Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Feb 1.
Published in final edited form as: Dev Psychol. 2019 Dec 19;56(2):208–220. doi: 10.1037/dev0000878

Helping Me Helps Us: The Role of Natural Mentors in the Ethnic Identity & Academic Outcomes of Latinx Adolescents

Bernadette Sánchez 1, Amy J Anderson 1, Jocelyn Smith Carter 1, Alison L Mroczkowski 2, Lidia Monjaras-Gaytan 3, David L DuBois 4
PMCID: PMC6980915  NIHMSID: NIHMS1064240  PMID: 31855014

Abstract

This research investigated the potential contributions of natural mentoring relationships and ethnic identity to the academic attitudes and performance of Latinx high school students. In a sample of 347 urban Latinx students in grades 9 and 10, analyses examined the presence and quality of mentoring relationships as well as ethnic identity exploration and affirmation as predictors of changes in grade point average and the perceived economic value of education. Analyses included tests for a hypothesized role of ethnic identity in mediating associations of mentoring measures with the academic outcomes. The presence of a mentoring relationship was not significantly related to ethnic identity or change in academic outcomes over time. However, mentoring relationship quality was associated with a more positive ethnic identity, and support was found for the hypothesis that ethnic identity mediates the association between the quality of mentoring relationships and change over time in the economic values towards education among Latinx adolescents. Study findings suggest the importance of supportive adults in the ethnic identity and academic outcomes of Latinx adolescents.

Keywords: mentoring, ethnic identity, Latinos, adolescents, academic outcomes, values toward education

People of Hispanic origin are currently the largest racial/ethnic minority group in the U.S, making up over 17% of the U.S. population in 2016 (U.S. Census Bureau, Population Division, 2017), and this population is expected to increase 86% by 2050 (U.S. Census Bureau, Population Division, 2017). Unfortunately, their educational attainment is lower compared to other racial/ethnic groups in the U.S. Reports indicate that only 70% of Hispanic adults over the age of 25 had a high school degree compared to about 90% of White adults (U.S. Census Bureau, 2017). Thus, it is imperative to find ways to improve academic outcomes among Latinx youth, particularly during the transition to high school when adolescents are most vulnerable for academic failure and dropout (Reyes, Gillock, Kobus & Sánchez, 2000). The current study examines the roles of mentoring and ethnic identity in the academic outcomes of urban, low-income Latinx1 high school students. Both mentoring and ethnic identity have been indicated to play important roles in positive youth development (e.g., Umaña-Taylor, Tynes, Toomey, Williams, & Mitchell, 2014; Van Dam et al., 2018).

According to Rhodes’ (2005) model of youth mentoring, identity development is one of the processes through which mentoring relationships may promote positive outcomes for youth. Rhodes argues that youths’ internalizations of themselves may change as they identify with their mentors and see them as role models. Consistent with Cooley’s (1902) construct of the looking-glass self, youths’ identity may also be influenced by the views that they perceive their mentors to have of them (Rhodes, 2005). Further, consistent with the concept of future possible selves (Oyserman, Destin, & Novin, 2015), the social opportunities that mentors provide may facilitate youths’ identity development by affording experiences (e.g., exposure to possible future careers) on which youth can draw in formulating their emerging sense of self (Rhodes, 2005). The present study focuses on natural mentoring relationships. Natural mentoring relationships refer to supportive relationships with nonparental adults that emerge organically in youth’s social networks, in contrast to those that may be established through pairing with an adult volunteer in a program (Zimmerman, Bingenheimer, & Behrendt, 2005).

An important aspect of identity is ethnic and racial identity, which refers to the “social and psychological experiences associated with identifying with an ethnic or racial group” (Rivas-Drake et al., 2014, p. 42). Research indicates that ethnic/racial identity plays a significant role in healthy adolescent development, particularly for youth of color (e.g., DuBois, Burk-Braxton, Swenson, Tevendale, & Hardesty, 2002). In a comprehensive review of research on ethnic identity and adolescent well-being, Wakefield and Hudley (2007) found that a stronger ethnic identity was related to more positive academic, psychological, and social outcomes among ethnic minority youth. Similarly, a meta-analysis (Grant, 2008) revealed small to medium associations of ethnic identity with indices of both positive psychological functioning (e.g., satisfaction, self-worth; r = .20 to .32) and psychological problems (e.g., depression, anxiety; r = −.04 to − .12), across childhood (age under 13 years; 14 studies) and adolescence (age between 13 and 18 years; 105 studies). Race/ethnicity did not moderate the association between ethnic identity and outcomes, but there were significant associations in studies targeting Latinx ethnic groups (Grant, 2008). A more recent literature review similarly concluded that ethnic/racial identity was related to more favorable academic, psychological and health outcomes among diverse adolescents of color (Rivas-Drake et al., 2014).

A critique of the literature on ethnic/racial identity is that some researchers examine this construct as a composite score, when in it is actually multidimensional construct (Rivas-Drake et al., 2014). It is suggested that researchers examine ethnic identity multidimensionally in order to understand their relative contributions to adjustment (Rivas-Drake et al., 2014). Two aspects of ethnic identity are investigated in this study: a) exploration, which capture the behaviors that individuals engage in to explore and be involved with their ethnic group, and b) ethnic identity affirmation, which reflects individuals’ belonging to and positive feelings and affect about their ethnic group (Phinney, 1993; Roberts et al., 1999). Because so little research has been conducted on mentoring and ethnic/racial identity, we do not limit the following review to only mentoring and ethnic identity exploration or affirmation.

Mentoring and Ethnic Identity

Research conducted on the role of mentoring in adolescents’ ethnic/racial identity is responsive to past calls for research on contextual factors that influence ethnic/racial identity (Seaton, Quintana, Verkuyten, & Gee, 2017). In a study of 541 African American older adolescents (mean age = 17.8 years), the presence of a natural mentor was associated with a more positive racial identity, including public regard (positive perceptions of how others view one’s race), private regard (feelings about one’s own race), and racial centrality (how central race is to one’s identity; Hurd, Sánchez, Zimmerman, & Caldwell, 2012). Another study (Yancey, Siegel, & McDaniel, 2002) of 744 ethnically diverse 12- to 17-year-old adolescents similarly found that reporting a role model was related to a stronger ethnic identity. Further, in line with the importance of mentoring, adolescents who indicated that they personally knew their role models reported the most favorable ethnic identity. There is also evidence that receiving mentoring through a formal program is associated with a stronger ethnic/racial identity (Kaplan, Turner, Piotrowski, & Silbert, 2009; Gordon, Iwamoto, Ward, Potts, & Boyd, 2009). In these studies, the majority (if not all) of the mentors were of the same race/ethnicity as the youth. Consistent with the above discussed possible mechanisms linking mentoring to identity development, having a similar-identity mentor may enable youth to see themselves in their mentor, which may become incorporated into their own identities and sense of self. Further, mentees’ imagined perceptions of how their mentors view them may influence how adolescents of color see themselves. Similar-identity mentors may also directly or indirectly communicate messages to mentees about their ethnic/racial identity and help to socialize them (Hurd et al., 2012) about what it means to be of their race/ethnicity. Finally, it is possible that similar-identity mentors are likely to provide social opportunities and experiences that allow youth to explore their ethnic/racial identities. Overall, these studies suggest that mentoring may promote adolescents’ ethnic/racial identity.

Previous research primarily addresses the presence of a natural mentor or participation in a mentoring program, rather than the quality of the mentoring relationship. This is a significant limitation in view of well-documented variation in characteristics of mentoring relationships among adolescents, including those that emerge naturally (e.g., DuBois & Silverthorn, 2005), and the finding of a recent meta-analysis that indices of natural mentoring relationship quality (e.g., closeness, social support) had stronger associations with socioemotional (e.g., social skills, life satisfaction) and academic and vocational functioning (e.g., school attendance, academic engagement) for youth than simply the presence of a natural mentor (Van Dam et al., 2018).

