Imagining Her Future: Diversity in Mothers’ Socialization Goals for Their Adolescent Daughters

Abstract

According to developmental niche theory, members of different cultural and ethnic groups often have distinct ideas about what children need to become well-adapted adults. These beliefs are reflected in parents’ long-term socialization goals for their children. In this study, we test whether specific themes that have been deemed important in literature on diverse families in the United States (e.g., Strong Black Woman [SBW], marianismo, familismo) are evident in mothers’ long-term socialization goals. Participants included 192 mothers of teenage daughters from a low-income city in the United States (58% Latina, 22% African American, and 20% European American [EA]/White). Socialization goals were assessed through a q-sort task on important traits for a woman to possess and content analysis of open-ended responses about what values mothers hoped they would transmit to their daughters as they become adults. Results from ANCOVAs and logistic regression indicate significant racial/ethnic differences on both tasks consistent with hypotheses. On the q-sort task, African American mothers put more importance on women possessing traits such as independence than mothers from other racial/ethnic groups. Similarly, they were more likely to emphasize self-confidence and strength in what they hoped to transmit to their daughters. Contrary to expectation, Latina mothers did not emphasize social traits on the q-sort; however, in open-ended responses, they were more likely to focus on the importance of motherhood, one aspect of marianismo and familismo. Overall, results suggest that these mothers’ long-term socialization goals incorporate culturally relevant values considered important for African American and Latino families.

Culture has been conceptualized as a set of beliefs, values, and practices shared by a group of people that organizes daily life and persists, with adaptations, over time (e.g., Bornstein, 2012; Le et al., 2008). There are several contextual factors, such as historical events or changes in widely accepted social norms, which influence the passage of culture between generations. Despite these other factors, the family context, and particularly parenting, is the primary mechanism through which cultural beliefs and values are passed down to the next generation (Keller, 2003). There is longitudinal evidence that the transmission of cultural values can be a bidirectional process such that adolescent endorsement of values predicts parental endorsement of values in some circumstances. Despite the bidirectionality, there is clear evidence of the importance of parental transmission of culture to their children (Perez-Brena, Updegraff, & Umaña-Taylor, 2015; Umaña-Taylor, Zeiders, & Updegraff, 2013). Developmental niche theory (Super & Harkness, 1986) provides one framework for identifying these cultural variations in the family context. A central construct in developmental niche theory is parental ethnotheories, which are the implicit and explicit beliefs that group members hold about what parents can and should do to influence their child’s development. The concept of ethnotheories is based on the assumption that most parents strive to raise their children to be well-adjusted and successful within the niche in which the family resides; however, definitions of what constitutes a “good child” or a “successful adult” vary depending on context, as do assumptions about what parents can do to effect these outcomes (Edwards, Knoche, Aukrust, Kumru, & Kim, 2006; Harkness & Super, 1992; Rosenthal & Roer-Strier, 2001).

Theoretically, parental ethnotheories are assumed to be instantiated in daily life and reflected in long-term socialization goals (Greenfield, Keller, Fuligni, & Maynard, 2003; Super & Harkness, 1999). The majority of research aiming to measure potential group differences in parental ethnotheories has focused on daily life, including the meaning parents give to family routines, parenting practices, and children’s behaviors. In contrast, fewer studies have examined differences in long-term socialization goals, in other words, what parents are working toward in how they raise their children. This is surprising because these desired developmental outcomes are a key component of parental ethnotheories. Limited cross-cultural studies have found group differences in parents’ long-term socialization goals (Harwood, Schoelmerich, Schulze, & Gonzalez, 1999; Suizzo & Cheng, 2007), and images of what constitutes “an adaptive adult” (Roer-Strier & Sands, 2004; Rosenthal & Roer-Strier, 2001). More research in this area is needed, however, particularly because differences in long-term socialization goals can influence current parenting behaviors. For example, parents choose more controlling and authoritarian parenting strategies (e.g., prohibiting a child from leaving the home without an explanation) when responding to events that may threaten the values they view as most important for their children (Padilla-Walker & Thompson, 2005). In this way, the nature and personal importance of parents’ long-term socialization goals may play a role in current parenting practices and parent–child relationships.

Drawing from these studies and developmental niche theory, the purpose of the current study was to examine racial and ethnic differences in parents’ long-term socialization goals within a diverse sample of mothers living in the same low-income city in the United States. Although the concepts of race and ethnicity are fraught with theoretical, definitional, and measurement problems, racial and ethnic–based categorization is ubiquitous in American society. Group membership to specific racial and ethnic categories is reified throughout one’s life through public documents (e.g., birth certificates, school records), governmental institutions (e.g., census counts, reported crime statistics), and mass media. Indeed, while race and ethnicity are by no means synonymous with culture, they are often defining elements of a child’s sociocultural context within the United States because of ongoing differences in one’s life experiences associated with ethnic/racial group membership (Hill, McBride Murry, & Anderson, 2005; McLoyd, 2004). Parents of different racial or ethnic groups may have distinct values or practices that reflect shared history, commonalities in their current context, and beliefs about what the future holds for their children. The challenge for researchers interested in these variations is to move beyond post hoc cultural explanations for mean group differences to instead generate a priori hypotheses grounded in relevant social science and humanities literature (Bornstein, 2012; Hill et al., 2005).

