The Gendered Division of Sterilization “Fertility Work:” The Role of Educational and Racial/Ethnic Heterogamy

Abstract

Drawing on data from the 2006 to 2010 and 2011 to 2013 rounds of the National Survey of Family Growth, this study examines the division of sterilization “fertility work” among couples who have completed intended childbearing. Results show that use of female versus male sterilization varies by couples’ racial/ethnic composition, but not by couples’ educational composition. Couples in which the man is non-Hispanic White and the woman is of minority background were found to have an increased likelihood of relying on male versus female sterilization. This finding supports the gender perspective, as it shows the male partner to be most willing to participate in sterilization “fertility work” when he is a member of the dominant racial/ethnic group and his partner is of minority background.

Keywords

education family demography fertility gender and family household labor quantitative race/ethnicity

Research on fertility and contraceptive use has traditionally focused on women (Almeling, 2015; Grady, Klepinger, Billy, & Cubbins, 2010; Thomson, 1997). Women are the ones who experience the physical processes of pregnancy, childbirth, and breastfeeding, and most contraceptive methods are female-controlled. Sterilization offers an interesting exception; both the female and male partner can get sterilized and, importantly, tubal ligation and vasectomy are about equally effective in preventing pregnancy (Grimes, 2009). Yet, even with regard to this method of contraception, the female partner is much more likely to do the work; 33% of married or cohabiting female contraceptive users rely on female sterilization for fertility control, as compared with 16% who rely on male sterilization (Eeckhaut & Sweeney, 2016). Such patterns have led researchers to argue that the work associated with navigating a couple’s fertility can be considered another form of household work that “often conforms to a gendered division of labor, with women primarily in charge” (Fennell, 2011, p. 496; see also, e.g., Bertotti, 2013; Dereuddre, Buffel, & Backe, 2017) and to coin this type of work “fertility work” (e.g., Bertotti, 2013, p. 13).

Clear parallels can be drawn between sterilization “fertility work” and other forms of household labor; most important, the benefits of sterilization—highly effective, cost-effective, long-acting protecting against pregnancy—are shared by both partners, but the physical and emotional labor is largely shouldered by the partner undergoing the procedure (Bertotti, 2013; Sandlow, Westefeld, Maples, & Scheel, 2001). In addition, the extent to which women are in charge of sterilization “fertility work” is not uniform across sociodemographic groups. Paralleling household labor, research has found persistent and gendered differentials in the use of sterilization by education and race/ethnicity. Higher levels of education tend to be associated with a more equal division of household labor (Bianchi, Milkie, Sayer, & Robinson, 2000). Similarly, the gap in reliance on female versus male sterilization is largest among married and cohabiting female contraceptive users with less than high school education—54% to 7%, respectively—and smallest (and even reversed) among married and cohabiting female contraceptive users with a bachelor’s degree—15% to 20%, respectively (Eeckhaut & Sweeney, 2016). Black couples tend to have a more and Hispanic couples tend to have a less equitable division of household labor, when compared with White couples (Bolzendahl & Gubernskaya, 2016; Wight, Bianchi, & Hunt, 2012). Conversely, the gap between female versus male sterilization is smallest among Whites, when compared with Hispanics and Blacks (Chandra, 1998; Daniels, Daugherty, Jones, & Mosher, 2015).

Despite the clear parallels, however, the sterilization literature has yet to incorporate the rich theoretical insights of the literature on the division of household labor (exceptions are as follows: Bertotti, 2013; Bumpass, Thomson, & Godecker, 2000; Forste, Tanfer, & Tedrow, 1995). Of particular relevance is the latter field’s long tradition of recognizing the importance of couples’ relative characteristics in shaping the gendered division of labor. Since the mid-20th century, theories such as resource theory have emphasized the role of comparative advantages between partners in the labor market (Blood & Wolfe, 1960). To the extent that characteristics such as education and race/ethnicity are connected to higher relative resources, differences in these characteristics are believed to shape the division of labor within couples. This is because such differences would enable the partner with the most resources to negotiate his or her way out of less attractive activities, such as household work. Feminist critiques of these theories have introduced opposing predictions regarding the effect of partners’ relative characteristics. In gender identity theory, women’s dominant responsibility for unpaid housework is explained by gender norms being socialized from an early age and then being displayed in daily life and continuously redefined in the process of “doing gender” (West & Fenstermaker, 1995; West & Zimmerman, 1987). Overall, men would engage in masculine activities, and women would engage in feminine activities to strengthen their gender identity. Individuals who cross boundaries compensate by doing gender elsewhere in the partnership; for example, men at the lower end of the dependency curve have been shown to do less housework than those in the middle (Brines, 1994). By providing a couple-level perspective on gendered inequalities in “fertility work,” resource theory and gender identity theory could contribute to the literature on contraceptive use. Such a couple-level perspective could be particularly relevant to research on contraceptive sterilization, as the vast majority of sterilization operations occur within the context of a coresidential partnership—94% of male operations and 83% of female operations (Eeckhaut, 2015)—meaning that partners had both a female and a male sterilization alternative available to them.

This study draws on data from the female and male samples from the 2006 to 2010 and 2011 to 2013 rounds of the National Survey of Family Growth (NSFG) to consider the role of couples’ relative sociodemographic characteristics for the gendered division of sterilization “fertility work.” The analysis is limited to married and cohabiting couples who have completed their intended childbearing, as sterilization should be considered a permanent method. A specialized dyadic technique, diagonal reference models (DRMs), is used to examine if and how a couple’s educational and racial/ethnic composition—that is, couples’ educational and racial/ethnic heterogamy—matters to the gendered division of sterilization “fertility work.” The focus on education and race/ethnicity is motivated by the large body of research (e.g., Borrero et al., 2009; Borrero et al., 2011; Bumpass & Presser, 1972; Chandra, 1998; Daniels et al. 2015; Eeckhaut & Sweeney, 2016; Shreffler, McQuillan, Greil, & Johnson, 2015) that identifies these two sociodemographic factors as key correlates of sterilization.

Background

Sterilization as a method of contraception increased in the United States during the second half of the 20th century, mainly between the late 1960s and early 1980s (Chan & Westhoff, 2010; Presser & Bumpass, 1972). Whereas 16% of married women aged 15 to 44 years relied on sterilization for fertility control in 1965, this number had risen to 42% by 1988, but remained relatively stable thereafter (Chandra, 1998; Jones, Mosher, & Daniels, 2012). The dramatic increase in the prevalence of sterilization was mainly driven by a rise in the use of female sterilization, as the prevalence of tubal ligation versus vasectomy shifted from being comparable in 1965, to being more than 1.5 times higher in 1995 (Chandra, 1998) and thereafter (Jones et al., 2012).

