Mexican-Origin Adolescent Mothers’ Beliefs and Practices Concerning Children’s School Readiness

Abstract

Parents’ academic socialization of their young children is a critical yet understudied area, especially in the context of vulnerable parent-child dyads. The current longitudinal study examined factors that informed mothers’ beliefs and practices concerning children’s kindergarten readiness in a sample of 204 Mexican-origin adolescent mothers (M_age = 19.94). Adolescent mothers’ individual characteristics and assets (i.e., parental self-efficacy, educational attainment, educational utility beliefs, knowledge of child development) and sources of stress (i.e., economic hardship, coparenting conflict) were related to the importance they placed on children’s social-emotional and academic readiness for kindergarten, their provision of cognitive stimulation and emotional support to their children in the home, and their enjoyment of literacy activities with their child. Moreover, adolescents’ perception of parenting daily hassles emerged as a mediator in this process. Findings underscore the importance of considering Mexican-origin adolescent mothers’ strengths and assets along with their unique contextual stressors as they relate to beliefs and practices that could have implications for their children’s school success.

With the increasing demands placed on today’s kindergarteners, enhancing children’s school readiness has received considerable attention within the academic community. Children who enter kindergarten with fewer prerequisite skills and competencies are more likely to encounter social and behavioral difficulties (Briceno et al., 2013) and educational maladjustment (Kern & Friedman, 2009). Prior work, including that with diverse samples that included Latinx children, suggests that parents’ academic socialization, including their beliefs about the importance of school readiness skills and practices that promote children’s readiness are important predictors of children’s early school success (e.g., Barbarin et al., 2008; Belfield & Garcia, 2014; Puccioni et al., 2019; Taylor et al., 2004). Less is known about the factors that may inform parents’ beliefs and practices concerning kindergarten readiness and the mechanisms through which these academic socialization processes may unfold (Taylor et al., 2004). Such work is particularly important for populations with complex support needs, including adolescent parents and their children (Fagan & Lee, 2013). In the present study, we focused on Mexican-origin adolescent mothers, who have the highest birthrate of all ethnic groups in the United States (National Vital Statistics Report; Martin et al., 2019), and are at risk for low educational attainment (U.S. Census Bureau; McElrath & Martin, 2021). Guided by theories on the determinants of parenting (Belsky, 1984) for Latinx adolescent mothers (Contreras et al.,2002) and parents’ academic socialization of their young children (Taylor et al., 2004), the current study examined whether Mexican-origin adolescent mothers’ individual characteristics and assets (i.e., parental self-efficacy, educational attainment, educational utility beliefs, knowledge of child development) and sources of stress (i.e., economic hardship; coparenting conflict) when their children were 3 years old informed their beliefs about the importance of children’s school readiness (i.e., self-reported endorsement of the academic and social-emotional skills important for kindergarten), and early educational practices (i.e., observed and reported home cognitive and emotional stimulation and literacy enjoyment) during their children’s transition to kindergarten at age 5 years. Moreover, we also examined the role of adolescent mothers’ reports of daily hassles (perceptions of the challenges associated with parenting; Belsky et al., 1995) when children were 4 years old as a mediator of the above links.

A focus on the education-related beliefs and practices of Mexican-origin adolescent mothers is of great importance. As of 2018, Latinx children comprised 25% of the U.S. K-12 school population, and thus represent an important group to study regarding successful kindergarten transitions (U.S. Census Bureau; Bauman & Cranney, 2020). Nationally representative data that includes Latinx families suggests that children of adolescent mothers are at disproportionate risk for academic underachievement in school (Fagan & Lee, 2013). Such disparities may be related to the context of adolescent parenthood including limited resources, the challenge of negotiating child rearing with individual demands, and subsequent under-preparedness to provide developmental stimulation compared to adult parents (Briceno et al., 2013). Importantly, there is much variability in these processes and the current study aimed to bring a strengths-based perspective to understand how adolescent mothers’ assets, along with contextual stressors, informed their beliefs and practices related to children’s school readiness (Coley & Chase-Lansdale, 1998; Luster et al., 2004; Taylor et al., 2004).

In addition, examining factors that inform school readiness beliefs and practices may be of particular relevance for Mexican-origin families, given the emphasis on the cultural value of educación or being bien educado/a that has been noted in Latinx families (Bridges et al., 2012; Valdés, 1996; Valenzuela, 2010). This cultural value emphasizes the importance of childrearing and instruction not only in academic but also non-academic (e.g., social) domains, including, for example, knowing right from wrong, learning respect for parents, and engaging in moral behaviors (Reese et al., 1995). Also, the strong endorsement by Mexican-origin families of familism cultural values (i.e., familismo; Knight et al., 2010; Calzada et al., 2012), which emphasize the role of the family as a source of support, may inform Mexican-origin adolescent mothers’ academic socialization practices with their children as well as the context of supports available to mothers as they negotiate the stressors associated with parenting alongside other potential ecological challenges that Latinx families disproportionately face (e.g., high rates of poverty, language barriers, and discrimination; Cabrera et al., 2019; Edwards, 2019).

Parents’ Beliefs and Practices Concerning Children’s School Readiness

School readiness is a multidimensional concept consisting of the pre-academic (e.g., language, pre-literacy, pre-mathematics) and social-emotional (e.g., peer interactions; emotion regulation; Pelletier & Brent, 2002) skills that are related to children’s successful kindergarten transitions. The majority of research on parents’ conceptions of school readiness is based on samples comprised largely of White participants or samples of pooled ethnic-racial groups, precluding the ability to understand potential heterogeneity within Latinx families and thus representing a gap in our understanding of these processes in Latinxs (e.g., Barbarin et al., 2008). Prior work based on ethnic-racially diverse samples that include Latinx families underscores the importance of understanding parents’ beliefs about the importance of such skills and education-related practices which constitute their academic socialization (e.g., Taylor et al., 2004; Puccioni et al., 2019). Parents who endorse the importance of early academic and asocial-emotional skills may be more likely to socialize or facilitate the skills they believe to be important, which ultimately shapes their children’s early development and success in the classroom (Luster et al., 2004; Rodriguez & Tamis-LeMonda, 2011). Data from the Early Childhood Longitudinal Study (ECLS-B and ECLS-K), a nationally representative study that includes Latinx families, supports this notion; parents’ school readiness beliefs were positively associated with their education-related practices, which in turn were associated with children’s academic and social-emotional competence (Puccioni, 2018; Puccioni et al., 2019). Findings from the limited work on school readiness in Latinx parents suggests that these parents placed greater importance on children’s academic readiness, social-emotional and behavioral attributes in comparison to White parents (Puccioni, 2018). Similarly, qualitative work with Latinx immigrant parents suggests that parents emphasize the importance of children’s development of morality and responsibility prior to kindergarten (Reese, et al., 1995). More information is needed on the factors that shape academic socialization among adolescent mothers who may have fewer resources to promote school readiness (Apiwattanalunggarn & Luster, 2005).

