Variation in Attitudes toward Being a Mother by Race/Ethnicity and Education among Women in the United States

Abstract

Do differences in experiences of motherhood (e.g., number of children, age at first child, and relationship type) by race/ethnicity and social class mean that attitudes toward motherhood also vary by social location? We examine attitudes toward being a mother among black, Hispanic, Asian, and white women of higher and lower socioeconomic status (SES, as measured by education). Results using the National Survey of Fertility Barriers (N = 4,796) indicate that, despite fertility differences, attitudes toward being a mother differ little between groups. White and Asian women have higher positive attitudes toward being a mother than black and Hispanic women. Only black women appear to distinguish between having and raising children; surprisingly, lower educated Hispanic women are less likely to think that they would be a mother, see motherhood as fulfilling, and think that it is important to have and to raise children compared with higher educated, white women.

Keywords

mothering motherhood intersectionality race/gender/class reproduction

Changes in family life in recent decades—including fertility and employment patterns—raise questions about how attitudes toward being a mother may differ among women in the United States. Among women in the United States, fertility rates for non-Hispanic white women (1.77 children per woman) are lower than those for non-Hispanic black women (1.92) and for Hispanic women (2.24; Martin et al. 2013). The largest proportion of births to white and black women are to those with at least some college (63 percent and 46 percent, respectively), whereas the largest proportion of births to Hispanic women are to those with less than a high school diploma (Livingston and Cohn 2010).

Despite differences in the patterns of motherhood by race/ethnicity and education, much of the discussion of “motherhood” is generic, with the experiences of white, middle-class women as the presumed norm and the standard against which all mothering is judged. Motherwork in the United States is shaped by cultural schemas about what it means to be a “good mother,” and dominant ideologies and social policies often support some mothers and denigrate others (Collins 1991; Segura 1994; Taylor 2011). Certain kinds of motherhood—for example, single motherhood, immigrant motherhood, welfare motherhood—are marked, and, therefore, seem unique or, often, deviant. Thus, motherhood stands as an organizing principle in society, which is simultaneously generic and specific, depending upon the context, as well as the value placed on the women enacting it. Variation in the experience and depiction of motherhood raises several questions: Do differences in social value placed on different kinds of mothers, or differences in the experience of mothering, based on intersecting social categories of race/ethnicity and education, lead to differences in attitudes toward being a mother between women in those groups? Or, despite differential social meanings and experiences, do attitudes toward being a mother vary among women, but in ways that do not depend upon race/ethnicity or education?

The idea that motherhood is in many ways synonymous with adulthood for women, and that women expect and want to be mothers, is ubiquitous. Therefore, it is possible that all women, regardless of social location, place high importance on being a mother. Yet prior research on American women shows that women do vary with regard to the importance of motherhood in their lives, with white women placing a higher importance on motherhood than black and Hispanic women (McQuillan et al. 2008). We advance research on subgroup differences in experiences of motherhood by assessing the impact of race/ethnicity and education simultaneously, using eight combinations of four race/ethnicity (white, black, Hispanic, and Asian) and two education (“higher”: 16 or more years, and “lower”: less than 16 years) groups. We anticipate that different cultural schemas about motherhood for subgroups created by intersections of race/ethnicity and education could lead to different attitudes toward being a mother (Ridgeway and Kricheli-Katz 2013). We are guided by insights into variations in mothering experiences generated by feminist and health research that use an intersectionality perspective (Choo and Ferree 2010; Collins 1991; Vespa 2009).

Understanding variation in attitudes toward being a mother by race/ethnic/education subgroups is important for several reasons. First, although attitudes are not always consonant with behavior, they may anticipate future trends in behavior, such as shifts in fertility rates or other changes in the meanings surrounding (or enactment of) motherhood, and these might vary by race/ethnic/education location. Second, variations in enactments of motherhood may be interpreted as differences in attitudes among women, when instead they reflect differences in resources, opportunities, support, or sense of entitlement. For example, having (more) children could seem to indicate highly positive attitudes toward motherhood, but, instead, reflect differential pressures on fertility, or opportunities for fertility control. Furthermore, understanding the value women place on various dimensions of motherhood could help to explain a range of motherhood behaviors, including more women choosing not to become mothers. It may also lead to policies that could support women doing the work of mothering across a wide range of social contexts. Therefore, exploring whether and how attitudes toward being a mother vary by social location is an important contribution to scholarship on motherhood.

Theoretical Framework

Intersectionality and Differences in the Mothering Experience

Social class, race/ethnicity, and culture shape mothering experiences and influence the meanings that mothers can assign to them (Collins 1994; Landry 2002). Contemporary discourse about motherhood in the United States is shaped by the (white middle-class) ideology of “intensive mothering” (Hays 1996), which suggests that mothers should invest all their material and emotional resources in their children. This ideology represents a “hegemonic form of mothering” (Taylor 2011:898), in that it dominates cultural understandings and descriptions of mothering, and obscures subcultural differences as well as inequalities in the material conditions under which motherwork occurs. This dominant perspective views “good” mothering as expert-guided, emotionally absorbing, and labor intensive. Many feminist scholars have argued that this ideal sets up unrealistic expectations for women that may be beyond the temporal and material resources of most mothers, particularly if they are employed outside the home (Douglas and Michaels 2005; Hays 1996).

Furthermore, “hegemonic mothering” is implicitly pronatalist; it assumes that all women can (and should want to) be mothers, but these pressures are differentially felt by, and exerted on, various groups of women. Poor and nonwhite women, for example, are often depicted as hyper-fertile and sexually irresponsible (Collins 1991; Roberts 1997)—particularly immigrant women and those receiving welfare payments (Bloch and Taylor 2014; Gutierrez 2008). Differences in the depictions of poor and nonwhite (compared with middle-class and white) women lead to social policies that discourage births among poor women and women of color, but promote births among white and middle-class women (Bell 2010).

Attitudes toward being a mother, then, are likely shaped by the distinct histories and experiences of race/ethnic/class groups in the United States, which are marked by a range of push/pull factors. First, women of color have often viewed domestic labor, including childbearing and child-rearing, as a form of resistance to racial oppression (Espiritu 2008), yet they must often perform this motherwork under difficult circumstances. For example, just over half of all African American households are headed by a single parent (Cherlin 2010). Black mothers also face greater unemployment than do white mothers (Hull et al. 2008), and for many women of color, blocked opportunities at work, as well as low pay and job flexibility, may force them into full-time motherhood (Damaske 2011).

Cultural discourse on mothering in recent decades has focused on the pressure felt by (white middle-class) women as they have moved into the paid labor force in larger numbers and struggled to meet the hegemonic expectations of intensive mothering (J. Williams 2000). Combining motherhood and paid work, however, has long been the reality of most lower socioeconomic status (SES) women and women of color in the United States (Landry 2002; Ornelas et al. 2009), and the demands implied by hegemonic mothering have not been accepted by all racial/ethnic groups (Collins 1991; Sutherland 2010; Taylor 2011). Black women, for example, have long embraced both mothering and paid work as part of normal femininity (Landry 2002). They do not focus exclusively on parenting their biological children; family, friends, and neighbors may serve as “other mothers” to children in their communities (Collins 1991). This “norm of solidarity and collective survival through community mothering practices” cuts across class lines (McDonald 1997:776). Indeed, data from the Current Population Survey (Kreider and Ellis 2011) reveal that, in 2009, black children were more likely to live with neither parent (8.7 percent) than either white (3.1 percent) or Hispanic children (4.0 percent).

