Individuals of Mexican origin constitute the largest proportion of all Latinx living in the United States (U.S. Census Bureau 2011). About 63% of U.S. Latinxs are of Mexican origin; one third of these individuals were born in Mexico, and two thirds were born in the U.S. to at least one Mexican immigrant parent (Gonzalez-Barrera and Lopez 2013). Mexican Americans’ cultural beliefs are distinct from those of other Latinx groups (De Von Figueroa-Moseley et al. 2006). Furthermore, within the Mexican American population, individuals differ on educational and psychological outcomes and on degree of acculturation according to their generation status (Chávez-Reyes 2010; Pagan Rivera 2015). First-generation Mexican Americans tend to report stronger attachment to Mexican cultural beliefs than second-generation, U.S.-born individuals (Lui 2015). Despite our increasing knowledge of the nuances underlying intergenerational differences, our understanding of how generation status may influence parenting practices and goals remains limited (Bornstein and Cote 2006). Parent-child relationships constitute one of the most direct pathways through which cultural beliefs and values are transmitted across generations (Keller and Greenfield 2000). Expanding our knowledge of intergenerational variations in parent-child relationships may therefore shed light on a key contributor to processes of cultural change.

To investigate the specific cultural values promoted by Mexican American parents, we utilized ecocultural theory as our theoretical framework. According to ecocultural theory, parents socialize their children according to ethnotheories or cultural models, defined as the interrelated beliefs, goals, and strategies shared in their cultural group and passed on through generations (Keller et al. 2006; Super and Harkness 1986; LeVine et al. 1994; Weisner 2002). Because parents’ practices are guided by these cultural models, the meanings of these practices vary across cultures. The same practice, such as reading to a child, may have different meanings in different cultures if the beliefs and goals associated with the practice vary (Bornstein 1995; García Coll and Pachter 2002). Thus, studying the beliefs and goals embedded in parents’ cultural models provides insights into cultural variations in meanings associated with parents’ actual practices.

Research on parental socialization across cultures has suggested the existence of two broad cultural models, independence and interdependence (Markus and Kitayama 1991). While all parents want their children both to function autonomously (independence) and to build and maintain relationships (interdependence), cultural beliefs and economic concerns shape the relative importance of each of these models (Kagitçibasi 2005; Tamis‐LeMonda et al. 2008). Parents in rural and less affluent regions of the world tend to prefer an interdependence cultural model that discourages autonomy and promotes relatedness, as their future livelihood depends in part on their children providing material resources when they become adults (LeVine et al. 1991). Parents in more affluent regions, where children are viewed more as economic costs, utilize an independence cultural model that promotes agency, self-sufficiency, and even separation from family to achieve one’s goals (Greenfield et al. 2003).

Although often presented as such, the cultural models of independence and interdependence are not dichotomous, but orthogonal. Kagitçibasi (1996) argued that these broad cultural models contain distinct dimensions and that parents across cultures promote dimensions of both in their children. Independence includes the dimensions of agency (autonomy and self-reliance) and separateness; and interdependence contains the dimensions of relatedness (prioritizing relationships) and conformity (submission to outside authority) (see Table 1). Kagitcibasi proposed that parents may choose to promote agency but not separateness in their children, even though both are dimensions of independence. Similarly, they may choose to promote close family relationships but not conformity, even though both of these dimensions are associated with interdependence. She identifies this combination as a cultural model of emotional interdependence in which children may develop a strong sense of agency and pursue their own goals, while still remaining close to, and somewhat dependent on, their families. Kagitçibasi suggested that this model describes the socialization beliefs of highly educated parents living in urban regions of societies such as Turkey (Kagitçibasi 2007) and Cameroon (Keller et al. 2004).

Table 1 Conceptual map of relations between cultural models, dimensions, mexican cultural values, and parents’ practice beliefs and long-term goals
Full size table

Mexican American parents’ socialization beliefs have often been classified as collectivist due to the value placed on promoting interdependence through close family relations and respect for elders (Church et al. 2012; Hofstede 1980; Shavitt et al. 2016). Yet, increasing research has suggested that they also promote independence and agency, which are associated with individualistic cultures, providing additional evidence that Individualism and Collectivism are not dichotomous, but orthogonal and multidimensional (Oyserman et al. 2002; Suizzo 2007; Tamis‐LeMonda et al. 2008).