Mentoring relationship quality is a multidimensional construct that comprises relationship characteristics (e.g., frequency of contact, duration) and perceptions of what is occurring in the relationship (e.g., closeness, satisfaction, social support; Nakkula & Harris, 2005). Nakkula and Harris (2005) conceptualize relationship quality as including subjective experiences in the relationship, specifically the mentor’s and mentee’s perception of the relational and instrumental quality of the relationship. Relational quality reflects the individual’s experience in the relationship, such as mentors’ and mentees’ feelings of closeness, satisfaction and engagement in the relationship. Instrumental quality refers to the goal orientation in the relationship, specifically experiencing the relationship as supportive of the mentee’s growth and as a means to achieving particular goals, such as improved academic achievement (Nakkula & Harris, 2005). Assessing relationship quality in mentoring research is important as the relationship is viewed to be at the heart of what makes mentoring effective (Rhodes, 2005).

A search of the literature revealed only one prior investigation that addressed the role of the quality of natural mentoring relationships in ethnic/racial identity. A cross-sectional study of 663 Black adolescents found a positive association between reported levels of closeness in natural mentoring relationships and racial pride (Wittrup et al., 2016). It is possible that the effect on ethnic/racial identity might vary by different components mentoring relationship quality (e.g., relational vs. instrumental). However, the recent meta-analysis of natural mentoring relationships among youth did not find that different aspects of relationship quality had a significant moderating effect on youth outcomes (VanDam et al., 2018).

Similar associations have been reported in the broader social support and ethnic/racial identity literature. In a study of 227 African-American 7th- and 8th-graders, perceived availability and quality of social support from peers and family was associated with stronger self-reported ethnic identity affirmation (Gaylord-Harden, Ragsdale, Mandara, Richards, & Petersen, 2007). A cross-sectional study of 521 African-American 12th-grade students found that greater self-reported emotional, problem-solving, and moral support in maternal relationships were positively associated with stronger private regard and centrality, two components of racial identity (Caldwell, Zimmerman, Bernat, Sellers, & Notaro, 2002). Similarly, a longitudinal study of 323 Latino teens (M = 15.31 years old) found that reported levels of support from fathers (i.e., warmth, affection, sense of being valued) predicted more ethnic identity exploration over time among boys (Umaña-Taylor & Guimond, 2010). In sum, research is consistent with the idea that feeling valued and accepted by others, such as parents, can indirectly communicate to adolescents of color that it is safe to explore their ethnic identities and environments (Umaña-Taylor & Guimond, 2010) and may promote positive feelings about their racial/ethnic group, but do not directly address the potential of supportive relationships with nonparental adults to facilitate ethnic identity among Latinx adolescents.

Role of Ethnic Identity in the Academic Outcomes of Latinx youth

There is considerable evidence that ethnic identity plays a positive role in the academic outcomes of Latinx youth. A cross-sectional study using self-report data obtained from 233 Latino 6th- through 12th-graders, for example, found positive associations between indices of ethnic identity affirmation and exploration with academic values, as well as ethnic identity resolution with academic efficacy and values (Umaña-Taylor et al., 2014). Other studies have reported similar cross-sectional associations between ethnic identity and more favorable academic attitudes or performance (Brown & Chu, 2012; Supple, Ghazarian, Frabutt, Plunkett, & Sands, 2006). In a notable exception, a longitudinal study of Latino adolescents found that higher reported ethnic identity affirmation in 7th grade predicted lower GPAs in 8th grade for boys, but the association was not significant for girls (Umaña-Taylor, Wong, Gonzales, & Dumka, 2012). The authors explained the unexpected finding about boys by stating that those who feel more positively about their ethnic identity may also experience stereotype threat about their academic potential, which would undermine their academic performance. Overall, although available research suggests that a stronger ethnic identity may be conducive to more favorable academic outcomes among Latinx youth, there is a need for further investigation within longitudinal research as well as with older samples.

As Latinx youth progress in their schooling, another important consideration may be the implications that their ethnic identity has for their values towards education, given its association with academic achievement (Colón & Sánchez, 2010). An important value is the perceived economic value of education, defined as the belief that earning an education will lead to job opportunities and economic success (Mickelson, 1990), and it constitutes both abstract and concrete attitudes towards education (Mickelson, 1990). Abstract attitudes about education (or the perceived economic benefits of education) is rooted in the American Dream ideology that one can gain upward social and economic mobility through hard work and education. In contrast, concrete attitudes (or the perceived economic limitations of education) reflect persons’ lived experiences with the societal opportunity structure, particularly the experiences of individuals who are marginalized (Mickelson, 1990). Researchers have found that low-income youth of color can uphold both abstract and concrete attitudes of education (Murdock et al., 2000). With regard to ethnic identity, in a previous study utilizing the same sample of urban, low-income, Latinx high school students (N = 396) as the current study, a more positive ethnic identity in 9th grade predicted higher perceptions of the economic benefits of education in 10th grade, controlling for perceived economic benefits of education in 9th grade (Mroczkowski & Sánchez, 2015). Among males, a more positive reported ethnic identity in 9th grade also predicted lower perceived economic limitations of education in 10th grade. A limitation of that previous study is that ethnic identity was examined as a composite score rather than investigating the effects of different dimensions of ethnic identity. Although not a Latinx sample, a study of 72 Asian-American high schoolers found that both ethnic identity belonging/affirmation and ethnic identity exploration were positively correlated with the perceived utility value of school (Kiang, Supple, Stein, & Gonzalez, 2012), operationalized as the belief that school is useful for one’s current everyday life and in the future. Another study of African American adolescents revealed that positive views towards their race was associated with the belief that school was important to their future success (Hurd et al., 2012). In general, these investigations suggest that ethnic/racial identity plays a role in the beliefs that adolescents develop about the value of education.

Ethnic/Racial Identity as a Mediator between Mentoring and Academic Outcomes

Existing research, as summarized above, is consistent with the potential for ethnic/racial identity development to serve as a mechanism linking mentoring with academic outcomes for youth of color. Yet, it appears that only two studies have directly tested this type of mediational pathway. Hurd et al. (2012) found that African American adolescents’ reported racial identity (i.e., private regard)partially mediated the association between the presence of a natural mentor and their beliefs regarding the importance of school for future success, controlling for gender, age and previous academic performance. Perhaps their mentors helped them feel positively about their race, which helped adolescents believe that school would benefit them as well as their own ethnic group, given the obstacles that African Americans have faced in education (Hurd et al., 2012). Another investigation of Black adolescents similarly reported that private regard as a facet of racial identity mediated the association between mentoring relationship closeness and academic engagement (Wittrup et al., 2016). Consistent with Rhodes’ (2005) model of youth mentoring, these findings suggest that ethnic/racial identity serves as an important developmental mechanism by which mentoring may promote positive outcomes in adolescents. For youth of color, the support from their natural mentors may encourage them to develop positive feelings about their ethnic/racial identity, which would purportedly lead to better academic outcomes, such as their values towards education. Researchers have yet to examine other components of ethnic/racial identity (e.g., ethnic identity exploration) as potential mediators of the association between mentoring and academic outcomes. Further, the previous findings are cross-sectional, however, and their applicability to Latinx youth is unclear.

Role of Gender

Research has also suggested that the mediational role of ethnic identity in mentoring and academic outcomes may be moderated by gender. As reported previously, ethnic identity affirmation predicted lower grades for Latino boys but not for Latina girls (Umaña-Taylor et al., 2012). Further, a more positive ethnic identity predicted lower perceived economic limitations of education among Latino boys, but not among Latina girls (Mroczkowski & Sánchez, 2015). Research also suggests that boys and girls may benefit differently from mentoring relationships and that they have different experiences in these relationships. For example, a study found that Latino boys were less likely to have close and growth-oriented natural mentoring relationships compared to Latina girls (Liao & Sánchez, 2016). Gender socialization patterns may also influence the aspects of mentoring relationships from which boys and girls benefit. Specifically, girls may benefit from relational closeness while boys may benefit from instrumental support from mentors (Bogat & Liang, 2014). This past research suggests that gender may influence the associations among mentoring, ethnic identity, and academic outcomes. One study tested the moderating role of gender in the mediating pathway of mentor presence, racial identity, and school beliefs among African American adolescents, but did not find a significant moderating effect for gender (e.g., Hurd et al., 2012). Given the limited research on gender as a moderator in this mediated path, further exploration is warranted.