Like black women, many Hispanic women have also seen employment as part of mothering (Segura 1994). Hispanic women are often characterized as being especially family-focused (Bengston 2001; Skogrand, Barrios-Bell, and Higginbotham 2009); they have a lower employment rate, higher fertility rate, and have children at a younger age than other racial/ethnic groups (Landivar 2013). Yet, there is a great deal of diversity among Hispanic mothers in terms of cultural background, national origin, and economic circumstances. Many Hispanic women struggle to parent with few resources. Recent immigrant mothers also face acculturation challenges (Horwitz et al. 2007; Ornelas et al. 2009) and may be unable to access the extended care networks to which they were accustomed in their native countries (Christopher 2013).

Motherhood for Asian women is also shaped by distinct cultural influences. For example, many Asian American women face great pressure to achieve occupational success, raise similarly high achieving offspring, and meet other (e.g., transnational) family obligations (Kang 2010b; B. K. Kim, Li, and Ng 2005; U. Kim and Park 2006; Lee 2009). Asian women working in low-wage jobs, however, often struggle to meet these demands, as hours spent away from home leave mothers with little time to care for their children, or to support them in their schooling. Mothers in difficult circumstances can feel like “failures,” and children are often resentful (Kang 2010a). Asian mothers also appear to face more conservative gender expectations than other racial/ethnic groups in that they are less likely to think that mothers of young children should be employed (Goldberg et al. 2012) and have lower labor force participation rates than blacks or whites when married with husbands present (Lee 2009). Of course, “Asian” is a socially constructed term that does not capture the diverse historical, cultural, and political variations between groups (e.g., Korean, Chinese, Japanese, Thai, Filipino, and Vietnamese) and within groups (e.g., regional, religious, cultural, political). Therefore, heterogeneity among women may outweigh any shared “Asian” attitudes toward motherhood (Glenn 1998).

Other research provides little evidence of racial/ethnic variation in mothering, emphasizing class differences instead. For example, motherhood appears to be central to the lives of low-income black, white, and Hispanic women because, unlike the vagaries of marriage, children provide both stability and meaning (Edin and Kefalas 2005). Moreover, women receiving welfare payments often strongly identify with the expectations of hegemonic mothering to neutralize the negative stereotypes of being a “welfare mother” (McCormack 2005). Lareau (2005) also concluded that class was more important than race in determining attitudes toward parenting. Black and white middle-class mothers practiced “concerted cultivation,” a parenting style that required pouring tremendous time and financial resources into children and their activities, in consort with the expectations of hegemonic mothering (and in contrast to the “natural growth” approach of black and white working-class and poor parents, which focused on allowing children more unstructured time with peers).

Even though many studies emphasize either class or race/ethnicity, feminist scholarship has recognized that the social organization of family and motherhood is dependent simultaneously on race, class, and gender structures; therefore, it is important to be sensitive to the unique social locations generated by the intersection(s) of these social forces (Baca Zinn and Thornton Dill 1996). Even seemingly similar outcomes are often the result of differing social pressures and opportunities. For example, over the last several decades of deindustrialization, there have been increases in the proportion of female-headed households in the United States, but the dynamics that create them differ by race and class. For working class and poor women, as well as women of color, decreasing job opportunities for potential male partners has increased the likelihood that many women will parent alone; for more affluent white women, increases in their own economic prospects have fueled the ability to choose parenthood without a partner (Baca Zinn 1990).

Given the wide range of experiences and pressures faced by women in different social locations, we examine the joint influence of race/ethnicity and class (as measured by education) on attitudes toward being a mother. The gender schema and intersectionality perspectives lead us to competing hypotheses:

Hypothesis 1: The dominance of cultural expectations of hegemonic mothering for women will contribute to consistently positive attitudes toward being a mother across all subgroups.

Hypothesis 2: The different cultural framings of women in racial/ethnic and education subgroups will contribute to more positive attitudes toward being a mother for women in groups in which motherhood is valued, celebrated, or supported, and less positive attitudes for women in groups in which motherhood is less valued, celebrated, or supported.

Data and Method

Sample

The National Survey of Fertility Barriers (NSFB)¹ is a national, population-based, random-digit-dialing (RDD) telephone survey with measures of social and health factors related to reproductive and fertility experiences among U.S. women (Johnson et al. 2009). The survey started with questions that determined eligibility for the survey. The response rate to these questions was 53 percent, typical for contemporary RDD surveys (McCarty et al. 2006). The overall response rate was lower (37 percent). These modest response rates are consistent with other RDD telephone surveys in the early 2000s (Keeter et al. 2006). Extensive comparisons of the NSFB data with other population data with in-person surveys that have higher responses rates (e.g., the National Survey of Family Growth and the Current Population Survey) show minimal bias in the NSFB (see Johnson et al. 2009 for additional details). Relevant to the current study, there is underrepresentation of less educated women and overrepresentation of more educated women. We, therefore, use survey weights to more accurately approximate population patterns.

The full sample includes completed interviews with 4,794 women aged 25 to 45 in the United States, collected between September 2004 and January 2007. Women from racial/ethnic minority groups, women who have experienced infertility, and women who are at higher risk for experiencing infertility were oversampled. The survey weight variable also adjusts for oversampling. The survey used a “planned missing” design that involved randomly assigning two thirds of the scale items to each participant to efficiently incorporate all necessary measures and minimize respondent burden. Because the scales were highly reliable and the data were missing completely at random, there was very little loss of information (Allison 2002). A single imputation using the Stata Imputation by Chained Equations (ICE) program (Royston 2005) was used to impute responses for the respondents on the items of the following missing scales, as well as for other instances of missing data. Methodological information, including the methodology report, introductory letters, interview schedules, interviewer guides, data imputation procedures, and a detailed description of the planned missing design is available at: http://sodapop.pop.psu.edu/codebooks/nsfb/wave1/. Ideally, the survey would explicitly ask about sexual minority status, but in the NSFB, women are coded as “lesbian” only if they volunteered this information; we excluded the 34 women who self-identified as lesbian for this analysis.

Concepts and Measures

The NSFB questionnaire included five items intended to measure attitudes toward the “importance of motherhood.” Four items are measured using Likert-type scales (strongly agree to strongly disagree): (1) “Having children is important to my feeling complete as a woman,” (2) “I always thought I would be a parent,” (3) “I think my life will be or is more fulfilling with children,” and (4) “It is important for me to have children.” A fifth item is measured on a scale ranging from very important to not important: (5) “How important is each of the following in your life . . . raising children?” The question about raising children was asked differently from the other importance of motherhood items. Because this question asks about whether raising children is important in one’s life, it is possible that women without children might say that children are not important in their lives even though they intend to have children in the future. To alleviate this concern, we control for number of children.

Race/ethnicity questions were asked using standard U.S. Census wording. We then constructed binary indicator variables for white, black, Hispanic, and Asian compared with white women. Women who reported “Other race” (n = 144) were excluded from the analysis. Individuals who reported more than one race were classified according to coding rules that gave first priority to identification as “Hispanic” and second priority to identification as “black.” White/higher education is the reference category.

We use education (measured in years) as an indicator for social class, a common practice among family demographers (cf. Cherlin 2010; Furstenberg 2009). To facilitate the creation of interaction terms, we dichotomized education into “higher” (16 or more years of formal schooling) and “lower” (less than 16 years of formal schooling). This use of college education as a marker of social (middle) class has been employed by other scholars investigating women’s employment and family formation/fertility patterns from an intersectional perspective (Damaske 2011; Dean, Marsh, and Landry 2013; see also Brand and Davis 2011; Musick, Brand, and Davis 2012, for reviews). Formal tests of interactions using both continuous and categorical measures of education revealed no differences in conclusions regarding race/ethnicity, education, and attitudes toward being a mother. To maintain our focus on how intersecting social locations matter for women’s attitudes toward being a mother, we present the results with education dichotomized. We use indicator variables for the joint race/ethnicity and education groups that capture meaningful subgroups of women who vary in claims to motherhood as a privileged status. In addition, in the multivariate analyses, we include measures of factors that are associated with race/ethnicity, education, and the importance of being a mother in women’s lives (McQuillan et al. 2008)—including economic, cultural, and life-course variables. We describe these control variables in Table 1.