Parents’ practices in the home environment are another important aspect of their academic socialization. Behaviors such as the provision of emotional support (e.g., sensitivity and responsiveness) and cognitive resources in the home (e.g., learning materials) have been linked to children’s positive developmental and educational outcomes (e.g., Apiwattanalunggarn & Luster, 2005; Arevalo et al., 2014; Jahromi et al., 2016; Luster et al., 2004). It is important to understand factors that promote or undermine parents’ provision of cognitive stimulation and emotional support as such qualities of children’s learning environment have been found to be varied in the homes of children born to low-income adolescent mothers and Latinx, Spanish-speaking mothers (Fagan & Lee, 2013; Rodriguez & Tamis-LeMonda, 2011) and positively associated with the academic outcomes of children in Mexican American adult parent households (Arevalo et al., 2014). In addition, it is important to examine adolescent mothers’ “literacy interface” (i.e., engagement and interest/enjoyment in literacy activities) as prior work suggests that, in general, Latinx families and adolescent mothers provide their children with fewer literacy-relevant opportunities, compared to non-Latinx (Briceno et al., 2013) and adult mother-headed families (Burgess, 2005; Fagan & Lee, 2013). Conceptually, parents’ enjoyment of reading with their children has been posited to model positive attitudes around literacy and encourage the development of children’s intrinsic value of literacy skills and self-engagement in literary activities (Burgess, 2005). Much of the above-mentioned work is not specific to Latinx families (see Arevalo et al., 2014 for an exception), reflecting a gap in our understanding of these processes specific to this important population. Latinx, African American, and Asian American parents have all been shown to engage in fewer education-related practices at home than European American parents from similar socioeconomic backgrounds (Puccioni et al., 2019). Furthermore, some work suggests that early reading activities were perceived by Latinx immigrant parents to have less importance prior to kindergarten such that parents believed children may be too young to benefit from literacy activities before the age of 5, and instead prioritized moral development (Gallimore & Goldenberg, 2001). Given the limited research on Latinx families, coupled with findings from the few existing studies indicating potential disparities, it is important to examine factors related to these education-related practices in Mexican-origin adolescent mothers.

Determinants of Adolescent Mothers’ School Readiness Beliefs and Practices

Scholars have called for research that explores the factors and mechanisms that inform variations in parents’ beliefs and practices (Puccioni et al., 2019). The current study was guided by a theoretical framework on the determinants of Latinx adolescent mothers’ parenting that emphasizes the unique contributions of mothers’ individual characteristics and assets and sources of stress as they inform aspects of their parenting (Belsky, 1984; Contreras et al., 2002). We also considered parenting hassles to be a mediator (linking these determinants to mothers’ beliefs and practices (Belsky et al., 1995) in our conceptual model (Figure 1).

Figure 1.

Conceptual Model

Adolescent Mothers’ Individual Characteristics and Assets

In the study of characteristics that may inform Mexican-origin adolescent mothers’ kindergarten readiness beliefs and practices, it is important to understand the role of mothers’ own educational attainment (Fagan & Lee, 2013; Rodriguez & Tamis-LeMonda, 2011; Taylor et al., 2004). Work with ethnic-racially diverse samples suggests that mothers with greater educational experiences may be better equipped to promote their children’s school readiness, more likely to provide stimulating home environments, and to harbor realistic expectations for skill development, compared to mothers with lower educational attainment (Burgess, 2005; Luster et al., 2004). Moreover, notions of success in school may inform mothers’ academic socialization of their children (Taylor et al., 2004). Work with Latinx adolescent mothers indicates that children of mothers who complete high school have more positive academic outcomes (Briceno et al., 2013), and research specifically focused on Latinx mothers suggests that mothers’ educational attainment may confer personal and economic resources (e.g., access to school-related social networks) that inform mothers’ academic readiness beliefs and facilitate their academic-related practices with their children (Palermo et al., 2019).

In addition, it is important to examine the cognitions (i.e., attitudes and knowledge) that may shape adolescent mothers’ understanding of the importance of their children’s school readiness (Burchinal et al., 2002). Mothers’ educational utility, or beliefs about important aspects of education, for instance, have been linked to children’s development of school readiness skills in ethnic-racially diverse samples (Barbarin et al., 2008). Children of adolescent mothers of varied ethnic-racial backgrounds whose mothers believe that education is important are more likely to be provided higher quality learning environments (Burgess, 2005). Similarly, research on predominantly White mothers or those with mixed ethnic-racial backgrounds suggests that those with greater knowledge about children’s development (i.e., accuracy concerning developmental milestones) may be more likely to structure developmentally appropriate experiences (Bornstein et al., 2010) and provide greater cognitive support in the home context (Benasich & Brooks-Gunn, 1996). In contrast, adolescent mothers with academic shortcomings may be less likely to expose their children to such contexts because they may not understand how their children will benefit from it (Burgess, 2005; Way & Leadbeater, 1999).

A final individual asset to consider is adolescent mothers’ parenting efficacy, or their perceptions of themselves as capable and effective parents. Self-efficacy has been linked to more positive parenting beliefs and practices among Latinx adolescent mothers (Contreras et al., 2002) and reports of greater involvement in children’s preschool education and promotion of school readiness among ethnically diverse parents (Pelletier & Brent, 2002).

Sources of Stress

Adolescent parenthood is often marked with socioeconomic disparities including an increased risk for low income (Briceno et al., 2013). Adolescent mothers’ experiences with economic hardship may be negatively linked to school readiness beliefs and practices as families who incur greater financial hardships may have less access to the resources needed to prepare their children for kindergarten (Barbarin et al., 2008). Past work with ethnically diverse families also suggests that parents from higher socioeconomic status backgrounds were more likely to endorse the importance of children’s school readiness (Puccioni, et al., 2019).

Prior theoretical work also emphasizes the critical role of family relationships as determinants of Latinx adolescent mothers’ parenting beliefs and practices (Contreras et al., 2002); thus, in the present study we consider the potential role of coparenting relationships with the biological fathers and grandmothers (i.e., mothers of adolescents). Specifically, due to their particular reliance on their mothers (i.e., children’s grandmothers) and children’s fathers as coparents (Contreras et al., 2002), Latinx adolescent mothers’ experiences with coparenting conflict in these relationships may impede their capacities to invest time and energy to endorse skills for kindergarten, provide a healthy learning environment, and engage in literacy activities. Parenting Hassles as a Mediator

Because the daily challenges associated with child-rearing can be perceived as particularly stressful, especially among adolescent mothers (Coley & Chase-Lansdale, 1998; Contreras et al., 2002), we examined parenting hassles as a mediator of this process. Parenting hassles, or one’s appraisal of the cumulative impact of the daily stresses of parenting, is an important index of parent stress (Crnic & Greenberg, 1990) that has been linked to a greater risk of neglectful parenting among Latinx adolescent mothers (East & Barber, 2014) and low-income parents of varied ethnic-racial backgrounds (Mathis & Bierman, 2015). Parenting stress may inhibit interactions between parents and their children that promote social and emotional development (Crnic & Greenberg, 1990; Crnic et al., 2005; East & Barber, 2014; Palermo et al., 2019). Among diverse groups of parents (i.e., African American, Caucasian), greater parenting stress was linked to parents’ views of their children’s lower perceived competencies (e.g., abilities, skills) and academic performance (Soltis et al., 2013).

Belsky et al. (1995) propose that parenting hassles may represent a “proximal” determinant of parenting that connects “distal” determinants with parents’ behaviors. Given work on the association between coparenting support (i.e., child care assistance) and less parenting hassles among Latinx adolescent mothers (East & Barber, 2014) and those of diverse ethnic-racial backgrounds (Pittman & Coley, 2011), it is possible that higher coparenting conflict may be associated with adolescent mothers’ perception of more parenting hassles, which undermine their abilities to engage in supportive parenting behaviors (Pittman & Coley, 2011). Moreover, Latinx adolescent mothers who experience more economic hardship may experience parenting as more of a hassle (East & Barber, 2014), which may impede academic socialization. Finally, mothers’ individual characteristics and assets (i.e., parenting efficacy, educational attainment, educational utility beliefs, knowledge of child development) may inform their school readiness beliefs and practices by promoting an experience of childrearing as less of a hassle, which has been found to promote mothers’ school readiness beliefs and practices in ethnic-racially diverse samples (East & Barber, 2014; Mathis & Bierman, 2015; Pelletier & Brent, 2002).

