Black adolescents\' relationships with natural mentors: associations with academic engagement via social and emotional development

Cultural Diversity and Ethnic Minority Psychology 2013, Vol. 19, No. 1, 76 – 85

© 2013 American Psychological Association 1099-9809/13/$12.00 DOI: 10.1037/a0031095

Black Adolescents’ Relationships With Natural Mentors: Associations With Academic Engagement via Social and Emotional Development Noelle M. Hurd

Robert M. Sellers

University of Virginia

University of Michigan

This study aimed to identify types of natural mentoring relationships that Black adolescents may experience and to connect these relationship types to academic outcomes via social and emotional development. Data were collected from 259 adolescents attending 3 middle schools in a Midwestern metropolitan area. Adolescents reported on whether or not they had a relationship with a natural mentor and answered several questions about characteristics of the relationship, as well as reporting on a number of indicators of positive development. Cluster analyses suggested two different types of natural mentoring relationships among study youth: less connected and more connected. We found that in comparison with youth without a natural mentor, youth who had a more connected natural mentoring relationship had greater student-reported academic engagement via higher social skills and psychological well-being and greater teacher-reported academic engagement via elevated social skills. We found that youth with less connected natural mentoring relationships did not differ from their counterparts without natural mentors on social skills, psychological well-being, or student- or teacher-reported academic engagement. Findings from the current study suggest that characteristics of the mentoring relationship may determine whether these relationships contribute to more positive outcomes among Black youth. Further, findings suggest that social and emotional development may transmit the positive effects of more connected natural mentoring relationships on Black youths’ academic engagement. Keywords: natural mentors, Black adolescents, academic engagement, social and emotional development

positively toward themselves, more confident in their abilities, and more focused on their future (Rhodes et al., 2006; Kogan, Brody, & Chen, 2011). These gains in social and emotional development likely lead to improved academic attitudes and behaviors (Durlak, Weissberg, Dymnicki, Taylor, & Schellinger, 2011). In the current study, we explored whether mentoring relationships may promote improved academic engagement via increased social skills and psychological well-being among a sample of Black adolescents. We also investigated whether relationship characteristics, such as relationship length and closeness, may influence these associations.

Natural mentors are caring and supportive adults in adolescents’ preexisting social networks, such as extended kin, neighbors, coaches, or community members. Despite increasing evidence that relationships with natural mentors may contribute to improved educational outcomes among Black youth (Hurd, Sanchez, Zimmerman, & Caldwell, 2012; Klaw, Rhodes, & Fitzgerald, 2003; Zimmerman, Bingenheimer, & Notaro, 2002), little is known about the mechanisms through which relationships with natural mentors influence these outcomes. Rhodes (2005) has proposed a model of youth mentoring that outlines potential processes through which mentoring relationships may affect youth outcomes and points to the significance of relationship characteristics in determining the potential influences of these relationships on youth outcomes. According to this model, one pathway from mentoring relationships to more positive youth outcomes is through social and emotional development (Rhodes, 2005; Rhodes, Spencer, Keller, Liang, & Noam, 2006). It may be that relationships with mentors increase youths’ interpersonal skills and ability to connect with others (Rhodes, Grossman, & Resch, 2000). Further, it is possible that these relationships may lead youth to feel more

Background Adolescence is a time of increased biological, social, cognitive, and emotional changes (Arnett, 1999). During this developmental period, youth are struggling to balance a growing need for autonomy with their enduring need for relatedness (Marcia, 1980). For this reason, nonparental adults may begin to play an increasingly important role in adolescents’ lives. Adolescents may elect, for example, to seek out nonparental adults for help and guidance rather than their parents because nonparental adults may be able to meet youths’ attachment needs without threatening youths’ sense of autonomy (Allen & Land, 1999; Zimmerman, Bingenheimer, & Behrendt, 2005). Mentors are nonparental adults who take a special interest in youths’ lives and form caring and supportive relationships with them (Rhodes et al., 2006). Research on mentoring relationships suggests that these relationships may fill a void not filled by peer or parental relationships.

Noelle M. Hurd, Department of Psychology, University of Virginia; Robert M. Sellers, Department of Psychology, University of Michigan. Correspondence concerning this article should be addressed to Noelle Hurd, 102 Gilmer Hall, P. O. Box 400400, Department of Psychology, University of Virginia, Charlottesville, VA 22904. E-mail: nh3v@ virginia.edu 76

NATURAL MENTORS & ACADEMIC ENGAGEMENT

Specifically, mentors may be trusted adults who youth can turn to for advice, support, and adult perspectives without the potential of negative consequences (e.g., punishment) that may come with disclosing intimate information to parents (Rhodes, 2002). Further, youth may see mentors as wiser than peers and consequently feel more comfortable seeking advice and guidance from these trusted adults (Beam, Chen, & Greenberger, 2002). Additionally, youth may allow themselves to be more vulnerable with their mentors than with their parents, teachers, or peers (Darling, Hamilton, Toyokawa, & Matsuda, 2002), leading them to discuss their true feelings with their mentors. These intimate conversations may foster mentees’ personal growth. In light of the substantial changes youth experience during adolescence, they may benefit tremendously from the additional support provided by a mentor. This support may both bolster and enhance youths’ positive social, emotional, and academic development (Rhodes & Lowe, 2009). Mentors who emerge from youths’ preexisting social networks and organically form mentoring relationships with youth are considered natural or informal mentors (in contrast to formal mentors who are paired with youth through programs; Zimmerman et al., 2005). Previous studies of natural mentoring relationships among Black adolescents suggest that these relationships are fairly prevalent among Black youth (Hurd & Zimmerman, 2010a, 2010b; Rhodes, Ebert, & Fischer, 1992; Klaw et al., 2003). This is consistent with research documenting a greater emphasis on intergenerational relationships among Black Americans and a more central role of extended kin and fictive kin (individuals with family like bonds formed in the absence of blood or legal ties) in the Black family system (Stack, 1974; Stewart, 2007). Beyond documenting the prevalence of these relationships, there remains a need to better understand the ways in which these relationships may shape Black adolescents’ development.