Current Study and Hypotheses

The current short-term, longitudinal study examines whether ethnic identity mediates the association between natural mentoring relationships and academic outcomes in a sample of Latinx adolescents over their 9th and 10th grades. It was hypothesized that reporting a natural mentor in 9th grade would be associated with greater perceived economic value of education and grades in 10th grade, controlling for grade 9 levels of the corresponding measure. It was also expected that these associations would be mediated by ethnic identity as reported in 9th grade. It was hypothesized that, among participants with natural mentors, the quality of these relationships, specifically relational and instrumental quality, would be similarly associated with improved economic values toward education and grades and these associations would be mediated by ethnic identity. Finally, we explored whether gender moderated the mediated pathway among ethnic identity and academic outcomes. Although gender differences have been found between aspects of ethnic identity and academic outcomes (e.g., Umaña-Taylor et al., 2012) and in mentoring relationships (Liao & Sanchez, 2016) among Latinx adolescents, it is unclear whether gender would moderate this mediated pathway so this served as an exploratory research question.

Method

Setting

Participants were recruited from two high schools in Chicago. Both schools enrolled predominantly Latinx students (95% and 91%), and the majority of enrolled students qualified as low-income (95% and 86%). At the time of the study both schools received federal School Improvement Status, which means that the schools did not meet adequate yearly progress for two consecutive years under the No Child Left Behind Act (Illinois State Board of Education [ISBE], 2018). Eighteen percent and 15% of students at each school met or exceeded state standards for math and reading, compared to the state average of 53% (ISBE, 2018). The average ACT scores at each school (16 and 15) were below state average ACT score of 21 (ISBE, 2018).

Participants

Trained graduate and undergraduate research assistants recruited participants through informational presentations about the study during 9th-grade homeroom classes. Bilingual research assistants were available to answer questions in both English and Spanish. Students were provided with parental active consent and youth assent forms (in English & Spanish). Students who returned a signed parental active consent form, regardless of their parent allowing them to participate in the study, were given a candy bar and entered in a raffle to win one of five prizes (e.g, movie tickets, iPod) in order to increase the number of returned consent forms.

There were 463 students who participated in the study, but the current investigation includes 347 youth who completed the survey at both 9th and 10th grades and self-identified as a Latinx ethnicity. Of the 116 participants who were excluded, 39 participants were excluded because they did not report a Latinx ethnicity, and 77 Latinx participants were excluded because they completed a survey only in 9th or 10th grade. Of the 347 participants in this sample, 53% were female (n = 183) and 47% were male (n = 164). The mean age of participants was 15.04 (SD = .58) in 9th grade. Participants could check multiple races/ethnicities, and reported their race/ethnicity as Mexican/Mexican American (n = 324; 93%), Puerto Rican (n = 19; 6%), White (n = 6; 2%), African American/Black (n = 7; 2%)2, and/or Other Latino(a) ethnicity (n = 18; 5%; e.g., Guatemalan). Most of the participants were either first-generation immigrants (participant born outside of U.S.; 23%; n = 78) or second-generation (at least one parent born outside U.S.; 62%; n = 215), with the remaining being third-generation and beyond (10%; n = 36).

Attrition analyses.

Chi-square tests of independence and independent samples t tests were conducted to compare participants in the current study (n = 347) to Latinx participants from the larger study who were excluded (n = 77). Chi-square analyses indicated no significant association between gender and study inclusion, χ2 (1, N = 424) = .083, ns. There was a significant association between study inclusion and generational status, χ2 (2, N = 396) = 14.308, p < .01, Cramer’s V = .19; more first-generation students were excluded from the sample than expected. Independent samples t tests indicated that the two groups differed significantly on age in 9th grade, t(390) = 2.57, p < .05, d = .39, with excluded participants being older (M = 15.27; SD = .64) than participants who were included in this study (M = 15.04; SD = .58).

The two groups were compared on study outcomes. They significantly differed on perceived economic limitations of education in 9th grade, t(392) = 2.57, p < .05, d = .39, and 9th-grade GPA, t(387) = −5.07, p < .001, d = .75. Excluded participants significantly reported more perceived economic limitations of education (M = 2.51; SD = .78) and had lower GPA (M = 1.80; SD = 1.04) than participants in this study (limitations of education: M = 2.23; SD = .67; and GPA: M = 2.54; SD = .93). The two groups did not significantly differ on 9th-grade ethnic identity exploration, t(391) = −.652, ns, 9th-grade ethnic identity affirmation, t(53.03) = −1.307, ns, 10th grade limitations of education, t(375) = .175, ns, benefits of education in 9th grade, t(392) = −.555, ns, or in 10th grade, t(375) = −.003, ns, or 10th-grade GPA t(50.1) = −.469, ns.

Procedure

Participants completed the surveys in their classrooms and could choose to complete the survey in either English or Spanish. Research assistants read the instructions and survey questions aloud to ensure accessibility across diverse student abilities. Research assistants reviewed surveys to make sure participants had completed the survey in its entirety and to reduce missing data. Survey completion took approximately 45 to 60 minutes, and participants received a $10 gift card for completing each survey. This research was approved by DePaul University’s Institutional Review Board (Project UNO #BS080508PSY).

Measures

Demographics.

In 9th grade, participants were asked to self-report race/ethnicity, gender, age, mother and father’s educational attainment, and generational status.

Presence of natural mentor.

In 9th grade, each participant was asked to indicate whether they had a mentor. Instructions were:

“Is there anyone in your life who is older (age 18 or older) and more experienced than you and you go to for support and guidance? This person is not a parent or the person who raised you or a boy/girlfriend. This person is someone who: a) you can count on to be there for you, b) who believes in you and cares deeply about you, c) who inspires you to do your best, and d) who has really influenced what you do and the choices you make?” (Rhodes, Contreras, & Mangelsdorf, 1994; Sánchez, Esparza, & Colón, 2008).

Participants indicated yes (1) or no (0) and were able to report up to three mentors. If students reported more than one mentor, then they were asked to rank order them from 1 (most important) to 3 (least important).

Mentor characteristics.

Participants with at least one natural mentor were asked questions about each mentor. They reported each mentor’s race/ethnicity by selecting one or more categories from a predetermined list. Mentor role type was assessed with the question “What is your relationship with this person?” Participants checked the role type from a list of options (e.g., teacher, sibling, pastor, grandparent) and then the categories were coded into three categories: (a) extended family (e.g., aunt/uncle), (b) sibling, and (c) non-familial (e.g., pastor). Participants reported the educational attainment of each mentor by answering, “How far did this person go in school?” Response options were: (1) less than a high school graduate, (2) high school graduate or GED, (3) technical school or two-year college (associate’s degree), (4) four-year college (bachelor’s degree), (5) more than a four-year college degree (e.g., Master’s, doctoral, law), and (6) I don’t know.

Mentoring relationship quality.

Each participant who reported having a natural mentor completed the Youth Mentoring Survey (YMS) to assess mentoring relationship quality (Harris & Nakkula, 2010). Students who reported more than one mentor were asked to think about all of them together as they answered the questions. The YMS has two broad subscales, and the wording was modified to state “important adult(s)” rather than “mentor.” The relational quality subscale (16 items) assesses the degree to which the responding youth feels happy, close, and satisfied in their relationship with their mentor (e.g., “My important adult(s) knows what is going on in my life”). The instrumental quality subscale (8 items) measures the extent to which the mentee perceives there to be instrumental benefits from the mentoring relationship (e.g., “I have learned a lot from my important adult(s)”). Responses are on a four-point scale (1 = Not at all true to 4 = Very true). A mean score is computed for both subscales such that higher scores indicate greater relational and instrumental relationship quality. Both subscales demonstrated good internal consistency (relational: α = .84; instrumental: α = .86).

Ethnic identity.

In 9th grade, ethnic identity was assessed using the Revised Multigroup Ethnic Identity Measure (Roberts et al., 1999), which consists of two subscales. The affirmation subscale (7 items) assesses the sense of belonging to one’s ethnic group as well as a sense of pride and positive feelings towards one’s ethnic group (e.g., “I am happy that I am a member of the group I belong to”). The exploration subscale (5 items) measures the extent to which the respondent explores, learns about and is actively involved in their ethnic group (e.g., “To learn more about my ethnic background, I have often talked to other people about my ethnic group”). Respondents indicate agreement or disagreement for each item on a four-point scale (1 = Strongly disagree to 4 = Strongly agree). For both subscales, a mean score was computed such that higher scores indicate higher affirmation and exploration. The measures demonstrated acceptable internal consistency in this study (affirmation: α = .85; exploration: α = .65).