In this study, we examine potential differences in the long-term socialization goals of Latina/Hispanic, African American/Black, and European American (EA)/White mothers of adolescent girls living within the same city in the United States.¹ Several studies have identified distinct value orientations and related parenting practices in different racial and ethnic groups within the United States that exist beyond potential socioeconomic differences (e.g., Harwood et al., 1999; Milan, Snow, & Belay, 2007; Wray-Lake, Flanagan, & Maggs, 2012). However, research on potential variations in parental socialization goals is limited, particularly among the parents of adolescents. Relative to earlier developmental periods, normative aspects of adolescence have distinct implications for parental beliefs and values. For youth from all backgrounds, adolescence involves physical maturation and increases in autonomy; these changes may heighten the salience of specific parental values (e.g., differences in the relative emphasis on independence vs. interdependence; Greenfield et al., 2003). In addition, adolescents are exposed to more outside influences (e.g., peers) and experiences (e.g., sexual involvement, exposure to community violence) that may be viewed by parents as a threat to their socialization goals. As a result, implicit values and goals may become more explicitly recognized and discussed during this developmental period. We focused specifically on mothers and daughters in this study because of the high rate of single motherhood in the area and evidence of gender-specific processes in parent-to-child socialization efforts in families of color (Chuang & Tamis-LeMonda, 2009; Sharp & Ispa, 2009; Varner & Mandara, 2014). We hypothesized that the long-term socialization goals of mothers from different racial or ethnic backgrounds would incorporate specific values deemed important in existing literature, as detailed below.

Important Cultural Values for Latina Adolescents

Research on parenting in Latino families in the United States has demonstrated the importance of socially related values and defined social roles (for reviews, see Halgunseth, Ispa, & Rudy, 2006 and Umaña-Taylor, Alfaro, Bamaca, & Guimond, 2009). The concepts of familismo and marianismo may be particularly relevant for understanding social values influencing parenting of girls in Latino families.

Familismo reflects the importance of strong family ties, shared daily activities and living, and an expectation of family as a primary source of support (Calzada, Tamis-LeMonda, &Yoshikawa, 2013). Familismo may be reflected in Latino parent and adolescent behavior and attitudes. For example, Latina mothers report greater family cohesion and stronger collectivism values than EA White mothers (Varela, Sanchez-Sosa, Biggs, & Luis, 2009). Similarly, Latino adolescents report spending more time with family members and a higher sense of family obligation compared with EA White adolescents (Hardway & Fuligni, 2006). Latino parents also tend to engage in more “gendered” parenting than parents from other backgrounds, meaning that their behaviors and expectations differ more for sons versus daughters (Azmitia, Cooper, Garcia, & Dunbar, 1996). For girls, marianismo is often emphasized, which includes proper and virtuous behavior, nurturing others, and a need for protection from potentially dangerous outside elements (Castillo, Perez, Castillo, & Ghosheh, 2010). While these concepts are at times overstated in depictions of Latino culture, they remain influential values in the upbringing of children (Halgunseth et al., 2006), and may influence adolescents’ decisions about sexual activity, motherhood, femininity, and domestic roles (Marston, 2004). Given the values of familismo and marianismo, we expected mothers’ socialization goals for their adolescent daughters to emphasize family connections, social relatedness, and virtuous behavior.

Important Cultural Values for African American Adolescents

Research on African American families has highlighted the influence of minority status, economic stress, and kinship networks on parenting values and practices (Hill et al., 2005). The “Strong Black Woman” concept, which emphasizes independence, confidence, resisting vulnerability, and obligation to help others, is seen as an influential gender image (Abrams, Maxwell, Pope, & Belgrave, 2014; Collins, 2005). The SBW role is thought to have developed in response to ongoing marginalization resulting from racism, sexism, and poverty. Throughout history, limited access to resources and support has required African American women to take on multiple roles to provide for their families and communities, sometimes at the expense of their own needs and psychological well-being (Beauboeuf-Lafontant, 2007; Woods-Giscombé, 2010). Sociocultural scholars have also highlighted the SBW gender image as an intentional contrast to prevailing images of White women as fragile and in need of protection (e.g., Beauboeuf-Lafontant, 2003).

Examples of the SBW concept have emerged in research with African American mothers. For example, Sharp and Ispa (2009) found that when asked about child-rearing goals, African American mothers emphasize education and self-respect as particularly important, as well as hoping their daughters will be strong and self-reliant. Similarly, African American adolescent females have reported admiring their mothers and grandmothers for independence and emotional strength (Kerrigan et al., 2007). In light of the existing literature, we expected socialization goals of African American mothers to emphasize independence, confidence, and strength in the face of adversity, consistent with the SBW concept.

Important Cultural Values for White Adolescents

Many of the assumptions about the values that influence White families in the United States are inferred from cross-national studies, in which predominantly White, middle-class American mothers of European descent are contrasted to mothers from other countries. In these studies, being White and middle class is often taken as the norm for American parenting, which is a problematic aspect of this body of research (Lansford, 2012; Tamis-LeMonda & McFadden, 2010). In delineating values that influence how American parents raise their children, Tamis-LeMonda and McFadden (2010) highlighted the importance of individual achievement and personal choice in how American parents raise their children. Other reviews have also described an American emphasis on independence over interdependence values in child rearing, although both values are concurrently held (Greenfield et al., 2003; Raeff, 2010). In studies comparing parenting beliefs in American families from different racial/ethnic groups, there is some evidence that autonomy and personal choice are indeed more valued by EA White mothers (e.g., Richman & Mandara, 2013). More broadly, however, there is surprisingly little to draw from in hypothesizing how the experience of being White in the United States may specifically affect socialization goals, particularly in low-income families.