Permanent sterilization is typically used during the last stage of the reproductive life course, when couples want to limit future childbearing (e.g., Daniels et al., 2015). During the early stages of the reproductive life course, U.S. women tend to rely on pills or condoms for fertility control—in 2006-2010, 54% of nulliparous contracepting women relied on pills and 23% relied on condoms (Jones et al., 2012). After a first birth, many women switch to long-acting reversible methods (e.g., intrauterine devices [IUDs]), or continue to rely on pills or condoms—In 2006-2010, 10% of contracepting women with one child relied on IUDs, 29% relied on pills, and 22% relied on condoms. It is not until women aim to limit future childbearing that they overwhelmingly start relying on either female or male sterilization—In 2006-2010, fully 44% of contracepting women who intended no more births relied on female sterilization and 16% relied on male sterilization.

Sterilization is a highly effective, cost-effective method of contraception that provides couples with long-acting protecting against pregnancy. However, compared with other contraceptive methods, the method also requires significant upfront physical and emotional “labor” of the partner undergoing the procedure. Both tubal ligation and vasectomy are medical procedures, but vasectomy tends to be simpler, more economical, and it has lower rates of minor and major complications (Rind, 1989; Shih, Turok, & Parker, 2011). The main reason is that vasectomy involves only small incisions (or punctures if the no scalpel method is used) on each side of the scrotum which can be done under local anesthesia (Schwingl & Guess, 2000), whereas tubal ligation generally requires general anesthesia and entry into the abdominal cavity (Shih et al., 2011). On the other hand, a significant portion of tubal ligations are performed in conjunction with a C-section (Whiteman et al., 2012), and newer transcervical techniques—such as Essure—provide a less invasive option to surgical tubal ligation, but these techniques have also been associated with a growing number of complications, including persistent pain, tubal or uterine perforation, and device migration (U.S. Food and Drug Administration, 2016). Sterilization is also associated with anxiety, such as pain anxiety, regret anxiety, and sexual anxiety (Groat, Neal, & Wicks, 1990; Sandlow et al., 2001). The person undergoing the procedure likely performs most of this emotional “labor,” though the partner may deal with some anxiety as well. In other words, the benefits of sterilization—highly effective, cost-effective, long-acting protecting against pregnancy—are shared by both partners, but the physical and emotional labor is largely shouldered by the partner undergoing the procedure. In that sense, a clear parallel can be drawn between sterilization “fertility work” and household labor—even though there are also clear differences in terms of the frequency of both types of “labor.”

In line with the broader literature on fertility and contraceptive use (Almeling, 2015; Grady et al., 2010; Thomson, 1997), the sterilization literature has traditionally focused on individual-level inequalities and on individual-level explanations for these inequalities. Most important, studies have found that the higher use of female sterilization and lower use of male sterilization among lower educated and minority women is partly explained by their accelerated childbearing schedules, higher achieved parity, and less stable union histories (for education: Chandra, 1998; Eeckhaut & Sweeney, 2016; Shreffler et al., 2015, for race/ethnicity: Borrero et al., 2009; Borrero et al., 2011; Bumpass et al., 2000; Godecker, Thomson, & Bumpass, 2001; Shreffler et al., 2015). Even after adjusting for individual-level covariates, however, individual-level inequalities tend to persist; for example, even after accounting for differences in the abovementioned covariates, lower educated, and Black and Hispanic women rely more heavily on female sterilization, and less heavily on male sterilization.

Much less is known about whether couples’ educational or racial/ethnic composition matters to the gendered division of sterilization “fertility work.” Few studies (Bertotti, 2013; Bumpass et al., 2000; Forste et al., 1995) have adopted a couple-level perspective when studying educational inequalities in sterilization, and only one previous study has examined the importance of couples’ racial/ethnic composition for sterilization (Forste et al., 1995). Moreover, with only one exception (Bertotti, 2013), these studies have relied on older data (20+ years old) from married couples only (e.g., Bumpass et al., 2000; Forste et al., 1995). Research is sorely needed to update and expand our knowledge of the importance of couple-level effects on gendered use of sterilization by education and race/ethnicity. Improving on previous studies, this research should include cohabiting couples (Bumpass et al., 2000; Forste et al., 1995), exclude couples who were sterilized prior to the current union (Bertotti, 2013), exclude couples who are still intending future births (Forste et al., 1995), and distinguish between different types of racially/ethnically heterogamous couples (Forste et al., 1995). Finally, this research should rely on analysis techniques that allow for the examination of heterogamy effects over and above both individual-level effects of partners’ sociodemographic characteristics. Previous research has relied on standard analysis techniques (e.g., Bertotti, 2013; Bumpass et al., 2000; Forste et al., 1995), which are ill-suited to dealing with the strong collinearity that commonly arises from the simultaneous estimation of individual-, partner-, and couple-level effects of partners’ sociodemographic characteristics (see Eeckhaut, Van de Putte, Gerris, & Vermulst, 2013, for a brief discussion of this issue). Specialized dyadic analysis technique are needed to ascertain that reported couple-level effects are not simply an artifact of the individual-level effects of partners’ sociodemographic characteristics.

The Current Research

Taken together, the rich theoretical literature on the division of household labor suggests a great potential for couple-level explanations to shed fresh light on persistent and gendered inequalities in contraceptive sterilization. Using nationally representative data from the female and male samples from the 2006 to 2010 and 2011 to 2013 rounds of the NSFG, this study considers the role of couples’ educational and racial/ethnic composition for the gendered division of sterilization “fertility work.” A specialized dyadic technique, DRMs, is used to analyze data on married and cohabiting couples who have completed their intended childbearing. Following Bumpass et al. (2000), the analysis models the choice for sterilization, and the choice for female versus male sterilization as one decision. In contrast to household work—which can be partly outsourced but is hard to abandon entirely—couples have the option to forego sterilization, in addition to being able to choose between female and male sterilization. Research suggests that these choices are linked, as each partner considers his or her own willingness to undergo sterilization when considering whether to opt for sterilization. For example, studies have shown that the partner undergoing sterilization tends to dominate the sterilization decision-making process (Lodewijckx, 1989; Terry & Braun, 2011) and to feel more favorable about the sterilization decision (Jamieson et al., 2002; Shain, Miller, & Holden, 1986). Thus, three potential outcomes are considered in the analysis: female sterilization, male sterilization, and use of a reversible method.