Research on Mexican American families has identified four cultural values that are associated with two dimensions of interdependence: Simpatía and familismo are associated with relatedness, and respeto and educación are associated with conformity. Simpatía includes qualities such as seeking and maintaining harmonious relationships, agreeableness, empathy, and concern for others (Griffith et al. 1998). These qualities may be seen in daily practices such as teaching children prosocial skills (e.g., sharing, turn-taking), and holding long-term goals associated with kindness and benevolence toward others. Familismo includes maintaining a strong attachment to one’s family, as well as feelings of reciprocity, loyalty, and solidarity (Marín and Marín 1991; Santiago-Rivera et al. 2002). Familismo also includes obligation to one’s family and conforming to traditions and rules established by elders, thereby also representing the conformity dimension of interdependence (Steidel and Contreras 2003). Respeto is also associated with conformity and refers to obedience, politeness, and high regard for adults and elders (Arcia and Johnson 1998; Durand 2011; Rodriguez and Olswang 2003). Finally, the cultural value of educación represents conformity through its emphasis on teaching children proper manners, morality, and responsibility (Delgado and Ford 1998; Okagaki and Sternberg 1993).

There is much less research on dimensions of independence among Mexican American parents. A few studies have shown that Mexican American parents promote autonomy and self-reliance, corresponding to the agency dimension of independence (Delgado and Ford 1998; Yau and Watkins 2018). In a recent study of Mexican American immigrant mothers of young children, Calzada et al. (2012) found that these mothers promoted the values of both autonomy and respeto in their young children, thereby combining dimensions of independence and interdependence. In their meta-analysis of studies measuring individualism and collectivism across cultures, Oyserman et al. (2002) found that Latinxs and European Americans did not differ on measures of individualism. These studies suggest that Mexican American parents do promote independence dimensions, but much more research is needed to identify which specific dimensions are most valued. Furthermore, research on the effects of acculturation and generation status on parenting beliefs suggests that independence and interdependence dimensions may vary across generations of Mexican American parents.

The limited research on generation and acculturation differences in the socialization beliefs of Mexican American parents shows varied results with regard to their preferences for dimensions of independence and interdependence. For example, Sabogal et al. (1987) found that more acculturated Latinxs (including Mexican Americans) valued obligation to family (conformity) less than less acculturated Latinxs. However, feelings of family support and closeness (relatedness) did not vary with acculturation level. In fact some research has found that feelings of family closeness increase with acculturation (Luna et al. 1996). Research examining the importance of educación (conformity) across generations of Mexican American mothers is extremely sparse, but a recent study of immigrant Latinx parents found that teaching children good manners and politeness was considered very important regardless of immigration status (Durand and Perez 2013). In one of the only studies of acculturation differences in Mexican American parents’ beliefs about young children’s agency, Savage and Gauvain (1998) found that more acculturated parents believed their young children capable of engaging in planning at earlier ages than did less acculturated parents. Delgado-Gaitan (1993) conducted a study comparing two generations of parents: Mexican immigrants who were born in Mexico and immigrated as adults, and “first-generation” Mexican Americans, defined in the study as parents who had been educated from childhood in the U.S., She found that the immigrant parents emphasized respect for elders (conformity) more than did first-generation parents, who focused more on developing critical thinking skills and autonomy (agency) in their children. First-generation parents still highly valued conformity, but did not view children’s autonomy and individual expression as disrespectful. Similarly, Okagaki and Sternberg (1993) found that Mexican immigrant parents taught their children conforming behaviors, such as doing neat and orderly work, more than independent thinking (agency) and creativity, while U.S.-born Mexican Americans showed the reverse pattern.

An important limitation of current research on acculturation effects is the focus on European American culture as the mainstream and only culture to which immigrants are exposed. Consequently, most of the literature we reviewed compared Mexican Americans to European Americans. As immigrant and ethnocultural minority groups come into contact with a society’s dominant culture, as well as with other ethnic minority groups who share their environments, the beliefs and practice of each cultural group influence the other through a process of acculturation (Berry 2003). Though acculturation is a bidirectional process, most often the dominant culture has inordinate power to influence, control, or even oppress the less powerful groups (Organista 2007). While generation status is not always associated with cultural incorporation (Fox et al. 2017) and does not fully capture the multidimensional, non-linear process of change over a lifetime (Lopez-Class et al. 2011), U.S. nativity and years lived in the U.S. are associated with exposure to European American cultural beliefs and values.