In the current sample, mothers who identified as EA White were lower income, and several were first- or second-generation immigrants from Eastern European countries, such as Poland and Bosnia. Thus, we did not have clear a priori hypotheses about specific values we expected to be present, only that there might be differences from African American and Latina mothers.

Current Study

The goal of the current study was to test whether racial/ethnic differences consistent with themes identified as culturally relevant in existing literature were evident in mothers’ long-term socialization goals for their daughters. One of the challenges to studying aspects of parental ethnotheories, including long-term socialization goals, is measurement. Studies that aim to capture the complexity of parental ethnotheories often use qualitative methods to derive meaning from open-ended interviews. These studies have provided rich insights, but do not allow for statistical comparisons of group differences. In contrast, studies that compare group responses on self-report measures of socialization goals may not capture important variations in why parents hold certain beliefs or what they feel their role is in meeting these goals. For example, parents may have a particular long-term goal for their child, but expect that it may be achieved through outside social influences (e.g., school). In addition, self-report measures limit the potential for eliciting new ideas from participants.

To measure long-term socialization goals in the current study, we drew from cross-cultural research on the “adaptive adult,” a concept initially proposed by Rosenthal and Roer-Strier (2001). The “adaptive adult” concept reflects the implicit, desired outcomes parents have for their children, which underlie and organize their socialization goals. We used two approaches designed to measure different aspects of maternal socialization goals: (a) The characteristics mothers believe young women growing up today need to be successful adults within their community using q-sort methods, and (b) what mothers hope to teach or transmit to their daughters as they become adults using quantified content analysis. These measures represent complementary but distinct aspects of socialization goals. Specifically, the q-sort task measures variations in mothers’ general beliefs about the most important traits in becoming an “adaptive adult.” The open-ended question taps into mothers’ more specific goals for their own daughter as she becomes an adult, and what values they believe they can and should transmit in this developmental process. We hypothesized that mother’s responses on these measures would reflect racial/ethnic differences consistent with culturally relevant (SBW, marianismo) concepts.

Method

Participants and Procedures

Study participants included 192 adolescent girls and their mothers residing in a midsized, low-income city in the Northeast United States (population 72,939). In 96% of dyads, the caretaker was the biological mother. In other cases, the primary caretaker was either a biological relative (aunt, grandmother, older sister) or stepmother. Families were participating in a larger National Institutes of Health (NIH)-funded study aimed at understanding the cultural and relational context of health disparities among adolescent girls to tailor health intervention programs in the region. All adolescent girls entering ninth through 11th grade within the city were eligible for participation, with the average age of participants at 15.1 years (SD = 1.05, range = 13–17). Caregivers ranged from 22.83 to 66.42 years (M = 41.58, SD = 8.04). Fifty-eight percent of participants identified as Latina (primarily Puerto Rican), 22% as African American/Black, and 20% as non-Hispanic, EA White. In 27% of families, at least one parent was born outside the United States, and another 18% were born in Puerto Rico (a U.S. territory). Thirty percent of homes included both biological parents at the time of participation. Forty-two percent of women had become mothers by the age of 18. Educationally, 22% of mothers had not completed high school, 67% had a high school degree, and 11% had a bachelor’s degree. The majority of adolescents (87%) qualified for free or reduced lunch at school. Racial/ethnic and socioeconomic characteristics of the sample are consistent with public data about the city and high school demographics. Based on census data, 25% of families live below the poverty level, 65% are single-parent households, 60% identify with a racial/ethnic minority group, and 20% are foreign-born. There is a low graduation (65%) and high teen pregnancy (17%) rate in the city. There is one large public high school serving the city, and the vast majority of adolescents in the study attended this school. Although efforts were made to achieve a sample representative of the city, it is a convenience sample.

Participants were recruited from city schools, community centers, health centers, Young Women’s Christian Association (YWCA), local media outlets, and word-of-mouth. Interviews were conducted in English (80%) and Spanish (20%) based on participant preference. Interviews were available in Polish to accommodate the largest immigrant group in the area, although no mothers chose this option. All measures were translated and back-translated and then piloted with local residents in an iterative process, following guidelines established by the World Health Organization (see www.who.int/substance_abuse/research_tools/translation/en/) (Process of Translation, n.d.). Mothers and daughters participated separately in face-to-face, semistructured interviews, which were audiotaped and later transcribed verbatim. Spanish interviews were translated and transcribed by bilingual, native Spanish speakers with each translated interview reviewed by a second reader. After the audio-recorded interview, mothers participated in the q-sort task, and then completed survey measures via Audio Computer Assisted Survey Instruments (ACASI) programmed in their preferred language. Interviews took approximately 2 hr, and participants were paid US$40 each for their time. All procedures were approved by the University of Connecticut Institutional Review Board.