Method

Data

Data for this study were drawn from the female and male samples from the 2006 to 2010 and 2011 to 2013 rounds of the NSFG. The data are representative of the U.S. civilian noninstitutionalized population aged 15 to 44 years when properly weighted, and include oversamples of teens, Blacks, and Hispanics. Face-to-face interviews were held with a total of 12,279 women and 10,403 men in 2006-2010, and 5,601 women and 4,815 men in 2011-2013. Response rates were 78% and 73% for women and 75% and 72% for men, respectively (Martinez, Daniels, & Chandra, 2012; U.S. Department of Health and Human Services, 2014). All analyses and descriptive statistics were adjusted for the NSFG’s complex sample design.

The analytic sample is limited to respondents aged 18 to 44 years who are married or cohabiting with an opposite-sex partner because questions regarding sterilization of the partner are limited to respondents in coresidential partnerships (N = 13,468). Sterilization is generally considered a permanent method so the analytic sample is limited to respondents who have completed their intended childbearing and are currently using contraception (including sterilization) for fertility control (N = 5,077). More specifically, respondents were included if they indicated that they and their partner do not intend to have a(nother) child, are not currently pregnant, and used a reversible method of contraception (including traditional methods) at last sex in the past 3 months, or if they indicated that either partner had a sterilization operation for contraceptive reasons during the course of the current coresidential relationship, regardless of later reversal, and omitting the small number of couples in which both partners had an operation. The latter restriction was based on the month and year of sterilization and the month and year of cohabitation or marriage, with the exception of female procedures in the male sample. Male respondents did not provide information on the date of female sterilization procedures, so instead couples in which the male respondent indicated that he relied on female sterilization at first sex with the current partner were omitted. Male respondents were also not asked about the reason for their partner’s tubal ligation, meaning that it is impossible to ascertain that all female procedures reported in the male sample were performed for nonmedical reasons. However, sensitivity analysis including all sterilization operations (i.e., contraceptive or other) in both the female and male samples did not lead to substantively different conclusions. Finally, the analytic sample excludes respondents who indicated that it would not be physically possible for themselves or their partner to have a child of their own for reasons other than surgical sterilization (e.g., accident, illness, menopause; n = 5,026), as well as respondents who had a missing value on any of the variables included in the analysis (n = 4,969).

Variables

The dependent variable in this research is contraceptive use status: female sterilization, male sterilization, or any reversible method (including traditional methods). First, respondents were identified who had, or whose partner had a sterilization operation for contraceptive reasons during the course of the current coresidential relationship. Next, for respondents in couples in which neither partner had a sterilization operation, reports of contraceptive method used at last sex in the past 3 months were considered. Sensitivity analyses excluding respondents who(se partner) had a sterilization operation, but reported not having had sex in the past 3 months, did not lead to substantively different conclusions.

The primary independent variables are education, race/ethnicity, educational heterogamy, and racial/ethnic heterogamy. Education was categorized as follows: (a) less than high school education, (b) high school education, (c) some college, and (d) bachelor’s degree or higher. Race/ethnicity was categorized as follows: (a) non-Hispanic White, (b) Hispanic, (c) non-Hispanic Black, and (d) non-Hispanic other or multiple race. Two types of heterogamy measures were used. A first type distinguished between two groups of respondents: (a) those in homogamous couples and (b) those in heterogamous couples. A second type distinguished between three types of respondents; for education: (a) those in educationally homogamous couples, (b) those in hypergamous couples (i.e., male partner has more education than the female partner), and (c) those in hypogamous couples (i.e., female partner has more education than the male partner), and for race/ethnicity: (a) those in couples in which both partners are non-Hispanic White or both partners are of minority background (i.e., Hispanic, non-Hispanic Black, or non-Hispanic other or multiple race; referred to hereafter as racially/ethnically homogamous couples), (b) those in couples in which the male partner is non-Hispanic White and the female partner is of minority background, and (c) those in couples in which the female partner is non-Hispanic White and the male partner is of minority background. Sensitivity analyses limiting the heterogamy category (two-category variable) or categories (three-category variable) of the educational heterogamy variable to couples in which one partner has a bachelor’s degree or higher and the other partner has no college degree did not lead to substantively different conclusions.

Based on their association with sterilization (e.g., Chandra, 1998), a number of other background measures were included. For each partner, a measure of age (18-29; 30-34; 35-39; 40-44 years), parity (0−1; 2; 3+), and union history (no previous marriage; a previous marriage) was included. For each couple, a measure of union status (married; cohabiting), whether the couple has any mutual biological children, religious affiliation (based on respondent’s current affiliation, categories are as follows: Protestant; Catholic; none; other), and woman’s age at last birth (<25; 25-29; 30-34; 35+ years) was included. In cases where the respondent did not report any mutual biological children with the current partner, the latter variable indicated woman’s age at the start of the current partnership, rather than at last birth, as this is the youngest age at which sterilization could have entered the couple’s contraceptive choice set.

Analysis Technique

The analysis was conducted in two steps. First, chi-square tests were conducted to examine bivariate associations of contraceptive use with education and race/ethnicity (and the other independent variables). The gendered division of sterilization “fertility work” was examined based on the female and the male partner’s characteristics.

Second, a specialized dyadic technique, DRMs, was used to examine variation in the use of female sterilization, male sterilization, or a reversible method by education, race/ethnicity, educational heterogamy, and racial/ethnic heterogamy. DRMs were designed by Sobel (1981) to study the effects of social mobility, but the technique has also been favorably judged in relation to other models for examining status inconsistency and heterogamy effects (Eeckhaut et al., 2013; Hendrickx, de Graaf, Lammers, & Ultee, 1993). DRMs adhere to the theoretical idea that individuals in homogamous unions represent the core of their group. More specifically, DRMs estimate the value on the dependent variable for individuals in each type of homogamous couple (e.g., couple in which both partners have: less than high school education, high school education, some college, bachelor’s degree or higher). The value on the dependent variable for individuals in heterogamous couples ij is expected to lie in-between the values as estimated for individuals in homogamous couples ii and individuals in homogamous couples jj. The relative impact of these diagonal referents is estimated by means of a salience parameter—p—with a value between 0 and 1. Covariates, such as control variables, can be added to the baseline model (in Equation 2: c variables X), meaning that the extended DRMs combine the advantages of log-linear models (analyzing nonlinear and interaction effects), with the flexibility of multivariate regression (incorporating multiple control variables; de Graaf & Heath, 1992). The multinomial logistic transformation of the DRMs was used as the dependent variable, contraceptive use status, was measured by means of three categories:

B_{k s} = \frac{\exp (θ_{k s})}{\sum_{s = 1}^{3} \exp (θ_{k s})}

(1)

θ_{k s} = p * μ_{i i s} + (1 - p) * μ_{j j s} + \sum_{c = 1}^{C} β_{s c} * X_{k c}

(2)

where $B_{k s}$ is the probability that respondent k is situated in category s of the dependent variable. $θ_{k s}$ stands for the log odds that respondent k is situated in category s.