As a result, generation status has often been used to gain insight into this acculturation process in the US (Plunket and Bámaca-Gómez 2003). Mexican immigrants who came to the U.S. as adolescents or adults are more likely to retain the beliefs of their culture of origin, whereas those born in the U.S. or who immigrated as children are more likely to have beliefs that resemble those of European American cultures (Schwartz et al. 2006; Umaña-Taylor and Yazedjian 2006). Some second and later generation individuals, however, choose to retain or re-claim the cultural beliefs and practices of their ancestors, even as they integrate mainstream cultural beliefs, and become bicultural (LaFromboise et al. 1993). Recent studies of second generation immigrants, those born in the U.S. with at least one parent born outside the U.S., have shown that many factors such as race, social status, and community or regional ecology uniquely impact the experiences of this generation (A. Portes et al. 2005; A. Portes and Hao 2004; Suarez-Orozco and Suarez-Orozco 2001). As parents, their beliefs and goals for their third-generation children may differ from those of first and third generation mothers, yet our knowledge of these mothers’ parenting beliefs remains very limited. In an observational study of first and second generation Mexican American mothers interacting with their four-year olds, Livas-Dlott et al. (2010) reported a tendency for second generation mothers’ compliance demands to be less power-assertive than those of first generation mothers. In this study, however, the second generation mothers also had higher education levels than the first generation mothers, making it difficult to identify the sources of variation in their different styles.

Research has shown that generation status is positively related to education level among immigrants in the U.S. (Zhou and Lee 2017). Research has also demonstrated a positive relationship between parental education, especially maternal education, and children’s health and well-being throughout the world (LeVine et al. 2001). Schooling positively affects girls’ self-confidence, self-efficacy, and aspirations for their future (LeVine et al. 2001). When they become mothers, educated women apply and transmit these belief and goals to their children. A few studies of Mexican-origin mothers have found that maternal education level influences their socialization beliefs, practices, and goals for their children (LeVine et al. 1991; Richman et al. 1992). However, there is little research on how generation status and education level interact to shape parents’ socialization beliefs and goals for their children.

Is the influence of heritage culture on parents’ socialization beliefs as strong in later generation parents as in first generation parents, even if later generation parents have obtained more education than their parents’ generation? In a study of Mexican American parents of adolescents, Buriel (1993) found that maternal education level increased with each generation, and that maternal parenting styles differed according to generation status. Ng et al. (2012) found that Latinx mothers with higher education levels valued self-maximization (agency) more, and proper demeanor and manners (conformity) less, than mothers from less educated groups. These findings suggest that belief in children’s agency may increase with maternal education, however, because of the association between maternal education level and acculturation, it is difficult to isolate the effects of each on parenting beliefs. In this study, we investigated the effects of both generation status and parents’ education level (years of schooling) to address this research gap and determine the unique effects of both on parents’ socialization beliefs and goals. We examined whether Mexican American mothers who are exposed to diverse cultural contexts, including their heritage culture, European American cultures, and other ethnic groups, also combine dimensions of independence and interdependence in their socialization beliefs and goals.

We investigated whether Mexican American mothers’ socialization practice-beliefs and long-term goals differ according to generation status, and if so, how the importance attached to dimensions of agency, conformity, and relatedness within those beliefs and goals varies across generations and with maternal education levels. We examined the extent to which each of these dimensions was represented in two components of parents’ cultural models, beliefs about practices and long-term goals. This study is part of a larger study of parents of young children across U.S. ethnic groups in which we found that Latinxs placed more importance on agency and relatedness than on conformity, although they valued conformity more than European Americans (Suizzo et al. 2008). The current study further explores these findings by selecting only Mexican origin mothers from the Latinx group, and investigating differences within this group by generation statuses.

We investigated three hypotheses in this study. First, we hypothesized that later generation mothers’ beliefs and goals would reflect greater importance attributed to agency and lower importance attributed to conformity than earlier generation mothers. The agency dimension was operationalized as the practice of promoting autonomy, and the long-term goal of agency and self-direction (see Table 1). The conformity dimension was operationalized as the practice of teaching obedience and manners and the long-term goal of tradition and conformity. Second, we hypothesized that the parents’ beliefs and goals reflecting relatedness (teaching prosocial skills, and the long-term goals of close family relationships and benevolence) would be high in all three generations. Third, we hypothesized that mothers’ generation status would moderate relations between maternal education and one or more of the beliefs and goals. Due to prior research showing the distinctness of second generation youth, we explored the possibility that first and third generation mothers might attach similar levels of importance to one or more of the beliefs and goals but would contrast with those of second generation mothers. However, because the research on second generation Mexican American mothers is extremely limited, we did not formulate a specific hypothesis.