Measures

Demographics

Mothers provided detailed demographic information on age, race/ethnicity, country of origin and immigration history, marital status and family structure, childbearing history, and various indicators of socioeconomic status (SES). Although all families would be considered “low-income” based on definitions by the National Center for Children in Poverty (see http://www.nccp.org/profiles/US_profile_6.html), an SES risk composite was computed to capture variations in economic disadvantage within this group. The SES risk composite was a count of three indicators: no high school degree, living in public housing, and receipt of free (as opposed to reduced or full-cost) lunch. Race/ethnicity classification for subsequent group comparisons was based on how the mother self-identified.²

Important traits

Participants sorted cards listing 19 positive traits with their definitions according to how important they believed each characteristic was for an adult woman to possess. Within these 19 traits, 10 were drawn from Antill, Russell, Goodnow, and Cotton’s (1993) Personality Characteristics Scale to represent either agentic/individualistic (e.g., ambitious, independent) or communal/social (e.g., caring, helpful) traits. The remaining words were selected by the research team from a list of the most common positive trait descriptors (from www.englishclub.com) that did not directly reflect either individualistic or social themes (e.g., smart, honest). On each card, definition/descriptions for each term were written in the same form directly under the word (“This is the sort of person who …”) consistent with the Antill et al. (1993) measure. To pilot the measure, eight undergraduate students were asked to first match words with their definition and then choose five words that best reflect individualistic and five words that best reflect social characteristics. Responses indicated that words and definitions were described clearly (over 95% correct matching), and that the 10 words from the Antill et al. measure reflecting individualistic or social traits were overwhelmingly viewed that way (each of the target words was selected correctly by at least six of the eight respondents; none of the other nine words was selected more than twice). The same process was conducted with three mothers from similar sociodemographic background as study participants with similar accuracy.

For the study, mothers were told to imagine a young woman growing up in today’s world and then instructed to sort the cards based on how important they believed each of the traits were: “For that woman to make a good life for herself and to be okay in the world she lives in, based on your own experiences and opinions.” Sorting was done by placing the cards on a large poster board with a forced-choice, normal distribution. The distribution ranged from 1 (least important) to 7 (most important), allowing one word to be placed in the most/least important spots, two words in the next spots (Values 2 and 6 on the 1–7 scale), four words in the next spots (3 and 5 on the 1–7 scale) and five words in the middle spot (Value 4 on the 1–7 scale). Mothers were allowed to change the position of any card during the process and again after being asked to make a final review. The mean placement values for the five individualistic (α = .70) and five socially (α = .73) focused words were computed to reflect the relative emphasis placed on these two domains. On this task, an emphasis on individualistic over social traits would be consistent with the SBW image, and an emphasis on social over individualistic traits would be consistent with familismo and social connection values.

Lessons learned

As part of the semistructured interview, caregivers were asked, “Now I am going to ask you to look ahead to the future. When your daughter is an adult, what would you like her to say that she learned from you, her mother?” If the caregiver was not the adolescent’s mother, terms were changed accordingly. Interviewers clarified questions and/or prompted mothers for elaboration when appropriate (e.g., Anything else? Can you say a little more about that?). Transcripts were coded for the presence of four themes that Tamis-Lemonda and colleagues (Ng, Tamis-LeMonda, Godfrey, Hunter, & Yoshikawa, 2012) found in spontaneous responses about socialization goals from African American and Latino mothers in the Northeast United States. These themes include (a) individual agency and achievement (e.g., independence, confidence, working hard), (b) social connection (e.g., tolerance for others and differences between people, social concern and kindness, family connectedness), (c) proper demeanor (e.g., respectfulness, sexual modesty), and (d) fulfillment or self-maximization (e.g., being happy, following dreams). Although only the first three themes relate to research hypotheses about SBW, familismo, and marianismo, we included all four themes for descriptive purposes. Four transcripts from each racial/ethnic group of mothers were randomly selected and reviewed by the authors to further refine examples of these themes and allow for emerging themes. At this stage, “understanding mother’s actions” was added as a social connection theme separate from family connectedness and religiosity was added to the proper demeanor theme. Examples of themes and categories within each theme with respondent quotes are presented in Table 1. A group of three undergraduate students blind to the racial/ethnic group of the respondents were trained to reliability and coded all transcripts. Responses could receive multiple content categories. Kappa coefficients, a measure of interrater agreement controlling for chance, were high for the four broad themes (between .65 and .78), and lower but still acceptable for individual examples within themes (.49-.71).

Table 1.

Themes and Examples of Mother’s Open-Ended Responses.

Theme	Example quote
Individual agency
Independence; able to take care of herself	“How to be independent and not rely on anyone, just rely on yourself to get ahead.” (ID 217)
Confidence; strength, self-belief	“That she would learn to keep her head up and be proud of who she is. Don’t let anybody make you feel less than what you are; hopefully, she is learning that from me now.” (ID 52)
Importance of working hard; never give up	“I think if something seems difficult you’re very capable of pushing on and just put one foot in front of the other and keep going and you can accomplish what you want to accomplish, you don’t have to give up.” (ID 220)
Social connection
Respect others; tolerance	“Just tolerance; you know that everybody is different and differences don’t mean that you are bad …” (ID 132)
Social concern; kindness, compassion	“I always taught her to be nice and kind, don’t do to somebody that you don’t want them to do to you. Don’t hurt people.” (ID 184)
Commitment to family	“We have a huge family on both sides, and I’m the one always planning big Thanksgiving dinners, planning big birthday parties. I plan so much so that the family is around … I’m always working to keep the family together. I want her to learn and keep that family thing going.” (ID 178)
Understanding mother’s action; maternal role	“My mom was not trying to be mean, she meant well for me. Yeah, she meant well and wanted the best for me. That would be nice, because I don’t think sometimes they get that.” (ID 225)
Proper demeanor
Responsibility; good behavior	“To make wise choices, smart decisions, and always have a good head on her shoulders.” (ID 8)
Self-respect; morality, religiosity	“How to live her life based around trust in God and never settle for anything less. Whatever He tells her to do I want her to do.” (ID 163)
Fulfillment
To enjoy life, be happy	“Just to live your dreams and you know, be happy. Don’t sweat the little stuff in life. I tell her that all the time.” (ID 80)

Data analytic plan

Repeated measures ANCOVA was used to examine whether mothers in the three racial/ethnic groups differed in where they placed the individual versus social words on the q-sorting task. In this analysis, race/ethnicity served as a between-person factor, type of words was a within-person factor, and sociodemographic factors were covariates. Next, chi-square tests were run to examine the association between race/ethnicity group and the frequency of different themes in open-ended responses. Significant results were followed up with logistic regression analysis with sociodemographic factors controlled.