Current Study

The current study examined factors that informed Mexican-origin adolescent mothers’ beliefs about the skills necessary for children’s school readiness and early education-related practices during their children’s transition to kindergarten at age 5. In line with theoretical and empirical work on the determinants of parenting among Latinx adolescent mothers, parents’ academic socialization of their young children, and the importance of parents’ perceptions of daily hassles (Belsky et al., 1995; Contreras et al., 2002; Taylor et al., 2004), we expected that sources of stress and adolescent mothers’ individual assets when their children were 3 years old would be linked to their academic socialization when children were 5 years old, and that adolescent mothers’ reports of parenting daily hassles at child age 4 would mediate these associations. We controlled for adolescent mothers’ age and nativity given previous findings suggesting that immigrant-born and adolescent mothers may have access to fewer school readiness resources, compared to U.S.-born and adult mothers (Fagan & Lee, 2013). We also controlled for adolescent mothers’ employment status, school enrollment status, and total number of children in the home, given work suggesting that children’s developmental outcomes may be a function of early maternal employment (Youngblade, 2003), adolescent mothers’ preoccupation with their own school responsibilities (Contreras et al., 2002), and the number of children in the home (Fagan & Lee, 2013). We controlled for co-residence with grandmothers and biological fathers given evidence that co-residence may be linked to indices of school readiness for children in Latinx families (Pilkauskas, 2014). Lastly, given evidence that ethnic-racially diverse mothers’ school readiness beliefs and practices may vary based on children’s enrollment in childcare or preschool (Ansari & Winsler, 2013) and children’s gender in a sample of Latinx parents (Simons et al., 2021), analyses also controlled for these variables.

Finally, we explored whether pathways linked to Mexican-origin adolescent mothers’ academic socialization differed based on mothers’ nativity and child gender; these tests were exploratory in nature given the limited nature of past work. Specifically, there is some evidence that cultural values and immigration may play a role in the importance Latinx parents place on children’s academic readiness and social-emotional development (Puccioni, 2018; Reese et al., 1995) and on the links between Latinx adolescent parents’ behaviors and their children’s school readiness (Briceno, et al., 2013), although this work has yielded mixed findings. Similar, past work has revealed gender differences favoring Latinx girls’ school readiness in terms of vocabulary development, social skills, and attention (Ansari & Winsler, 2012; Cabrera & Hennigar, 2019) and one study of Latinx families found that parents of boys were more likely to mention the importance of their children’s math school readiness (Simons et al., 2021).

Method

Participants

Data were from a longitudinal study of 204 Mexican-origin adolescent mothers, their infants, and their mother figures (Umaña-Taylor, et al., 2015). A majority (64.4%) of adolescent mothers were born in the United States. Those born in Mexico had lived in the U.S. from less than 4 years to 22 years (M = 11.79, SD = 4.60). At Wave 5 (W5) adolescent mothers were an average of 20.94 (SD = 1.00) years old and 13.3% were enrolled in school, 45.7% had graduated from high school or earned a GED, and 41.0 % had dropped out. Forty-four percent of adolescent mothers lived with their mother figures (e.g., biological mothers, aunts, boyfriend’s mother), 30% lived with the biological father of their child, and 8% lived with both their mother and the child’s biological father at W5. At W5, children were, on average, 48.39 months of age (SD = 1.33), 61.8% of adolescent mothers had other children, and 48% of adolescent mothers reported working for money with an average hourly wage of $8.86 (SD = $2.96; range = $1.75-$30.00). At W5, 59% of adolescent mothers reported that their children (58% male) attended preschool or childcare or were regularly cared for by another person. From the 204 adolescent mothers who participated at W1, 195 (96%) participated at W2, 173 (85%) participated at W3, 171 (84%) participated at W4, 173 (85%) participated at W5, and 172 (84%) participated at W6.

We also tested mean level differences on key study variables and found that there were no significant differences on demographic factors (i.e., adolescent mothers’ nativity, age, employment status, school enrollment, coresidency with mothers, coresidency with child’s father, child’s age, enrollment in childcare/ preschool, other children in the home, child gender) or key study variables (i.e., parenting self-efficacy, educational attainment, educational utility beliefs, knowledge of child development, economic hardship, coparenting conflict with grandmothers and biological fathers, socio-emotional and academic readiness beliefs) between participants with complete versus missing data. Results available upon request.

Procedure

Pregnant adolescents were recruited from high schools, health centers, and community agencies in a Southwestern metropolitan area. To be eligible for participation, adolescents had to be of Mexican origin, pregnant, between the ages of 15 and 18 years, and not legally married. Additionally, adolescents had to have a mother figure who was also willing to participate in the study. Parental consent and youth assent were obtained for participants younger than 18 years of age, and informed consent was obtained for those who were 18 years and older, in compliance with the Human Subjects Institutional Review Board. Participants were interviewed when the adolescent mother was approximately 30.87 weeks pregnant (W1), 10-months postpartum (W2), 24-months postpartum (W3), 36-months postpartum (W4), 48-months postpartum (W5), and 60-months postpartum (W6). Adolescent mothers participated in face-to-face interviews (2.5 hours in length), in which questions were read out loud to them. Participants received compensation for their participation ($25 at W1, $30 at W2, $35 at W3, $40 at W4, $50 at W5, $60 at W6). Interviews were conducted in participants’ preferred language (e.g., English or Spanish), with 61.5% of adolescents completing their interviews in English.

Measures

Parenting Efficacy (W4; independent variable)

Adolescent mothers’ parenting efficacy was assessed using the Toddler Care Questionnaire (TCQ; Gross & Rocissano, 1988). The TCQ is a 38-item scale that assesses parents’ perceived efficacy for tasks associated with raising a young child (e.g., behavior issues). Adolescent mothers responded to statements (e.g., How confident do you feel about knowing how to make your child feel better when he/she is upset) using a 5-point Likert scale with responses ranging from 1 (“you do not feel confident about this with your child”) to 5 (“you feel very confident about this with your child”). Items were averaged, with higher scores indicating greater confidence in parenting. At W4, Cronbach’s alpha was .96.

Educational Attainment (W4; independent variable)

Adolescent mothers reported on their highest level of education completed at W4. Responses were coded based on the number of years of completed schooling. The range for educational attainment at W4 was from “7^th grade” (7) to “2-years college,” “vocational-technical school,” or “Associate degree” (14).

Educational Utility (W4; independent variable)

Adolescent mothers’ achievement-oriented values were assessed using the educational utility scale (Fuligni et al., 2005). This scale assesses adolescents’ beliefs that education is essential for success. Adolescents rated items (e.g., “Doing well in school is the best way to get ahead in life”) on a 5-point scale ranging from 1 (“not at all true”) to 5 (“almost always true”). Higher values indicated that adolescents more strongly believed in the value of getting an education. Cronbach’s alpha in the current study was .81 at W4.

Accuracy of Child Development Knowledge (W4; independent variable)

A 19-item modified version of the Knowledge of Infant Development Inventory (KIDI; MacPhee, 2002) was used at W4 to assess adolescents’ knowledge about norms and milestones, child development principles, parenting strategies, and health and safety. Adolescents rated their accuracy of child development knowledge (e.g., “A five-year old child can read four or more words”) on a four-point scale with choices of 1 (“agree”), 2 (“younger”), 3 (“older”), and 4 (“not sure”). Each of the KIDI items was recoded as right (+1), wrong (−1), or not sure (0) based on the 4-point scale. A measure of accuracy was computed (# correct/# answered), with higher scores indicating greater accuracy in knowledge of child development. At W4, the Kuder-Richardson Formula 20 (KR20) reliability was .84.

Economic Hardship (W4; independent variable)

Adolescent mothers’ economic hardship was assessed using the 17-item Economic Hardship Measure (Barrera et al., 2001), which contains four subscales. The first three were 5-point Likert scales which included items assessing financial strain [e.g., “In the next three months, how often do you think that you or your family will experience bad times such as poor housing or not having enough food?”; scale ranged from 1 (“almost never”) to 5 (“almost always”)], inability to make ends meet [e.g., “Think back over the last 3 months and tell me how much difficulty (you/your family) had with paying your bills”; scale ranged from 1 (“no difficulty at all”) to 5 (“a great deal of difficulty”)], and not having enough money for necessities (e.g., “We had enough money to afford the kind of car we need”; scale ranged from 1 (“strongly disagree”) to 5 (“strongly agree”)]. The fourth subscale assessed economic adjustments or cutbacks (e.g., “In the last 3 months, has your family had to change food shopping or eating habits a lot to save money”) using “yes” or “no” options. Consistent with guidelines in Barrera et al. (2001), a composite score was created using participants’ standardized weighted scores on all four subscales, with higher values indicating greater perceived economic hardship. Cronbach’s alpha, computed using the unweighted standardized economic hardship subscale scores, was .85 at W4.