Method

Participants

A sample of 103 Mexican-origin mothers from three generations participated in the study. We recruited participants from two major cities in Texas, each located about 200 miles from the US-Mexican border. The populations of these cities contain large proportions of Mexican-origin families. To obtain a range of socioeconomic status levels, we recruited parents from a variety of sites, including day care centers, preschools, parent organizations, Mexican American associations, and churches. These sites were located across multiple urban neighborhoods that ranged in socioeconomic status, from upper middle class to low-income and poverty level. The neighborhoods were diverse with regard to ethnicity; some were predominantly Mexican American while some were predominantly European American. We asked parents to report their highest education level and current occupation status, which enabled us to measure variations in SES within the sample. Because we were interested in parents’ beliefs about young children, only parents with at least one child between one and six years old were included. Envelopes containing cover letters, IRB-approved informed consent forms, and questionnaires were distributed to parents at these sites. All documents were provided in both English and Spanish. In the letters, we asked the child’s primary caregiver to complete the questionnaires and return them to the site in the sealed envelope provided. All primary caregivers who returned questionnaires were mothers.

The mean maternal age was 32.33 years, SD = 5.51, and ranged from 21 to 52. There was no difference in the mean ages of first- and third-generation mothers, but second-generation mothers were slightly younger than both these groups (see Table 2). Maternal age was not significantly correlated with any of the scales and was therefore not controlled in subsequent analyses. The mean maternal education level was 4.62, SD = 2.03, or “some college,” and differed across generation groups. Most mothers (79.6%) were married and 11.7% were single. Of the full sample, 49% worked at least 20 h per week. The most frequent maternal occupations listed were homemaker (30%), clerical/sales (15.7%), doctor/lawyer/professional (13.7%), and nurse/teacher/technician (11.8%).

Procedures

Sociodemographic variables and generation status. Each participant provided socio-demographic information including her age, ethnic background, country of birth, parents’ and grandparents’ country of birth, age at immigration to the U.S. (if foreign-born), education level, and employment status. Generation status was determined from participants’ country of birth, their age of arrival in the U.S. if foreign-born, and their parents’ and grandparents’ countries of birth. Demographers identify three distinct generation statuses. The first generation includes foreign-born individuals who immigrated to the U.S. More recently, demographers have differentiated the first generation category into first and 1.5 generation. The 1.5 generation consists of foreign-born individuals who immigrated to the U.S. as children (Chiang-Hom 2004; A. Portes and Rumbaut 2001; Zhou 2004). Research has shown distinct differences in the experiences of individuals who grew up in Mexico and immigrated to the U.S. as adults (first generation), and those who immigrated as children and grew up in the U.S. (Lopez et al. 2004; P. R. Portes and Zady 2002; Zambrana and Powell 1992). The second generation includes U.S.-born individuals with at least one parent born outside the U.S. The third generation includes individuals who were born in the U.S. with both parents born in the U.S.

Accordingly, in our study, we coded mothers who were born in Mexico and had immigrated to the U.S. as adolescents or adults, after the age of 12, as first generation. The age of 12 is the average age of transition from childhood to adolescence, when significant cognitive, social-emotional, and neurological changes begin to occur (Koepke and Denissen 2012; Spear 2000; Steinberg 2005; Steinberg and Morris 2001). Mothers who were born in Mexico and immigrated to the U.S. as children (before 12) and had at least one parent born in Mexico were coded as 1.5 generation. We coded mothers who were born in the U.S., and had at least one Mexico-born parent as second generation. Finally, mothers who were born in the U.S. with both parents born in the U.S. and at least one grandparent born in Mexico were coded third generation. Because the sample of 1.5-generation mothers was small, we compared the means of the seven study variables between this group and the second generation group. We found only one significant difference (agency and self-reliance as a long-term goal), and therefore combined these two subgroups under the label “second generation.” The sample contained 38 first-generation mothers, 26 second-generation mothers, and 39 third-generation mothers.