Relationships With Natural Mentors and Youth Outcomes Researchers have documented links between natural mentoring relationships and more positive academic outcomes among mentees. Zimmerman et al. (2002) found that the presence of a natural mentoring relationship was associated with more positive school attitudes among a predominantly African American sample of late adolescents. Klaw, Rhodes, and Fitzgerald (2003) found lower drop-out and greater high school completion rates among African American adolescent mothers who had enduring relationships with natural mentors. Researchers have also found associations between receiving support from natural mentors and greater school belonging and expectations for success among Latino adolescents from low-income neighborhoods (Sánchez, Esparza, & Colon, 2008). Among a nationally representative sample of youth, DuBois and Silverthorn (2005b) found that youth who reported a natural mentoring relationship were more likely to have completed high school and enrolled in college by early adulthood. Similarly, Erickson, McDonald, and Elder (2009) found positive associations between the presence of a natural mentor and high school performance and overall educational attainment among a nationally representative sample of emerging adults. Of note, Erickson et al. found these benefits of mentoring relationships above and beyond parental, peer, school, and personal resources available to youth.

77

Despite increasing evidence supporting associations between natural mentoring and youths’ academic outcomes, limited research attention has been given to the processes through which mentoring may promote more positive outcomes among mentees. According to Rhodes’ (2005) model of youth mentoring, close mentoring relationships may facilitate improvements in youth outcomes via their contributions to mentees’ social and emotional development. Supportive and caring mentoring relationships may help youth to be more accepting of themselves and to feel more confident in their ability to develop positive relationships with others. Positive relationships formed with mentors may generalize to other relationships in youths’ lives, thus promoting more positive interpersonal relationships overall (Rhodes, 2005). Further, empathic mentors may help youth to feel more understood, and through modeling may help youth learn how to be more empathic in their relations with others (Rhodes, 2005; Rhodes et al., 2006). By listening to youth discuss their problems, offering advice and guidance, and modeling effective communication and problemsolving skills, adult mentors may help youth to better understand, express, and modulate their emotions (Pianta, 1999; Rhodes, 2005). Moreover, youth can learn new strategies to manage conflict, problems, and impulsivity from their older and more experienced adult mentors. For some youth, mentors may serve as alternate attachment figures who may increase youths’ sense of security, thus allowing them to safely explore their environment and to advance socially and emotionally (Bowlby, 1988). Yet even among youth who already have a number of secure attachment figures, mentors may be used as additional supports to help youth traverse the challenging terrains of adolescence. To date, few researchers have investigated the potential of natural mentoring relationships to influence social and emotional development; however, the few studies that have investigated this possibility have yielded promising results. DuBois and Silverthorn (2005b) found, for example, that youth with natural mentoring relationships were more likely to report higher levels of selfesteem and life satisfaction. Other researchers have also found associations between relationships with natural mentors and greater self-esteem and more positive self-concept among youth (Ahrens, DuBois, Richardson, Fan, & Lozano, 2008; Sánchez, Reyes, & Singh, 2006). In addition, Kogan, Brody, and Chen (2011) found that protective natural mentoring relationships were associated with greater self-regulation and future orientation among African American emerging adults. For the most part, these studies have not considered social and emotional development as mediators between natural mentoring relationships and youth outcomes (for exception, see Kogan et al., 2011). In the current study, we seek to advance this area by investigating associations between natural mentoring relationships and Black adolescents’ academic outcomes via social and emotional development.

Significance of Relationship Characteristics In addition to its increased attention to mediating processes between mentoring relationships and youth outcomes, Rhodes’ (2005) model also addresses the role of mentoring relationship characteristics in shaping youth outcomes. Her model stipulates that a strong interpersonal connection is a necessary requirement for mentors to positively influence youth outcomes. Research findings from studies on both formal and natural mentoring rela-

HURD AND SELLERS

78

tionships support this claim and underscore the importance of considering how characteristics of mentoring relationships may influence the extent to which these relationships contribute to more positive youth outcomes. Researchers have found that relationship duration (DuBois & Rhodes, 2006; Grossman, Chan, Schwartz, & Rhodes, 2011; Grossman & Rhodes, 2002; Klaw et al., 2003), closeness (Chen, Greenberger, Farruggia, Bush, & Dong, 2003; DuBois & Silverthorn 2005a; Greenberger, Chen, & Beam, 1998; Spencer, 2006; Thomson & Zand, 2010), and frequency of contact may influence the degree to which youth are positively affected by mentoring relationships (DuBois, Holloway, Valentine, & Cooper, 2002; DuBois, Neville, Parra, & Pugh-Lilly, 2002; Rhodes, 2002). It may be that greater contact and longer relationship length create additional opportunities for mentor-youth involvement and closeness, as well as allow for relationship processes to unfold that may contribute to more positive youth outcomes (Rhodes, 2002). In the present study we were interested in exploring the role of relationship characteristics (i.e., relationship length, involvement, closeness, and frequency of contact) in shaping youths’ social, emotional, and academic outcomes. Previous work shows that relationship characteristics such as frequency of contact and closeness are often interrelated (Beam et al., 2002; DuBois & Silverthorn, 2005a; Klaw et al., 2003; Parra, DuBois, Neville, PughLilly, & Povinelli, 2002). This interrelatedness has made it difficult to examine the influence of separate mentoring relationship characteristics (DuBois & Silverthorn, 2005a). It is unclear, however, whether attempting to parcel out the competing effects of each relationship characteristic is the most appropriate approach to evaluating mentoring relationships. We argue that rather than examining the unique contribution of each mentoring relationship characteristic to youth outcomes, attention to the confluence of relationship characteristics may provide more information about the relative effectiveness of these relationships in promoting more positive youth outcomes. In an attempt to consider the collective contributions of mentoring relationship characteristics, we opted to use a profile approach to identify types of natural mentoring relationships based on the following four key relationship characteristics: relationship length, involvement, closeness, and frequency of contact. This analytic approach allowed us to identify types of natural mentoring relationships that Black youth may possess and link these types of relationships to youth outcomes.