Model 1 of the analysis estimates the multinomial logistic DRM for education and the multinomial logistic DRM for race/ethnicity, including control variables. Model 2 adds the heterogamy variables to these multinomial logistic DRMs for education and race/ethnicity (in Equation 4: h variables X) to examine associations between contraceptive use and educational and racial/ethnic heterogamy, respectively:

B_{k s} = \frac{\exp (θ_{k s})}{\sum_{s = 1}^{3} \exp (θ_{k s})}

(3)

θ_{k s} = p * μ_{i i s} + (1 - p) * μ_{j j s} + \sum_{c = 1}^{C} β_{s c} * X_{k c} + \sum_{h = 1}^{H} β_{s h} * X_{k h}

(4)

All DRMs were estimated using newly written code (“svygnm” in logmult package; Bouchet-Valat, 2016) that allows for the use of complex survey data—such as the NSFG data—when using the gnm command in R (GNM package; Turner & Firth, 2012).

Results

Table 1 shows that 31% of couples rely on female sterilization, 22% rely on male sterilization, and 47% rely on a reversible method. The majority of female and male partners are non-Hispanic White (67% and 66%, respectively), have some college education or a bachelor’s degree (65% and 59%, respectively), are age 30 years or older (84% and 90%, respectively), have two or more children (81% and 72%, respectively), and have not been previously married (84% and 84%, respectively). Most couples are married (85%), Protestant (47%), and had their last child when the woman was less than 35 years old (87%). A much larger share of couples are racially/ethnically homogamous (86%), as compared with educationally homogamous (52%, calculated based on Table 2).

Table 1.

Percentage of Respondents Relying on Female Sterilization, Male Sterilization, or a Reversible Method, by Education, Race/Ethnicity, and the Other Independent Variables: Female and Male Partner (N = 4,969).

	Female partner				Male partner
	Percentage of total	Female sterilization	Male sterilization	Reversible method	Percentage of total	Female sterilization	Male sterilization	Reversible method
Education		***				***
Less than high school	10.7	54.4	4.2	41.4	14.5	53.8	5.9	40.3
High school	24.6	41.4	15.5	43.1	26.4	36.6	17.3	46.1
Some college	30.3	29.9	22.1	48.0	27.4	31.4	19.9	48.7
Bachelor’s degree or higher	34.4	17.0	32.0	51.0	31.8	15.3	34.9	49.8
Race/ethnicity		***				***
Non-Hispanic White	67.0	26.3	28.1	45.7	65.6	24.9	28.4	46.7
Hispanic	18.7	43.9	9.2	46.9	19.1	43.8	8.5	47.7
Non-Hispanic Black	7.6	44.8	7.7	47.6	9.0	42.6	10.0	47.4
Non-Hispanic other or multiple race	6.7	25.6	12.2	62.2	6.3	37.6	12.7	49.7
Age		***				***
18-24 Years	4.3	9.3	1.2	89.5	2.6	8.7	1.1	90.2
25-29 Years	11.4	26.5	9.1	64.4	7.7	28.0	4.8	67.2
30-34 Years	21.5	32.5	15.0	52.5	18.4	29.0	17.5	53.5
35-39 Years	31.5	29.4	27.0	43.6	28.1	29.0	22.8	48.3
40+ Years	31.4	35.9	29.2	34.9	43.3	34.8	27.6	37.6
Parity		***				***
0	7.7	4.9	5.7	89.5	13.8	20.0	14.3	65.8
1	11.5	11.8	17.4	70.9	14.1	25.9	14.6	59.6
2	42.9	26.5	25.5	48.1	38.4	26.6	25.2	48.2
3+	38.0	47.0	22.7	30.3	33.7	42.4	24.5	33.1
Union history		***
No previous marriage	84.0	28.9	22.7	48.4	83.6	30.3	21.8	47.9
Had a previous marriage	16.0	41.5	18.2	40.4	16.4	34.1	22.8	43.2
Union status		***				***
Married	85.0	30.9	25.5	43.7	85.0	30.9	25.5	43.7
Cohabitation	15.0	31.3	2.2	66.6	15.0	31.3	2.2	66.6
Woman’s age at last birth		***				***
<25 Years	26.7	32.4	15.4	52.2	26.7	32.4	15.4	52.2
25-30 Years	31.9	33.3	24.6	42.1	31.9	33.3	24.6	42.1
30-34 Years	28.2	27.7	26.2	46.1	28.2	27.7	26.2	46.1
35+ Years	13.1	28.9	19.9	51.2	13.1	28.9	19.9	51.2
Any mutual children		***				***
Yes	87.0	32.0	24.0	43.9	87.0	32.0	24.0	43.9
No	13.0	23.4	7.9	68.6	13.0	23.4	7.9	68.6
Current religious affiliation		***				***
Protestant	46.9	33.1	26.5	40.4	46.9	33.1	26.5	40.4
Catholic	25.7	32.1	19.6	48.4	25.7	32.1	19.6	48.4
None	18.9	26.6	17.0	56.4	18.9	26.6	17.0	56.4
Other	8.6	25.0	15.1	59.9	8.6	25.0	15.1	59.9
Total	100.0	30.9	22.0	47.1	100.0	30.9	22.0	47.1

Note. Percentages/numbers may not sum to the total because of rounding. Significance testing of the association between contraceptive methods use and each of the covariates used chi-square tests.

***

p < .001.

Table 2.

Number and Total, Row, and Column Percentage of Respondents in Each Type of Educational and Racial/Ethnic Homogamous and Heterogamous Couple.