Beliefs about Infants and Young Children-Multicultural

The BIYC-MC measures parents’ beliefs about daily childrearing and socialization practices with children from birth to six years (BIYC-MC; Suizzo et al. 2008). This measure focuses on beliefs about practices with children, rather than on parents’ actual practices. Respondents indicated how important they believe specific parenting practices are for children in general, even if they did not themselves engage in those practices with their own children. All parents responded to all items regardless of the age of their child(ren).

In this study, we used three scales associated with the dimensions of agency, conformity, and relatedness. The first scale, promoting autonomy, measures agency and contains 6 items including encourage child to make own choices and let child feed him/herself, even if it causes a mess. The second scale, teaching obedience and manners, measures conformity and contains 8 items, including teach child to obey elders and teach child not to interrupt adult conversations. The third scale, teaching prosocial skills, measures relatedness and contains 4 items, including teach child to share his/her toys with others, and teach child to say “please” and “thank you”. Level of importance was measured using a five-point Likert scale ranging from “not important” to “extremely important,” with higher values representing greater importance. The original study using these scales across U.S. ethnic groups reported reliabilities ranging from .76 to .83 for teaching conformity, .51–.74 for promoting autonomy,.and 72 to .80 for teaching prosocial skills (Suizzo et al. 2008). The Cronbach’s alpha statistics for these scales in this sample were .84, .71, and .83 respectively. The Cronbach’s alpha statistics by generation status ranged from .58 to .95, with 20 of the 24 values greater than .70 (see Table 3).

Table 2 Sociodemographic characteristics of sample by generation status (N = 103)
Full size table

Long-term goals and values

The Goals and Values in Adulthood Questionnaire, (GVAQ; Suizzo 2007) was used to measure parents’ long-term goals and values for their children. The GVAQ asked parents to rate how important it is to them that their children have each of 45 goals and values as adults. Parents were asked to indicate the level of importance they attach to the goals and values using a ten-point Likert scale ranging from “not important” to “extremely important,” with higher values representing greater importance. This questionnaire contains five scales, four of which were used in this study: agency and self-direction (10 items), tradition and conformity (9 items), close family relationships (4 items), and benevolence (7 items). The first scale reflects the agency dimension, the second reflects the conformity dimension, and the last two scales reflect the relatedness dimension. This questionnaire has been used in studies of parents from various U.S. ethnic groups, and has obtained very good reliability indices, ranging from .80 to .87 for agency and self-direction, from .69 to .72 for tradition and conformity (Richman and Mandara 2013). The original study using these scales reported the following alpha coefficients for the sample of Mexican American parents: .91 for agency, .85 for conformity, .59 for relatedness, and .82 for prosocial (Suizzo 2007). In this study, the Cronbach’s alpha statistics for these scales were .92 for agency, .88 for conformity, .74 for relatedness, and .85 for prosocial.

Data Analysis

An analysis of missing values indicated that 5.43% of all possible data points was missing, and that these values were missing completely at random (Little’s MCAR test; χ2 = 3138, p = .104). All missing values were then replaced with estimated means for each item (Rovine and Delaney 1990). We conducted a multivariate analysis of variance to test for effects of generation status on the seven dependent variables. Bartlett’s (1950) test of sphericity, Chi square (21) = 408.41, p < .001, indicated that the dependent variables were intercorrelated, supporting the multivariate approach. Multivariate effects emerged for generation status, F(14,188) = 7.12, p < .001, hp2 = .35, indicating that mothers’ generation status differentiated means of at least one dependent variable. We then conducted univariate tests to identify the specific dependent variables affected by mother’s generation status. Post hoc comparisons were conducted to examine group differences in each dependent variable, controlling for maternal education. Finally, we conducted seven regression models, one predicting each scale, to test for interactions between maternal education and generation status, and identify the distinct effects of each variable on mothers’ socialization practice-beliefs and long-term goals.

Results

Generation Differences in Agency and Conformity

Univariate ANCOVA’s revealed a main effect of maternal education on one of the two agency scales, promoting autonomy, F(6,93) = 3.20, p = .007, hp2 = .17, and on both of the conformity scales: teaching obedience and manners, F(6,93) = 3.93, p = .002, hp2 = .20, and the long-term goal of tradition and conformity, F(6,93) = 4.21, p = .001, hp2 = .21. There was no main effect of maternal education on the long-term goal scale agency and self-direction.