Economic value of education.

In 9th and 10th grades, perceptions about the economic value of education were assessed using the Benefits and Limitations of Education scale (Murdock et al., 2000). Participants indicated their agreement on a five-point scale (1 = Agree Very Much to 5 = Disagree Very Much). The Benefits of Education subscale (5 items) assesses Mickelson’s (1990) concept of abstract attitudes, which refers to the belief that an education is necessary for future employment and economic opportunities (e.g., “If I do well in school, I will get a good job”). This subscale exhibited good internal consistency (9th grade: α = .88; 10th grade: α = .89). The Limitations of Education subscale (10 items) taps into Mickelson’s (1990) concrete attitudes about education, which reflects individuals’ lived experiences with the opportunity structure and the realities of the experiences of members in their group. The subscale, thus, assessed participants’ doubts that education is necessary for future economic and job opportunities (e.g., “I could be successful without an education”). This scale also exhibited good internal consistency (9th grade: α = .78; 10th grade: α = .89). Scores on both scales were reverse coded, and then a mean score was computed for each subscale, such that higher scores on the benefits of education subscale indicated more positive abstract beliefs about education, and higher scores on the limitations subscale indicated more pessimistic beliefs.

Academic performance.

Participants provided consent for their school to release their weighted cumulative grade point average (GPA) at the end of 9th and 10th grades.

Data Analysis Plan

Primary analyses were conducted using path analysis in AMOS 13.0 (Arbuckle, 2014) with all constructs modeled as manifest (observed) variables (see Figures 1 and 2 for the models tested). The first model (Model 1a) included paths from report of one or more mentors in 9th grade (1=yes; 0=no) to 9th grade ethnic identity affirmation and exploration as well as paths from all of these variables to perceived economic benefits of and perceived limitations of education, respectively, in 10th grade; the 9th grade levels of the attitude toward education variables were also included in the model, with paths from each of these variables to the corresponding variable at grade 10. This model thus allowed for a test of whether there was an indirect (mediated) association of the report of a natural mentor with the attitudes toward education outcomes via its associations with the ethnic identity measures. Correlated disturbances were modeled between the two ethnic identity variables as well as the two attitudes toward education measures. A similar model examined GPA instead of perceived economic benefits and limitations (Model 1b). Next, models replacing the presence of a mentor variable with the two measures of relationship quality variables were examined (Model 2). Additional versions of this model were tested using only economic benefits of education (Model 2a); only economic limitations of education (Model 2b), and GPA (Model 2c) as outcome variables. All models included controls for the relevant academic outcome(s) at 9th grade as well as participant gender and age. Covariances were modeled among the mentoring relationship variable(s), age, and gender in all models.

Figure 1.

Figure 1.

Open in a new tab

Models testing the roles of presence of natural mentors and ethnic identity in academic outcomes.

Note: Models controlled for participant age and gender.

Figure 2.

Figure 2.

Figure 2.

Open in a new tab

Models testing the roles of relational and instrumental mentoring quality and ethnic identity in academic outcomes.

Note: Models controlled for participant age and gender.

Criteria for acceptable model fit were as follows: a comparative fit index value greater than .95 (CFI; Bentler, 1990); Tucker-Lewis Index value greater than .95 (TLI; Tucker & Lewis, 1973); and root mean square error of approximation value lower than .08 (RMSEA: Browne & Cudeck, 1993). The indirect (mediated) associations of mentoring variables at grade 9 with grade 10 academic outcomes via the ethnic identity variables were tested using bootstrapping. Bootstrap analyses were conducted in AMOS (IBM SPSS Amos, 2012) using only participants with complete data with 200 samples and bias-corrected confidence intervals.

Additionally, gender differences in the paths from mentoring variable(s) to ethnic identity variables and from the ethnic identity variables to the grade 10 academic outcome(s) were examined in a multiple group model. The multiple group model was tested using chi-square comparison statistics to compare unconstrained versus constrained models (Yuan & Chan, 2016). A significant change in chi-square between the unconstrained and constrained models indicate that worse model fit with the paths constrained to equality between groups and thus provide evidence of moderation.

Finally, because we examined the mentoring and ethnic identity variables in 9th grade in our models, we conducted supplemental analyses via a series of cross-lagged panel models in order to determine the extent of bidirectionality in the associations between mentoring presence/quality and the ethnic identity variables.

Results

Preliminary Analyses

Descriptive statistics for each variable and Pearson’s bivariate corrections are presented in Table 1. In general, participants reported levels of ethnic identity affirmation and exploration, and perceived economic benefits of education above the midpoint of the scale, and they reported levels of perceived economic limitations of education below the midpoint. Among participants who had a natural mentor, participants reported levels of relationship quality above the midpoint of the scale. Relational and instrumental relationship quality with mentors were significantly and positively associated with ethnic identity affirmation and exploration as well as perceived economic benefits of education. Perceived limitations of education was significantly and negatively associated with relational and instrumental quality and ethnic identity affirmation; however, there was no significant association with exploration. Further, GPA was significantly and positively associated with relational and instrumental relationship quality with mentors and perceived economic benefits of education, and significantly and negatively related to perceived limitations of education. However, there was no significant association between GPA and ethnic identity exploration.

Table 1.

Means, standard deviations, and Pearson bivariate correlations of study variables

Variable M (SD) 1 2 3 4 5 6 7 8 9
1. Relational Quality (9th grade) 3.13 (.45)
2. Instrumental Quality (9th grade) 3.23 (.56) .74**
3. Exploration (9th grade) 2.50 (.56) .21** .20**
4. Affirmation (9th grade) 3.12 (.53) .40** .37** .48**
5. EVE Benefits (9th grade) 4.36 (.80) .19** .24** .09 .11*
6. EVE Limitations (9th grade) 2.23 (.67) −.35** −.33** −.12* −.26** −.43**
7. EVE Benefits (10th grade) 4.27 (.84) .16* .18** .14* .22** .22** −.26**
8. EVE Limitations (10th grade) 2.30 (.68) −.21** −.20** −.09 −.16** −.27** .46** −.51**
9. GPA (9th Grade) 2.54 (.93) .15* .12 .11* .17** .16** −.14** .25** −.14**
10. GPA (10th grade) 2.67 (.86) .15* .16* .10 .15** .17** −.16** .27** −.14** .92**
Open in a new tab

Note. For each variable, higher levels indicate more of a construct; Relational and instrumental quality variables only apply to participants who reported a mentor;

*

p < .05;

**

p < .01.

Descriptive Statistics of Natural Mentors

In 9th grade, 262 (76%) participants reported having at least one mentor. Almost half (47%; n = 164) of participants reported having three mentors, 15% (n = 53) reported two mentors, and 13% (n = 45) reported one mentor. For mentor #1, 42% (n = 110) were extended family members, 38% (n = 100) were siblings, and 20% (n = 51) were non-familial adults. Similarly for mentor #2, 49% (n = 127) were extended family mentors, 18% (n = 48) were siblings, and 16% (n = 42) were non-familial adults. Participants reported mentor #3 as extended family members (n = 94; 36%), 8% (n = 21) were siblings, and 19% (n = 49) were non-familial adults. The majority of mentors #1 were Latinx (n = 252; 96%), while only 5% (n = 13) were White, 2% (n = 4) were African-American, and less than one percent were Asian/Pacific Islander (n = 1). Similarly, mentors #2 were primarily Latinx (n = 210; 97%), while only 4% (n = 10) were White, 1% (n = 3) were African American/Black, and less than 1% were Asian/Pacific Islander (n = 1) or American Indian/Native American (n = 1). Mentors #3 were also primarily Latinx (n = 158), while 2% (n = 8) were White, and less than 1% were American Indian/Native American (n = 2), or African American/Black (n = 3).

Participants reported the highest educational attainment for mentor #1 to be high school or GED (37%; n = 98), less than high school (16%; n = 41), four-year college (10%; n = 27), technical or two-year college (6%; n = 16), or more than a four-year degree (4%; n = 10). Similar patterns occurred for mentor #2, in which the highest educational attainment was most commonly high school or GED (30%; n = 79), less than high school (17%; n = 44), four-year college (8%; n = 20), technical or two-year college (6%; n = 16), or more than a four-year degree (3%; n = 9). Likewise, mentor #3 earned high school or GED (20%; n = 52), less than high school (14%; n = 37), four-year college (10%; n = 25), technical or two-year college (3%; n = 8), or more than a four-year degree (1%; n = 3). Some participants reported not knowing the educational attainment of mentor #1 (26%, n = 68), mentor #2 (18%, n = 48), or mentor #3 (15%; n = 39).