Table 1.

Descriptions of the Control Variables.

Name	Description
Economic resources
Employment	Indicator variables constructed for full-time employment (35+ hours per week) and part-time employment (>35 hours). Reference category is no employment.
Economic hardship	Summative scale using three questions, such as: “During the last 12 months, how often did it happen that you had trouble paying the bills?” (α = .82).
Family income	Total family income in 12 categories expressed in approximately $10,000 increments, logged.
Culture variables
Nonegalitarian gender attitudes	Scale comprised of two items: (1) “It is much better for everyone if the man earns the main living and the woman takes care of the home and family.” And (2) “If a husband and a wife both work full-time they should share household tasks equally.” This second question was reverse-coded. Higher values indicate less egalitarian attitudes.
Value leisure	“How important is having leisure to enjoy your own interests?” Responses range from (1) Not at all important to (4) Very important.
Value career	“How important is being successful in your line of work?” Responses range from (1) Not at all important to (4) Very important.
Religious behavior	“How often do you attend religious services?”
Life-course variables
Relationship satisfaction	Scale constructed by averaging responses to three questions: (1) “Taking all things together, how would you describe your relationship? Would you say that it is very happy, pretty happy, or not too happy?” (2) “Have you ever thought your relationship might be in trouble? Do you feel this way now?” and (3) “Have you and your partner discussed the possibility of ending your relationship any time in the last three years?” Because these items were measured on different scales, positive responses to yes/no questions responses were coded as “3’” and negative responses were coded as “1” before creating the scale, which was then mean-centered. Those not in relationships have a value of 0 (α = .57).
Age	Measured in years.
Parity	Indicator variable constructed so that those with no children have a value of “0” and those with any children have a value of “1”; therefore, the reference category is no children.
Identifies a fertility problem	An indicator variable measured by affirmative responses to either of the following questions: “Do you think of yourself as someone who has, has had, or might have trouble getting pregnant?” or “Do you think of yourself as someone who has or has had fertility problems?”
Close to stepchildren	An indicator variable created from the question “Do you think of this child or these children as if they were your own?” for women whose spouse/partner have at least one child. Those who responded that this was completely or somewhat true were coded as 1; all others coded as 0.
Parents want	An indicator variable created from the question “It is important to my parents that I have children,” 1 = strongly agree, 0 = all other answers.
Partner wants	An indicator variable created from the question “It is important to my partner that we have children,” 1 = strongly agree, 0 = all other answers.
Most have kids	An indicator variable created from the question “Thinking about your family and friends, would you say that all, most, some, few, or none of them have kids?” 1 = strongly agree, 0 = all other answers.

Analytic Strategy

We conducted preliminary analyses to assess whether the five indicators of attitudes toward being a mother loaded in a similar manner for each subgroup, and they did not (Raja, Laffitte, and Byrne 2002).² From a scale measurement perspective, the differences in loadings indicate that the scale is not consistent across racial/ethnic subgroups. We, therefore, did not combine the items into a single scale but, instead, conducted five separate analyses, using each attitude item as a separate measure.

Because the dependent variables are ordinal, ordinary least squares (OLS) regression was not an appropriate analytic tool. We needed a method that would accommodate the ordinal level of measurement. We also needed to assess whether all of the categories of the dependent variable were distinguishable from one another. Ordinal logistic regression is appropriate only when the model meets the parallel lines (proportional odds) assumption (Winship and Mare 1984). This assumption requires that the slopes predicting values of the dependent variable are parallel for every level of the dependent variable. Using the Rollin Brant (1990) technique, we determined that the parallel lines assumption did not hold for this set of variables. Therefore, we used stereotype logistic regression (Long and Freese 2006), an alternative form of ordinal logistic regression that does not require the proportional odds assumption. We determined that the categories of the dependent variables were distinguishable from each other; the phi test showed that all of the terms operated as ordinal variables. Therefore, stereotype logistic regression is the most appropriate method for this analysis. In the stereotype logistic regression results, we use “white/higher education” as the reference category because this is the implicit, but often “unmarked,” focal group in motherhood research. Because with multiple comparisons, it is possible to have significant associations by chance, some scholars use Bonferroni corrections. We follow the guidance of John D. Williams (1971), who argues that a multiple regression approach is equivalent to the Dunnett test, a more appropriate adjustment strategy.

Results

Descriptive and Bivariate Statistics

We first examined descriptive statistics for each race/ethnic/educational subgroup in Table 2. We present means and standard deviations for quantitative variables and proportions for categorical variables. We also tested for differences among the race/ethnicity/education subgroups using analysis of variance for mean differences and chi-square tests for the categorical variables. The bivariate associations between independent variables and the dependent variables provide information about which control variables should be included in the final analyses.

Table 2.

Descriptive and Bivariate Statistics for the Variables in the Analyses by Race/Ethnicity and Education.

Variables	White				Black				Hispanic				Asian
	Education LT 16yrs		Education 16yrs+		Education LT 16yrs		Education 16yrs+		Education LT 16yrs		Education 16yrs+		Education LT 16yrs		Education 16yrs+
	(n = 1,876)		(n = 1,005)		(n = 534)		(n = 167)		(n = 774)		(n = 131)		(n = 75)		(n = 171)
	M	SD	M	SD	M	SD	M	SD	M	SD	M	SD	M	SD	M	SD
Attitude toward being a mother measures
Feel complete as a woman	3.11	0.83	2.94	0.93	2.97	0.88	2.84	0.92	3.05	0.80	3.01	0.91	3.03	0.88	3.08	0.80
Always thought I’d be a mother	3.32	0.78	3.29	0.84	3.10	0.87	3.21	0.83	3.14	0.72	3.26	0.88	3.32	0.72	3.33	0.76
Life more fulfilling as a mother	3.52	0.69	3.43	0.78	3.36	0.69	3.19	0.86	3.35	0.64	3.37	0.70	3.44	0.50	3.46	0.72
Important to have children	3.34	0.75	3.29	0.83	3.06	0.77	3.03	0.85	3.19	0.64	3.18	0.83	3.45	0.58	3.36	0.80
Important to raise children	3.74	0.65	3.59	0.86	3.69	0.64	3.59	0.78	3.66	0.58	3.56	0.82	3.81	0.49	3.51	0.83
Economic
Not employed	0.36		0.22		0.36		0.15		0.48		0.30		0.30		0.21
Employed FT	0.48		0.58		0.51		0.76		0.35		0.60		0.57		0.68
Employed PT	0.16		0.20		0.12		0.09		0.17		0.09		0.13		0.11
Economic hardship	1.97	1.77	1.03	1.30	2.54	2.02	1.76	1.93	2.21	1.85	1.38	1.58	2.90	2.61	0.94	1.32
Family income	8.08	2.77	9.94	2.11	5.77	2.96	8.90	2.34	5.79	2.86	9.23	2.46	7.24	3.10	9.64	2.28
Culture
Nonegalitarian gender	1.96	0.55	1.77	0.53	1.89	0.52	1.74	0.53	2.12	0.48	1.82	0.60	1.97	0.53	1.97	0.54
Value leisure	3.11	0.88	3.24	0.83	3.29	0.82	3.44	0.77	3.24	0.78	3.27	0.80	3.29	0.88	3.14	0.75
Value career	3.15	0.89	3.23	0.84	3.52	0.71	3.51	0.70	3.42	0.70	3.54	0.66	3.58	0.57	3.30	0.60
Religious behavior	4.31	2.35	4.31	2.22	4.54	2.31	4.98	2.13	4.52	2.27	4.62	2.26	4.96	2.23	3.73	2.30
Life course
Not in a relationship	0.19		0.20		0.55		0.52		0.24		0.27		0.38		0.27
Has a partner	0.81		0.80		0.45		0.48		0.76		0.73		0.62		0.73
Relationship Satisfaction	0.05	0.46	0.02	0.45	−0.03	0.36	−0.02	0.38	−0.04	0.51	−0.06	0.58	−0.13	0.45	−0.01	0.36
Age	36.26	5.98	35.66	5.89	35.20	6.45	35.03	6.13	33.87	5.69	34.76	6.00	36.18	6.08	34.04	5.38
Has no children	0.16		0.33		0.13		0.32		0.07		0.28		0.10		0.45
Has 1 child	0.18		0.20		0.17		0.24		0.14		0.23		0.22		0.32
Has 2 children	0.34		0.33		0.29		0.27		0.29		0.25		0.67		0.19
Has 3+ children	0.32		0.15		0.40		0.17		0.50		0.23		0.01		0.05
No identification of a fertility problem	0.71		0.75		0.71		0.75		0.78		0.74		0.65		0.66
Identifies a fertility problem	0.29		0.25		0.29		0.25		0.22		0.26		0.35		0.34
No close relationship with stepchild	0.85		0.93		0.83		0.89		0.90		0.90		1.00		0.99
Close to stepchild	0.15		0.07		0.17		0.11		0.10		0.10		0.00		0.01
Parents want grandchildren	2.97	0.879	3.01	0.848	2.60	0.919	2.71	0.976	2.85	0.749	3.00	0.808	3.39	0.763	3.25	0.789
Partner wants	0.60	1.501	0.63	1.508	−0.64	1.591	−0.52	1.653	0.31	1.481	0.34	1.595	0.20	1.779	0.30	1.707
Most have kids	0.94	0.736	0.57	0.790	0.92	0.899	0.82	0.742	1.20	0.774	0.77	0.847	1.20	0.651	0.38	0.893