Current Study The current study aimed to advance our understanding of natural mentoring relationships among Black early adolescents. Our focus on early adolescents is consistent with previous findings that younger adolescents, in comparison with older adolescents, report greater disclosure with adults (Thomson & Zand, 2010) and may form more enduring relationships with mentors (Grossman & Rhodes, 2002). Moreover, to date, the bulk of studies on natural mentoring relationships among Black youth have focused on older adolescents and emerging adults (Hurd & Zimmerman, 2010a, 2010b; Hurd et al., in press; Klaw & Rhodes, 1995; Klaw et al., 2003; Kogan & Brody, 2010; Kogan et al., 2011; Rhodes et al., 1992; Zimmerman et al., 2002), leaving us with a limited understanding of these relationships in early adolescence. In addition, much of this work has focused on special subsets of Black youth such as adolescents from economically disadvantaged back-

grounds, emerging adults in resource-poor rural communities, and pregnant or parenting teen moms. In light of these unique samples, much prior research on mentoring has been conducted within a risk and resilience framework with a heavy emphasis on the potential of natural mentoring relationships to prevent or reduce negative youth outcomes. The current study is unique in its emphasis on positive developmental outcomes among a sample of Black early adolescents from diverse socioeconomic backgrounds. Using a strengths-based approach, we identified profiles of natural mentoring relationships and linked them to youths’ academic engagement via youths’ social skills and psychological well-being. We expected that natural mentoring relationships reflecting greater adult–youth connectedness would relate to higher student- and teacher-reported academic engagement via participants’ elevated social skills and psychological well-being.

Method Participants Two hundred fifty-nine seventh, eighth, and ninth grade students (58% female; mean age ⫽ 13.56, SD ⫽ .96) participated in the current study. Participants were recruited from three middle schools in a Midwestern metropolitan area, with 80 participants from the first school, 78 from the second school, and 101 from the third school. The three schools were in different school districts and varied in their socioeconomic and racial composition. Across the three schools, 22% (school 1), 29% (school 2), and 71% (school 3) of students were eligible for free or reduced-price lunch. Of the 259 participants, 220 self-identified as Black or African American and 39 participants self-identified as biracial or multiracial with at least 1 African American or Black parent. Seventy percent of participants reported two parents (biological or stepparents) who were primarily responsible for raising them. The average highest level of education of participants’ caregivers was a bachelor’s degree.

Procedure This study received approval from both the Institutional Review Board at the University of Michigan and the staff at the schools where data were collected. Participant recruitment was conducted at participants’ schools during the 2010 –2011 school year. Recruitment fliers were distributed during students’ lunch period and also mailed to participants’ homes and distributed at parent meetings. Parental consent and participant assent were collected before survey administration. Survey administration took approximately 45 minutes, and surveys were completed during participants’ class period, lunch hour, or after school. Participants’ were compensated for their participation with a $20 Visa gift card. In addition, participants’ social studies and English teachers were contacted and asked to complete evaluations of study participants. Teachers were e-mailed online consent forms, and those who consented were asked to complete online evaluations of the students who they had in their classroom. Teachers were compensated with a $10 Visa gift card for every student assessment they completed.

Measures Natural mentors. We used one item to assess whether participants had a natural mentor. Participants were asked, “Is there an

NATURAL MENTORS & ACADEMIC ENGAGEMENT

important adult in your life other than your parents or a person who raised you who has taken a special interest in you and who you can go to for support and guidance?” Participants who responded affirmatively were then asked a series of questions about their relationship with the identified adult. Specifically, participants were asked to indicate how they knew the adult (open-ended), how long the adult had been an important person in their life (4-point scale ranging from 1 to 4: less than a year to 5 or more years), how close they felt to the adult (5-point scale ranging from 1 to 5: not close at all to very close), and how frequently they had contact with the adult (7-point scale ranging from 1 to 7: less than once a year to almost every day; Udry, 2003). A 13-item checklist was used to assess mentor involvement. Adolescents checked the type of instrumentally and emotionally supportive activities (0 –13) their mentor did for them. Items included help with schoolwork, help with relationship problems, talking about the participant’s future, and listening. These 13 items were summed to yield a total involvement item. Social skills. We used the Social Skills Rating System – Student Form (Gresham & Elliott, 1990) to assess participants’ social skills. This measure includes 39 items assessing four dimensions of social skills: self-control (e.g., “I avoid doing things with others that may get me in trouble with adults.”), assertion (e.g., “I let friends know I like them by telling or showing them.”), empathy (e.g., “I try to understand how my friends feel when they are angry, upset, or sad.”), and cooperation (e.g., “I compromise with parents or teachers when we have disagreements.”). Participants were asked to indicate the frequency with which they did each of the items on a 3-point scale ranging from 0 (never) to 2 (very often). Cronbach’s alpha for participants’ self-reported social skills was .88. Psychological well-being. We used a 24-item modified version of the Ryff Scales of Psychological Well-being (Ryff, 1989) to assess participants’ psychological well-being. The measure includes assessments of six dimensions of psychological well-being including purpose (e.g., “I have a sense of direction and purpose in life.”), positive relationships with others (e.g., “I find it really difficult to open up and talk to others.”: reverse coded), personal growth (e.g., “I feel that I have developed a lot as a person over time.”), autonomy (e.g., “My decisions are not usually influenced by what everyone else is doing.”), environmental mastery (e.g., “In general, I feel I am in charge of my life.”), and self-acceptance (e.g., “In general, I feel confident and positive about myself.”). Adolescents were asked to indicate the extent to which they agreed or disagreed with each statement on a scale from 1 (strongly disagree) to 5 (strongly agree). Negative items were reverse coded so that higher scores on all items indicated higher psychological well-being. Cronbach’s alpha for these items was .82. Academic engagement. We assessed student-reported academic engagement using eight items from Wellborn’s (1991) academic engagement scale of behavioral engagement. This measure was intended to assess students’ effort, attention, and persistence in starting and persevering with academic tasks (Skinner & Belmont, 1993). The eight items included in the current study assessed students’ curiosity toward novel classroom material (e.g., “I work hard when we start something new in class.”) and persistence in the face of academic challenge (e.g., “If I can’t get a problem right the first time, I just keep trying.”; Neblett, Philip, Cogburn, & Sellers, 2006; Smalls, White, Chavous, & Sellers,

79

2007). Participants were asked to rate how true each statement was for them on a scale of 1 (not at all true) to 4 (very true). Cronbach’s alpha for these eight items was .79. A similar version of this scale was administered to teachers and they were asked to report on students’ academic curiosity (e.g., “This student participates when we discuss new material.”) and persistence (e.g., “If a problem is really hard, s/he keeps working at it.”). Teachers rated how true these statements were for each student based on the teacher’s personal experiences with the student in their classroom. Response options ranged from 1 (not at all true) to 4 (very true). Cronbach’s alpha for these 15 items was .97. Participant and natural mentor demographics. Adolescents reported on their and their natural mentors’ age, gender, and race/ethnicity. Participants also reported on their caregivers’ and their natural mentors’ educational attainment. In our analyses, we used the highest educational attainment of any of participants’ caregivers. Highest caregiver educational attainment ranged from junior high school or less to doctoral or professional degree (4.7% high school incompletion, 11.2% high school diploma, 19.7% some college, 31.6% college diploma, 4.6% some graduate school, 24.7% master’s degree, and 3.5% doctoral or professional degree). We also had data on which of the three schools included in this study participants attended.