Results

Preliminary analyses tested for racial/ethnic group differences on demographic characteristics. The groups did not differ by age, biological mother status, or marital status. However, EA White mothers reported fewer SES risk factors than Latina and African American mothers (χ² = 34.69, df = 6, p < .001; 0.4 vs. 1.2 and 1.3, respectively, on 0–3 scale) and were less likely to have been a teen mother (χ² = 17.34, df = 2, p < .001; 17% vs. 53% and 35%, respectively). These two variables were controlled in subsequent analysis.

On the q-sort task, racial/ethnic differences were evident in mothers’ endorsement of the importance of both individualistic, F(2, 183) = 3.54, p < .05, and social, F(2, 183) = 386, p < .05, traits, controlling for SES factors. Mean scores by group are presented in Table 2. In post hoc contrasts, African American mothers placed significantly more importance on individual traits compared with both Latina and EA White mothers. Latina mothers placed more importance on social traits compared with African American mothers. Next, repeated measures ANCOVA was used to examine whether the relative importance mothers put on the two types of traits differed for Latina, African American, and EA White women. There was a main effect of type of word, F(1, 181) = 13.62, p < .0001, with mothers across the sample putting more importance on individualistic traits. However, this result was qualified by a statistically significant interaction between race/ethnicity and type of word, F(2, 181) = 4.37, p < .05. As shown in Figure 1, only African American mothers placed greater importance on words reflecting individualistic traits relative to social traits. For Latina and EA White mothers, individualistic and social traits were rated with relatively equal importance. In other words, African American mothers emphasized the need for traits such as independence and ambition, consistent with hypotheses. Contrary to expectation, Latina mothers did not rate social traits as more important than individualistic traits although they did rank these traits higher than their African American counterparts.

Table 2.

Means and Standard Deviation Scores on Importance Placed on Individualistic and Social Traits on the q-Sort Task.

	Total	Latina (n = 108)	African American (n = 39)	EA White (n = 37)	F value
Individualistic traits	4.25 (.79)	4.18b (.75)	4.51a (.88)	4.16b (.76)	3.54*
Social traits	4.03 (.60)	4.13a (.61)	3.83b (.54)	3.99a,b (.59)	3.86*

Note. Values with different superscripts are statistically significantly different in post hoc testing.

^*p < .05.

Figure 1.

Mean scores for individualistic versus social traits on q-sort task by race/ethnicity.

Note. EA = European American.

Table 3 presents the frequency of different themes evident in mothers’ responses to the open-ended question about lessons they hoped their daughter would learn from them. At the broad level, social connection themes were the most common (64%), followed by individual agency themes (44%), proper demeanor (20%), and fulfillment (2%). One third of the responses received more than one of these four broad theme codes, so it was relatively common for mothers’ responses to incorporate multiple themes. This coexistence of individualism and social relatedness was particularly evident in mothers’ responses. For example, one mother said she hoped her daughter would be able to say,

What I learned from my mom is to be strong, be a leader and not a follower, and that goes with being strong also. Be caring, you know, treat people the way you want to be treated. That’s the big motive of my children. And respect. (EA White mother, ID 57)

Table 3.

Frequency of Different Themes in Maternal Responses About What They Hoped to Teach Their Daughter for the Sample and by Race/Ethnicity.

Theme	Total %	Latina % (n = 108)	African American % (n = 39)	EA White % (n = 37)	χ2 (df = 2, N = 184)
Any individual focused	46.2	40.7	64.1	43.2	6.92*
Independence	13.6	10.2	23.1	13.5	4.06
Confidence	16.3	11.1	28.2	18.9	6.37*
Importance of hard work	26.1	29.6	25.6	16.2	2.57
Any socially focused	64.1	66.7	61.5	59.5	0.77
Respect for others	9.8	6.5	12.8	16.2	3.48
Social concern/kindness	16.8	7.4	30.8	29.7	16.65***
Commitment to family	10.3	11.1	5.1	13.5	1.61
Understand mother	39.1	50.0	28.2	18.9	13.66**
Any proper demeanor	20.7	20.4	25.6	16.2	1.04
Responsibility	15.2	16.7	15.4	10.8	0.73
Self-respect, morality, religiosity	6.5	5.6	10.3	5.4	1.12
Any emotional fulfillment	2.2	2.7	2.6	1.9	0.13

Note. EA = European American.

^*p < .05.

^**p < .01.

^***p < .001.