Note. NSFB Women ages 25 to 45; excluding women in the “Other” race/ethnicity category. LT = less than; FT = full-time; PT = part-time; NSFB = National Survey of Fertility Barriers; ANOVA = analysis of variance.

All ANOVA and chi-square tests comparing the eight race/ethnicity-education groups are significant at the .001 level.

Within each race/ethnicity group, attitudes toward being a mother differ by level of education. Among white women, less educated women have more positive attitudes than higher educated women. The pattern is similar for black women, with the exception that more educated black women had higher attitude scores than less educated women for the item always thought that I’d be a mother. For Hispanic women, the attitude scores for less and more educated women are similar for complete as a woman, motherhood fulfilling, and important to have children. For the question always thought that I’d be a mother, however, more educated women have higher attitude scores, while for important to raise children, less educated women have higher attitude scores. Among Asian women, there are differences by education for only two questions—important to have children and important to raise children. For both, less educated women have higher attitude scores. These patterns suggest that it is important to attend to the interaction between race/ethnicity and education in trying to understand attitudes toward being a mother among U.S. women. All of the other independent variables also differ by race/ethnicity and education. Because many of these are likely to be associated with attitudes toward being a mother, we include these variables in the stereotype logistic regression analyses that we summarize next.

Stereotype Logistic Regression Results

Table 3 provides the stereotype regression results of the race/ethnicity by education groups regressed on the measures of attitudes toward being a mother, adjusted for economic, culture, and life-course measures. There are differences by race/ethnicity/education compared with white women with higher education for each of the measures of attitudes toward being a mother. Which groups have significant differences are measure-specific. For example, white women with less education have higher adjusted complete as a woman scores than more educated white women. Hispanic women with less education have lower scores on all of the other measures: thought I would be a mother, motherhood fulfilling, important to have children, and important to raise children. Black women of any level of education have lower scores on the motherhood fulfilling and important to have children measures than more educated white women. Black women with more education have lower scores on motherhood fulfilling than more educated white women.

Table 3.

Stereotype Logistic Regression Models of Measures of Importance of Motherhood by Race/Education Interactions.

Variables	Complete as a woman			Thought would be a mother			Motherhood fulfilling			Important to HAVE children			Important to RAISE children
Variables	B	OR	p > z	B	OR	p > z	B	OR	p > z	B	OR	p > z	B	OR	p > z
Race/education interactions
White, 16 plus	R			R			R			R			R
White, less than 16	0.55	1.73	.002	0.07	1.08	.618	−0.04	0.96	.850	−0.12	0.89	.593	.34	1.40	.135
Black, 16 plus	0.15	1.16	.551	−0.10	0.91	.647	−0.90	0.41	.001	−0.56	0.57	.057	0.38	1.46	.194
Black, less than 16	0.24	1.28	.296	−0.18	0.83	.347	−0.63	0.53	.018	−0.89	0.41	.001	−0.02	0.98	.950
Hisp., 16 plus	0.19	1.21	.499	−0.34	0.71	.158	−0.55	0.57	.090	−0.66	0.52	.060	−0.09	0.92	.801
Hisp., less than 16	−0.14	0.87	.556	−0.92	0.40	.000	−1.36	0.26	.000	−1.69	0.18	.000	−1.20	0.30	.000
Asian, 16 plus	0.82	2.26	.080	0.09	1.09	.820	0.10	1.11	.841	0.36	1.43	.532	0.50	1.65	.335
Asian, less than 16	0.75	2.11	.500	0.17	1.19	.839	−0.54	0.58	.659	0.24	1.27	.862	3.74	41.94	.126
Rational choice and economic
Not employed	R			R			R			R			R
Employed FT	0.06	1.06	.713	−0.06	0.94	.645	−0.11	0.89	.525	0.05	1.05	.791	−0.23	0.79	.303
Employed PT	0.17	1.18	.436	0.05	1.05	.772	0.10	1.10	.698	0.11	1.12	.662	−0.12	0.89	.719
Economic hardship	0.01	1.01	.742	−0.01	0.99	.814	−0.02	0.98	.720	0.06	1.07	.169	−0.06	0.94	.256
Family income	−0.04	0.96	.176	0.03	1.03	.217	0.03	1.03	.291	−0.07	0.93	.029	−0.03	0.98	.494
Culture and identity
Nonegalitarian gender attitudes	0.62	1.87	.000	0.05	1.05	.643	−0.37	0.69	.007	0.23	1.26	.123	0.53	1.70	.001
Value leisure	−0.15	0.86	.061	−0.16	0.85	.014	−0.25	0.78	.009	−0.26	0.77	.007	−0.10	0.90	.388
Value career	0.15	1.17	.056	0.15	1.17	.023	0.03	1.03	.769	0.27	1.31	.007	0.66	1.93	.000
Religious behavior	0.06	1.06	.028	0.09	1.10	.000	0.15	1.16	.000	0.10	1.10	.004	0.24	1.27	.000
Life course and situational
Relationship satisfaction	0.11	1.11	.498	−0.11	0.89	.372	0.09	1.10	.617	0.07	1.07	.726	0.32	1.37	.210
Age	−0.02	0.98	.034	−0.01	0.99	.162	−0.02	0.98	.050	−0.06	0.94	.000	−0.10	0.91	.000
Has 1 child or more	2.00	7.41	.000	1.12	3.07	.000	3.93	50.97	.000	3.39	29.62	.000	5.46	235.95	.000
Identifies a fertility problem	0.48	1.62	.000	0.29	1.34	.010	0.76	2.14	.000	0.73	2.07	.000	0.83	2.30	.000
Close to stepchild	−0.02	0.98	.920	−0.16	0.85	.331	0.15	1.17	.514	−0.67	0.51	.005	0.27	1.31	.381
Parents want grandchildren	0.88	2.41	.000	0.86	2.37	.000	1.10	3.01	.000	1.94	6.95	.000	0.62	1.86	.000
Partner wants child	0.27	1.31	.000	0.18	1.20	.000	0.38	1.46	.000	0.85	2.33	.000	0.46	1.58	.000
Most friends/family have kids	0.05	1.05	.525	0.06	1.06	.408	0.20	1.22	.027	0.10	1.11	.287	0.12	1.12	.232

Note. Data are the NSFB. All continuous variables are mean centered. FT = full-time; PT = part-time; OR = odds ratio; NSFB = National Survey of Fertility Barriers.