Data Analysis We used estimation maximization imputation to deal with a small amount of missing data across study variables (⬍8%). We then conducted K-means cluster analyses using our four characteristics of mentoring relationships to identify mentoring relationship profiles/types. We used the profiles that emerged from these analyses in our structural equation model. We created composite variables for each subscale of the Social Skills Rating System by summing items. These composites were used as indicators of a social skills latent factor in our structural model. We created composite variables for each subscale of the Ryff Scales of Psychological Well-Being by averaging the subscale items. We used the composite variables as indicators of a psychological well-being latent factor in our structural model. We also created composites by averaging the academic curiosity and persistence subscales for the student- and teacher-reported items assessing academic engagement. All of these composites were used as indicators of their respective latent constructs in our measurement and structural models. We used M-plus 6 software (Muthén & Muthén, 2010) to test our measurement and structural models. In our structural models we assessed associations between mentoring relationship profiles and participants’ social skills, psychological well-being, and student- and teacher-reported academic engagement accounting for participants’ age, gender, school, and caregiver education. We also investigated whether social skills and psychological wellbeing may transmit the benefits of certain mentoring relationship profiles on student- and teacher-reported academic engagement. In our analyses, we correlated all exogenous variables with each other. We also correlated the disturbances of the two intervening variables (social skills and psychological well-being) and disturbances of the two outcome variables (student- and teacher-reported academic engagement). We relied on the ␹2 statistic, comparative fit index (CFI), the Tucker Lewis Index (TLI), and the root-meansquare error of approximation (RMSEA) to evaluate model fit. To

HURD AND SELLERS

80

assess the significance of indirect effects, we generated bootstrapped confidence intervals of the indirect effects. If the 95% confidence interval of the standardized specific indirect effect did not include 0, we concluded that there was a significant indirect effect.

Results Natural Mentors

LC Mentoring Rela
onship

MC Mentoring Rela
onship

15 1.5 1 0.5 0 -0.5 -1

Almost 75% (n ⫽ 195) of participants reported the presence of a natural mentor in their lives. Most (65%; n ⫽ 126) mentors were related to participants (of related mentors, 37% were aunts or uncles, 33% were grandparents, and 17% were older siblings). Unrelated mentors (n ⫽ 69) were primarily family friends (53%), church members or pastors (21%), and teachers or coaches (6%). Sixty-three percent of mentors were female, 61% of participants identified a gender-matched natural mentor, and 80% of mentors were Black or African American (an additional 8% were bi- or multiracial with one Black or African American parent). Approximately half of participants’ mentors were between 18 and 39 years old (10% of mentors were 18 –19 years old, 31% were 20 –29 years old, 13% were 30 –39 years old) and 38% were 40 years old or older (8% of participants with a mentor reported that they did not know the age of their mentor). According to participants’ reports, 2% of mentors had less than a high school diploma, 13% had completed high school or a GED, 19% had received some college or vocational training, 23% completed college, and 15% completed graduate or professional school after college (28% of participants with a mentor reported that they did not know the highest educational attainment of their mentor). On average, participants reported that their mentor had been an important person in their lives for 3– 4 years, that they saw their mentor about once a week, and that they felt quite close (4 on a scale of 1–5) to their mentors. Average reported mentor involvement also was high (10.73 of 13, SD ⫽ 2.26).

connected mentor). Participants did not differ in terms of participant gender, age, or caregivers’ educational attainment across mentor group status. We also did not find differences in mentoring group status as a function of the school participants attended, ␹2(4) ⫽ 4.64, ns. Youth with more connected mentoring relationships had higher social skills and psychological well-being than youth with no mentor or less connected mentors. Participants did not differ in student- or teacher-reported academic engagement as a function of mentor group status.

Natural Mentoring Relationship Cluster Analysis

Structural Model

We standardized the four mentoring relationship variables (i.e., relationship length, involvement, closeness, and frequency of contact) and then conducted cluster analyses using a k-means approach to identify distinct mentoring relationship profiles. The k-means approach was selected because it maximizes betweencluster differences and minimizes within-cluster variance (Hartigan, 1975). We requested two-, three-, and four-cluster solutions. We retained the two-cluster solution to maximize group differences while also maintaining sufficiently large cluster sizes for our analyses. As can be seen in Figure 1, the first cluster (n ⫽ 72) was characterized by shorter relationship length, lower involvement and closeness, and less frequent contact; therefore, we labeled this cluster less connected. The second cluster (n ⫽ 123) was characterized by longer relationship length, greater involvement and closeness, and more frequent contact; consequently, we labeled this cluster more connected. Table 1 includes descriptive and comparative statistics for less connected mentoring relationships and more connected mentoring relationships. Table 2 reports descriptive characteristics of participants as a function of mentor group status (i.e., no mentor, less connected mentor, or more

In our structural model, we examined the direct and indirect effects of less connected and more connected mentoring relationships (in comparison with no mentoring relationship) on participants’ student- and teacher-reported academic engagement via adolescents’ social skills and psychological well-being controlling for participants’ age, gender, school, and caregiver education. Our structural model yielded a statistically significant ␹2 value, ␹2(df ⫽ 141) ⫽ 249.23, p ⬍ .01; however, all other fit statistics indicated acceptable model fit to the data [CFI ⫽ .95, TLI ⫽ .93, RMSEA ⫽ .06 (95% CI for RMSEA ⫽ .05, .07)]. Results of our model (see Figure 2) indicated that having a more connected mentoring relationship (in comparison with no mentoring relationship) was associated with greater social skills, which related to higher levels of student- and teacher-reported academic engagement. Similarly, having a more connected mentoring relationship (in comparison with no mentoring relationship) was associated with increased psychological well-being, which was associated with greater student-reported academic engagement. Bootstrapped confidence intervals of standardized indirect effects suggest that more connected natural mentoring relationships indirectly affected partici-

-1.5 Length of Rela
onship

Involvement

Closeness

Frequency of Contact

Figure 1. Summary of natural mentoring profiles (standardized means). LC ⫽ less connected; MC ⫽ more connected.