Controlling for maternal education, generation status had a main effect on both of the agency scales and both of the conformity scales. Results of post-hoc pairwise mean comparisons with Bonferonni adjustment mostly confirmed the first hypothesis that third generation mothers preferred practice-beliefs and long-term goals associated with agency more, and those associated with conformity less, than first and second generation mothers (see Table 2). Third generation mothers attached more importance to promoting autonomy than both first and second-generation mothers (see Fig. 1). Regarding long-term goals, third generation mothers valued agency and self-direction more than second-generation mothers, but as much as first generation mothers (see Fig. 2). As for conformity, confirming our hypothesis, first-generation mothers attached more importance to teaching obedience and manners than later generation mothers. First-generation mothers also valued the long-term goal of tradition and conformity more than second generation mothers, but the same as third generation mothers.

Fig. 1
figure 1

Mean importance of Mexican American Mothers’ Practice-Beliefs by generations status (N = 103)

Full size image
Fig. 2
figure 2

Mean imporTAnce of Mexican American Mothers’ Long-term Goals by generations status (N = 103)

Full size image

High Value Placed on Relatedness across Generations

Univariate ANCOVA’s revealed no main effect of maternal education on any of the three scales associated with relatedness. However, generation status had a main effect on all three of these scales: teaching prosocial skills, F(2,99) = 14.59, hp2 = .22, p < .001, the long-term goal of benevolence, F(2,99) = 8.98, p < .001, hp2 = .15, and the long-term goal of close family relationships, F(2,99) = 8.27, p < .001, hp2 = .14. Post-hoc pairwise means comparisons provided mixed support for our second hypothesis. First-generation mothers regarded teaching prosocial skills as more important than did both second and third generation mothers. First-generation mothers also valued the long-term goals of benevolence and close family relationships more than second-generation mothers. However, there were no differences in the importance attached to benevolence and close family relationships between first and third generation mothers. These results suggest that controlling for maternal education, the importance attached to these goals decreased from first to second generation mothers, but appeared to increase again among third generation mothers.

Interactions between Maternal Education and Generation Status

We conducted seven multiple regression analyses, one predicting each of the seven scales (three practice-beliefs and four goals), to test possible interaction effects between maternal education and generation status. Our hypothesis that mothers’ generation status would moderate relations between maternal education and beliefs and goals was supported by the results of one model: Generation status moderated the relation between maternal education and the practice belief of teaching obedience and manners to young children (see Fig. 3). Higher levels of schooling showed a stronger effect on teaching obedience and manners among third generation mothers than among second-generation mothers, b = −.123, F(3,99) = 16.911, p < .001. The R2 for this model was .346, indicating that the interaction between generation status and maternal education level explained 35% of the variation in the importance of teaching obedience and manners. We found no interactions for any other practice belief or long-term goal, suggesting that the magnitudes of relations between these variables were the same across all three generations.

Fig. 3
figure 3

Interaction between maternal education and generation status in predicting importance of the socialization practice of teaching obedience and manners (N = 103)

Full size image

Discussion

This study compared three generations of Mexican American mothers on three scales of beliefs about practices with one- to six-year-old children, and four scales of long-term goals for adults. Only a handful of studies have investigated generation differences in Mexican American mothers’ socialization beliefs and long-term goals for young children, and even fewer have examined the distinct contributions of maternal education and generation status. We found that when controlling for maternal education, all seven scales varied according to mothers’ generation status. We also found evidence to suggest that later generation Mexican American mothers’ cultural models may represent an orientation toward emotional interdependence. These mothers value beliefs and goals associated with agency more, and those associated with conformity less, than earlier generations. Yet they continued to place importance on relatedness, with the exception of close family relationships.

Agency and Conformity: Switching Places with Increasing Generations

We found that agency was highly valued across all three generation groups, as reflected in both beliefs and goals. The importance of the practice-belief of promoting autonomy increased across the three generations, from “important” to “very important,” while the long-term goal of agency and self-direction was considered “extremely important” by all three groups. The difference in importance attributed to the practice-belief versus the long-term goal by first generation mothers may be at least partially explained by concern for safety for one’s child. First generation mothers may hope that their children will become autonomous, self-reliant, and independent as adults, and they likely chose to immigrate to the U.S. so that their children would have a better future. But as recent immigrants, in their day to day lives, they may be concerned with monitoring their children closely to keep them safe in this new and unpredictable environment with negative outside influences (Ceballo et al. 2012; Kim et al. 2018; Suárez-Orozco and Suárez-Orozco 2015).