Coefficients in bold are statistically significant (p < .05). All tests are two-tailed.

The analyses also provide the associations for all of the economic resource, culture, and life-course variables with the measures of attitudes toward being a mother. Among the economic measures, only family income is associated with a measure of importance of being a mother: important to have children. Greater family income is associated with slightly lower scores on important to have children.

Several culture variables are associated with measures of attitudes toward being a mother. Women with nonegalitarian gender attitudes have higher complete as a woman and important to raise children scores than women with egalitarian gender attitudes, but lower scores on motherhood fulfilling. Therefore, less egalitarian women may see motherhood as important even if they find motherhood less personally rewarding. Women who value leisure have lower scores on thought would be a mother, motherhood fulfilling, and important to have children, and women who place higher value on career success have higher scores on thought would be a mother, important to have children, and important to raise children. Consistent with prior research, higher levels of religiosity are associated with more positive attitudes toward being a mother; the association is significant for the always thought that I’d be a mother, motherhood fulfilling, and raising children is important measures.

Several life-course variables are associated with attitudes toward being a mother. Greater age is associated with lower scores for complete as a woman and important to have children. Having children is associated with higher scores on all of the measures of attitudes toward being a mother. Women who perceive a fertility problem also have higher scores on all of the measures of attitudes toward being a mother compared with women who do not identify a fertility problem. Women who perceive that their parents want grandchildren and that their partner wants a child have higher scores than those who do not on all of the measures. Women who say that most of their friends have children score higher on motherhood fulfilling.

Comparisons of Adjusted Means

Table 4 reports adjusted means (predicted scores after control variables are taken into account) and standard errors for the eight combinations of race/ethnicity and education. To determine which groups were significantly different from one another, we compared all of the groups to each other (rather than to just one reference category) using the post-hoc probing simple slopes approach (Holmbeck 2002). To do this, we estimated a series of eight OLS regression models for each dependent variable, rotating the reference category to determine which groups were significantly different from each other.

Table 4.

Summary of Adjusted Means with Standard Errors by Race/Ethnicity and Education for Importance of Motherhood Variables.

Variables	1		2		3		4		5
	“Having children is important to my feeling complete as a woman”		“I always thought that I’d be a mother”		“I think my life is or will be more fulfilling with children”		“It is important for me to have children”		“How important is each of the following in your life: Raising children”
	(1 = SD to 4 = SA)		(1 = SD to 4 = SA)		(1 = SD to 4 = SA)		(1 = SD to 4 = SA)		(1 = not important to 4 = very important)
	Adjusted M	SE	Adjusted M	SE	Adjusted M	SE	Adjusted M	SE	Adjusted M	SE
White Higher (W Hi)	2.94	.20	3.28	.02	3.43	.02	3.27	.02	3.66	.02
White Lower (W Low)	3.05	.25	3.31	.06	3.44	.05	3.27	.04	3.66	.05
Black Higher (B Hi)	2.92	.26	3.24	.07	3.26	.06	3.12	.04	3.65	.06
Black Lower (B Low)	2.98	.25	3.23	.07	3.35	.06	3.13	.04	3.65	.06
Hispanic Higher (H Hi)	2.98	.29	3.19	.08	3.36	.06	3.19	.04	3.65	.07
Hispanic Lower (H Low)	2.94	.40	3.12	.07	3.29	.06	3.11	.04	3.62	.06
Asian Higher (A Hi)	3.09	.22	3.32	.11	3.45	.09	3.30	.04	3.61	.09
Asian Lower (A Low)	3.04	.20	3.30	.21	3.32	.17	3.24	.04	3.52	.18
	*p < .05Differences:		*p < .05Differences:		*p < .05Differences:		*p < .05Differences:		*p < .05Differences:
	W Low > W Hi, B Hi, H Low		W Hi, W Low, B Hi, B Lo, A Hi > H Low;		W Hi, W Low > B Hi, B Low, H Low		W Hi, W Low > B Hi, B Low, H Hi, H Low;		W Hi, W Low, B Hi, B Low, A Hi > H Low
			W Low > H Hi				W Low > H Hi

Note. NA=strongly agree: SD=strongly disagree; Significance test of the differences in the adjusted means between race/education groups were estimated by rerunning the model with each group taking a turn as the reference group. All tests are two-tailed

Table 4 summarizes significant differences between groups on each attitude toward being a mother measure. We summarize the groups that are significantly different from each other by using shorthand descriptions for the groups and symbols. For example, “W Hi, W Low, B Hi, B Low, A Hi > H Low” indicates that white women with more education, white women with less education, black women with more education, black women with less education, and Asian women with more education all have significantly higher means than Hispanic women with less education on a particular variable, adjusted for other variables in the model. To some degree, the patterns depended upon the specific measure, justifying our decision to not simply create a combined scale.

For the measure complete as a woman, less educated white women have significantly higher scores than more educated white women and less educated Hispanic women. For the measure always thought that I’d be a mother, white women, black women, and more educated Asian women have higher scores than less educated Hispanic women. For the measure motherhood fulfilling, white women and more educated Asian women have higher scores than more educated black women and less educated Hispanic women. Less educated black women and more educated Hispanic women also have higher scores on this measure than less educated Hispanic women. For the measure important to have children, white women had higher scores than less educated black or Hispanic women. Black women, more educated Hispanic women, and more educated Asian women had higher scores than less educated Hispanic women. Finally, for the measure important to raise children, white and black women and more educated Hispanic and Asian women have significantly higher scores than less educated Hispanic women.

Overall, Asian women had the highest scores. It is interesting that white women with more education had one of the highest scores for having children but one of the lowest scores for raising children, ahead of only less educated Hispanic women. Consistent with the idea that some women face more challenges raising children (e.g., black and Hispanic mothers must help their children confront racism in the United States), black and Hispanic women of all education levels had lower scores for motherhood fulfilling.

Discussion and Conclusion

We explored attitudes toward being a mother among subgroups of women created from the intersection of the structures of social class (as measured by education) and race/ethnicity. We found support for both the hypothesis that attitudes toward motherhood scores would be relatively high across groups and the hypothesis that scores would vary by race/ethnicity/education groups. Accounting for cultural, life-course, and economic measures, we highlight three main findings. First, as we hypothesized, scores across the indicators of attitudes toward being a mother are fairly high for all subgroups of women. This is a valuable finding because it suggests that, despite whatever differences exist between race/ethnic/education groups in how motherhood is viewed, valued, and enacted, women across these groups tend to have positive attitudes toward being a mother. Second, our results suggest that some of the assumptions embedded in the ideology of intensive mothering are less salient for some women, which may suggest future shifts in the ideology itself or, perhaps, in mothering practices. Third, the differences in attitudes toward motherhood scores among the groups are not large, and are not consistent across race/ethnic class groups. Of particular note are the high scores for less educated white women, the low scores for less educated Hispanic women, the high scores for Asian women, and the differences between having and raising children for black and white women.