Primary Analyses

Role of mentoring and ethnic identity in academic outcomes.

First, the model (1a) that tested whether the presence of mentors (1=yes; 0=no) predicted perceived economic benefits of education and economic limitations of education and whether ethnic identity affirmation and exploration mediated those associations did not fit the data well (χ2= 27.17, df = 11, (24) p < .001; RMSEA = .07, CFI = .88, TLI = .70); hence, no additional examination of path coefficients was conducted. A similar model (Model 1b) with GPA as an outcome was tested but did not fit the data well (χ2= 39.83, df = 10, (25) p < .001; RMSEA = .10, CFI = .96, TLI = .89). No additional path coefficients were examined.

The model in which instrumental and relational mentoring quality were the predictors, ethnic identity affirmation and exploration were the mediators, and perceived economic limitations and benefits of education were the outcomes (Model 2) was a poor fit to the data (χ2 = 233.347; CFI = .691; TLI = .26; RMSEA = .16). We thus also examined models in which consideration was limited to economic benefits (Model 2a) or economic limitations (Model 2b). With perceived economic benefits of education as the outcome, the overall model provided a good fit to the data (χ2= 31.33, df = 11, (33) p < .001; RMSEA = .07, TLI = .83, CFI = .95; see Figure 3). In this model, relational mentoring quality positively predicted both ethnic identity affirmation and exploration. Instrumental mentoring quality positively predicted ethnic identity affirmation, but did not significantly predict exploration. Affirmation positively predicted perceived economic benefits of education. The direct paths from relational mentoring quality, instrumental mentoring quality, gender, and age to economic benefits of education were not significant, although 9th grade perceived economic benefits did significantly predict 10th grade benefits. The beta coefficients for the significant models are provided in Table 2.

Figure 3.

Figure 3.

Open in a new tab

Structural model showing associations among relational and instrumental mentoring quality, ethnic identity affirmation and exploration, and the perceived economic benefits of education in 10th grade.

Note: We also controlled for participant age and gender in the model. *p < .05, **p < .01, ***p < .001. Dotted arrows are non-significant paths.

Table 2.

Standardized estimates (standard errors in parentheses) of good fitting models predicting perceived economic benefits of education (Model 2a), economic benefits of limitations (Model 2b) and GPA (Model 2c) in 10th grade

Parameter estimated Model 2a Model 2b Model 2c
Instrumental Quality → EI-Affirmation .16 (.05)** .15 (.08) .16 (.06)*
Instrumental Quality →EI-Exploration .03 (.06) .08 (.09) .03 (.07)
Relational Quality → EI-Affirmation .31 (.07)*** .29 (.10)*** .31 (.08)***
Relational Quality → EI-Exploration .20 (.08)** .14 (.11) .20 (.09)**
EI-Affirmation → EVE Benefits (10th grade) .16 (.10)*
EI-Exploration → EVE Benefits (10th grade) .04 (.09)
EI-Affirmation → EVE Limitations (10th grade) −.02 (.07)
EI-Exploration → EVE Limitations (10th grade) −.02 (.08)
Instrumental Quality → EVE Benefits (10th grade) .08 (.13)
Instrumental Quality → EVE Limitations (10th grade) −.03 (.10)
Relational Quality → EVE Benefits (10th grade) −.01 (.17)
Relational Quality → EVE Limitations (10th grade) −.03 (.13)
EVE Limitations (9th grade) → EVE Limitations (10th grade) .45 (.05)***
EVE Benefits (9th grade) → EVE Benefits (10th grade) .19 (.05)***
Sex → EVE Limitations (10th grade) .04 (.07)
Sex → EVE Benefits (10th grade) −.01 (.09)
Age → EVE Limitations (10th grade) .01 (.01)
Age → EVE Benefits (10th grade) .02 (.01)
Instrumental Quality → GPA (10th grade) .06 (.06)
Relational Quality → GPA (10th grade) −.02 (.07)
EI-Affirmation → GPA (10th grade) −.01 (.04)
EI-Exploration → GPA (10th grade) −.01 (.04)
GPA (9th grade)→ GPA (10th grade) .93 (.02)***
Sex → GPA (10th grade) .02 (.04)
Age → GPA (10th grade) −.01 (.01)
Open in a new tab
*

p < .05,

**

p < .01,

***

p < .001:

EI – ethnic identity; EVE – economic value of education

Results of the bootstrapping analyses for this model (Model 2a) indicated that the model pathway representing the indirect effect of relational mentoring quality on economic benefits of education was statistically significant (B = .06, 95% CI = .03 - .13, p < .01). This bootstrapped indirect effect estimate is equivalent to the standardized indirect effect found with the larger data set (B = .06).

The model (Model 2b) in which the mentoring relationship quality subscales were the predictors, the ethnic identity subscales were the mediators, and economic limitations of education was the outcome was a poor fit of the data (χ2 = 76.75; CFI = .87; TL = .68, RMSEA = .11); hence, the individual path coefficients were not examined for significance.

The model (Model 2c) in which the mentoring relationship quality subscales were the predictors, the ethnic identity subscales were mediators, and GPA was the outcome fit the data well (χ2= 31.33 df = 11, (33) p < .001; RMSEA = .07, CFI = .95), but no mentoring or ethnic identity variables significantly predicted GPA.

Gender as a Moderator.

Gender was examined as a moderator of the best-fitting model (Model 2a) that tested for the direct and indirect effects of affirmation and exploration on the economic benefits of education. The multiple group analysis compared the unconstrained model (paths for boys and girls estimated separately) and the constrained model (paths for boys and girls constrained to equal each other). The unconstrained model provided an excellent fit to the data (χ2= 41.29 df = 14, (56) p < .001; RMSEA = .08, CFI = .92). The results of the chi-square nested model comparison showed that the structural weights and intercepts could be constrained to equality across each gender (χ2Δ change = 18.63, df = 13, ns) suggesting that separate models are not needed for boys and girls and moderation is not present. However, an examination of the individual path coefficients for each gender in the unconstrained model showed minor differences. For girls, relational mentoring quality positively predicted ethnic identity affirmation (β = .26, p < .05). No other paths were significant. In contrast, for boys, relational mentoring quality significantly predicted both ethnic identity affirmation (β = .32, p < .001) and exploration (β = .28, p < .01). Additionally, affirmation positively predicted perceived economic benefits of education (β = .23, p < .01).

Supplemental Analyses

Supplemental analyses were conducted to determine the extent of endogeneity bias in the associations between the ethnic identity variables and the presence/quality of mentoring. A series of three cross-lagged path analyses were modeled to examine the reciprocal effects of ethnic identity exploration and affirmation on (a) instrumental mentoring quality, (b) relational mentoring quality, and (c) the presence of natural mentors in grades 9 and 10 while including the auto-correlation paths between the same variable measured at both time points. Ethnic identity affirmation in 9th grade significantly predicted instrumental mentoring quality in 10th grade (β = .15, p < .05). Instrumental mentoring quality in 9th grade significantly predicted both 10th grade ethnic identity affirmation (β= .18, p < .01) and ethnic identity exploration (β= .16, p < .01). All auto-correlations were significant. Thus, reciprocal associations between ethnic identity affirmation and instrumental mentoring quality exist.

Ethnic identity exploration (β = −.13, p < .05) and ethnic identity affirmation (β= .22, p < .01) in 9th grade significantly predicted 10th grade relational mentoring quality. Relational mentoring quality in 9th grade significantly predicted 10th grade ethnic identity affirmation (β = .12, p < .05). All auto-correlations were significant. Hence, reciprocal associations between ethnic identity affirmation and relational mentoring quality exist.

Presence of a natural mentor in 9th grade did not significantly predict either type of 10th-grade ethnic identity. The auto-correlations between both types of ethnic identity were significant, but the path from natural mentor presence in 9th grade to natural mentor presence in 10th grade was not significant.