Co-parental Conflict (W4; independent variable)

The interparental conflict scale, a subscale from Ahrons’ (1981) Co-parental Communication Scale, was used to measure the degree of hostility, conflict, tension, and disagreement between coparents when discussing parenting issues. A 4-item modified version was adapted to examine co-parental relationships between adolescent mothers and their own mother (referred to hereafter as grandmothers) and between adolescent mothers and their child’s biological father (Madden-Derdich, 2002). Items (e.g., “When you and your mother discuss parenting issues, how often is there hostility and anger between you?”) were scored on a 5-point Likert scale, ranging from 1 (“never”) to 5 (“always”). At W4 of the current study, Cronbach’s alphas for adolescent mothers’ co-parental conflict with biological fathers and grandmothers were .89 and .86, respectively.

Parenting Daily Hassles (W5; mediator)

At W5, the degree to which adolescent mothers perceived daily parenting hassles was assessed with a 24-item modified version of the Parenting Daily Hassles (PDH) scale (Crnic & Greenberg, 1990; Crnic & Booth, 1991). Adolescents rated the intensity of stress (“How much of a hassle or bother is it when your child is constantly hanging on you and getting in the way”) on a 5-point scale ranging from 1 (“not a hassle at all”) to 5 (“it’s a very big hassle”). The intensity scale was created by summing all 24 items. For the current study, Cronbach’s alpha was .91 at W5.

Endorsement of the Importance of Academic and Social Emotional Readiness for Kindergarten (W6; dependent variable)

At W6, adolescent mothers’ beliefs regarding the importance of various skills for their children’s academic and social emotional readiness for kindergarten were assessed using the Early Childhood Longitudinal Study-Kindergarten Cohort (ECLS-K) Kindergarten Teacher Questionnaire (Fall) Part B (U.S. Dept. of Education, National Center for Education Statistics, 2010). Endorsement of academic readiness (e.g., “Importance of knowing colors and shapes”) and social emotional readiness (e.g., “Importance of being sensitive to others”) were measured with 4 and 9 items, respectively, on a 5-point Likert scale with responses ranging from 1 (“not at all”) to 5 (“very much”). Mean scores were calculated for each subscale with higher scores indicating greater endorsement of the importance of academic and social-emotional skills for kindergarten. At W6, Cronbach’s alphas for endorsement of academic and social-emotional readiness skills were .79 and .87, respectively.

Provision of Emotional Support and Cognitive Stimulation in the Home (W6; dependent variable)

Adolescent mothers’ provision of emotional support and cognitive stimulation in the home at W6 were assessed with the HOME-Short Form, an adapted version of the Home Observation for Measurement of the Environment (HOME-EC; Caldwell & Bradley, 2003). The HOME-Short Form consists of 25-items that were answered by adolescent mothers, as well as items that were scored by the interviewer from her observation of the child’s physical environment. Both adolescent-response items and interviewer observations were used to assess adolescent mothers’ provision of “cognitive stimulation” (14 items, e.g., Please tell me if you have helped or are helping your child learn numbers) and “emotional support” (11-items; e.g., Mother caressed, kissed or hugged child at least once). Statements about the home environment were scored as either “not applicable” if there was no opportunity to observe, or the adolescent did not discuss this issue during the course of the visit, “yes”, or “no”. A count of all “yes” items was computed with higher scores on the HOME observation short form indicating a home environment marked with greater cognitive stimulation and emotional support. According to the authors, measures of internal consistency are not appropriate for this observation because it is an index consisting of causal indicators (Bradley, 2015).

Literacy Enjoyment (W6; dependent variable)

At W6, the degree to which adolescent mothers perceived literacy activities with their child to be enjoyable was assessed using a 4-item subscale from the parent literacy beliefs (PLB) inventory, a measure developed for the current study. Adolescent mothers reported on their experiences reading with their children (e.g., “I enjoy reading pictures or storybooks to my child”) with a 5-point Likert scale ranging from 1 (“strongly disagree”) to 5 (“strongly agree”). Mean scores were calculated with higher scores reflecting more enjoyment engaging in literacy activities. Cronbach’s alpha was .81 at W6.

Demographic and Contextual Covariates

The current analyses controlled for adolescent mothers’ reports of W1 nativity status (0 = foreign-born, 1 = U.S.-born), as well as age, employment status (0 = no, 1 = yes), school enrollment (0 = no, 1 = yes), coresidency with their mother figure (0 = no, 1 = yes), coresidency with the child’s biological father (0 = no, 1 = yes), other children (0 = no, 1 = yes), child age, child gender (0 = male, 1 = female), and children’s enrollment in childcare/preschool at W5 (0 = no, 1 = yes).

Analytic Strategy

Path analysis within a structural equation modeling framework in Mplus version 7.11 (Muthén & Muthén, 2010) was used to examine study hypotheses. Control variables included adolescent mothers’ nativity, age, employment status, school enrollment, coresidency with mother and child’s biological father, having other children, child age, child gender, and children’s enrollment in childcare/preschool at W5. All exogenous variables were allowed to covary.

Missing data were accounted for using full information maximum likelihood (Enders, 2010). As recommended, good model fit was determined based on fit indices with a value of less than or equal to 0.05 for the root mean square error of approximation (RMSEA; Kline, 1998), greater than or equal to 0.95 for the comparative fit index (CFI), and less than 0.05 for the standardized root mean square residual (SRMR; Hu & Bentler, 1999). Mediation was tested using the bias-corrected bootstrap method (with 2000 bootstrapped samples) to compute confidence intervals for significant mediational pathways. (Mackinnon et al., 2004). Mediation is significant if the confidence intervals do not contain zero (Fritz & MacKinnon, 2007).

Sensitivity analyses were then conducted such that multi-group models were tested for significant differences by children’s gender and adolescent mothers’ nativity status. These tests compared nested models and examined the difference between them using the Satorra-Bentler scaled (mean adjusted; Satorra, 2000) chi-square difference test, the ΔCFI, and model fit. Using this method, the first model allowed path estimates to be freely estimated across groups (i.e., unconstrained model) and the second model constrained the path estimates to be equal across groups (i.e., fully constrained model). If the Satorra-Bentler chi-square difference test was significant or the ΔCFI was greater than .01, this suggested significant differences between groups. Subsequent models were then tested by sequentially constraining paths to identify which paths differed between groups.

Results

The current study first examined the associations among adolescent mothers’ individual characteristics and assets (i.e., parental self-efficacy, educational attainment, educational utility beliefs, knowledge of child development), sources of stress (i.e., economic hardship; coparenting), and their beliefs and practices concerning kindergarten readiness (see Table 1 for descriptive data and bivariate associations). To examine potential differences by children’s gender and adolescent mothers’ nativity, a series of t-tests were conducted (See Table 2). Only two significant mean level differences emerged. First, adolescent mothers of male children reported significantly higher endorsement of emotional support in the home at W4, compared to those of female children. Next, U.S. born adolescent mothers reported significantly higher levels of parenting efficacy at W4, compared to Mexico-born adolescent mothers. Adolescent mothers’ characteristics and assets and sources of stress were hypothesized to be indirectly related to their beliefs and practices, such that perceived parenting hassles would mediate this association. Fit indices met our criteria for good model fit, χ² (10) = 11.51, p = .32, χ²/df ratio = 1.15, CFI = 1.00, RMSEA= .03 (90% C.I. = 0.00, 0.08), and SRMR = .01 (see Figure 2).

Table 1.