The relatively high scores across the groups on all measures suggest that attitudes toward being a mother are generally positive, and that the ideology of intensive mothering remains influential in women’s thinking about motherhood. One set of scores, however, suggests that some of the assumptions embedded in this ideology are becoming less salient for women. The lowest attitudes toward motherhood scores across all groups are for the question about motherhood making one feel complete as a woman. Because motherhood has often been conflated with successful adult femininity, these results suggest a loosening of the link between motherhood and femininity for women across race/ethnic and education groups. Due to increased educational and occupational opportunities for women over the last several decades, women in all of the groups may be less likely to see motherhood as their primary source of adult identity. Comparisons of attitudes among women across several cohorts would help to clarify whether there has been change over time.

The variations in attitudes toward motherhood scores by group are sometimes surprising and raise interesting questions. For example, we did not expect white women with less education to be near the top of the scores for all outcomes. Only for one measure (important to have children) did white women with more education have higher scores than white women with less education, but the difference was not significant. Less educated white women also score higher on complete as a woman than most other groups. Several factors may help explain this finding. We know that some lower-income women across race/ethnic groups enjoy fewer opportunities for educational and occupational success, and that some feel strongly that motherhood “saved” them; without their children, they would be dead or in jail (Edin and Kefalas 2005). For less educated white women, it may also be easier for them to “pass” as middle class (McCormack 2005), thus, avoiding the combined class stigmatization and racialized stereotypes facing poor women of color. Less educated white women have also been spared the harshest application of restrictive birth control and sterilization policies imposed on poor Hispanic and black women (Gutierrez 2008; Roberts 1997). Therefore, they may be able to use race privilege and motherhood to create a more “successful” feminine identity (complete as a woman).

Another rather surprising finding is that less educated Hispanic women had lower attitudes toward motherhood scores compared with other groups across all five indicators. These findings challenge the notion of Hispanic women as especially oriented toward motherhood, as does the fact that Hispanic women’s fertility rates declined most dramatically during the recent economic recession (Livingston 2012). These patterns suggest that being a mother may be especially difficult for less educated Hispanic women for a number of reasons. First, Hispanic women are likely to work in occupations that are the most poorly paid and provide the least flexibility for meeting family obligations (Landivar 2013). Second, it may be that the expected extended care networks that facilitate good mothering in some Hispanic cultural traditions (or native countries) are not available to contemporary U.S. Hispanic women, making motherhood both less desirable and less manageable for less educated Hispanic women in the United States at this time (Christopher 2013). Third, the anti-immigrant rhetoric of recent decades, and negative stereotypes of Mexican women (in particular) as “breeders,” combined with policies to block their access to reproductive care (Gutierrez 2008), may lower their scores on these measures of attitudes toward being a mother. Fourth, these results may indicate that motherhood is less closely linked to Hispanic women’s identities, in comparison with white, black, and Asian women. Fifth, it could be that the measures we have of the importance of being a mother do not capture the way that Hispanic women think about motherhood. These possible explanations raise more questions than they answer, thus, indicating the need for further investigation to understand the meanings behind the finding that, despite higher fertility rates, Hispanic women tend to have lower scores on these measures of attitudes toward being a mother.

For most outcomes, white women and more educated Asian women had the highest scores. We expected white women to have positive attitudes toward motherhood because white women’s motherhood has been most highly valued in the United States. Finding more positive attitudes among more educated Asian women was surprising. We expected somewhat lower scores for Asian women, because Asians in the United States are “racially triangulated” between blacks and whites, superior to the former and inferior to the latter (Espiritu 2008:124). Asian women have often been negatively stereotyped as either promiscuous and exotic, or diminutive and submissive. They have not, however, been attacked as bad mothers or hyper-fertile in the same way as Hispanic and black women (Gutierrez 2008; Roberts 1997), and this may account for Asian women’s scores being higher than those of black and Hispanic women. In addition, Asian women are highly successful in the labor market. They are more likely to work in higher status occupations and less likely to be unemployed than women in other racial/ethnic groups (U.S. Department of Labor, Bureau of Labor Statistics 2010), and in our own data, valuing career success is associated with higher scores on attitudes toward being a mother. Both Asian and white women are also most likely to work in managerial or professional occupations, which means they may be in a better position to control at least some of the terms or conditions of their employment. This may then have a positive impact on attitudes toward being a mother.

Another striking difference between groups is that, across both education groups, both white and black women place a high importance on raising children, yet white women place a greater importance on having children than do black women. This distinction between having and raising children could be interpreted to mean that black women expect to raise children that may not be “their own.” These findings may be related to both unique structural constraints and cultural expectations faced by black women. It is possible that attitudes toward having and raising children may be related to differential rates of success in matching fertility outcomes to fertility desires (Littlejohn 2012). College-educated black women, in particular, face a range of barriers in family formation (marriage and childbearing). They have a more difficult time finding romantic partners and are, therefore, more likely to remain celibate and unmarried and to spend a smaller number of their reproductive years in marriage, compared with their white and Hispanic peers, all of which depresses fertility (Clarke 2011). Our findings also appear consistent with prior evidence that black women expect (or are expected) to engage in more communal mothering by serving as “other mothers” to the children of other women (Collins 1991; Dean et al. 2013; Roberts 1997). This valuing of communal mothering has not been central to the ideology of intensive mothering (e.g., Hays 1996) or to social policies designed to support families.

The potential uncoupling of having and raising children is important because it could challenge the notion of motherhood (and fatherhood) as private activities. Valuing, emphasizing, and encouraging more collective understandings of societal obligations toward children has social and policy implications. As most federal, state, and employer policies are currently written, family benefits—particularly family leave policies—accrue to women who bear (or adopt) children. This means that women who serve as “other mothers” to the children of their relatives, friends, and neighbors, who are stepmothers, or who care for children whose parents are incarcerated or incapacitated, are not entitled to the same cultural legitimacy or practical support as birth or adoptive mothers, which has a negative impact on the well-being of both caregivers and children. Recent conversations about diversity in family life have contributed to policy shifts, most notably the legal recognition of same-sex marriage in the United States. Paying closer attention to the attitudes toward (and experiences of) mothering across race/ethnic/class groups (i.e., “shifting the center” of motherhood discussions; Collins 1994) could generate a similar policy shift that would provide more meaningful support for all mothers and families.

As with all research, however, there are limitations to this study, and we are cautious about interpreting some of our results. For example, we may be incorrect in our interpretations of the meaning of “having” and “raising” children among our respondents. We also are unable to address the full range of variation by race/ethnic/class groups. With a larger sample, we could have explored finer gradations of education, which could be important for racial/ethnic subgroups in which the proportions of women earning various levels of education differ considerably. We ran analyses, however, using different cutoff points for level of education (e.g., less than a high school degree, or some college or less) and education as a continuous measure, and the overall patterns were similar. Yet, the education differences are fairly small. In addition to oversimplifying education, we necessarily oversimplified racial/ethnic groups. We expect that if we could explore more homogeneous subgroups within the broader categories, we would find valuable information based upon history, culture, and routes to the United States among Hispanic (e.g., Puerto Rican, Cuban, Mexican, etc.), black (e.g., West Indies, Caribbean, Somali, etc.), and Asian (e.g., Japanese, Korean, Chinese, etc.) women.

Even with these limitations, this exploration of race/ethnic/class differences in women’s attitudes toward being a mother provides several valuable insights. In attempting to explain these differences, we have drawn on qualitative and quantitative research that examines variation in the contexts and experiences of motherhood, as well as employment and family formation patterns. Future research on attitudes toward being a mother should bring together these various explanatory threads by simultaneously examining greater diversity in racial/ethnic/class groups, the structural constraints faced by them in family formation and maintenance, and the cultural depictions of, and expectations for, mothering across these race/ethnic/class locations (Collins 1994).