As shown in Table 3, chi-square analyses were conducted to test for racial/ethnic differences in themes present in responses. Group differences were evident for the presence of any individualistic theme, χ²(df = 2, N = 184) = 7.99, p < .05, with 64% of African American mothers having this kind of theme present, compared with 41% of EA White and 43% of Latina mothers. Follow-up logistic regression indicated that this difference held when SES risk and teen motherhood were included as covariates, with EA White (adjusted odds ratio [AOR] = .24, 95% confidence interval [CI] = [.08, .71]) and Latina (AOR = .31, 95% CI = [.13, .77]) mothers significantly less likely to give a response with an individualistic theme compared with African American mothers. Of the different types of individualistic categories, the overall group difference was most evident in the “confidence, belief in self” example, χ²(df = 2, N = 184) = 6.37, p < .05, with 28% of African American mothers versus 19% of EA White mothers and 11% of Latina mothers giving a response with this goal included. An example with this type of theme is given below:

Smartness, ambition, you know the desire to go far in goal-reaching. Don’t give up, don’t let people tell you that you can’t do it when you know you can do it. Because I pushed hard, I push her hard. People tell me “you know you can’t do that,” prove them wrong … prove them wrong all the time. Because now they call me the lucky one, no I’m not lucky. I just didn’t want to hear what people say. When people say “oh no you can’t do that,” say “I know I can do it.” (African American mother, ID 149)

Contrary to expectation, the frequency of responding with a socially focused theme did not differ by racial/ethnic group, χ²(df = 2, N = 184) = .77, p < .68. However, review of the different categories within the social connectedness theme did suggest racial/ethnic differences. Specifically, significant group differences were evident in social concern or kindness, χ²(df = 2, N = 184) = 16.65, p < .001 (7% of Latinas, 31% of African American, 30% of EA White mothers) and understanding of maternal role, χ²(df = 2, N = 184) = 13.66, p < .01 (50% of Latinas, 28% of African American, 19% of EA White mothers). These findings indicate that the lack of a racial/ethnic group difference in social connectedness themes when considered broadly obscured two group differences in the opposite direction at the subcategory level. After controlling for SES differences, Latina mothers were significantly less likely to mention social concern or kindness than EA White (AOR = .12, 95% CI = [.03, .38]) or African American (AOR = .13, 95% CI = [.04, .37]) mothers. In contrast, they were more likely to respond with understanding of the maternal role as part of what they hoped their daughter would say she learned from them compared with EA White (AOR = 3.85, 95% CI = [1.4, 10.2]) and African American (AOR = 2.43, 95% CI = [1.1, 5.5]) mothers.

No racial/ethnic group differences were evident for proper demeanor, χ²(df = 2, N = 184) = 1.04, p = .59, or fulfillment, χ²(df = 2, N = 184) = .13, p = .94, themes. Overall, these themes, particularly fulfillment, were less common in mothers’ responses. One unexpected topic that emerged in open-ended responses was how mothers hoped their daughters would relate to men, as illustrated in these two examples:

I want her to know the responsibility of taking a job and you know working on herself. You know, not relying on any man to take care of her. I teach her that she shouldn’t rely on a man, even if he does have work. She should always have her own mind, her own jobs … Of course when she has kids she might be home, but I want her to have a background where she can take care of herself. (Latina mother, ID 33)

To take care of herself, to work independently. I really want her to grasp that you don’t necessarily need a mate to make it. I want her to be secure. And I think myself and her older sisters give her that example already—that you don’t have to depend on a man for your completion. I would really like her to get that and not feel dumb, not feel like her existence is tied up in her relationship with a guy … (African American mother, ID 193)

Although not initially part of hypotheses or coding, this message was relatively common, occurring in 20% of responses. Messages about men were embedded within responses with individualistic and social relatedness themes, and were equally common across the three racial/ethnic groups.

Discussion

Parental ethnotheories are theorized as one way the prevailing values of a specific sociocultural context guide parental behavior. Although there are well-documented differences in parental ethnotheories in cross-cultural studies, there are few studies investigating how aspects of ethnotheories differ among families from different racial or ethnic backgrounds within the United States (e.g., Raghavan, Harkness, & Super, 2010). Overall, results from the two tasks used in this study suggest that mothers’ long-term socialization goals incorporate multiple values, as emphasized in literature on American parenting belief systems (e.g., Edwards et al., 2006; Tamis-LeMonda & McFadden, 2010). However, we also found evidence of racial and ethnic group differences consistent with existing literature on important values among families of color. Below we discuss findings for mothers in general, and then turn to results for three specific racial/ethnic groups.

Within the United States, parental socialization goals vary based on socioeconomic context (Wray-Lake, Flanagan, & Maggs, 2012). In this sample, mothers were generally lower income, single parents of color interviewed during a time of economic recession. These contextual factors were evident in the open-ended responses. For example, the need to be self-reliant from men emerged across racial/ethnic lines, which may be a distinct value transmitted to daughters in settings in which single motherhood is the community norm. Terms such as “sacrifice” and “survival” were also common in responses emphasizing both individualistic and social connectedness values. Given socioeconomic stressors, mothers may have felt it was particularly important that their daughters develop the same attributes that helped them within this type of environment, such as the ability to work hard to overcome obstacles or developing positive relationships for support during difficult times. As one woman stated,

I want her to look back and say “Mom, as I was growing up, I learned to watch the rough situation you’ve been in, being out of work, and how you handled it. You didn’t break down. You kept on going and getting up every day; you kept on going and going. And I respect not giving up and making it happen for me.” That’s what I want her to say. And the qualities of understanding and respect I give her, and loving. (African American mother, ID 142)