Means, Standard Deviations, and Correlations among Key Study Variables (N = 204)

	1	2	3	4	5	6	7	8	9	10	11	12	13
1. Economic hardship W4	-
2. Biological father conflict W4	.01	-
3. Grandmother conflict W4	.29***	.30***	-
4. Parenting efficacy W4	−.26**	−.18*	−.15	-
5. Educational attainment W4	−.21**	.07	−.06	.03	-
6. Educational utility W4	.08	.05	.01	.24**	.08	-
7. Knowledge of Child Devel. W4	−.17*	.02	−.07	.08	.24**	−.03	-
8. Parenting hassles W5	.26**	.30***	.37***	−.30***	−.12	−.12	.02	-
9. Social-emot. readiness belief W6	−.30***	−.14	−.12	.33***	.29***	.21**	.15	−.30***	-
10. Academic readiness belief W6	−.34***	−.01	−.11	.18*	.31***	.15	.20*	−.23**	.75***	-
11. Cognitive stimulation W6	−.22*	−.23*	−.06	.17	.18*	.09	.12	−.20*	.28**	.27**	-
12. Emotional support W6	−.09	−.12	.02	.24**	.17	.17	.13	−.14	.32***	.22*	.16	-
13. Literacy enjoyment W6	−.24**	−.07	−.10	.34***	.08	.20*	.20*	−.21**	.41***	.37***	.28**	.08	-
Mean	.00	2.56	2.20	168.43	11.12	4.53	0.56	39.80	4.18	4.23	11.10	7.27	4.59
Standard Deviation	2.39	1.04	0.87	18.12	1.54	0.53	0.15	11.50	0.66	0.75	1.90	1.40	0.49
Percent missing	16.7%	34.8%	17.6%	18.1%	17.6%	16.7%	18.1%	15.2%	15.7%	15.7%	34.3%	35.3%	16.7%

Note. W4 = Wave 4; W5 = Wave 5; W6 = Wave 6; Devel = Development; Social-emot = Social-emotional.

^*p < .05

^**p < .01

^***p < .001

Table 2.

Group Differences in Study Variables by Nativity and Child Gender

	Mothers’ Nativity				Children’s Gender
	US-Born		Mexico-Born		Female		Male
	M	SD	M	SD	M	SD	M	SD
1. Economic hardship W4	−0.33	2.23	0.59	2.55	0.28	2.31	−0.20	2.44
2. Biological father conflict W4	2.66	1.08	2.37	0.94	2.53	1.02	2.57	1.06
3. Grandmother conflict W4	2.23	0.83	2.16	0.93	2.21	0.84	2.19	0.89
4. Parenting efficacy W4	170.88a	15.26	163.95a	21.88	167.76	20.28	168.92	16.48
5. Educational attainment W4	11.39	1.50	10.66	1.51	11.11	1.62	11.14	1.49
6. Educational utility W4	4.53	0.56	4.53	0.48	4.47	0.60	4.58	0.48
7. Knowledge of Child Devel. W4	0.56	0.15	0.56	0.17	0.55	0.15	0.57	0.16
8. Parenting hassles W5	40.03	11.20	39.38	12.12	40.60	10.70	39.23	12.06
9. Social-emot. readiness belief W6	4.33	0.60	3.89	0.68	4.14	0.67	4.20	0.65
10. Academic readiness belief W6	4.38	0.70	3.95	0.76	4.17	0.74	4.27	0.75
11. Cognitive stimulation W6	11.15	1.85	11.00	2.01	11.27	1.98	10.96	1.82
12. Emotional support W6	7.31	1.45	7.13	1.30	7.02a	1.27	7.49a	1.48
13. Literacy enjoyment W6	4.57	0.49	4.53	0.51	4.56	0.52	4.55	0.48

Note. Sample sizes for analyses by mother nativity ranged from 87 to 131 for US-born and 38 to 73 for Mexico-born; Sample sizes for analyses by child gender ranged from 59 to 86 for female and 70 to 118 for male; W4 = Wave 4; W5 = Wave 5; W6 = Wave 6; Devel = Development; Social-emot = Social-emotional. Means within rows with the same superscript are significantly different from one another at p < .05.

Figure 2:

Hypothesized Study Model. Model fit, χ² (10) = 11.51, p = .32, χ² /df ratio = 1.15, CFI = 1.00, RMSEA= .03 (90% C.I. = 0.00, 0.08), and SRMR = .01. Unstandardized coefficients displayed. Note. W1 = Wave 1, W2 = Wave 2, W4 = Wave 4, W5 = Wave 5, W6 = Wave 6. Nativity (0 = foreign born, 1 = U.S.-born), employment status (0 = no, 1 = yes), school enrollment (0 = no, 1 = yes), coresidency (0 = no, 1 = yes), child’s enrollment in childcare/preschool (0 = no, 1 = yes), other children (0 = no, 1 = yes), child gender (0 = male, 1 = female). Bolded paths indicate significant mediation. ** p < .01, *p < .05, +p = .05.

Hypothesized Associations

With respect to the role of individual characteristics and assets, all direct paths described below were in the hypothesized direction and statistically significant (unstandardized coefficients reported). As adolescent mothers reported higher parenting efficacy at W4, they also tended to report higher endorsement of social-emotional skills as important to children’s school readiness (B = .01, SE = .003, p = .045), greater provision of emotional support in the home (B = .02, SE = .01, p = .001), and higher literacy enjoyment (B = .01, SE = .002, p = .001) at W6. Adolescent mothers who reported higher levels of W4 educational attainment tended to report higher endorsement of social-emotional (B = .09, SE = .03, p = .009) and academic skills (B = .08, SE = .04, p = .05) as important to children’s school readiness, and a greater provision of emotional support in the home (B = .21, SE = .10, p = .04) at W6. Next, higher levels of educational utility beliefs at W4 were positively associated with a greater endorsement of social-emotional skills as important for school readiness (B = .19, SE = .09, p = .04) and greater literacy enjoyment (B = .15, SE = .07, p = .03) at W6. Additionally, adolescent mothers who reported higher levels of W4 knowledge of child development reported a greater endorsement of academic skills as important to kindergarten (B = .73, SE = .35, p = .04) and greater literacy enjoyment (B = .70, SE = .23, p = .003) at W6. Concerning sources of stress, adolescent mothers who reported higher levels of economic hardship at W4 tended to report lower levels of endorsement of the academic skills for children’s kindergarten readiness (B = −.07, SE = .03, p = .01), literacy enjoyment (B = −.04, SE = .02, p = .04), and cognitive stimulation at home (B = −.17, SE = .08, p = .04) at W6.

Mediation of Hypothesized Model

Eight mediation paths were tested (i.e., W4 economic hardship, W4 coparental conflict with grandmother, W4 coparental conflict with child’s biological father, and W4 parenting efficacy to W6 socioemotional readiness and W6 academic readiness via W5 parenting hassles). Mediation analyses indicated that W4 coparenting conflict with grandmothers, economic hardship, and parenting efficacy indirectly predicted adolescent mothers’ endorsement of social-emotional school readiness beliefs at W6 via W5 perceived parenting hassles. Also, W4 coparenting conflict with grandmothers indirectly predicted adolescent mothers’ endorsement of academic skills as important for readiness at W6 via W5 perceived parenting hassles. All other mediation paths were non-significant (See Table 3).

Table 3.

Tested Mediation Effects between W4 Predictors and W6 Outcomes via W5 Mediator (i.e., Parenting Hassles)

	Dependent Variable
	W6 Endorsement of Socio-emotional readiness for kindergarten			W6 Endorsement of Academic readiness for kindergarten
Predictors	Indirect Effect	SE	95% CI	Indirect Effect	SE	95% CI
W4 Economic Hardship	−0.031	0.020	[−0.076, −0.001]		−0.027	0.019	[−0.072, 0.001]
W4 Coparental Conflict with Biological Father	−0.033	0.022	[−0.082, 0.000]		−0.029	0.021	[−0.078, 0.002]
W4 Coparental Conflict with Grandmother	−0.045	0.023	[−0.097, −0.007]		−0.039	0.024	[−0.093, −0.001]
W4 Parenting Efficacy	0.033	0.021	[0.001, 0.081]		0.029	0.021	[−0.001, 0.077]

Note. W4 = Wave 4, W5 = Wave 5, W6 = Wave 6; SE = Standard error; CI = Confidence Interval. Bolded font indicates significant mediated effect.