Nonetheless, the finding that Mexican American mothers highly value agency at all is significant in that the goals of self-direction and autonomy are not generally found to be as important to people of Mexican-origin. Thus while second-generation status and time in the U.S. are associated with lower achievement and lower aspirations for the future (Suarez-Orozco and Suarez-Orozco 2001), our results suggest that regardless of their aspirations for themselves, second-generation parents have high hopes for their young children and wish for them to become independent, self-directed adults. Furthermore, our study indicates that mothers value proactively teaching their young children independence skills, corroborating the findings of Calzada et al. (2010) that low-income Mexican immigrant mothers promoted the values of both autonomy and respeto in their young children. Further research is needed to investigate the significance of the dimension of agency in the cultural models of second and later generation Mexican-origin parents.

Table 3 Means and standard deviations of practice beliefs and long-term goals scales by generation status (N = 103)
Full size table

While agency remained important across all three generations, the importance of beliefs and goals associated with conformity (teaching obedience and manners and tradition and conformity) decreased among second and third generation mothers. This decline in the importance of teaching children conformity through obedience and manners corroborates the findings of previous researchers (Delgado-Gaitan 1993; Okagaki and Sternberg 1993), although overall, conformity remains “somewhat important” even among later generations. In the long-term, however, Mexican American mothers believed it “very important” that their children maintain traditions and respect for elders as adults. This finding is consistent with previous research showing that duty and obligation to family are highly valued by Mexican Americans and Latinxs across generations (Marín and Gamba 2003; Perez and Padilla 2000). Our study extends this research to mothers’ aspirations for their children as adults, offering evidence for the purposeful transmission of this cultural value across generations.

Relatedness: Continuity across Generations

We expected that beliefs and goals scales associated with relatedness, such as teaching prosocial skills, benevolence and close family relationships, would be highly valued by all mothers due to the importance of close family relationships among Mexican Americans. We found that the benevolence and teaching prosocial skills were rated “extremely important” by first-generation mothers, and decreased slightly in later generation mothers. The finding that behaviors and beliefs associated with kindness toward others remain highly valued across generations provides further evidence for the cultural script of simpatia that includes warmth and friendliness toward others, identified among Latinxs by previous researchers (Griffith et al. 1998). Interestingly, in their multigenerational study of Mexican American cultural scripts, Cuellar et al. (1995) found that personalismo, similar to simpatia, was the only cultural script that did not vary with acculturation or generation status. Our study confirms that simpatia remains an important value across generations of Mexican Americans and provides evidence that one way this value is maintained is through parents socializing their children to express kindness, empathy, and warmth toward others.

We found that close family relationships were considered “very important” by first-generation mothers and decreased to “important” among second and third generation mothers. This finding is consistent with findings of Sabogal et al. (1987) showing that the relatedness dimension of familismo remains strong across generations. The slight downward trend we identified, however, was somewhat surprising given the reported importance of family closeness among Mexican Americans. In a more recent study of four aspects of familismo across two generations of Puerto Ricans, Steidel and Contreras (2003) also found that family interconnectedness and support, associated with the relatedness dimension, declined in importance with generation status. Our findings provide evidence that this trend may also exist among Mexican origin parents.

Despite the intergenerational trends identified in this study, it important to note that across all generations, Mexican-origin mothers highly valued aspects of both agency and conformity as well as relatedness. In a study of first-generation Mexican immigrant mothers, Domenech-Rodríguez et al. (2009) identified a distinct Latinx parenting style they called protective parenting. These parents combined strictness with warmth and conveying messages about the importance of achievement goals to strengthen their children against stressors of poverty and discrimination. The promotion of goals and practices associated with both relatedness and agency by Mexican American mothers across all three generations found in our study provides additional evidence for this parenting style. These findings also provide support for the family socialization model of emotional interdependence identified by Kagitcibasi (2005), and contribute to growing evidence that Latino immigrant parents promote values associated with both their home and host cultures, adopting a “bicultural” cultural model (Aldoney and Cabrera 2016; McCabe et al. 2016).

Generation Differences in Influence of Maternal Education

We found that maternal education level had a stronger influence on mothers’ belief in teaching obedience and manners to their children depending on the mother’s generation status. Among second- and third-generation mothers, those with a college education were much less likely to teach obedience and manners than mothers who had not completed high school. However, among first-generation mothers, regardless of their amount of schooling, mothers believed in the importance of teaching obedience and manners to their children. This finding suggests that even among mothers who were born in the U.S. and whose parents were born in the U.S., years of schooling makes a difference in how they think about socializing their children.