Discussion

The aim of this short-term longitudinal study was to examine the role of mentoring in Latinx adolescents’ ethnic identity and to investigate the mediating role of ethnic identity in the association between mentoring and academic outcomes. Partial support was found for study hypotheses. Report of a mentoring relationship was not related to ethnic identity or academic outcomes over time. Support was found, however, for the hypothesis that ethnic identity mediates the association between the quality of mentoring relationships and change over time in values towards education among Latinx adolescents. Results thus support Rhodes’ (2005) framework that identity development is a mechanism by which mentoring promotes positive youth development and offers the first demonstration of this, to our knowledge, with a sample of Latinx youth. Given the current negative climate in the U.S. towards immigrants, Mexicans and other Latinx groups, promoting the positive ethnic identity of adolescents from this population is essential to their healthy development.

Mentoring relationship quality was significantly associated with greater reported levels of ethnic identity affirmation and exploration in 9th grade. Specifically, relational quality of mentoring relationships was associated positively with both subscales of ethnic identity, whereas their instrumental quality was significantly associated with only the affirmation subscale. These findings are consistent with research indicating contributions of mentoring relationship closeness (Wittrup et al., 2016), parental warmth and support (Caldwell et al., 2002; Umaña-Taylor & Guimond, 2010) and peer and family support (Gaylord-Harden et al., 2007) to ethnic/racial identity among adolescents of color. A possible explanation for why mentoring relationship quality was associated with ethnic identity is that perhaps natural mentors implicitly communicated messages to their mentees about the meanings of their race/ethnicity through their interactions (Hurd et al., 2012). The majority of the mentors in the current study was family members (e.g., siblings, aunts/uncles) and was of the same ethnicity (e.g., Mexican/Mexican American). Thus, having mentoring relationships with family members characterized by more closeness, satisfaction, and support may make adolescents feel that they can see themselves in their mentors; the relationships may encourage Latinx adolescents to explore their ethnicity and feel pride and an affinity towards their ethnic group.

As noted, findings also are consistent with an indirect contribution of relatively higher relational quality of Latinx adolescents’ mentoring to increased perceived benefits of education in 10th grade via ethnic identity affirmation/belonging. Similarly, past researchers found that private regard mediated the association between mentoring and academic outcomes among Black adolescents (Hurd et al., 2012; Wittrup et al., 2016). Our study went a step further, compared to past research, by examining Latinx students’ academic outcomes (i.e., beliefs about the economic benefits of education) over time. At first glance, the perceived economic benefits of education may appear to be individualistic (e.g., I will make more money someday if I do well in school). Feeling positively and a sense of belonging to their ethnic group combined with the Latinx cultural value of collectivism, in which the ingroup is emphasized over the individual, may have motivated Latinx adolescents in this study to believe that gaining an education will not only lead to more job opportunities and earnings in the future for themselves but that it would enable them to help their ethnic group overall. In a qualitative study of Latino college students at a predominantly White university, Rivas-Drake (2008) found that a subset of participants had a strong affiliation with other Latinos while at the same time having a collectivist perspective about their achievement motivation. In other words, they believed that doing well in school and moving up the economic ladder would allow them to give back to their own ethnic community (Rivas-Drake, 2008). Similarly, having close relationships with mostly familial adults and a stronger ethnic identity may have encouraged our participants to believe that earning an education is a way to give back to their ethnic community in the future.

Our findings suggest the distinctive characteristics of instrumental versus relational mentoring quality, given that instrumental quality was not indirectly associated with academic outcomes. It was further surprising that instrumental relationship quality did not have direct effects on neither the economic values of education or GPA. The instrumental relationship quality subscale includes items about the support that mentors provide in school (e.g., I am doing better at school because of my mentor’s help) and in learning and growing overall (e.g., I have learned a lot from my mentor). Perhaps the relatively lower educational levels of natural mentors in this study (only 20% of mentors had at least a 4-year college degree) limited the amount or kind of instrumental support that they could provide to their mentees. A previous study of natural mentoring relationships of Latino high school students revealed that nonfamilial mentors had higher levels of education compared to their familial mentors (Sánchez et al., 2008). Further, having nonfamilial natural mentors during adolescence predicted higher levels of education compared to familial mentors in early adulthood in a nationally representative sample (DuBois & Silverthorn, 2005). Perhaps if participants in the present study had more nonfamilial mentors with higher levels of education, then instrumental quality might have had significant effects on the academic outcomes.

As stated earlier, we did not find evidence in support of our hypotheses regarding the role of the presence of natural mentors, which contrasts past findings that the presence of natural mentors is related to better academic outcomes (e.g., Van Dam et al., 2018) and ethnic/racial identity (e.g., Hurd et al, 2012). However, a recent natural mentoring meta-analysis revealed that relationship quality had bigger effects on academic and socioemotional functioning compared to the presence of natural mentors (Van Dam et al., 2018). Another study found that having at least a “somewhat close” mentoring relationship was important in producing better academic outcomes rather than simply having a mentoring relationship (Bayer, Grossman, & DuBois, 2015). Again, it is possible that for our sample, having mostly familial mentors with less than a college-level education limits how much they can help with Latinx adolescents with school given their limited experience with education. Further, the majority (85%) of participants came from immigrant families, which also means that many of their familial mentors might not have been educated in the U.S.; familial, immigrant mentors might have limited knowledge about how to navigate the educational system in this country. Coming from immigrant families may also explain why relationship quality was only related to the perceived economic benefits of education rather than also the perceived economic limitations of education. Perhaps familial mentors are discussing the economic and educational opportunities that are available to mentees in the U.S. compared to their home countries, and mentors may not have yet experienced the racial and class inequalities in education and jobs. Moreover, the presence of these mostly familial, natural mentors and the quality of their mentoring relationships might be impacting other developmental outcomes in Latinx youth, such as their psychological well-being.

Although we did not find evidence of gender moderation in the associations among mentoring, ethnic identity and academic outcomes, gender differences emerged with respect to the significance of several individual paths for Latinx boys versus girls, providing preliminary evidence that gender may be an important consideration. While higher relational quality was associated with higher ethnic identity affirmation and exploration for boys, relational quality was only associated with affirmation for girls. A possible explanation is that ethnic socialization processes for Latino boys supports opportunities for independence, whereas Latina girls may be socialized to stay closer to home (Raffaelli & Ontaai, 2004). This gendered ethnic socialization may contribute to boys being able to explore their ethnic identity and perhaps with extended family members and nonfamilial adults. Further, ethnic identity affirmation was associated with more perceived economic benefits of education for boys, but not for girls. It is possible that positive feelings about their ethnic group may be associated with perceived economic benefits of education for boys as they may be socialized to pursue education and economic success compared to their female counterparts.

Limitations, Future Directions, and Implications

There are several limitations in the current study. First, the measure of mentoring relationship quality asked participants with multiple mentors to think about all of their mentors at the same time when answering the questions. Thus, we were unable to examine the relationship quality with each individual mentor. Perhaps relationship quality varied between individual mentors, or by mentor role type (i.e., familial vs. non-familial). Second, although we had two time-points, a third time point is necessary to increase support of our mediational findings. Relatedly, we examined the mentoring and ethnic identity variables in 9th grade, which introduces the possibility that ethnic identity might lead to mentoring relationship quality rather than vice versa. We investigated the extent of endogeneity bias in our study and found that there were reciprocal associations between ethnic identity affirmation and instrumental and relational mentoring quality across 9th and 10th grades. Hence, it is possible that Latinx youth who have more positive views about their ethnic group may seek out and bond with their natural mentors, who are also of the same race/ethnicity. However, we also found that mentoring quality in 9th grade was significantly correlated with 10th grade ethnic identity variables in support of our hypotheses, and so ethnic identity and mentoring quality are likely both to influence one another. A third limitation is that attrition analyses indicated sampling bias between the current sample and those excluded from the larger sample. Specifically, excluded participants were slightly older and reported higher perceived limitations and lower GPAs in 9th grade. Thus, findings may reflect youth who were less pessimistic and more motivated about school.