Additional Analyses: Moderation by Gender and Nativity

To explore whether associations in the hypothesized model varied by children’s gender (i.e., 0 = boys vs., 1 = girls) or adolescent mothers’ nativity status (i.e., 0 = “Mexico-born” vs., 1 = “U.S.-born”) we tested multi-group models. In the first model, we compared an unconstrained model in which all paths were freely estimated across children’s gender to a nested model in which all estimates were constrained to be equal. The chi-square difference test was significant [Δ χ² (Δ df = 92) = 121.153, p = .02], suggesting that there were significant differences based on children’s gender. Individual paths, including study controls, were sequentially constrained to be equal to identify which paths significantly differed by children’s gender.

The final partially constrained model is shown in Figure 3. The links between individual characteristics and assets (i.e., W4 parenting efficacy, educational attainment, educational utility beliefs, accuracy of child development knowledge) and W6 study outcomes did not vary by gender, but sources of stress did (i.e., W4 coparenting conflict with biological fathers and coparenting conflict with grandmothers, W5 parenting hassles). In particular, children’s gender moderated the link between W4 coparenting conflict with biological fathers and W6 cognitive stimulation in the home, such that among girls W4 coparenting conflict with biological fathers was negatively linked with W6 cognitive stimulation in the home (B = −.87, SE = .22, p < .001), but among boys, this link was not significant (B = .04, SE = .22, p = .84). Next, the link between W4 coparenting conflict with grandmothers and W5 parenting hassles was also moderated by children’s gender, such that W4 coparenting conflict with grandmothers was positively associated with W5 parenting hassles among boys (B = 4.42, SE = 1.31, p = .001), whereas there was no significant link among girls (B = 0.98, SE = 1.58, p = .54). Additionally, a trend emerged with W4 coparenting conflict with grandmothers positively linked to W6 academic readiness among boys (B = .16, SE = .09, p = .06) and negatively linked among girls (B = −.09, SE = .09, p = .32); this finding may be attributed to multicollinearity and should therefore be interpreted with caution given negative associations at the bivariate level (r _boys = −.06, p = .57; r _girls = −.17, p = .18). Children’s gender also moderated the link between W5 parenting hassles and W6 endorsement of social emotional readiness for kindergarten, with a significant negative association for girls, (B = −.02, SE = .01, p < .001), and a non-significant link for boys (B = −.00, SE = .01, p = .43).

Figure 3:

Multi-group Analysis by Child’s Gender (i.e., girls vs. boys). Model fit, χ² (df = 100) = 89.35, p = .77, χ² /df ratio = 0.89, CFI = 1.00; RMSEA = .00 (.00-.04), SRMR < .05. Unstandardized coefficients displayed for all trending or significant paths, with estimates for boys presented within parentheses.

Note. W1 = Wave 1, W2 = Wave 2, W4 = Wave 4, W5 = Wave 5, W6 = Wave 6. Study controls (i.e., adolescent nativity, age, employment status, school enrollment, coresidency with mothers, coresidency with child’s father, child age, other children, child enrollment in childcare/preschool) included in model constraint testing within the partially constrained model. Bolded paths indicate significant mediation.

Moreover, one key study association varied by children’s gender, but was not significant in the model and therefore excluded from the figure. Specifically, the link between W4 economic hardship and W6 emotional support in the home significantly differed by gender, such that the association was stronger among girls (B = −.09, SE = .08, p = .25), compared to boys (B = −.02, SE = .09, p = .85). All other key study associations did not differ significantly by children’s gender, as the ΔCFI was less than .01 and the adjusted chi-square difference tests were not significant (all p > .05). The final partially constrained model demonstrated good model fit [χ² (df = 100) = 89.35, p = .77; CFI = 1.00; RMSEA = .00 (.00-.04), SRMR < .05].

In the second multi-group model, we then tested for differences by adolescent mothers’ nativity status. The chi-square difference test was significant [Δ χ² (Δ df = 92) = 165.08, p <.001], indicating significant differences based on nativity status. Thus, the unconstrained model was used to determine which paths significantly differed by nativity. With respect to individual characteristics and assets, adolescent mothers’ nativity status moderated several associations (See Figure 4). Notably, the link between W4 educational attainment and W6 social emotional readiness for kindergarten was positively associated among Mexico-born adolescents (B = .18, SE = .05, p < .001), but only marginally significant among U.S. born-adolescents (B = .06, SE = .03, p = .07); the link between W4 educational attainment and W6 emotional support in the home was also positively associated among Mexico-born adolescents (B = .50, SE = .15, p = .001) but only marginally significant among U.S. born-adolescents (B = .21, SE = .11, p = .06); the link between W4 educational attainment and W6 cognitive stimulation in the home was positively associated among Mexico-born adolescents (B = .68, SE = .21, p = .001) but not significant among U.S. born-adolescents (B = .08, SE = .13, p = .55).

Figure 4:

Multi-group analysis by Adolescent Mothers’ Nativity Status (i.e., U.S.-born vs. foreign born). Model fit, χ² (89) = 92.10, p = .39, χ² /df ratio = 1.03, CFI = 0.99, RMSEA= .02 (90% C.I. = 0.00, 0.06), and SMR = .06. Unstandardized coefficients displayed for all trending or significant paths, with estimates for foreign-born adolescent mothers presented within parentheses. Note. W1 = Wave 1, W2 = Wave 2, W4 = Wave 4, W5 = Wave 5, W6 = Wave 6. Study controls (i.e., adolescent age, employment status, school enrollment, coresidency with mothers, coresidency with child’s father, child age, other children, child gender, child enrollment in childcare/ preschool) included in model constraint testing within the partially constrained model. Bolded paths indicate significant mediation. *** p < .001, ** p < .01, *p < .05, +p < .10.

Next, links between W4 accuracy of child development knowledge and W6 study outcomes also varied by adolescents’ nativity status, such that W4 accuracy of child development knowledge was positively associated with W6 endorsement of socio-emotional readiness for kindergarten among U.S.-born adolescents (B = .75, SE = .31, p = .01) but negatively associated and not significant among Mexico-born adolescents (B = −.66, SE = .41, p = .11); further, W4 accuracy of child development knowledge was negatively associated with W6 cognitive stimulation in the home among Mexico-born adolescents (B = −3.35, SE = 1.62, p = .04) but positively associated and not significant among U.S.-born adolescents (B = .90, SE = 1.41, p = .52); W4 accuracy of child development knowledge was positively associated with W6 literacy enjoyment among U.S.-born adolescents (B = 1.06, SE = .30, p < .001), but not significant among Mexico-born adolescents (B = 0.05, SE = .36, p = .89); last, W4 accuracy of child development knowledge was negatively associated with W6 provision of emotional support in the home among Mexico-born adolescents (B = −2.01, SE = 1.08, p = .06) and positively associated among U.S.-born adolescents (B = 1.77, SE = 1.05, p = .09), albeit trending significance across groups, consistent with bivariate associations (r _boys = −.06, p = .57; r _girls = −.17, p = .18). Additionally, adolescents’ nativity moderated associations between W4 parenting efficacy and W6 cognitive stimulation in the home, such that W4 parenting efficacy was positively linked with W6 cognitive stimulation among U.S.-born adolescents (B = 0.04, SE = .01, p = .003), but negatively and marginally associated among Mexico-born adolescents (B = −0.02, SE = .01, p = .08).

