores.
We also find a negative relationship between partner structural commitment and husbands’ perceived fairness. This finding is not easily explained. Below, we offer some explanations, but do so tentatively. We issue this caveat as our explanations are speculative, and as there is no evidence for these processes in the data.
We speculate that the negative relationship between husbands’ perceived fairness in chores and wives’ structural commitment reflects not only the degree of wives’ bargaining power in negotiating their fair share but also husbands’ confidence in the wife’s marital commitment. That is, husbands may be more likely to respond to the division as fair if their wives’ structural commitment is lower and vice versa. Because wives with lower structural commitment experience little appreciation for the work they do around the house (Lee & Waite, 2010), they may be inclined to improve fairness in chores by negotiating a better division with their husbands. However, wives’ success in efforts toward household equity should be greater in a highly committed marriage in which spouses are confident in placing their own welfare at each other’s hands and are responsible to each other’s needs (Murray & Holmes, 2011). However, when husbands doubt the wives’ intention to stay in the marriage, they may withdraw from wives’ demands of equity. Marital risk may then ensue given wives’ low structural commitment and husbands’ low support of wives’ needs. In contrast, wives with higher structural commitment might accept the disproportionate division given few alternatives (Lennon & Rosenfield, 1994) or to avoid a worse consequence (e.g., divorce).
Greater Actor Moral Commitment, Greater Perceptions of Fairness for Husbands
This study also reveals the importance of understanding husbands’ perceptions of fairness in chores from the perspective of actor moral commitment, an internal constraint defined by religious activity and beliefs about marriage. For husbands in the current study, greater actor moral commitment is associated with a higher likelihood of responding to the division as fair to both spouses. Here, husbands with higher moral commitment may be more likely to have traditional gender attitudes. In a post hoc analysis (not shown here), we found that husbands’ moral commitment was not associated with their own perceptions of fairness unless we control for gender ideology from spouses. That is, net of spouses’ gender ideology, husbands with high moral commitment in the current study are more likely to protect their marriage and respond to their spouse’s needs because they emphasize the long-term perspective of marriage. Other studies (DeMaris, Mahoney, & Pargament, 2010) also found that the sanctification of marriage (i.e., the spiritual meaning of marriage) and religious practice can buffer perceptions of inequality. Nevertheless, it could also mean that husbands with higher moral commitment are less sensitive to individual experience of inequity in the relationship and more concerned about the communal goal of marriage. Husbands and wives in the current study were similar in their moral commitment and in gender attitudes. It is likely that they both emphasize communal goal—family functioning, for example, and have fewer concerns about if housework is shared equally. Consequently, husbands with higher moral commitment may not be aware of the inequality because they receive few complaints from their wives (e.g., Ellison & Bartkowski, 2002). These explanations, like the ones above for structural commitment, are also speculative. Our aim in offering these explanations is that future researchers will be able to more directly assess these study constructs and processes.
Adding to the existing literature, this study shows the distinct patterns between each type of marital commitment and perceptions of fairness in chores for each gender. Studies have pointed out that partner’s appreciation and responsiveness matter to women’s perceptions of fairness in chores whereas the emotional dependence (i.e., degree of dependence on the marriage) accounts for men’s perceptions of fairness (Hochschild & Machung, 2003; Sanchez & Kane, 1996). Unique from previous research, the current study emphasizes the importance of interdependence in understanding both genders’ perceptions of fairness in chores. Here, women’s perceptions of fairness are linked to both the affection given and received in the marriage. Similarly, men’s perceptions of fairness are linked to the alternatives evaluated by both themselves and their spouse. It is both their own and their partner’s evaluation of union continuity, rather than his or her commitment to the marriage, that accounts for the perceptions of fairness in chores.
Limitations, Strengths, and Suggestions for Future Research
Several limitations may constrain the interpretation and application of our study. Most important, we acknowledge the NSFH data we use here are almost two decades old. Like others who have used the NSFH data recently (e.g., Lee & Waite, 2010), we realize that issues pertaining to our study variables are older than we would prefer. However, we note that few other national studies collect data on perceptions of housework on both spouses. At the same time, changes in gender ideology do not seem to improve fairness in division of housework given that patterns of housework in recent years continue to show that women spend more hours per day on housework than do men (e.g., U.S. Bureau of Labor Statistics, 2010). Given these continued patterns over time, we believe that using NSFH data is still valuable and that the findings are relevant.
Results from the current study were also limited by the nature of the cross-sectional data. Because marital commitment and perceptions of fairness were measured simultaneously, patterns from the current study indicate only the degree of correlation rather than causality from marital commitment to perceptions of fairness. It is quite possible that perceptions of fairness affect the different kinds of commitment. To address this limitation, future studies could consider using longitudinal data to examine the direction of the relationship between marital commitment and perceptions of fairness. For example, because marital commitment can be influenced by life events (Kapinus & Johnson, 2003), examining changes in marital commitment and perceptions of fairness before and after such a life event can help us to understand more about the direction and strength of the relationships between marital commitment and perceptions of fairness. Furthermore, the study was specific to housework. Future research may consider examining perceptions of fairness in other aspects (e.g., childrearing or emotional work).
We also note that the reliability for moral commitment is lower than desired, which could limit the interpretation of the study findings. However, a low reliability value is also common in survey studies and when measuring a multidimensional psychological construct (Kline, 1999). That is, the low internal consistency of moral commitment may result from different aspects of moral attitudes (i.e., religiosity versus moral obligations of a spouse and parent). Future assessment of items that better conceptualize moral commitment in terms of higher internal consistency are still needed.
Several strengths are evident. We used a national data set that provided a large sample size. Furthermore, we examined dyadic data, both actor and partner effects, and influences of gender in associations of marital commitment and perceptions of fairness in chores. Typically studies of perceptions of chores rely on the reports of one spouse, mostly the wives, or only test actor effects, without consideration of partner effects. Moreover, we moved beyond commonly assessed variables associated with perceptions of fairness in chores and identified how different types of marital commitment explain perceptions of fairness in chores between spouses.
In sum, previous research indicates that employed wives continue to do the disproportionate amount of housework and that conflicts arise from negotiating the division of housework, which can be detrimental to the marriage. Based on our results from the current study, we argue that our findings offer an important key in understanding perceptions of fairness in chores as linked with different types of marital commitment. Furthermore, the inclusion of actor and partner effects in our study, and the investigation of gender differences, reveals that men may judge equity based on what they and their wives are “supposed” to get and give in a marriage or their beliefs about marriage. In contrast, women may judge equity based on marital quality and being happy in their relationships with their husbands.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.