Education
		Male partner
		Less than high school	High school	Some college	Bachelor’s degree or higher	All
Female partner	Less than high school	308	142	64	16	530
		6.2%	2.9%	1.3%	0.3%	10.7%
		58.1%	26.8%	12.0%	3.1%	100.0%
		42.7%	10.8%	4.7%	1.0%	10.7%
	High school	246	562	294	121	1,224
		5.0%	11.3%	5.9%	2.4%	24.6%
		20.1%	45.9%	24.0%	9.9%	100.0%
		34.2%	42.9%	21.6%	7.7%	24.6%
	Some college	135	432	604	334	1,505
		2.7%	8.7%	12.2%	6.7%	30.3%
		8.9%	28.7%	40.1%	22.2%	100.0%
		18.7%	32.9%	44.4%	21.2%	30.3%
	Bachelor’s degree or higher	32	175	399	1,105	1,711
		0.6%	3.5%	8.0%	22.2%	34.4%
		1.9%	10.3%	23.3%	64.6%	100.0%
		4.4%	13.4%	29.3%	70.1%	34.4%
	All	721	1,312	1,360	1,577	4,969
		14.5%	26.4%	27.4%	31.7%	100.0%
		14.5%	26.4%	27.4%	31.7%	100.0%
		100.0%	100.0%	100.0%	100.0%	100.0%
Race/ethnicity
		Male partner
		Non-Hispanic White	Hispanic	Non-Hispanic Black	Non-Hispanic other or multiple race	All
Female partner	Non-Hispanic White	2,991	157	67	116	3,331
		60.2%	3.2%	1.4%	2.3%	67.0%
		89.8%	4.7%	2.0%	3.5%	100.0%
		91.8%	16.5%	14.9%	37.0%	67.0%
	Hispanic	134	763	19	14	930
		2.7%	15.4%	0.4%	0.3%	18.7%
		14.4%	82.1%	2.0%	1.5%	100.0%
		4.1%	80.5%	4.2%	4.4%	18.7%
	Non-Hispanic Black	13	10	347	5	377
		0.3%	0.2%	7.0%	0.1%	7.6%
		3.6%	2.8%	92.2%	1.5%	100.0%
		0.4%	1.1%	77.3%	1.7%	7.6%
	Non-Hispanic other or multiple race	120	17	16	178	332
		2.4%	0.4%	0.3%	3.6%	6.7%
		36.2%	5.2%	4.8%	53.7%	100.0%
		3.7%	1.8%	3.5%	56.8%	6.7%
	All	3,259	948	449	314	4,969
		65.6%	19.1%	9.0%	6.3%	100.0%
		65.6%	19.1%	9.0%	6.3%	100.0%
		100.0%	100.0%	100.0%	100.0%	100.0%

Table 1 also shows that education is negatively associated with reliance on female sterilization, and positively associated with reliance on male sterilization or a reversible method. Gradients are similar when looking at the female versus the male partner’s education, and confirm that the gap in reliance on female versus male sterilization is largest among those with less than high school education (54% vs. 4% among female partners, 54% vs. 6% among male partners, respectively), smallest among those with some college (30% vs. 22% among female partners, 31% vs. 20% among male partners, respectively), and reversed among those with a bachelor’s degree or higher (17% vs. 32% among female partners, 15% vs. 35% among male partners, respectively).

Turning to bivariate results for race/ethnicity, Table 1 shows that non-Hispanic Whites are somewhat more likely to rely on male versus female sterilization, whereas Hispanics and non-Hispanic Blacks are much more likely to rely on female versus male sterilization. Again, patterns are similar when looking at the male versus the female partner’s characteristics, and confirm that the gap between female versus male sterilization is smallest among non-Hispanic Whites (26% vs. 28% among female partners, 25% vs. 28% among male partners, respectively), when compared with Hispanics (44% vs. 9% among female and male partners, respectively) and non-Hispanic Blacks (45% vs. 8% among female partners, 43% vs. 10% among male partners, respectively).

Next, differences in contraceptive use by education and race/ethnicity are considered in a multivariate context. Table 3 shows relative risk ratios (RRR) for relying on male sterilization or a reversible method, when compared with female sterilization, as estimated for respondents in educationally (Table 3A) and racially/ethnically (Table 3B) homogamous couples (i.e., Model 1). For education, a positive association is observed with reliance on male sterilization or a reversible method, when compared with female sterilization. Among respondents in educationally homogamous couples, those with less than high school education are less likely to rely on male versus female sterilization (RRR = 0.03) and less likely to rely on a reversible method versus female sterilization (RRR = 0.17), when compared with those with a bachelor’s degree or higher. For race/ethnicity, results confirm the higher reliance on male versus female sterilization among non-Hispanic Whites, as compared with all other racial/ethnic groups (Table 3B). Among respondents in racially/ethnically homogamous couples, Hispanics (RRR = 0.41), non-Hispanic Blacks (RRR = 0.21), and non-Hispanic others (RRR = 0.32) are less likely to rely on male versus female sterilization, when compared with non-Hispanic Whites. These results by and large confirm that couples with lower levels of education and minority couples tend to rely more heavily on female sterilization, and less heavily on male sterilization or a reversible method (only for education).

Table 3.

Results of the Multinomial Logistic DRMs.

	Male sterilization	Reversible method
A: Education
Respondents in homogamous couples with educational level i (µ_ii’s)
Less than high school	0.03 [0.01, 0.05]	0.17 [0.08, 0.26]
Completed high school	0.19 [0.08, 0.30]	0.26 [0.13, 0.39]
Some college	0.29 [0.09, 0.48]	0.43 [0.20, 0.65]
Bachelor’s degree or higher (ref)	1.00	1.00
Salience parameter	0.44 [0.22, 0.66]
B: Race/ethnicity
Respondents in homogamous couples of race/ethnicity i (µ_ii’s)
Non-Hispanic White (ref)	1.00	1.00
Hispanic	0.41 [0.18, 0.65]	1.14 [0.66, 1.61]
Non-Hispanic Black	0.21 [0.07, 0.36]	0.76 [0.46, 1.07]
Non-Hispanic other or multiple race	0.32 [0.05, 0.60]	1.06 [0.31, 1.81]
Salience parameter	0.51 [−0.68, 1.69]

Note. DRM = Diagonal reference model. Exponentiated coefficients (ref = female sterilization), 95% confidence intervals (between square brackets), and salience parameters for education and race/ethnicity. Boldface indicates coefficient differs significantly from reference group, at p < .05 level. The results for the control variables are omitted from this table. As described in the text, the following control variables were entered in the model: union status, any mutual biological children, religious affiliation, woman’s age at last birth, woman’s race/ethnicity (DRMs for education), woman’s education (DRMs for race/ethnicity), woman’s age, woman’s parity, woman’s union history, man’s race/ethnicity (DRMs for education), man’s education (DRMs for race/ethnicity), man’s age, man’s parity, man’s union history.