Regarding sources of stress, adolescents’ nativity status moderated associations such that W4 economic hardship was positively linked with W5 parenting hassles (B = 1.47, SE = .45, p = .001) among U.S.-born adolescents, but negative and not significant among Mexico-born adolescents (B = −.42, SE = .61, p = .49). Next, W4 coparenting conflict with biological fathers was positively linked to W5 parenting hassles among U.S.-born adolescents (B = 2.16, SE = 1.01, p = .03), yet negatively associated and not significant among Mexico-born adolescents (B = −1.95, SE = 1.84, p = .29). Further, W4 coparenting conflict with grandmothers was positively linked to W5 parenting hassles among Mexico-born adolescents (B = 6.34, SE = 1.79, p < .001), yet not significant among U.S.-born adolescents (B = 1.61, SE = 1.24, p = .20). Finally, adolescents’ nativity status also moderated links between W5 parenting hassles and W6 endorsements of socio-emotional readiness and W6 academic readiness for kindergarten. Specifically, the association between W5 parenting hassles and W6 socio-emotional readiness was negative among U.S-born adolescents (B = −.02, SE = .005, p < .001) and this path was not significant among Mexico-born adolescents (B = .00, SE = .006, p = .96). Finally, W5 parenting hassles was negatively linked with W6 academic readiness among U.S.-born adolescents (B = −.02, SE = .006, p = .001) and not significant among Mexico -born adolescents (B = .00, SE = .008, p = .95). All other key study associations did not differ significantly by adolescent mothers’ nativity status (i.e., ΔCFI was less than .01 and adjusted chi-square difference tests were not significant). The final partially constrained model demonstrated good model fit [χ² (df = 89) = 92.10, p = .39; CFI = 0.99; RMSEA = .02 (.00-.06), SRMR = .06].

Discussion

Guided by conceptual frameworks on the factors that inform Latinx adolescent mothers’ parenting (Contreras et al., 2002) and parents’ academic socialization efforts (Taylor et al., 2004), the current study examined Mexican-origin adolescent mothers’ beliefs and practices concerning children’s kindergarten readiness. As expected and consistent with theoretical notions on the determinants of parenting among Latinx adolescent mothers which stress the importance of mothers’ individual characteristics and assets and sources of stress as they inform their parenting (Belsky, 1984; Contreras et al., 2002), Mexican-origin adolescent mothers’ educational assets, self-efficacy, and knowledge of children’s development were positively linked to their beliefs and practices concerning children’s kindergarten readiness, whereas stressors (i.e., economic hardship and coparenting conflict) were negatively related to their academic socialization of their young children. Furthermore, supporting theoretical notions that conceptualize parenting hassles as a mediator linking determinants to parenting (Belsky et al., 1995), mothers’ experiences of daily parenting hassles emerged as a mediator in the link between economic hardship, coparenting conflict, and parental self-efficacy with their academic socialization. These findings have implications for programs to targeting adolescent mothers’ academic socialization toward the benefit of their offspring’s successful kindergarten transition.

Adolescents’ Individual Characteristics and Assets

Contreras and colleagues (2002) posit that adolescent mothers’ individual characteristics and assets may serve as important resources to inform the academic socialization of their young children; consistent with these ideas, our study found that adolescents’ educational utility beliefs and educational attainment were positively linked to their endorsement of academic and social-emotional skills as important for children’s kindergarten readiness. Mothers with higher levels of educational attainment may have greater access to information about the skills needed to be successful in school (Briceno et al., 2013; Rodriguez & Tamis-LeMonda, 2011), as well as accurate expectations for their offspring’s academic and social-emotional skill development, compared to mothers with lower educational attainment (Luster et al., 2004). Young mothers who believe that education is key to success in life, and who have higher levels of education themselves (e.g., high school completion), may also have a greater awareness of the social skills needed to function in academic settings (e.g., sharing, communication) or access to resources that directly informs their educational-related beliefs and practices with their children (Palermo et al., 2019). Interestingly, results from the multi-group analyses found that the link from educational attainment to endorsement of social-emotional readiness were significant for Mexican-born adolescent mothers, but trending for US-born mothers. It may be that because educational values are often related to immigrant families’ reasons for migration (e.g., Langenkamp, 2019), these factors are particularly salient to such mothers’ academic socialization as compared to US-born adolescent mothers (Simons et al., 2021). Moreover, these findings may reflect the emphasis on social and moral domains of development for Latinx parents’ conceptions of children’s instruction, or educación (i.e., parents’ socialization of children’s good behaviors, manners, and respect; Reese et al., 1995).

Likewise, adolescents’ educational assets were also linked to their practices. Mothers with higher educational attainment were more likely to provide emotional support in the home. It may be that those mothers who continued their education were better equipped to provide an emotionally-sensitive home environment as prior work has linked young mothers’ educational resources with responsive parent-child interactions (East & Barber, 2014). Again, these pathways were found to be significant for adolescents born in Mexico, and trending or non-significant for US-born mothers. It is possible that Mexican-born mothers are more strongly connected to traditional cultural values (Gonzales et al., 2008; Updegraff & Umaña-Taylor, 2010), and had a stronger focus on their children being bien educado as a cultural value.

Adolescent mothers with higher educational utility beliefs also reported greater literacy enjoyment suggesting that those mothers who believe that education is important may also appreciate the importance of their child’s emerging literacy, and find literacy behaviors with their child more reinforcing. Although past work has found Mexican-origin adolescent mothers to be less inclined to expose their children to literacy activities due to perceptions of minimal benefits (Burgess, 2005), our study points to educational utility beliefs as an important factor that can introduce variability into their beliefs regarding literacy activities.

Adolescent mothers’ knowledge of child development was positively linked with their beliefs regarding the importance of academic skills for kindergarten readiness and their literacy enjoyment. Importantly, the paths between mothers’ knowledge of child development and their beliefs about the importance of social emotional readiness and literacy enjoyment was found to be significant for US-born mothers and not Mexican-born mothers. In contrast, the link between knowledge of child development and the provision of cognitive stimulation in the home was significant for Mexican-born mothers but not US-born mothers. It is not clear why nativity might moderate the association between knowledge of child development and educational-related beliefs and practices, as one would generally expect knowledge to afford mothers a better appreciation of requisite skills children may need prior to school entry (Benasich & Brooks-Gunn, 1996; Bornstein et al., 2010). It may be that for Mexican-born mothers, their beliefs about the importance of social-emotional readiness reflects a cultural value that is not necessarily reliant on their knowledge of child development. Moreover, as past work found that Latinx immigrant parents reported believing that children younger than 5 were too young to benefit from literacy activities, and instead prioritized moral development (Gallimore & Goldenberg, 2001), it may be that mothers’ cultural priority was on children’s social-emotional development for Mexican-born mothers, whereas beliefs about the importance of literacy were less salient.

Finally, adolescent mothers’ parenting efficacy was positively linked to their belief that academic and social-emotional skills are important for kindergarten readiness, their provision of emotional support in the home context, and their literacy enjoyment, and a link emerged between self-efficacy and cognitive stimulation for US-born mothers. It is possible that young mothers may be more likely to endorse beliefs about the importance of strengthening their child’s capacity to build social relationships and regulate emotions in preparation for entry to kindergarten when they have greater confidence in their ability to promote those aspects of children’s development (Pelletier & Brent, 2002). Consistent with Contreras and colleagues’ (2002) model of parenting, it is possible that the more efficacious that adolescent mothers feel about their parenting skills, the more competent they are at engaging in nurturing behaviors with their children. It also may be that adolescent mothers with greater confidence in their parenting skills report literacy activities with their children as more enjoyable as a result of having more positive attitudes toward supporting this and other aspects of their children’s development. It is important to note that Mexican-born mothers in our sample reported significantly lower parenting self-efficacy than US-born mothers, highlighting the importance for future work to target this aspect of parental wellbeing among immigrant adolescent mothers in particular. Together, these findings bring a strengths-based approach to our understanding of Mexican-origin adolescent mothers and their young children by highlighting several key assets that are linked to mothers’ beliefs and practices related to their children’s kindergarten readiness.