A somewhat unexpected finding across the sample was the infrequency of responses that included fulfillment or happiness as what mothers hoped their daughter would learn from them. Within America, there is a common belief that parents want their kids to be happy first and foremost as evidenced by countless parenting books, magazine articles, and blogs dedicated to this topic (e.g., Greenspan, 2007). This desire may be fairly global: In a survey of 3,500 mothers in six countries including the United States, mothers from all countries stated “happiness” as the number one wish for their child’s future (The Fisher Price Moms’ Hopes & Wishes Study, 2015). Yet, mothers’ responses to our question about what they hoped their daughter would say she learned from them did not incorporate this theme. One possibility is that mothers want their children to be happy, but do not see this as something they will transmit. In a study by Tulviste, Mizera, and De Geer (2012), mothers were asked about values for their child now, in the future, and what they hoped to pass on to them. Mothers commonly endorsed wanting happiness in their child’s future, but were much less likely to indicate identifying their happiness as a value they hoped to pass on. Instead, benevolence (e.g., kindness to others) was endorsed most in response to this question. Together, these findings suggest that mothers may hold feeling fulfilled or happy as long-term goals for their children, but not believe this attribute will be learned directly from them. The construct of ethnotheories incorporates both what parents want for their children and what they believe they can and should do to effect this outcome. Mothers may believe what they personally can do to help ensure their child becomes a happy adult is model, teach, and support the characteristics they deem necessary for this outcome, such as self-reliance or maintaining social ties. It is also possible that for mothers facing multiple stressors associated with economic disadvantage, emotional fulfillment or happiness simply may be less salient and thus not generated spontaneously in an open-ended task.

One interesting aspect of our results is the difference in responses between the two tasks. Across the sample, mothers viewed social traits as equally or less important than individualistic traits in rating characteristics they believe are most important for a woman. In contrast, they emphasized social connection more than individual agency as what they hoped their daughter would learn from them, as has been found in other studies (e.g., Tulviste et al., 2012). Thus, mothers may believe there are several traits that young women need to acquire on the road to adulthood, but see the transmission of certain social values as part of the maternal role.

Responses From Latina Mothers

In understanding Latino culture, considerable attention has been given to concepts of familismo and marianismo as reflections of values related to social connection and specified social roles. Findings in the current study are mixed in providing support for these values. When asked to identify the traits of an adaptive adult woman, Latina mothers placed more importance on social traits than did African American mothers but not EA White mothers; however, they placed relatively equal importance on both social and individualistic traits (i.e., they did not favor social traits in the forced-choice sorting task). These findings are in line with meta-analytic results showing no difference between Latino and EA White Americans in the domain of interdependence (Oyserman, Coon, & Kemmelmeier, 2002). Latina mothers’ responses to the question about what they would like their daughter to learn from them were also mixed in regard to social values. Specifically, Latina mothers were less likely to give a response that included general social concern and kindness, but were more likely to give responses that reflected the importance of the maternal role. Thus, there may be benefit to refining ideas about social values in cultural research from a conceptualization of generalized social relatedness or interdependence to focus instead on specific relationships.

We initially conceptualized responses about the maternal role as an aspect of family connection and familismo in our coding scheme. The frequency with which mother-specific responses came up in piloting led us to code this as a distinct subcategory, highlighting the benefit of allowing for emerging concepts in the task. For some mothers, responses that received this code emphasized their daughters learning how to be a good mother. For example, one Latina mother wanted her daughter to learn

how to clean, how to cook, how to be a good mom … I got friends, people and my family they always tell me “you’re a wonderful mother.” I want her, when she got her kids, to be the same way too and to say “I’m glad I am like that because my mother was like that with me … I got this from my mother.” (Latina mother, ID 61)

These responses may be reflections of the marianismo theme. Although marianismo often is used as the opposite of machismo in describing how women are expected to approach sexual and romantic relationships, the concept is rooted in the Catholic notion of Mary as both a virgin and a mother (Castillo et al., 2010). While the concept may be grounded in Catholicism, the emphasis of women as mothers extends beyond specific religious groups and is perpetuated in popular Spanish-language media in the United States and Latin America (Rivadeneyra, 2011). Empirical studies have also found that the maternal role has special weight in Latino culture and that Latina mothers may prioritize motherhood over other aspirations for their children (e.g., Driscoll, Biggs, Brindis, & Yankah, 2001; Roehling, Jarvis, & Swope, 2005).

In contrast, a number of responses emphasizing the maternal role were less focused on daughters learning how to be a mother and instead expressed the hope that daughters would understand their mothers’ actions as they became adults. As one mother stated, she hoped

that she learned how to take care of herself … She doesn’t really understand why I look after her so much, why do I prohibit her from doing some things … But in the future, she’s going to thank me for doing that. She’s going to say “Thank you mom. I avoided doing so many things. Look at my friends, what they’ve gone through.” So I know that in the future, she’s going to thank me for the way I’m with her now. (Latina mother, ID 116)

These types of responses may be tied to acculturation and acculturative stress in relation to socialization goals if efforts to teach certain values cause conflict among Latino families (Umaña-Taylor & Yazedjian, 2006). Belief that their daughter will understand maternal behaviors later may help justify difficult decisions to their daughters in the present (e.g., why the daughter has to eat dinner with the family or attend a church service instead of going out with friends). Similarly, if the maternal role is particularly important to Latinas, then normative parent–adolescent conflict may be more distressing. If so, these responses may reflect mothers’ desires for their daughters to ultimately understand reasons for conflict within the family from the mother’s perspective.