Finally, does educational or racial/ethnic heterogamy matter to the gendered division of sterilization “fertility work?” Table 4 displays results of four parallel multivariate analyses; for education and race/ethnicity, for the two- and the three-category heterogamy variables (i.e., Model 2). Results show no significant effect of educational heterogamy (Table 4A), regardless of whether it is measured by means of the two- or the three-category heterogamy variable. In contrast, both the two- and the three-category heterogamy variables suggest a significant effect of racial/ethnic heterogamy (Table 4B). The two-category variable shows that racially/ethnically heterogamous couples are more likely to rely on male versus female sterilization, as compared with racially/ethnically homogamous couples (RRR = 1.62). The three-category variable suggests that this overall difference between homogamous and heterogamous couples is driven entirely by the higher likelihood of relying on male versus female sterilization among couples in which the male partner is non-Hispanic White and the female partner is of minority background (RRR = 3.52). Overall, this indicates that the type of racial/ethnic heterogamy matters, rather than the presence of heterogamy as such, illustrating the gendered ways in which partners’ sociodemographic characteristics shape sterilization “fertility work” among contemporary U.S. couples.

Table 4.

Results of the Multinomial Logistic DRMs.

	Male sterilization	Reversible method
A: Education
Two-category heterogamy variable
Homogamous (ref)	1.00	1.00
Heterogamous	1.07 [0.79, 1.46]	1.13 [0.85, 1.48]
Three-category heterogamy variable
Homogamous (ref)	1.00	1.00
Hypogamous	0.80 [0.37, 1.72]	0.92 [0.55, 1.51]
Hypergamous	1.40 [0.68, 2.88]	1.39 [0.82, 2.36]
B: Race/ethnicity
Two-category heterogamy variable
Homogamous (ref)	1.00	1.00
Heterogamous	1.62 [1.01, 2.58]	0.99 [0.68, 1.44]
Three-category heterogamy variable
Homogamous (ref)	1.00	1.00
Woman White, man minority	0.72 [0.40, 1.29]	0.84 [0.53, 1.33]
Man White, woman minority	3.52 [1.77, 7.00]	1.15 [0.62, 2.13]

Note. DRM = Diagonal reference model. Exponentiated coefficients (ref = female sterilization), and 95% confidence intervals (between square brackets) for educational and racial/ethnic heterogamy. Boldface indicates coefficient differs significantly from reference group, at p < .05 level. Results for the control variables are omitted from this table. As described in the text, the following control variables were entered in the model: union status, any mutual biological children, religious affiliation, woman’s age at last birth, woman’s race/ethnicity (DRMs for education), woman’s education (DRMs for race/ethnicity), woman’s age, woman’s parity, woman’s union history, man’s race/ethnicity (DRMs for education), man’s education (DRMs for race/ethnicity), man’s age, man’s parity, man’s union history.

Discussion

Drawing on the literature on the gendered division of labor, this study advanced understanding of persistent and gendered differentials in contraceptive sterilization. Sterilization “fertility work” was considered as just another form of household labor that needs to be negotiated within couples—though there are important differences as well, such as in terms of the frequency of labor. Results confirmed that certain types of sociodemographic heterogamy matter to the division of sterilization “fertility work”—though others do not. Most important, racial/ethnic heterogamy was found crucial to sterilization decision making: couples in which the male partner is non-Hispanic White and the female partner is of minority background were found to have an increased likelihood of relying on male versus female sterilization. This pattern is in line with the gender perspective on the division of labor, which asserts that men tend to avoid feminine tasks and engage in more masculine activities, particularly if the couple crosses gender boundaries. Vice versa, if the couple conforms to traditional gender norms—such as when the man enjoys higher racial/ethnic privilege because he is a member of the dominant racial/ethnic group and the women is of minority background—the male partner may be more willing to participate in tasks that are traditionally viewed as a woman’s responsibility, such as sterilization “fertility work.” This indicates that women’s lower relative racial/ethnic privilege may have different implications for the division of sterilization “fertility work” than women’s lower absolute racial/ethnic privilege, which has typically been linked to the male partner sharing less of the burden (e.g., results show lower levels of male versus female sterilization among minority, as compared to non-Hispanic White women).

Results for educational heterogamy were not in line with the gender perspective, nor with resource theory. Educational heterogamy may be too common to lead to a compensatory “doing gender”-response (48% of the analytic sample is educationally heterogamous), or its impact could be weakened by the fact that women commonly receive lower pay for equal work (O’Reilly, Smith, Deakin, & Burchell, 2015), and are underrepresented in typically male study fields (Charles & Bradley, 2002) and in elite positions in the labor market (Mandel & Semyonov, 2006). Further (qualitative) research is needed to confirm that racially/ethnically homogamous couples’ motivations and negotiations are in line with the “doing gender” interpretation, and to examine why these patterns do not extend to education.

The lack of a significant effect of educational heterogamy in this study contradicts the results of Bertotti (2013) and Bumpass et al. (2000), who found women’s higher relative education to be associated with a reduced likelihood of relying on male versus female sterilization. Similarly, results for racial/ethnic heterogamy in this study contradict those by Forste et al. (1995), who reported racial heterogamy to reduce the overall reliance on sterilization, but matter little for the use of female versus male sterilization. Differences in study period, analytic sample, and the operationalization of heterogamy can explain some of the differences. Another key factor is likely the use of nondyadic analysis techniques in these previous studies, as these techniques are ill-equipped to dealing with the specific challenges that arise from the dyadic nature of heterogamy effects. Most important, strong collinearity between terms has generally prevented researchers from including all individual-level effects of partners’ sociodemographic characteristics, in addition to the couple-level effect (e.g., Bumpass et al., 2000). In fact, if the current analyses would have relied on multinomial logistic regression and adjusted for the female, but not the male partner’s education (in addition to the other covariates), results would have confirmed that hypogamous couples are less likely to rely on male versus female sterilization, when compared with educationally homogamous couples (results available on request from the corresponding author). This illustrates the need to include both main effects of partners’ sociodemographic characteristics, in addition to the heterogamy effects, as this is the only way to ascertaining that reported heterogamy effects are not simply an artifact of these main effects. The current study adopted a specialized dyadic technique to examine effects of sociodemographic heterogamy over and above the individual-level effects of partners’ sociodemographic characteristics, thus providing the first reliable estimate of the role of educational and racial/ethnic heterogamy for decisions regarding the use of (female vs. male) sterilization. Future research on sterilization—and contraceptive use and fertility more generally—could benefit from this dyadic approach to analyzing couple-level effects.