Correlations and Measurement Model Table 3 shows correlations among study variables. The measurement model indicated appropriate fit of our latent constructs. Despite a statistically significant ␹2 value, our other indicators of model fit (CFI ⫽ .96, TLI ⫽ .94, and RMSEA ⫽ .05) all suggested acceptable fit. Factor loadings for the indicators of latent factors ranged from .46 to .94.

NATURAL MENTORS & ACADEMIC ENGAGEMENT

81

Table 1 Descriptive Characteristics of Mentoring Relationships and Natural Mentors Variable

LC mentoring relationship (n ⫽ 72)

MC mentoring relationship (n ⫽ 123)

Relationship length Involvement Closeness Frequency of contact % Familial mentors % Female mentors Mentor age range Mentor educational attainment % Racially-matched mentors % Gender-matched mentors

2.56 (.98) 9.22 (2.45) 3.71 (.72) 4.71 (1.6) 53% 56% 2.72 (1.12) 3.44 (1.05) 81% 64%

3.94 (.25) 11.62 (1.59) 4.86 (.37) 6 (1.3) 72% 67% 3.22 (1.19) 3.53 (1.12) 91% 60%

Note.

t(193) ⫽ ⫺14.89, p ⬍ .01 t(193) ⫽ ⫺8.28, p ⬍ .01 t(193) ⫽ ⫺14.76, p ⬍ .01 t(193) ⫽ ⫺6.10, p ⬍ .01 ␹2(1) ⫽ 6.99, p ⫽ .01 ␹2(1) ⫽ 2.39, ns t(193) ⫽ ⫺2.92, p ⬍ .01 t(193) ⫽ ⫺.48, ns ␹2(1) ⫽ 4.48, p ⬍ .05 ␹2(1) ⫽ .39, ns

LC ⫽ less connected; MC ⫽ more connected.

pants’ student-reported academic engagement via higher social skills (95% CI: .01, .17) and greater psychological well-being (95% CI: .01, .18). There was also a significant indirect effect from more connected natural mentoring relationships to participants’ teacher-reported academic engagement via participants’ social skills (95% CI: .02, .21). Though not displayed in our final model, we found that caregiver educational attainment predicted greater social skills (␤ ⫽ .16, p ⬍ .01) and higher teacher-reported academic engagement (␤ ⫽ .16, p ⬍ .05). We also found that being female was associated with greater social skills (␤ ⫽ .15, p ⬍ .05) and higher teacher-reported academic engagement (␤ ⫽ .32, p ⬍ .05).

Discussion Overall, our findings support Rhodes’ (2005) model of youth mentoring by underscoring the importance of mentoring relationship characteristics and the potential of social and emotional development to transmit positive effects of mentoring relationships on youths’ academic outcomes. Specifically, our findings suggest that mentoring relationships characterized by longer duration (5 or more years), more frequent contact (contact at least 2–5 times a week), greater levels of closeness (mentees reported feeling quite or very close to their mentor), and more mentor–mentee involvement (average of 11.62 on scale of 0 –13 activities that mentor does with mentee) have the potential to positively influence Black adolescents’ outcomes, whereas relationships without these attributes may not contribute to more positive outcomes among Black

youth. These findings are consistent with theory (Rhodes, 2005) and previous research findings emphasizing the role of mentor– mentee relationship characteristics in shaping youth outcomes (DuBois & Silverthorn, 2005a; Kogan et al., 2011). Our findings are also in line with other work that has only found positive mentoring influences among youth in longer-lasting mentoring relationships (Rhodes, Reddy, & Grossman, 2005). These findings suggest that the mere presence of a mentoring relationship may not be enough to confer benefits to youth, but rather a strong interpersonal connection may need to be formed in order for these relationships to contribute to more positive developmental outcomes. Moreover, the results of this study suggest that youth with mentoring relationships characterized by moderate levels of connectedness did not differ from youth without relationships with natural mentors across any study outcomes. We labeled our mentoring groups less versus more connected because these labels were relative. Descriptive analyses indicated that the mentoring relationships in the less connected group still displayed moderate levels of connectedness. Thus, our findings suggest that there may be discernible differences in youth outcomes as a function of moderate versus high levels of connectedness. This distinction may have particular relevance to early adolescents who may be particularly reliant on strong nonparental bonds as they transition from childhood to adolescence. Early adolescents may be most strongly influenced by natural mentors who they see often, interact with frequently, have known for a long time, and feel especially

Table 2 Descriptive Characteristics of Participants as a Function of Mentor Group Status

Variable

No mentoring relationship (n ⫽ 64)

LC mentoring relationship (n ⫽ 72)

MC mentoring relationship (n ⫽ 123)

Participant gender (% female) Participant age Caregiver educational attainment Social skills Psychological well-being Student-reported academic Engagement Teacher-reported academic Engagement

64% 13.67 (.82) 5.31 (1.48) 51.09a (10.55) 3.72a (.43) 3.03 (.49) 2.77 (.76)

58% 13.56 (.96) 5.05 (1.58) 51.10a (10.10) 3.72a (.54) 3.07 (.58) 2.95 (.69)

55% 13.5 (1.03) 5.05 (1.62) 56.30b (9.56) 3.98b (.49) 3.14 (.50) 2.93 (.69)

Note.

␹2(2) ⫽ 1.33, ns F(2, 256) ⫽ .72, ns F(2, 256) ⫽ .67, ns F(2, 256) ⫽ 3.34, p ⬍ .05 F(2, 256) ⫽ 3.38, p ⬍ .05 F(2, 256) ⫽ 1.15, ns F(2, 256) ⫽ 1.28, ns

Means with differing subscripts differ significantly based on Tukey post hoc comparisons. LC ⫽ less connected; MC ⫽ more connected.