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) disclosed receipt of the following financial support for research, authorship, and/or publication of this article: This research was supported in part by grant R01-HD044144 “Infertility: Pathways and Psychosocial Outcomes” funded by the National Institute of Child Health and Development (NICHD). Dr. Lynn White (The University of Nebraska–Lincoln) and Dr. David R. Johnson (The Pennsylvania State University) were co-PIs (principal investigators) on the first wave of data collection.

Notes

Author Biographies

Veronica Tichenor is associate professor of sociology at the State University of New York (SUNY)–Polytechnic Institute. Her research interests include the importance of motherhood and fatherhood (identities) in contemporary society, as well as a wide range of power dynamics in families, including the impact of Adverse Childhood Experiences (ACE) on a range of behaviors and health outcomes, such as bullying and pregnancy complications.

Julia McQuillan is professor and chair of the Department of Sociology at the University of Nebraska. She was co-principal investigator on the ADVANCE-Nebraska National Science Foundation (NSF) grant, is helping to lead a team studying social dimensions fertility and infertility, and collaborates with a multidisciplinary group studying gender and race inequities in science identities.

Arthur L. Greil is professor of sociology at Alfred University. He has written extensively on psychosocial aspects of infertility. He has also written on fertility intentions, pregnancy loss, and the importance of motherhood. In addition, he does research in the sociology of religion.

Andrew V. Bedrous is assistant professor of sociology at Kansas Wesleyan University in Salina, Kansas. He has been working with other scholars to study fertility intentions and motherhood. He is also interested in studying factors that influence international environmental attitudes.

Amy Clark is a visiting associate professor in the Social Work Department at Nebraska Wesleyan University. Currently, she is a researcher for University of Nebraska Medical Center on the Dementia Care Ecosystem Project, a three-year, $10 million Healthcare Innovations Award from the Centers for Medicare & Medicaid, in partnership with the University of California–San Francisco.

Karina M. Shreffler is an associate professor of human development and family science at Oklahoma State University. She studies decisions and behaviors regarding childbearing and the relationships among childbearing, work, and well-being, including a focus on fertility barriers.

References

Allison

Paul D.

2002. Missing Data. Thousand Oaks, CA: Sage Publications.

André

Stefanie

Gesthuizen

Maurice

Scheepers

Peer

. 2013. “Support for Traditional Female Roles across 32 Countries: Female Labour Market Participation, Policy Models, and Gender Differences.” Comparative Sociology 12(4):447–476.

Baca Zinn

Maxine

. 1990. “Family, Feminism, and Race in America.” Gender & Society 4(1):68–82.

Baca Zinn

Maxine

Dill

Bonnie Thornton

. 1996. “Theorizing Difference from Multiracial Feminism.” Feminist Studies 22(2):321–331.

Bell

Ann V.

2010. “Beyond (Financial) Accessibility: Inequalities within the Medicalisation of Infertility.” Sociology of Health & Illness 32(4):631–646.

Bengston

Vern L.

2001. “Beyond the Nuclear Family: The Increasing Importance of Multigenerational Bonds.” Journal of Marriage and Family 63(1):1–16.

Bloch

Katrina

Taylor

Tiffany

. 2014. “Welfare Queens and Anchor Babies: A Comparative Study of Stigmatized Mothers in the United States.” Pp. 199–210 in Mothering in the Age of Neoliberalism, edited by Giles

Melinda Vandenbeld

. Toronto, Ontario, Canada: Demeter Press.

Brand

Jennie

Davis

Dwight

. 2011. “The Impact of College Education on Fertility: Evidence for Heterogeneous Effects.” Demography 48(3):863–887.

Brant

Rollin

. 1990. “Assessing Proportionality in the Proportional Odds Model for Ordinal Logistic Regression.” Biometrics 46(4):1171–1178.

10.

Cherlin

Andrew

. 2010. “Demographic Trends in the United States: A Review of Research in the 2000s.” Journal of Marriage and Family 72(3):403–419.

11.

Choo

Hae Yeon

Ferree

Myra Marx

. 2010. “Practicing Intersectionality in Sociological Research: A Critical Analysis of Inclusions, Interactions and Institutions in the Study of Inequalities.” Sociological Theory 28(2):129–149.

12.

Christopher

Karen

. 2013. “African-Americans’ and Latinas’ Mothering Scripts: An Intersectional Analysis.” Pp. 187–208 in Notions of Family: Intersectional Perspectives, edited by Kohlman

Marla

Krieg

Dana

Dickerson

Bette

. Bingley, England: Emerald Group.

13.

Clarke

Averil

. 2011. Inequalities of Love: College-educated Black Women and the Barriers to Romance and Family. Durham, NC: Duke University Press.

14.

Collins

Patricia Hill

. 1991. Black Feminist Thought: Knowledge, Consciousness and Empowerment. Boston, MA: Unwin Hyman.

15.

Collins

Patricia Hill

. 1994. “Shifting the Center: Race, Class, and Feminist Theorizing about Motherhood.” Pp. 56–74 in Representations of Motherhood, edited by Basin

Donna

Honey

Margaret

Kaplan

Meryle Mahrer

. New Haven, CT: Yale University Press.

16.

Damaske

Sarah

. 2011. For the Family: How Class and Gender Shape Women’s Work. New York: Oxford University Press.

17.

Dean

Paul

Marsh

Kris

Landry

Bart

. 2013. “Cultural Contradiction or Integration? Work-family Schemas of Black Middle Class Mothers.” Pp. 137–158 in Notions of Family: Intersectional Perspectives, edited by Kohlman

Marla

Krieg

Dana

Dickerson

Bette

. Bingley, England: Emerald Group.

18.

Douglas

Susan J.

Michaels

Meredith W.

2005. The Mommy Myth. New York: Free Press.

19.

Edin

Kathryn

Kefalas

Maria

. 2005. Promises I Can Keep: Why Poor Women Put Motherhood Ahead of Marriage. Berkeley, CA: University of California Press.

20.

Espiritu

Yen Le

. 2008. Asian American Women and Men: Labor, Laws, and Love. 2nd ed. New York: Rowman & Littlefield.

21.

Furstenberg

Frank F.

2009. “If Moynihan Had Only Known: Race, Class, and Family Change in the Late Twentieth Century.” The ANNALS of the American Academy of Political and Social Science 621:94–110.

22.

Glenn

Evelyn Nakano

. 1998. The Social Construction and Institutionalization of Gender and Race: An Integrative Framework. Pp. 3–45 in Revisioning Gender: New Directions in the Social Sciences, edited by Ferree

Myra Marx

Hess

Beth B.

Lorber

Judith

. Los Angeles, CA: Sage Publications.

23.

Goldberg

Wendy A.

Kelly

Erin

Matthews

Nicole L.

Kang

Hannah

Weilin

Sumaroka

Mariya

. 2012. “The More Things Change, the More They Stay the Same: Gender, Culture, and College Students’ View about Work and Family.” Journal of Social Issues 68(4):814–837.

24.

Gutierrez

Elena

. 2008. Fertile Matters: The Politics of Mexican-origin Women’s Reproduction. Austin, TX: University of Texas Press.

25.

Hays

Sharon

. 1996. The Cultural Contradictions of Motherhood. New Haven, CT: Yale University Press.

26.

Holmbeck

Grayson N.

2002. “Post-hoc Probing of Significant Moderational and Mediational Effects in Studies of Pediatric Populations.” Journal of Pediatric Psychology 27(1):87–96.

27.

Horwitz

Sarah McCue

Briggs-Gown

Margaret J.

Storfer-Isser

Amy

Carter

Alice S.