Several factors are limitations to this study. First, given the focus on permanent sterilization, the analytic sample was limited to respondents who did not intend future childbearing. Yet childbearing intentions are often dynamic (hence, sterilization reversal), and couples are likely to vary in their level of agreement and certainty about their future intentions (Thomson, 1997). Future research could benefit from examining how such variation may affect contraceptive use patterns by education and race/ethnicity. Moreover, because use of sterilization tends to be restricted to the last stage of the reproductive life course, the decision on who gets sterilized typically follows many other decisions related to “fertility work.” As later decisions (i.e., who gets sterilized) are likely shaped by earlier decisions (e.g., contraception used prior to sterilization), an important direction for future research will be to examine the division of other types of “fertility work” across the reproductive life course. Such research might also benefit from extending the focus on “fertility work” related to avoiding pregnancy (i.e., contraceptive use), to include “fertility work” related to seeking pregnancy (e.g., navigating infertility treatment) or maintaining fertility (e.g., STD testing).

Second, as most research analyzing nationally representative data on contraceptive use, this study relied on a measure of education at the time of interview, rather than at the time of contraceptive decision making. Given the focus on married and cohabiting individuals who have completed childbearing, education is likely more stable in this sample, as compared with more general samples of individuals using contraception. Nevertheless, future research could benefit from data collection efforts aimed at measuring individuals’ characteristics at the time of contraceptive decision making.

Third, this study relied on a broad categorization of racial and ethnic background which, while commonly used in research on contraceptive use, distinguishes between only three broadly defined racial/ethnic groups. Moreover, this measure includes a heterogeneous “other” category (i.e., “non-Hispanic other or multiple race”), meaning that patterns could not be examined for different groups of individuals of “non-Hispanic other or multiple race.” Finally, all couples in which both partners are a member of a minority group were considered as racially/ethnically homogamous. This operationalization was motivated by the limited representation of certain combinations of race/ethnicity in the analytic sample (see Table 2), and the theoretical focus on the contrast between the dominant (or privileged) non-Hispanic White group on the one hand and the Hispanic and non-Hispanic Black minority groups on the other hand. Future research should examine if results are sensitive to the level of detail of the racial/ethnic heterogamy measure.

Finally, the analysis was limited to couples who are using contraception because the relative number of nonusers, as well as the relative importance of reasons for not using contraception (e.g., cost of contraceptives, access to contraceptives) likely varies by education and race/ethnicity (Frost, Singh, & Finer, 2007; Grady, Dehlendorf, Cohen, Schwartz, & Borrero, 2015). In addition to facilitating the interpretation of findings, it should be kept in mind that excluding couples who are currently not using contraception also limits the generalizability of the current findings.

In spite of these limitations, this study provides an important contribution to the literature on persistent and gendered inequalities in the use of sterilization, while at the same time expanding the literature on the division of household work to include sterilization “fertility work.” Future research should examine other types of “fertility work,” and consider how the division of fertility work might vary across the reproductive life course. In addition, future research should consider if couples’ sociodemographic composition at the time of sterilization is associated with the likelihood of later sterilization regret. High levels of sterilization regret remain an issue in the contemporary United States—in 2006-2010, one-in-four sterilized women (25%) expressed a desire to have their tubal ligations reversed (Eeckhaut, Sweeney, & Feng, 2018)—and our understanding of this issue could greatly benefit from research examining the context surrounding decisions regarding contraceptive sterilization.

Footnotes

Acknowledgements

The author would like to thank Milan Bouchet-Valat and Heather Turner for their help with the estimation of the multinomial logistic diagonal reference models in R.

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Research reported in this publication was supported by the University of Delaware General University Research fund.

References

Almeling

(2015). Reproduction. Annual Review of Sociology, 41, 423-442.

Bertotti

A. M.

(2013). Gendered divisions of fertility work: Socioeconomic predictors of female versus male sterilization. Journal of Marriage and Family, 75, 13-25.

Bianchi

S. M.

Milkie

M. A.

Sayer

L. C.

Robinson

J. P.

(2000). Is anyone doing the housework? Trends in the gender division of household labor. Social Forces, 79, 191-228.

Blood

R. O.

Wolfe

D. M.

(1960). Husbands and wives: The dynamics of married living. Glencoe, IL: Free Press.

Bolzendahl

Gubernskaya

(2016). Racial and ethnic homogamy and gendered time on core housework. Socius, 2, 1-15.

Borrero

Abebe

Dehlendorf

Schwarz

E. B.

Creinin

M. D.

Nikolajski

Ibrahim

(2011). Racial variation in tubal sterilization rates: Role of patient-level factors. Fertility and Sterility, 95(1), 17-22.

Borrero

Moore

C. G.

Qin

Schwarz

E. B.

Akers

Creinin

M. D.

Ibrahim

S. A.

(2009). Unintended pregnancy influences racial disparity in tubal sterilization rates. Journal of General Internal Medicine, 25, 122-128.

Bouchet-Valat

(2016). logmult: Log-multiplicative models, including association models, R package (Version 0.6.3) [Computer software]. Retrieved from http://CRAN.R-project.org/

Brines

(1994). Economic dependency, gender, and the division of labor at home. American Journal of Sociology, 100, 652-688.

10.

Bumpass

L. L.

Presser

H. B.

(1972). Contraceptive sterilization in the U.S.: 1965 and 1970. Demography, 9, 531-548.

11.

Bumpass

L. L.

Thomson

Godecker

A. L.

(2000). Women, men, and contraceptive sterilization. Fertility and Sterility, 73, 937-946.

12.

Chan

L. M.

Westhoff

C. L.

(2010). Tubal sterilization trends in the United States. Fertility and Sterility, 94, 1-6.

13.

Chandra

(1998). Surgical sterilization in the United States: Prevalence and characteristics, 1965-95. Vital and Health Statistics, 23(20), 1-33.

14.

Charles

Bradley

(2002). Equal but separate? A cross-national study of sex segregation in higher education. American Sociological Review, 67, 573-599.

15.

Daniels

Daugherty

Jones

Mosher

(2015). Current contraceptive use and variation by selected characteristics among women aged 15-44: United States, 2011-2013. National Health Statistics Report, 86, 1-14.

16.

de Graaf

N. D.

Heath

. (1992). Husbands’ and wives’ voting behaviour in Britain: Class-dependent mutual influence of spouses. Acta Sociologica, 35, 311-322.

17.

Dereuddre

Buffel

Backe

(2017). Power and the gendered division of contraceptive use in Western European couples. Social Science Research, 64, 263-276.

18.

Eeckhaut

M. C. W.

(2015). Marital status and female and male contraceptive sterilization in the United States. Fertility and Sterility, 103, 1509-1515.

19.

Eeckhaut

M. C. W.

Sweeney

M. M.

(2016). Gender, class, and contraception in ten low-fertility countries: The perplexing links between sterilization and disadvantage. Population Studies, 70, 39-58.

20.