Despite these limitations, our findings make an important contribution to the research literature on mentoring and ethnic identity. Namely, we explicate the role of ethnic identity as a mechanism for natural mentoring to play a role in the educational attitudes of Latinx adolescents, and the important role of mentoring relationship quality in Latinx youth’s ethnic identity. Future research should also examine mentoring as a mechanism between ethnic identity and academic outcomes given the reciprocal associations between mentoring and ethnic identity. It is also recommended that researchers investigate components of ethnic identity separately when examining their role on youth outcomes. As demonstrated by our findings and prior research (e.g., Umaña-Taylor et al., 2014), the affirmation and exploration elements of ethnic identity may play different roles in academic outcomes, and it is important to consider their unique contributions. Future research should also consider mentor type and characteristics when examining the natural mentors of low-income, Latinx adolescents. Specifically, researchers should study not only whether they have familial mentors, but explicitly examine whether they also have non-familial mentors, particularly when examining academic outcomes. This is important because non-familial mentors may likely have higher educational levels and access to other networks and information (Stanton-Salazar, 2010). Access to mentors across networks might play a role in the current participants’ academic outcomes, wherein few participants reporting nonfamilial mentors or mentors with higher than college educational attainment.

Our findings have implications for mentoring practitioners and those concerned with youth development broadly. Given the influential role of ethnic/racial identity in the healthy development of adolescents of color and the current negative political climate towards the Latinx population in the U.S., it is important to recognize sources of support that may help facilitate ethnic identity development. Practitioners should consider ways in which programs and youth-focused settings can support high quality relationships with natural mentors and also deem a positive ethnic/racial identity as an important goal in their interventions. Natural mentors may be more likely to have a similar race/ethnicity, and those with a healthy ethnic/racial identity may serve as a positive role model for Latinx youth’s ethnic identity development. Additionally, practitioners should consider ways to facilitate youth attainment of a diverse mentoring network. Namely, low-income youth may not know adults in their existing networks who can help them with education and career development (Raposa, Erickson, Hagler, & Rhodes, 2018). For example, initiatives that seek to increase social capital by building youth’s capacity to seek natural mentoring relationships may be particularly useful for low-income youth (Schwartz et al., 2018). Mentoring practitioners can positively influence Latinx adolescents by supporting their existing natural mentoring relationships, providing them with opportunities to obtain more social capital, and promoting a positive ethnic/racial identity as part of programming.

Acknowledgments:

This research was supported by a grant awarded to the first author from the National Institute of Child Health and Human Development (1R03HD057343-01A2). The writing of this manuscript was in part supported by a Distinguished Fellowship award from the William T. Grant Foundation to the first author. We appreciate the feedback provided on earlier versions of this manuscript by the MIC-Psych research team. We would also like to thank the public schools that allowed us to conduct this research as well as for the participation of the high school students.

Footnotes

1

Latinx is a gender-neutral term sometimes used in lieu of Latino or Latina. Santos (2017) provides a discussion of this term. When citing previous research, we use the authors’ terms (e.g., Latino, Hispanic) in those studies.

2

Participants who self-identified as a non-Latinx ethnicity also reported identifying with at least one Latinx ethnicity.

Contributor Information

Alison L. Mroczkowski, Museum of Science & Industry

Lidia Monjaras-Gaytan, DePaul University.