2007. “Prevalence, Correlates and Persistence of Maternal Depression.” Journal of Women’s Health 18(5):637–645.

28.

Hull

Pamela

Kilbourne

Barbara

Reece

Michelle

Husaini

Baqar

. 2008. “Community Involvement and Adolescent Mental Health: Moderating Effects of Race/Ethnicity and Neighborhood Disadvantage.” Journal of Community Psychology 36(4):534–551.

29.

Johnson

David R.

McQuillan

Julia

Jacob

Mary Casey

Greil

Arthur L.

Lacy

Naomi

Scheuble

Laurie K.

2009. National Survey of Fertility Barriers Methodology Report. Retrieved June 22, 2016 (http://sodapop.pop.psu.edu/nsfb_page1.html).

30.

Kang

Miliann.

2010a. “Asian American Women and the ‘Opting Out’ Debate.” American Sociological Association Annual Meeting, Atlanta, GA.

31.

Kang

Miliann

. 2010b. The Managed Hand: Race, Gender, and Body in Beauty Service Work. Berkeley, CA: University of California Press.

32.

Keeter

Scott

Kennedy

Courtney

Dimock

Michael

Best

Jonathan

Craighill

Peyton

. 2006. “Gauging the Impact of Growing Non-response on Estimates from a National RDD Telephone Survey.” Public Opinion Quarterly 70(5):759–779.

33.

Kim

Brian K.

Lisa C.

Gladys F.

2005. “The Asian American Value Scale—Multidimensional: Development, Reliability and Validity.” Cultural Diversity & Ethnic Minority Psychology 11(3):187–201.

34.

Kim

Uichol

Park

Young-Shin

. 2006. “Indigenous Psychological Analysis of Academic Achievement in Korea: The Influence of Self-efficacy, Parents and Culture.” International Journal of Psychology 41(4):287–292.

35.

Kreider

Rose M.

Ellis

Renee

. 2011. Living Arrangements of Children, 2009. Current Population Reports, P70-126. Washington, DC: U.S. Census Bureau.

36.

Landivar

Liana Christin

. 2013. “Labor Force Participation among Asian, Black, Hispanic, and White Mothers in 20 Occupations.” Notions of Family: Intersectional Perspectives, Advances in Gender Research 17:263–286.

37.

Landry

Bart

. 2002. Black Working Wives. Berkeley, CA: University of California Press.

38.

Lareau

Annette

. 2005. Unequal Childhoods. Berkeley, CA: University of California Press.

39.

Lee

Sehwa.

2009. “Overcoming Invisible Whiteness in the Discourse of Asian American Women’s Motherhood, Marital Status, and Labor Force Participation.” American Sociological Association Annual Meeting, San Fransisco, CA.

40.

Littlejohn

Krystale

. 2012. “Hormonal Contraceptive Use and Discontinuation Because of Dissatisfaction: Differences by Race and Education.” Demography 49(4):1433–1452.

41.

Livingston

Gretchen

. 2012. In a Down Economy, Fewer Births. A Social & Demographic Trends Report. Pew Research Center Retrieved June 22, 2016 (http://www.pewsocialtrends.org/2011/10/12/in-a-down-economy-fewer-births/).

42.

Livingston

Gretchen

Cohn

D’Vera

. 2010. The New Demography of American Motherhood. A Social & Demographic Trends Report. Pew Research Center. Retrieved June 22, 2016 (http://www.pewsocialtrends.org/files/2010/10/754-new-demography-of-motherhood.pdf).

43.

Long

J. Scott

Freese

Jeremy

. 2006. Regression Models for Categorical Dependent Variables Using Stata. 2nd ed. College Station, TX: Stata Press.

44.

Martin

Joyce A.

Hamilton

Brady E.

Ventura

Stephanie J.

Osterman

Michelle J. K.

Matthews

T. J.

2013. Births: Final Data for 2011. National and Vital Statistics Reports, Vol. 62, no. 1. Hyattsville, MD: National Center for Health Statistics.

45.

McCarty

Christopher

House

Mark

Harman

Jeffery

Richards

Scott

. 2006. “Effort in Phone Survey Response Rates: The Effects of Vendor and Client-controlled Factors.” Field Methods 18(2):172–188.

46.

McCormack

Karen

. 2005. “Stratified Reproduction and Poor Women’s Resistance.” Gender & Society 19(5):660–679.

47.

McDonald

Katrina Bell

. 1997. “Black Activist Mothering: A Historical Intersection of Race, Gender, and Class.” Gender & Society 11(6):773–795.

48.

McQuillan

Julia

Greil

Arthur L.

Shreffler

Karina M.

Tichenor

Veronica

. 2008. “The Importance of Motherhood among Women in the Contemporary United States.” Gender & Society 22(4):477–496.

49.

Musick

Kelly

Brand

Jennie

Davis

Dwight

. 2012. “Variations in the Relationship between Education and Marriage: Marriage Market Mismatch?” Journal of Marriage and Family 74(1):53–69.

50.

Ornelas

India

Perreira

Kristina

Beeber

Linda

Maxwell

Lauren

. 2009. “Challenges and Strategies to Maintaining Emotional Health: Qualitative Perspectives of Mexican Immigrant Mothers.” Journal of Family Issues 30(11):1556–1575.

51.

Raja

Nambury S.

Laffitte

Larry J.

Byrne

Barbara M.

2002. “Measurement Equivalence: A Comparison of Methods Based on Confirmatory Factor Analysis and Item Response Theory.” Journal of Applied Psychology 87(3):517–529.

52.

Ridgeway

C. L.

Kricheli-Katz

Tamar

. 2013. “Intersecting Cultural Beliefs in Social Relations: Gender, Race, and Class Binds and Freedoms.” Gender & Society 27(3):294–318.

53.

Roberts

Dorothy

. 1997. Killing the Black Body: Race, Reproduction, and the Meaning of Liberty. New York: Pantheon.

54.

Royston

Patrick

. 2005. “Multiple Imputation of Missing Values: Update of ICE.” The Stata Journal 5(4):527–536.

55.

Segura

Denise A.

1994. “Working at Motherhood: Chicana and Mexican Immigrant Mothers and Employment.” Pp. 211–221 in Mothering: Ideology, Experience, and Agency, edited by Glenn

Evelyn Nakano

Chang

Grace

Forcey

Linda Rennie

. New York: Routledge.

56.

Skogrand

Linda

Barrios-Bell

Abril

Higginbotham

Brian

. 2009. “Stepfamily Education for Latino Families: Implications for Practice.” Journal of Couple & Relationship Therapy 8(2):113–128.

57.

Sutherland

Jean-Anne

. 2010. “Mothering, Guilt and Shame.” Sociology Compass 4(5):310–321.

58.

Taylor

Tiffany

. 2011. “Re-examining Cultural Contradictions: Mothering, Ideology, and the Intersections of Class, Gender, and Race.” Sociology Compass 10(5):898–907.

59.

U.S. Department of Labor, Bureau of Labor Statistics. 2010. The Editor’s Desk. Retrieved December 29, 2015 (http://www.bls.gov/opub/ted/2013/ted_20131104.htm).

60.

Vespa

Jonathan

. 2009. “Gender Ideology Construction: A Life Course and Intersectional Approach.” Gender & Society 23(3):363–387.

61.

Williams

Joan

. 2000. Unbending Gender: Why Family and Work Conflict and What to Do about It. New York: Oxford University Press.

62.

Williams

John D.

1971. “A Multiple Regression Approach to Multiple Comparisons for Comparing Several Treatments with a Control.” The Journal of Experimental Education 39(3):93–96.

63.

Winship

Christopher

Mare

Robert D.

1984. “Regression Models with Ordinal Variables.” American Sociological Review 49:512–525.