HURD AND SELLERS

82 Table 3 Correlations Among Study Variables Variable 1. 2. 3. 4. 5. 6. 7. 8. 9.

LC mentoring relationship MC mentoring relationship Social skills Psychological well-being Student-reported academic engagement Teacher-reported academic engagement Female Age Caregiver educational attainment

1

2

3

4

5

6

7

8

— ⫺.59ⴱ .09 .10 .04 .05 .01 .01 ⫺.02

— .16ⴱ .16ⴱ .09 .04 ⫺.06 ⫺.06 ⫺.04

— .52ⴱ .53ⴱ .35ⴱ .11 .03 .18ⴱ

— .54ⴱ .12 ⫺.04 .04 .08

— .31ⴱ .01 ⫺.04 .17ⴱ

— .37ⴱ ⫺.08 .17ⴱ

— .05 ⫺.01

— .01

Note. LC ⫽ less connected; MC ⫽ more connected. p ⬍ .05.

ⴱ

older than natural mentors in the less connected group. There were no differences between the two groups in the percentage of female or gender-matched mentors, nor did the mentors in each group differ in their average levels of educational attainment. These findings suggest that natural mentors who are older, racially matched, and related to their mentees may be more likely to develop strong mentoring bonds with Black early adolescents. Older adults may be more prepared to invest in these relationships, may feel more experienced and better able to support young people, and/or may have more time to devote to these intergenerational relationships in comparison with younger adults. Natural mentors who share their mentees’ race may feel more confident in their ability to support their mentees and may be better able to connect with their mentees given shared histories, life experiences, or values. It is important to note, however, that although these mentoring groups did differ by age and racial composition, the

close to. Relationships with such strong levels of connectedness may have the most potential to positively shape early adolescents’ social and emotional well-being. Further, given the strong tradition of intergenerational relationships in the Black community and the significant role of extended kin in the Black family (Stack, 1974; Stewart, 2007), our findings suggest that moderately strong levels of connectedness with nonparental adults may be more normative among Black early adolescents. Therefore, whereas moderately strongly connected mentoring relationships did not predict improved psychosocial outcomes, mentoring relationships characterized by particularly strong levels of connectedness were associated with more positive psychosocial outcomes. Of note, there was a greater representation of familial and Black natural mentors in the more connected mentoring relationship group in comparison with the less connected mentoring relationship group. Natural mentors in the more connected group were also

Self-Control

Empathy

Assertion

.58

.70

.67

.84

Social Skills

LC Natural Mentoring Relationship

student report R² = .65

Academic Curiosity

.80

Academic Persistence

.49 .46

.94

.18

teacher report R² = .38

Psychological Well-Being .56

Positive Relations

Academic Curiosity

Academic Engagement

.75

Purpose

.73

Academic Engagement

.43

.19

MC Natural Mentoring Relationship

Cooperation

.59

Personal Growth

.46

Autonomy

.76 .58

Environmental Mastery

.91

Academic Persistence P i t

S lf SelfAcceptance

Figure 2. Associations between mentoring relationship profiles (in comparison with no mentoring relationship) and youths’ social skills, psychological well-being, and student- and teacher-reported academic engagement adjusted for participants’ gender, age, school, and caregiver education. ␹2(df ⫽ 141; n ⫽ 259) ⫽ 249.23, p ⬍ .01; CFI ⫽ .95, TLI ⫽ .93, RMSEA ⫽ .06 (95% CI for RMSEA ⫽ .05, .07). LC ⫽ Less Connected; MC ⫽ More Connected. Model displays standardized coefficients for significant paths only. Hyphenated arrows represent nonsignificant paths. Not shown in the model are the correlated disturbance variances between social skills and psychological well-being (r ⫽ .63; p ⬍ .05) and between student-reported academic engagement and teacherreported academic engagement (r ⫽ .15; p ⬍ .05).

NATURAL MENTORS & ACADEMIC ENGAGEMENT

average differences appear to be relatively small. There was, however, a larger difference between the representation of familial mentors in each group with approximately half of the mentors in the less connected mentoring relationship group being familial and almost three fourths of the mentors in the more connected mentoring relationship group being familial. The greater representation of familial mentors in the more connected mentoring relationship group suggests that familial mentors may have greater opportunities to establish strong bonds with their mentees. This finding is consistent with DuBois’ and Silverthorn’s (2005a) finding that youth with familial mentors reported greater frequency of contact, closeness, and relationship duration in comparison with youth with nonfamilial mentors. Findings of our structural model suggest that strongly connected natural mentoring relationships may contribute to greater academic engagement via greater social skills and psychological well-being. This is one of the first studies to explicitly test Rhodes’ (2005) proposed social and emotional pathway from mentoring relationships to more positive youth outcomes. Of note, we did not find direct effects of strongly connected mentoring relationships on student- or teacher-reported academic engagement; however, we did find significant indirect effects via social skills and psychological well-being. These findings suggest that having a strong connection to a supportive nonparental adult may have positively contributed to adolescents’ interpersonal skills and self confidence. Through supportive interactions, natural mentors may have been helping youth learn how to effectively express and regulate their emotions. These adults also may have been reflecting back positive images of their adolescent mentees, helping them to feel more self-assured and capable. Being able to turn to these supportive nonparental adults rather than parents for assistance with schoolwork or personal problems may have helped build early adolescents’ sense of autonomy. Further, talking to their natural mentors about their future plans may have fostered greater goal-orientation among youth. As we expected, greater social skills and psychological wellbeing were associated with higher academic engagement. A growing body of work points to the critical role of social and emotional well-being in facilitating academic success (for a review see Durlak et al., 2011). Not surprisingly, we found that students who reported doing better socially and psychologically also reported greater academic engagement. We also found a positive association between students’ social skills and teacherreported academic engagement; however, students’ psychological well-being was not related to teacher-reported academic engagement. It is unclear why student-reported academic engagement was related to youths’ psychological well-being while teacher-reported academic engagement was unrelated to youths’ psychological well-being. Although student- and teacher-reported academic engagement were correlated (r ⫽ .31, p ⬍ .05), this moderate correlation suggests that students’ and teachers’ perceptions of students’ academic engagement varied considerably. We found that teacher-reported academic engagement was correlated with participants’ gender (association with female gender: r ⫽ .37, p ⬍ .05), whereas studentreported academic-engagement was not correlated with gender. Although teachers’ ratings may be informed by comparisons with other students, it is not clear whether these perceptions are more accurate than students’ own perceptions. Teachers enter