More years of schooling appears to decrease the importance attached to teaching children obedience, manners, and compliance with rules. These socialization practices have important implications for children’s social competence. There is little research that has investigated possible links between maternal education level, teaching prosocial skills, and young children’s social competence; however, a recent longitudinal study of low-income mothers (Harding 2015) found that children whose mothers obtained additional schooling the year prior to their kindergarten year were more likely to exhibit externalizing symptoms. The author suggests that mothers may experience additional stress due to being enrolled in school, which may negatively influence their young children. Our study points to an additional possible explanation: Mothers’ experiences in school may influence their daily practices with their children such that they may be less concerned with their children’s obedience and manners. More research is needed to investigate other possible consequences of schooling on mothers’ socialization practices, especially in Mexican American samples. Further, research that examines other within-group variations, such as occupation, income, and neighborhood context, would deepen our knowledge of how these factors may interact with generation status to shape mothers’ beliefs and practices.

Limitations and Future Research Directions

Our study has several limitations that should be addressed in future studies. First, although our study assessed mothers’ genealogies, we combined groups of mothers who may have had different experiences, for example mothers who immigrated from rural areas with those who came from urban centers. Mothers who were employed outside their homes may have different beliefs from those who were full-time homemakers. Collecting a larger sample would enable more differentiation between such groups of mothers. Second, in this study we focused on primary caregivers, all of whom were mothers, and did not collect data from fathers. Future research needs to include fathers, even if they are not seen as primary caregivers, due to the strong effects of their beliefs and behaviors on children’s development. Third, we did not measure individual differences such as personality, which might have explained additional variation in our sample. Mothers who choose to immigrate as adults are more likely to value agency and autonomy and therefore may already rate these scales higher than their U.S.-born peers. Other individual differences such as such as income level and neighborhood residence may explain additional within-group variations in our findings. Including these factors in future research would further elucidate the complexity of Mexican American mothers’ parenting beliefs. Fourth, investigating the factors associated with time in the U.S. that may influence cultural beliefs, such as exposure to media, contact with other culture groups, and experiences of discrimination, would increase our understanding of sources of variation in parents’ beliefs across generations. Finally, this study investigated group differences in cultural beliefs, and not culture change. Longitudinal investigations are needed to measure change in cultural models with increasing time in the U.S.

By measuring specific dimensions of cultural beliefs and examining how they are valued differently across generation statuses, our study provides further evidence of the complexity and multidimensionality of Mexican-origin cultural beliefs. We identified differences between first-generation and later generation mothers, contributing to research on the distinctness of U.S.-born versus Mexico-born individuals (Padilla and Gonzalez 2001). Future research should continue to consider generational differences when examining cultural and parenting beliefs. In particular, additional research is needed to explore whether, as our findings suggest, second-generation parents may constitute a distinct group, in some ways resembling first-generation mothers, in other ways resembling third-generation mothers.

A parental value that has not been widely reported as important among Mexican Americans is that of agency. Our study found that later generation mothers valued conformity and relatedness less, but agency more, than did earlier generations. More research is needed to explore how the value of agency, particularly among later generations of Mexican American mothers may be an important psychological strength.

Intervention studies that account for acculturation and generational differences may offer guidance to practitioners and policy-makers involved in the design and administration of programs to support Mexican American families. Successful intervention programs must effectively recruit program participants, engage participants in services, retain participants, produce measurable outcomes and foster satisfaction with services. Some research has shown that parents’ attendance and persistence in intervention programs may vary with their acculturation level, and that more acculturated parents demonstrate lower persistence rates (Kim et al. 2016). More research is needed to understand how generational differences in parental beliefs and goals may help explain such differences in treatment participation and success in Mexican American families.

In addition, formative research to plan intervention programs should consider the diversity in values among Mexican Americans. Research-based interventions with children that have been modified to be more culturally relevant for Latinos include the GANA program, which is a modification of Parent Child Interaction Therapy (McCabe et al. 2005), and Culturally Modified Trauma Focused Cognitive Behavioral Therapy (de Arellano and Danielson 2005). Both these programs include assessment of families’ beliefs and attributes (e.g., acculturation, beliefs about discipline) that may influence families’ perception of and participation in interventions. Additional research to plan other culturally relevant intervention programs could examine the role generational differences may play in values and beliefs relevant to specific interventions.