Responses From African American Mothers

The importance of independence and strength as long-term socialization goals among African American women was demonstrated in both tasks. On the q-sort, African American mothers placed more importance on possessing individualistic traits than did Latina and EA White mothers and they were the only group to clearly value individualistic traits over social traits. They also were more likely than other mothers to incorporate individualistic themes, and particularly self-confidence, in what they hoped their daughters would learn from them. This pattern of results across the two tasks suggests that African American mothers value independence, confidence, and strength for women in general, in themselves, and in their daughters, consistent with the SBW cultural image.

The SBW concept is thought to serve a protective function against the negative effects of discrimination and limited resources. Indeed, for African American girls, perceptions of maternal encouragement of independence and strength predicts higher self-esteem (Ridolfo, Chepp, & Milkie, 2013), which may be one reason African American girls consistently report higher self-esteem than members of other racial/ethnic groups (see Twenge & Crocker, 2002, for review). While belief in one’s self may serve an important and positive role in the lives of African American women, maternal socialization messages may incorporate caution and the unavailability of social support (e.g., Wray-Lake et al., 2012). For example, several African American mothers emphasized the need for self-reliance saying “don’t rely on anybody” or “make sure you can take care of yourself.” In this way, the SBW ideal could contribute to feelings of isolation for some African American women. Consistent with this possibility, findings about the SBW concept and mental health are mixed, with some studies indicating that the risk for poor mental health is highest among women who endorse the SBW ideal (e.g., Donovan & West, 2015; Harrington, Crowther, & Shipherd, 2010).

It is important to note that although African American mothers emphasized individualistic values in both tasks, they also frequently expressed a desire for their daughters to be socially concerned and kind. Several mothers responded that they hoped their daughters would be strong so as to help others, and thus focused on both independence and social connection. Indeed, the SBW concept, while encompassing individual strength, also values social concern in the form of providing for the family and for the community (Woods-Giscombé, 2010). Consequently, commitment to others and community may be viewed as part of what strong women should do within disadvantaged settings.

Responses From EA White Mothers

Our findings provide the least insight into the long-term socialization goals of low-income EA White mothers. The significant group differences that emerged across the analyses do not a provide clear pattern. EA White mothers were similar to Latina mothers in how much they emphasized individual traits or goals, with both groups putting less emphasis than African American mothers. In contrast, they were similar to African American mothers but not Latina mothers in hoping their daughters would learn kindness from them. These findings are consistent with research demonstrating parents in the United States highly value both interdependence and independence (Suizzo, 2007). Given limited research on the parenting values of low-income EA White mothers, we did not have specific a priori hypotheses regarding their responses. In studies of EA working-class communities, diverse individualistic socialization practices were emphasized by mothers, indicating heterogeneity in EA parenting values (Kusserow, 1999). Our analyses were also limited because the group of EA White mothers was small and included several first- or second-generation women.

The question of how one’s “whiteness” affects parenting, particularly within diverse communities, has not received much attention compared with how membership in other racial or ethnic groups may influence parenting. We are currently analyzing data from low-income EA White mothers in which they were asked explicitly to think how “being white” influences parenting in an attempt to garner some preliminary understanding. In the present study, however, we were only able to understand low-income EA White parents as an exploratory comparison group, rather than studying their unique values. The current political context in the United States highlights the need to better understand the specific challenges facing low-income EA White families. More broadly, the increased publicity of systemic racism in policing and the justice system makes it painfully clear that race and ethnicity matter when American mothers think about their child’s future.

Limitations

Our study findings should be interpreted in light of several limitations. First, generalizations to racial or ethnic groups do not accurately represent within-group variation. For example, the level of acculturation of each member of the dyad, as well as the family as a unit, may affect parents’ and adolescents’ cultural values (Knafo & Schwartz, 2001). In this study, generational level was not related to responses on either task, although analyses were limited by sample size. Future research should examine within-group differences, including place of origin, immigration history, U.S. region of residence, and SES. Second, this study focused only on mothers and daughters. Given the intersectionality of race/ethnicity and gender, our findings may not generalize to fathers or sons. Finally, cross-sectional studies cannot incorporate the transactional nature of parent–child relationships, the bidirectional transmission of values and beliefs, or the dynamic nature of cultural beliefs (Le et al., 2008; Roest, Semon Dubas, & Gerris, 2009). Asking adolescents what they have learned from their mothers or what values their mothers are implicitly and explicitly teaching them would provide more insight into the transmission of socialization goals and help clarify how much mothers’ socialization goals are actually reflected in their parenting. Longitudinal research is needed to examine whether socialization goals predict different developmental outcomes (e.g., early motherhood, college attendance), the ways children and parents mutually influence each other, and how parents’ values change over time.

Conclusion

Cultural and developmental scholars have called for researchers to better identify “constellations of American parental belief systems in their complexity” (Edwards et al., 2006, p.149) and to move toward grounded, a priori hypotheses about potential racial or ethnic group differences (Hill et al., 2005). Following these recommendations, we identified themes and concepts that have been deemed relevant in existing literature on diverse families in the United States and tested whether these themes were evident in mothers’ responses on two measures designed to tap into aspects of parental ethnotheories. Consistent with our hypotheses, we found evidence of culturally relevant values in the long-term socialization goals of both African American and Latina mothers. The next step for research in this area should be to delineate how these longer term goals may be enacted in daily life, and what the implications are of variations in long-term goals both between and within specific racial and ethnic groups. Understanding these variations can provide richer insight into family life in an increasingly diverse society.