83

classrooms with their own sets of biases and also may struggle to monitor the engagement of each student, particularly when class sizes are large. Continued investigation into associations between psychological well-being and student- and teacherreported academic engagement is needed to better understand factors contributing to these inconsistent findings. This study is limited by its observational nature and crosssectional design. As a result, we were unable to determine causal associations or establish temporal precedence of predictor variables. Further, we did not account for other assets or resources in youths’ lives that may have made them more likely to have a strongly connected natural mentoring relationship and more positive psychosocial outcomes. In addition, caution should be used when attempting to generalize study findings to Black youth in nonmetropolitan settings. Yet our study also possesses a number of strengths such as the inclusion of a sample of Black early adolescents from diverse socioeconomic backgrounds, the implementation of a cluster approach to identify mentoring relationship profiles, and the use of data from multiple sources. This study is one of the first to consider the role of natural mentoring relationships in promoting positive developmental outcomes among a socioeconomically diverse sample of Black early adolescents. Our findings suggest that natural mentoring relationships characterized by strong bonds and connectedness may promote Black early adolescents’ academic engagement via adolescents’ social skills and psychological well-being. The promotion of these types of relationships among youth who do not have a natural mentor and the strengthening of these bonds among youth with a moderately connected mentoring relationship may be effective strategies for the promotion of more positive academic outcomes among Black early adolescents. Though these relationships may lead to increased academic engagement among Black youth, it is not clear whether and how these relationships may help Black youth to overcome structural barriers (e.g., limited access to academic opportunities, lower teacher expectations) to academic success they may face. Future natural mentoring research with Black youth that explores the potential of these relationships to address some of these challenges is needed.

References Ahrens, K. R., DuBois, D. L., Richardson, L. P., Fan, M.-Y., & Lozano, P. (2008). Youth in foster care with adult mentors during adolescence have improved adult outcomes. Pediatrics, 121, e246 – e252. doi:10.1542/ peds.2007-0508 Allen, J. P., & Land, D. (1999). Attachment in adolescence. In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment theory and research (pp. 319 –335). New York, NY: Guilford Press. Arnett, J. J. (1999). Adolescent storm and stress, reconsidered. American Psychologist, 54, 317–326. doi:10.1037/0003-066X.54.5.317 Beam, M. R., Chen, C., & Greenberger, E. (2002). The nature of adolescents’ relationships with their “very important” nonparental adults. American Journal of Community Psychology, 30, 305–325. doi:10.1023/ A:1014641213440 Bowlby, J. (1988). A secure base: Parent-child attachment and healthy human development. New York, NY: Basic Books. Chen, C., Greenberger, E., Farruggia, S., Bush, K., & Dong, Q. (2003). Beyond parents and peers: The role of important non-parental adults (VIPs) in adolescent development in China and the United States. Psychology in the Schools, 40, 35–50. doi:10.1002/pits.10068

84

HURD AND SELLERS

Darling, N., Hamilton, S., Toyokawa, T., & Matsuda, S. (2002). Naturally occurring mentoring in Japan and the United States: Social roles and correlates. American Journal of Community Psychology, 30, 245–270. doi:10.1023/A:1014684928461 DuBois, D. L., Holloway, B. E., Valentine, J. C., & Cooper, H. (2002). Effectiveness of mentoring programs for youth: A meta-analytic review. American Journal of Community Psychology, 30, 157–197. doi:10.1023/ A:1014628810714 DuBois, D. L., Neville, H. A., Parra, G. R., & Pugh-Lilly, A. O. (2002). Testing a new model of mentoring. New Directions for Mental Health Services, 2002, 21–57. doi:10.1002/yd.23320029305 DuBois, D. L., & Rhodes, J. E. (2006). Youth mentoring: Bridging science with practice. Journal of Community Psychology, 34, 647– 655. doi: 10.1002/jcop.20121 DuBois, D. L., & Silverthorn, N. (2005a). Characteristics of natural mentoring relationships and adolescent adjustment: Evidence from a national study. The Journal of Primary Prevention, 26, 69 –92. doi:10.1007/ s10935-005-1832-4 DuBois, D. L., & Silverthorn, N. (2005b). Natural mentoring relationships and adolescent health: Evidence from a national study. American Journal of Public Health, 95, 518 –524. doi:10.2105/AJPH.2003.031476 Durlak, J. A., Weissberg, R. P., Dymnicki, A. B., Taylor, R. D., & Schellinger, K. B. (2011). The impact of enhancing students’ social and emotional learning: A meta-analysis of school-based universal interventions. Child Development, 82, 405– 432. doi:10.1111/j.1467-8624.2010 .01564.x Erickson, L. D., McDonald, S., & Elder, G. H. (2009). Informal mentors and education: Complementary or compensatory resources? Sociology of Education, 82, 344 –367. doi:10.1177/003804070908200403 Greenberger, E., Chen, C., & Beam, M. R. (1998). The role of “very important” nonparental adults in adolescent development. Journal of Youth and Adolescence, 27, 321–343. doi:10.1023/A:1022803120166 Gresham, F. M., & Elliott, S. N. (1990). The social skills rating system. Circle Pines, MN: American Guidance Service. Grossman, J. B., Chan, C., Schwartz, S., & Rhodes, J. (2012). The test of time in school-based mentoring: The role of relationship duration and re-matching on academic outcomes. American Journal of Community Psychology, 49, 43–54. doi:10.1007/s10464-011-9435-0 Grossman, J. B., & Rhodes, J. E. (2002). The test of time: Predictors and effects of duration in youth mentoring relationships. American Journal of Community Psychology, 30, 199 –219. doi:10.1023/A: 1014680827552 Hartigan, J. A. (1975). Clustering algorithms. New York, NY: Wiley. Hurd, N. M., Sanchez, B., Zimmerman, M. A., & Caldwell, C. H. (2012). Natural mentors, racial identity, and educational attainment among African American adolescents: Exploring pathways to success. Child Development. doi:10.1111/j.1467-8624.2012.01769.x Hurd, N. M., & Zimmerman, M. A. (2010a). Natural mentoring relationships among adolescent mothers: A study of resilience. Journal of Research on Adolescence, 20, 789 – 809. doi:10.1111/j.1532-7795.2010 .00660.x Hurd, N. M., & Zimmerman, M. (2010b). Natural mentors, mental health, and risk behaviors: A longitudinal analysis of African American adolescents transitioning into adulthood. American Journal of Community Psychology, 46, 36 – 48. doi:10.1007/s10464-010-9325-x Klaw, E. L., & Rhodes, J. E. (1995). Mentor relationships and the career development of pregnant and parenting African-American teenagers. Psychology of Women Quarterly, 19, 551–562. doi:10.1111/j.1471-6402 .1995.tb00092.x Klaw, E. L., Rhodes, J. E., & Fitzgerald, L. F. (2003). Natural mentors in the lives of African American adolescent mothers: Tracking relationships over time. Journal of Youth and Adolescence, 32, 223–232. doi: 10.1023/A:1022551721565