David L. DuBois, University of Illinois at Chicago

References

  1. Arbuckle JL (2014). Amos (Version 23.0) [Computer Program]. Chicago: IBM SPSS. [Google Scholar]
  2. Bayer A, Grossman J, & DuBois DL (2015). Using volunteer mentors to improve the academic outcomes of underserved students: The role of relationships. Journal of Community Psychology, 43(4), 408–429. doi: 10.1002/jcop.21693 [DOI] [Google Scholar]
  3. Bentler PM (1990). Comparative fit indexes in structural models. Psychological Bulletin, 107 (2), 238–46. doi: 10.1037/0033-2909.107.2.238 [DOI] [PubMed] [Google Scholar]
  4. Bogat GA, & Liang B (2014). Gender in mentoring relationships In DuBois DL & Karcher MJ (Eds.), Handbook of Youth Mentoring (pp. 205–217). Thousand Oaks, CA: Sage. [Google Scholar]
  5. Brown CS, & Chu H (2012). Discrimination, ethnic identity, and academic outcomes of Mexican immigrant children: Importance of school context. Child Development, 83(5), 1477–1485. doi: 10.1111/j.1467-8624.2012.01786.x [DOI] [PubMed] [Google Scholar]
  6. Browne M, & Cudeck R (1993). EQS Structural Equations Program Manual. Los Angeles: Multivariate Software Inc. [Google Scholar]
  7. Caldwell CH, Zimmerman MA, Bernat DH, Sellers RM, & Notaro PC (2002). Racial identity, maternal support, and psychological distress among African American adolescents. Child Development, 73(4), 1322–1336. doi: 10.1111/1467-8624.00474 [DOI] [PubMed] [Google Scholar]
  8. Colón Y, & Sánchez B (2010). Explaining the gender disparity in Latino youth’s education: Acculturation and economic value of education. Urban Education, 45(3), 252–273. doi: 10.1177/0042085908322688 [DOI] [Google Scholar]
  9. Cooley CH (1902). Human nature and the social order. New York, NY: Scribner’s. [Google Scholar]
  10. DuBois DL, Burk-Braxton C, Swenson LP, Tevendale HD, & Hardesty JL (2002). Race and gender influences on adjustment in early adolescents: Investigation of an integrative model. Child Development, 73(5), 1573–1592. doi: 10.1111/1467-8624.00491 [DOI] [PubMed] [Google Scholar]
  11. DuBois D, & Silverthorn N (2005). Characteristics of natural mentoring relationships and adolescent adjustment: Evidence from a national study. Journal of Primary Prevention, 26(2), 69–91. doi: 10.1007/s10935-005-1832-4 [DOI] [PubMed] [Google Scholar]
  12. Gaylord-Harden NK, Ragsdale BL, Mandara J, Richards MH, & Petersen AC (2007). Perceived support and internalizing symptoms in African American adolescents: Self-esteem and ethnic identity as mediators. Journal of Youth and Adolescence, 36, 77–88. doi: 10.1007/s10964-006-9115-9 [DOI] [Google Scholar]
  13. Gordon DM, Iwamoto DK, Ward N, Potts R, & Boyd E (2009). Mentoring urban Black middle school male students: Implications for academic achievement. Journal of Negro Education, 78, 277–289. [PMC free article] [PubMed] [Google Scholar]
  14. Grant JH (2008). The relationship between ethnic identity and psychological health: A meta-analytic review (Dissertation). Retrieved from ProQuest Dissertations and Theses database (UMI Number: 3341032). [Google Scholar]
  15. Harris JT, & Nakkula MN (2010). Youth mentoring survey (YMS) [Measurement instrument]. Fairfax, VA: Applied Research Consulting. [Google Scholar]
  16. Hurd N, Sánchez B, Zimmerman MA, & Caldwell CH (2012). Natural mentors, racial identity, and educational attainment among African American adolescents: Exploring pathways to success. Child Development, 83(4), 1196–1212. doi: 10.1111/j.1467-8624.2012.01769.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Illinois State Board of Education. (2018). AYP FAQS. Retrieved from https://www.isbe.net/Pages/AYP-FAQs.aspx#.
  18. Kaplan CP Turner SG, Piotrowski C & Silbert E (2009). Club Amigas: A promising response to the needs of adolescent Latinas. Child and Family Social Work, 2, 213–221. doi: 10.1111/j.1365-2206.2009.00625.x [DOI] [Google Scholar]
  19. Kiang L, Supple AJ, Stein GL, & Gonzalez LM (2012). Gendered academic adjustment among Asian American adolescents in an emerging immigrant community. Journal of Youth and Adolescence, 41(3), 283–294. doi: 10.1007/s10964-011-9697-8. [DOI] [PubMed] [Google Scholar]
  20. Liao CL & Sánchez B (2016). Mentoring relationship quality profiles and their association with urban, low-income youth’s academic outcomes. Youth & Society, 1–20. DOI: 10.1177/0044118X16668058 [DOI] [Google Scholar]
  21. Mickelson RA (1990). The attitude-achievement paradox among Black adolescents. Sociology of Education, 63(1), 44–61. doi: 10.2307/2112896. [DOI] [Google Scholar]
  22. Murdock TB, Anderman LH, & Hodge SE, (2000). Middle-grade predictors of students’ motivation and behavior in high school. Journal of Adolescent Research, 15, 327–351. doi: 10.1177/0743558400153002 [DOI] [Google Scholar]
  23. Nakkula MJ, & Harris JT (2013). Assessing mentoring relationships In DuBois D & Karcher M (Eds.), Handbook of Youth Mentoring (pp. 45–62). Thousand Oaks, CA: Sage. [Google Scholar]
  24. Oyserman D, Destin M, & Novin S (2015). The context-sensitive future self: Possible selves motivate in context, not otherwise. Self and Identity, 14(2), 173–188. doi: 10.1080/15298868.2014.965733 [DOI] [Google Scholar]
  25. Phinney JS (1993). A three-stage model of ethnic identity development in adolescence In Bernal ME & Knight GP (Eds.), SUNY series, United States Hispanic studies. Ethnic identity: Formation and transmission among Hispanics and other minorities (pp. 61–79). Albany, NY, US: State University of New York Press. [Google Scholar]
  26. Raffaelli M, & Ontai LL (2004). Gender socialization in Latino/a families: Results from two retrospective studies. Sex Roles, 50, 287–299. doi: 10.1023/B:SERS.0000018886.58945.06 [DOI] [Google Scholar]
  27. Raposa EB, Erickson LD, Hagler M, & Rhodes JE (2018). How economic disadvantage affects the availability and nature of mentoring relationships during the transition to adulthood. American Journal of Community Psychology, 61(1–2), 191–203. doi: 10.1002/ajcp.12228 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Reyes O, Gillock KL, Kobus K, & Sánchez B (2000). A longitudinal examination of the transition into senior high school for adolescents from urban, low-income status, and predominantly minority backgrounds. American Journal of Community Psychology, 28, 519–544. doi: 10.1023/A:1005140631988 [DOI] [PubMed] [Google Scholar]
  29. Rhodes JE (2005). A model of youth mentoring In DuBois DL & Karcher MJ (Eds.), Handbook of Youth Mentoring (pp. 30–43). Thousand Oaks, CA: Sage. [Google Scholar]
  30. Rhodes JE, Contreras JM, & Mangelsdorf SC (1994). Natural mentor relationships among Latina adolescent mothers: Psychological adjustment, moderating processes, and the role of early parental acceptance. American Journal of Community Psychology, 22, 211–227. doi: 10.1007/BF02506863 [DOI] [PubMed] [Google Scholar]
  31. Rivas-Drake D (2008). Perceived opportunity, ethnic identity, and achievement motivation among Latinos at a selective public university. Journal of Latinos and Education, 7(2), 113–128. doi: 10.1080/15348430701828632 [DOI] [Google Scholar]
  32. Rivas-Drake D, Seaton EK, Markstrom C, Quintana S, Syed M, Lee RM, Schwartz SJ, Umana-Taylor AJ, French S, & Yip T (2014). Ethnic and racial identity in adolescence: Implications for psychosocial, academic, and health outcomes. Child Development, 85(1), 40–57. doi: 10.1111/cdev.12200 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Roberts R, Phinney J, Masse L, Chen Y, Roberts C, & Romero A (1999). The structure of ethnic identity in young adolescents from diverse ethnocultural groups. Journal of Early Adolescence, 19, 301–322. doi: 10.1177/0272431699019003001 [DOI] [Google Scholar]
  34. Sánchez B, Esparza P, & Colón Y (2008). Natural mentoring under the microscope: An investigation of mentoring relationships and Latino adolescents’ academic performance. Journal of Community Psychology, 36(4), 468–482. doi: 10.1002/jcop.20250 [DOI] [Google Scholar]
  35. Santos CE (2017). The history, struggles, and potential of the term Latinx. Latina/o Psychology Today, 4(2), 7–14. [Google Scholar]
  36. Schwartz EO, Kanchewa SS, Rhodes JE, Gowdy G, Stark AM, Horn JP, Parnes M, & Spencer R (2018). “I’m having a little struggle with this, can you help me out?”: Examining impacts and processes of a social capital intervention for first-generation college students. American Journal of Community Psychology, 61, 166–178. doi: 10.1002/ajcp.12206 [DOI] [PubMed] [Google Scholar]
  37. Seaton EK, Quintana S, Verkuyten M, & Gee GC (2017). Peers, policies, and place: The relation between context and ethnic/racial identity. Child Development, 88, 683–692. doi: 10.1111/cdev.12787 [DOI] [PubMed] [Google Scholar]
  38. Stanton-Salazar RD (2010). A social capital framework for the study of institutional agents and their role in the empowerment of low-status students and youth. Youth & Society, 10(5), 1–44. doi: 10.1177/0044118X10382877 [DOI] [Google Scholar]
  39. Supple AJ, Ghazarian SR, Frabutt JM, Plunkett SW, & Sands T (2006). Contextual influences on Latino adolescent ethnic identity and academic outcomes. Child Development, 77(5), 1427–1433. doi: 10.1111/j.1467-8624.2006.00945.x [DOI] [PubMed] [Google Scholar]
  40. Tucker LR, & Lewis C (1973). A reliability coefficient for maximum likelihood factor analysis. Psychometrika, 38, 1–10. doi: 10.1007/BF02291170 [DOI] [Google Scholar]
  41. Umaña-Taylor A, & Guimond AB (2010). A longitudinal examination of parenting behaviors and perceived discrimination predicting Latino adolescents’ ethnic identity. Developmental Psychology, 46(3), 636–650. doi: 10.1037/a0019376 [DOI] [PubMed] [Google Scholar]
  42. Umaña-Taylor A, Wong JJ, Gonzales NA, & Dumka LE, (2012). Ethnic identity and gender as moderators of the association between discrimination and academic adjustment among Mexican-origin adolescents. Journal of Adolescence, 35(4), 773–786. doi: 10.1016/j.adolescence.2011.11.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Umaña-Taylor A, Tynes BM, Toomey RB, Williams DR, & Mitchell KJ (2014). Latino adolescents’ perceived discrimination in online and offline settings: An examination of cultural risk and protective factors. Developmental Psychology, 51(1), 87–100. doi: 10.1037/a0038432 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. U.S. Census Bureau (December, 2017). 2003–2017 Annual Social and Economic Supplement to the Current Population Survey. Retrieved from https://www.census.gov/library/visualizations/time-series/demo/cps-historical-time-series.html [DOI] [PMC free article] [PubMed]
  45. U.S Census Bureau, Population Division (June, 2017). Annual Estimates of the Resident Population by Sex, Age, Race, and Hispanic Origin for the United States and States: April 1, 2010 to July 1, 2016. Retrieved from https://factfinder.census.gov/faces/tableservices/jsf/pages/productview.xhtml?src=bkmk
  46. Van Dam L, Smit D, Wildschut B, Branje SJT, Rhodes JE, Assink M, & Stams GJJM (2018). Does natural mentoring matter? A multilevel meta-analysis on the association between natural mentoring and youth outcomes. American Journal of Community Psychology, 62, 203–220. doi: 10.1002/ajcp.1224 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Wakefield WD, & Hudley C (2007). Ethnic and racial identity and adolescent well-being. Theory Into Practice, 46(2), 147–154. doi: 10.1080/00405840701233099 [DOI] [Google Scholar]
  48. Wittrup AR, Hussain SB, Albright AN, Hurd NM, Varner FA, & Mattis JS (2016). Natural mentors, racial pride, and academic engagement among Black adolescents: Resilience in context of perceived discrimination. Youth & Society, 1–21. doi: 10.1177/0044118X16680546 [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Yancey AK, Siegel JM, & McDaniel KL (2002). Role models, ethnic identity, and health risk behaviors in urban adolescents. Archives of Pediatric and Adolescent Medicine, 156, 55–61. doi: 10.1001/archpedi.156.1.55 [DOI] [PubMed] [Google Scholar]
  50. Yuan KH, & Chan W (2016). Measurement invariance via multigroup SEM: Issues and solutions with chi-square-difference tests. Psychological Methods, 21(3), 405. [DOI] [PubMed] [Google Scholar]
  51. Zimmerman MA, Bingenheimer JB, & Behrendt DE (2005). Natural mentoring In: DuBois DL; Karcher MJ., editors. Handbook of Youth Mentoring (pp. 143–157). Thousand Oaks, CA: Sage. [Google Scholar]

RESOURCES