Eeckhaut

M. C. W.

Sweeney

M. M.

Feng

(2018). Desire for sterilization reversal among U.S. females: Increasing inequalities by educational level. Perspectives on Sexual and Reproductive Health, 50, 139-145. doi:10.1363/psrh.12076

21.

Eeckhaut

M. C. W.

Van de Putte

Gerris

J. R. M.

Vermulst

A. A.

(2013). Analysing the effect of educational differences between partners: A methodological/theoretical comparison. European Sociological Review, 29, 60-73.

22.

Fennell

J. L.

(2011). Men bring condoms, women take pills. Men’s and women’s roles in contraceptive decision making. Gender & Society, 25, 496-521.

23.

Forste

Tanfer

Tedrow

(1995). Sterilization among currently married men in the United States, 1991. Family Planning Perspectives, 27, 100-107 (+122).

24.

Frost

J. J.

Singh

Finer

L. B.

(2007). Factors associated with contraceptive use and nonuse, United States, 2004. Perspectives on Sexual and Reproductive Health, 39, 90-99.

25.

Godecker

A. L.

Thomson

Bumpass

L. L.

(2001). Union status, marital history and female contraceptive sterilization in the United States. Family Planning Perspectives, 33, 35-41 (+49).

26.

Grady

C. D.

Dehlendorf

Cohen

E. D.

Schwartz

Borrero

(2015). Racial and ethnic differences in contraceptive use among women who desire no future children, 2006-2010 National Survey of Family Growth. Contraception, 92, 62-70.

27.

Grady

W. R.

Klepinger

D. H.

Billy

J. O. G.

Cubbins

L. A.

(2010). The role of relationship power in couple decisions about contraception in the US. Journal of Biosocial Science, 42, 307-323.

28.

Grimes

D. A.

(2009). Forgettable contraception. Contraception, 80, 497-499.

29.

Groat

H. T.

Neal

A. G.

Wicks

J. W.

(1990). Sterilization anxiety and fertility control in the later years of childbearing. Journal of Marriage and the Family, 52, 249-258.

30.

Hendrickx

de Graaf

N. D.

Lammers

Ultee

W. C.

(1993). Models for status inconsistency and mobility: A comparison of the approaches by Hope and Sobel with the mainstream square additive model. Quality & Quantity, 27, 335-352.

31.

Jamieson

D. J.

Kaufman

S. C.

Costello

Hillis

S. D.

Marchbanks

P. A.

Peterson

H. B.

(2002). A comparison of women’s regret after vasectomy versus tubal sterilization. Obstetrics & Gynecology, 99, 1073-1079.

32.

Jones

Mosher

Daniels

(2012). Current contraceptive use in the United States, 2006-2010, and changes in patterns of use since 1995. National Health Statistics Reports, 60, 1-25.

33.

Lodewijckx

(1989). Recent trends in contraceptive sterilization in Flanders. Journal of Biosocial Science, 21, 59-70.

34.

Mandel

Semyonov

(2006). A welfare state paradox: State interventions and women’s employment opportunities in 22 countries. American Journal of Sociology, 111, 1910-1949.

35.

Martinez

Daniels

Chandra

(2012). Fertility of men and women aged 15-44 years in the United States: National Survey of Family Growth, 2006-10 (Report No. 51). Hyattsville, MD: National Center for Health Statistics.

36.

O’Reilly

Smith

Deakin

Burchell

(2015). Equal pay as a moving target: International perspectives on forty-years of addressing the gender pay gap. Cambridge Journal of Economics, 39, 299-317.

37.

Presser

H. B.

Bumpass

L. L.

(1972). The acceptability of contraceptive sterilization among U.S. couples: 1970. Family Planning Perspectives, 4(4), 18-26.

38.

Rind

(1989). Vasectomy proves to be preferable to female sterilization. Family Planning Perspectives, 21, 191. doi:10.2307/2135815

39.

Sandlow

J. I.

Westefeld

J. S.

Maples

M. R.

Scheel

K. R.

(2001). Psychological correlates of vasectomy. Fertility and Sterility, 75, 544-548.

40.

Schwingl

P. J.

Guess

H. A.

(2000). Safety and effectiveness of vasectomy. Fertility and Sterility, 73, 923-936.

41.

Shain

R. N.

Miller

W. B.

Holden

A. E. C.

(1986). Married women’s dissatisfaction with tubal sterilization and vasectomy at first-year follow-up: Effects of perceived spousal dominance. Fertility and Sterility, 45, 808-819.

42.

Shih

Turok

D. K.

Parker

W. J.

(2011). Vasectomy: The other (better) form of sterilization. Contraception, 83, 310-315.

43.

Shreffler

K. M.

McQuillan

Greil

A. L.

Johnson

D. R.

(2015). Surgical sterilization, regret, and race: Contemporary patterns. Social Science Research, 50, 31-45.

44.

Sobel

M. E.

(1981). Diagonal mobility models: A substantively motivated class of designs for the analysis of mobility effects. American Sociological Review, 46, 893-906.

45.

Terry

Braun

(2011). “It’s kind of me taking responsibility for these things”: Men, vasectomy and “contraceptive economies.” Feminism & Psychology, 21, 477-495.

46.

Thomson

(1997). Couple childbearing desires, intentions, and births. Demography, 34, 343-354.

47.

Turner

Firth

(2012). gnm: Generalized nonlinear models, R package (Version 1.0-7) [Computer software]. Retrieved from http://CRAN.R-project.org/

48.

U.S. Department of Health and Human Services. (2014). Public use data file documentation 2011-2013 National Survey of Family Growth: User’s guide. Retrieved from https://stacks.cdc.gov/view/cdc/26485

49.

U.S. Food and Drug Administration. (2016). Essure permanent birth control. Retrieved from http://www.fda.gov/MedicalDevices/ProductsandMedicalProcedures/ImplantsandProsthetics/EssurePermanentBirthControl/default.htm

50.

West

Fenstermaker

(1995). Doing difference. Gender & Society, 9, 8-37.

51.

West

Zimmerman

D. H.

(1987). Doing gender. Gender & Society, 1, 125-151.

52.

Whiteman

M. K.

Cox

Tepper

N. K.

Curtis

K. M.

Jamieson

D. J.

Penman-Aguilar

Marchbanks

P. A.

(2012). Postpartum intrauterine device insertion and postpartum tubal sterilization in the United States. American Journal of Obstetrics & Gynecology, 206, 127.e1-127.e7.

53.

Wight

V. R.

Bianchi

S. M.

Hunt

B. R.

(2012). Explaining racial/ethnic variation in partnered women’s and men’s housework: Does one size fit all? Journal of Family Issues, 34, 394-427.