Sources of Stress

In line with past work linking parents’ socioeconomic status with their academic socialization, adolescent mothers in the current study who reported greater economic hardship reported a lower endorsement of social-emotional skills as important for kindergarten, and lower provision of cognitive stimulation in the home environment. It may be that young mothers who experience high levels of economic hardship in the early years of parenting lack the financial and social capital to access resources (e.g., child care/preschool) that could facilitate a greater awareness of their child’s necessary preparation for kindergarten. It is also possible that challenges associated with poverty may reflect underlying disadvantages to endorsing readiness for kindergarten. Moreover, young mothers who experience economic difficulties may lack the financial resources to provide necessary learning materials (e.g., books, educational games) for their children (Barbarin et al., 2008). Together, these findings are in line with those of past work with diverse samples and suggest that, within our sample of Mexican-origin adolescent mothers, economic barriers play an important in mothers’ beliefs and practices concerning their children’s kindergarten readiness. Given the economic disparities faced by Mexican-origin families, future research efforts are needed to identify potential buffers against these economic stressors and opportunities for intervention with Mexican-origin adolescent mothers.

Interestingly, results of the multi-group analysis by child gender revealed a negative link between adolescent mothers’ coparenting conflict with children’s biological fathers and the provision of cognitive stimulation for girls but not boys. Moreover, there was a group difference in the sample as a whole, such that mothers of boys reported higher provision of emotional stimulation. There is some limited work with diverse samples that include Latinx families suggesting that as adolescent mothers report greater coparenting conflict with their children’s biological fathers, those fathers tend to be less engaged and involved with their children (Fagan & Lee, 2011). Furthermore, there is evidence from predominantly White samples that fathers of girls, compared to boys, are more likely to withdraw from their parenting role in the context of marital distress (Elliston et al., 2008). In the present study we cannot establish whether decreased provision of cognitive stimulation for girls was due to the reduced involvement or engagement of their fathers, thus further research is needed to explore this finding.

Mediating Role of Parenting Hassles

Parenting hassles is an important stressor faced by adolescent mothers (Coley & Chase-Lansdale, 1998) that has been linked to children’s social functioning (Crnic & Greenberg, 1990) and academic skill development (Farver et al., 2006). Less is known about whether parenting hassles may link adolescent mothers’ individual characteristics and sources of stress with their academic socialization. Findings from the current study partially supported our hypotheses of mediational links. Specifically, adolescents who perceived greater coparenting conflict with grandmothers reported greater parenting hassles, which were then negatively associated with beliefs concerning the importance of social-emotional skills as important to kindergarten readiness. Interestingly, parenting hassles did not emerge as a significant mediator between coparenting conflict with biological fathers. It may be that co-parental conflict between Mexican-origin adolescent mothers and grandmothers played a stronger role in adolescent mothers’ perceptions of the parenting role as a hassle, as young mothers are particularly likely to co-reside with and share in childrearing responsibilities with their own mothers (Contreras et al., 2002). Given that we recruited participants who were unmarried and able to participate with a female family member, the coparenting relationship with grandparents may have been particularly salient in our sample. It could also be that those adolescents who relied more heavily on grandmothers for coparenting were more likely to experience conflict within their shared parenting role, and that such conflict was a greater risk for their experiences of childrearing as a hassle. Finally, it is possible that an underlying third variable, such as aspects of their personality traits (Belsky et al., 1995), explain the association between adolescent mothers’ experiences of coparenting conflict and perceptions of daily hassles. Such an assertion would be consistent with theoretical perspectives on the determinants of parenting that informed this study (Belsky, 1984; Contreras et al., 2002) and should be explored in future work. Because prior work has demonstrated that it is possible to reduce mothers’ perceptions of parenting hassles via interventions (e.g., emotional disclosure through journal writing; Whitney & Smith, 2014), such efforts may help young mothers to lessen the hassles associated with parenting.

Child gender and adolescent nativity status were found to moderate some of the above mentioned links. Specifically, coparenting conflict with grandmothers was positively related to adolescents’ perceptions of parenting hassles for boys but not girls. This finding is consistent with limited work on predominantly White parents suggesting boys may be more vulnerable to the negative effects of parental conflict (Kolak & Vernon-Feagans, 2008). Additionally, whereas coparenting conflict with grandmothers was negatively link to parenting hassles for Mexico-born adolescents, coparenting conflict with children’s biological fathers was linked to adolescents’ perceptions of parenting hassles for US-born mothers, indicating there may have been greater reliance on children’s grandmothers, and a greater experience of stress resulting from conflict in such coparenting relationships, for immigrant adolescent mothers. Finally, parenting hassles was negatively linked to adolescents’ endorsement of social emotional readiness for girls but not boys and linked to lower endorsement of academic and social emotional readiness among US-born but not Mexico-born mothers. Limited information on gender differences in Latinx immigrant parents’ school readiness beliefs indicates that parents of boys were more likely to endorse the importance of mathematics (Simons et al., 2021). Although preliminary, our finding may suggest that factors linked to Mexican-origin adolescent mothers’ beliefs about social-emotional domains of school readiness may also differ according to gender-related expectations.

Regarding the link between individual-level assets and adolescent mothers’ school readiness beliefs and practices, only parenting efficacy emerged as a significant indirect predictor via parenting hassles. Specifically, higher levels of parenting efficacy were associated with lower levels of parenting hassles, which was then linked to mothers’ greater endorsement of social-emotional readiness for kindergarten. In line with prior empirical and conceptual work with young Latinx mothers, adolescents who report high levels of parenting efficacy may be more likely to endorse school readiness as they may be more equipped to balance and negotiate parenting hassles with positive practices (Pelletier & Brent, 2002). With respect to the non-significant indirect effects of other individual-level assets, it may be that parenting efficacy is a stronger link to parenting hassles than educational and knowledge-related assets, as it reflects adolescent mothers’ sense of confidence in their role as a parent, which may be more salient to their experience of stress and parenting than knowledge of developmental milestones. Together, results from our study suggest that knowledge and self-efficacy are both important to adolescent mothers’ outcomes and aspects of their parenting and should be the target of intervention efforts.

Limitations and Future Directions

The current study focused on Mexican-origin adolescent mothers’ beliefs and practices and offered important new insights into individual and contextual factors that can inform their beliefs and practices regarding their young children’s school readiness. The literature would benefit from understanding how the beliefs and practices of other central caregivers, such as fathers and grandparents, may inform these processes in this population of children. Second, the present study focused on school readiness beliefs and practices (i.e., literacy enjoyment, endorsement of kindergarten readiness, provision of cognitive and emotional support) 5 years post-partum. It will be informative for future research to extend this work by examining how these processes change over time after school entry, as academic socialization perspectives may change as parents and children interact with schools (East & Barber, 2014; Crnic et al., 2005). Future work should also examine the direct associations between these indices of academic socialization and children’s school readiness skills and behaviors. Additionally, analyses relied primarily on self-report data, introducing the potential problem of shared method variance for young mothers’ reported risks (e.g., co-parental conflict) and study outcomes (i.e., rater effects; Podsakoff, et al., 2003) as well as the potential that measures (e.g., endorsement of academic and social emotional readiness skills) were subject to positive bias. Future research would benefit from the use of observational methods to assess multidimensional features of parenting characteristics and school readiness practices. Finally, future work should incorporate repeated measures of these processes to examine mechanisms when controlling for prior measures.

Despite the above limitations, the current findings have important implications for future research. First, these findings contribute to a growing body of literature underscoring the need to investigate the context under which adolescent mothers’ academic socialization of their young children develops (Fagan & Lee, 2013; Luster et al., 2004). Professionals may benefit from considering the ecological context of stressors while also recognizing the unique role of adolescent mothers’ individual assets that inform their kindergarten readiness beliefs and practices. Furthermore, findings illustrate the importance of interventions seeking to support young Mexican-origin mothers’ academic socialization to target parenting hassles, as frustrations related to childrearing may jeopardize mothers’ abilities to endorse school readiness skills and engage in practices that that support their children’s development. Overall, ensuring the school readiness of Latinx children, who represent 25% of the K-12 U.S. population (Bauman and Cranney, 2020), is critical to society at large.