Kogan, S. M., & Brody, G. H. (2010). Linking parenting and informal mentor processes to depressive symptoms among rural African American young adult men. Cultural Diversity and Ethnic Minority Psychology, 16, 299 –306. doi:10.1037/a0018672 Kogan, S. M., Brody, G., & Chen, Y. (2011). Natural mentoring processes deter externalizing problems among rural African American emerging adults: A prospective analysis. American Journal of Community Psychology, 48, 272–283. doi:10.1007/s10464-011-9425-2 Marcia, J. E. (1980). Identity in adolescence. In J. Adeison (Ed.), Handbook of adolescent psychology (pp. 109 –137). New York, NY: Wiley & Sons. Muthén, L. K., & Muthén, B. O. (2010). Mplus user’s guide (6th edition). [Computer software and manual]. Los Angeles, CA: Muthén & Muthén. Neblett, E. W., Philip, C. L., Cogburn, C. D., & Sellers, R. M. (2006). African American adolescents’ discrimination experiences and academic achievement: Racial socialization as a cultural compensatory and protective factor. Journal of Black Psychology, 32, 199 –218. doi:10.1177/ 0095798406287072 Parra, G. R., DuBois, D. L., Neville, H. A., Pugh-Lilly, A. O., & Povinelli, N. (2002). Mentoring relationships for youth: Investigation of a processoriented model. Journal of Community Psychology, 30, 367–388. doi: 10.1002/jcop.10016 Pianta, R. C. (1999). Enhancing relationships between children and teachers. Washington, DC: American Psychological Association. doi: 10.1037/10314-000 Rhodes, J. E. (2002). Stand by me: The risks and rewards of mentoring today’s youth. Cambridge, MA: Harvard University Press. Rhodes, J. E. (2005). A model of youth mentoring. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 30 – 43). Thousand Oaks, CA: Sage. doi:10.4135/9781412976664.n3 Rhodes, J. E., Ebert, L., & Fischer, K. (1992). Natural mentors: An overlooked resource in the social networks of young, African American mothers. American Journal of Community Psychology, 20, 445– 461. doi:10.1007/BF00937754 Rhodes, J. E., Grossman, J. B., & Resch, N. L. (2000). Agents of change: Pathways through which mentoring relationships influence adolescents’ academic adjustment. Child Development, 71, 1662–1671. Rhodes, J. E., & Lowe, S. (2009). Youth mentoring and adolescent development. In R. M. Lerner & L. Steinberg (Eds.), Handbook of adolescent psychology (3rd ed.; pp. 152–189). New York: Plenum Press. Rhodes, J. E., Reddy, R., & Grossman, J. B. (2005). The protective influence of mentoring on adolescents’ substance use: Direct and indirect pathways. Applied Developmental Science, 9, 31– 47. doi:10.1207/ s1532480xads0901_4 Rhodes, J. E., Spencer, R., Keller, T. E., Liang, B., & Noam, G. (2006). A model for the influence of mentoring relationships on youth development. Journal of Community Psychology, 34, 691–707. doi:10.1002/jcop .20124 Ryff, C. D. (1989). Happiness is everything, or is it? Explorations on the meaning of psychological well-being. Journal of Personality and Social Psychology, 57, 1069 –1081. doi:10.1037/0022-3514.57.6.1069 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, 468 – 482. doi:10.1002/jcop.20250 Sánchez, B., Reyes, O., & Singh, J. (2006). A qualitative examination of the relationships that serve a mentoring function for Mexican American older adolescents. Cultural Diversity and Ethnic Minority Psychology, 12, 615– 631. doi:10.1037/1099-9809.12.4.615 Skinner, E. A., & Belmont, M. J. (1993). Motivation in the classroom: Reciprocal effects of teacher behavior and student engagement across the school year. Journal of Educational Psychology, 85, 571–581. doi: 10.1037/0022-0663.85.4.571

NATURAL MENTORS & ACADEMIC ENGAGEMENT Smalls, C., White, R., Chavous, T., & Sellers, R. (2007). Racial ideological beliefs and racial discrimination experiences as predictors of academic engagement among African American adolescents. Journal of Black Psychology, 33, 299 –330. doi:10.1177/0095798407302541 Spencer, R. (2006a). Understanding the mentoring process between adolescents and adults. Youth & Society, 37, 287–315. doi:10.1177/ 0743558405278263 Stack, C. (1974). All our kin: Strategies for survival in the Black community. New York, NY: Harper & Row. Stewart, P. (2007). Who is kin? Family definition and African American families. Journal of Human Behavior in the Social Environment, 15, 163–181. doi:10.1300/J137v15n02_10 Thomson, N. R., & Zand, D. H. (2010). Mentees’ perceptions of their interpersonal relationships: The role of the mentor-youth bond. Youth & Society, 41, 434 – 445. doi:10.1177/0044118X09334806

85

Udry, J. R. (2003). The National Longitudinal Study of Adolescent Health (Add Health). Wave III, 2001–2002 http://www.cpc.unc.edu/projects/ addhealth/codebooks/wave3 Wellborn, J. G. (1991). Engaged and disaffected action: The conceptualization and measurement of motivation in the academic domain. Unpublished doctoral dissertation, University of Rochester, Rochester, NY. Zimmerman, M. A., Bingenheimer, J. B., & Behrendt, D. E. (2005). Natural mentoring relationships. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 143–157). Thousand Oaks, CA: Sage. doi:10.4135/9781412976664.n10 Zimmerman, M. A., Bingenheimer, J. B., & Notaro, P. C. (2002). Natural mentors and adolescent resiliency: A study with urban youth. American Journal of Community Psychology, 30, 221–243. doi:10.1023/A: 1014632911622