Exploring Racial-Ethnic Pride & Perceived Barriers in PYD: A Latent Profile Analysis

Abstract

This study sought to identify profiles of positive youth development (PYD) integrating racial-ethnic factors, specifically racial-ethnic pride and perceived racial-ethnic barriers in a sample of African American (77%) and Latino (23%) children (N = 234, Mean age = 8). Using a latent profile analysis, we found three profiles: The High PYD, Proud & Optimistic (High PYD, racial-ethnic pride, and low perceived racial-ethnic barriers), the High PYD, Proud & Aware (high PYD, pride, and perceived barriers), and the Low PYD and Disconnected (low PYD, pride and high barriers). The Optimistic profile exhibited fewer overall adjustment problems and higher standardized achievement at Time 2 than both the Aware and the Disconnected profiles. The Aware and the Disconnected showed similar adjustments. This study highlights the critical role of helping youth to feel competent, caring, connected, and proud, further supporting the role of socio-cultural factors in the PYD of African American and Latino children.

Introduction

I have a dream that my four little children will one day live in a nation where they will not be judged by the color of their skin, but by the content of their character. Martin Luther King (theme for SRCD Special Themed Meeting on Character, 2018, Richard Lerner and Deborah Vandell, conveners).

The population of the United States has long been racially and ethnically diverse, comprised of immigrants from Europe, Asia, Africa, and Latin America, along with indigenous peoples. Recent demographic changes expand this diversity in that between the years of 2011 and 2012, more than half of the children born in the United States were African American, American Indian/Native American, Asian-Pacific Islander or Latino (United States Census Bureau, 2012). This ethnic diversification of the U.S. adds value to its social, economic and cultural vitality (Ottaviano & Peri, 2012; Sparber, 2010). Despite these contributions, African American and Latino youth in the U.S. are continuously challenged by poverty and systematic discrimination in employment, health and education (Berends, Lucas, & Penaloza, 2008; Darby & Rury, 2018; Darity, 2003; Darity & Hamilton, 2018; Giersch, Bottia, Mickelson, & Stearns, 2016; Gregory, Skiba, & Noguera, 2010; Pachter, Bernstein, Szalacha, & García-Coll, 2010; Potter & Morris, 2017; Skiba et al., 2011; Umaña-Taylor, 2016). In the face of these challenges, a disproportionate amount of research attends to risks and deficits, with far less focus on positive youth development (PYD) and pathways of desirable youth adjustment (Cabrera & the SRCD Ethnic-Racial Issues Committee, 2013; Leman, Smith, Petersen, & SRCD Ethnic–Racial Issues and International Committees, 2017; McLoyd, 2006; Smith, Osgood, & Witherspoon, 2017).

Despite this hardship, many racial-ethnic minority youth thrive, with recent trends in increased rates of high school graduation, college matriculation, economic development and wealth among minority youth; indicative of substantial progress in decreasing historical disparities between minority and White youth (Cabrera & the SRCD Ethnic-Racial Issues Committee, 2013; McLoyd, 2006; Leman et al., 2017; Spencer, Harpalani, & Dell’Angelo, 2002; Spencer, Fegley, & Harpalani, 2003). However, more research is needed that uses PYD perspectives to empirically chart the developmental protective factors that support resilience – that is, desirable adjustment in the face of adversity, which is another important concept in the PYD literature (Masten, 2014; Oshri, Topple, & Carlson, 2017). In particular, the process of resilience among minority youth would benefit from an examination of socio-cultural factors that significantly influence youth adjustment (Spencer, 2006; Spencer & Spencer, 2014, Lerner et al., 2017). Therefore, this article attends to the factors that contribute to desirable youth adjustment. We posit that for African American and Latino youth, socio-cultural factors, specifically, racial-ethnic pride and perceived racial-ethnic barriers, and PYD all play a role in shaping adjustment.

Positive Youth Development and Youth Adjustment

The positive youth development (PYD) framework was developed to advance the study of strengths in youth and their attendant ability to thrive (Damon, 2004; Larson, 2000; Lerner et al., 2005). Over the past two decades, research has been focused on pathways of PYD and the developmental assets that foster PYD (Hamilton, 1999; Larson, 2000; Lerner et al., 2005, 2016, 2017; Scales et al., 2001; Smith et al., 2017). Lerner et al. (2005, 2016, 2017) propose that PYD can be conceptualized through the “5 C’s of PYD”, including competence, confidence, caring, connection, and character. Lerner and colleagues describe relational development systems (RDS, Lerner et al., 2016, 2017; Overton, 2015) metatheory as recognizing the mutually beneficial exchanges between the young people and their ecological context that evolve in systematic or idiosyncratic ways over time. This description is supported by empirical findings from the 4-H Study of PYD, a longitudinal study that links ecological contexts to PYD by investigating how exchanges between youth and their community are associated with their developmental adjustment (Lerner et al., 2005, 2016, 2017). Findings in this study have shown that developmental assets in youth’s developmental contexts, such as parental involvement and family support, as well as strengths within youth, such as self-regulatory ability, can foster PYD (Lewin-Bizan, Bowers, & Lerner, 2010; Schmid & Lopez, 2011; Theokas & Lerner, 2006; Urban, Lewin-Bizan, & Lerner, 2010; Zaff, Boyd, Li, Lerner, & Lerner, 2011). The study also has found that higher PYD is related to youth’s contribution to their developmental contexts, such as school and family, and fewer adjustment problems, such as risky behaviors, anxiety and depressive symptoms (Holsen, Geldhof, Larsen, & Aardal, 2017; Jelicic, Bobek, Phelps, Lerner, & Lerner, 2007).

Extant literature shows that PYD mitigates the effects of risk on youth, but most of this body of research is predominantly based on European American samples in the 4-H study (Lerner et al., 2017). A paucity of research on PYD was done with racial and ethnic minority youth; nonetheless, emerging research suggests that PYD is relevant across ethnic groups. Studies done by other teams among youth from diverse racial-ethnic backgrounds have demonstrated that PYD is related to fewer risky sexual behaviors, externalizing and internalizing problems, and higher self-esteem (Holsen et al., 2017; Murry et al., 2014; Smith et al., 2013; Sun & Shek, 2013; Williams, Anderson, Francois, Hussain, & Tolan, 2014; Witherspoon, Schotland, Way, & Hughes, 2009).

Research suggests that African American and Latino youth often need to cope with unique interpersonal and structural challenges to thrive in their settings. Thus, there may be additional racial-ethnic and cultural dimensions salient to PYD (Spencer, 2006; Spencer & Spencer, 2014; Lerner et al., 2017). Spencer et al.’s Phenomenological Variant of Ecological System Theory (PVEST, 1990, 1995, 1997, 2002) points out that for African American and Latino youth, structural challenges, such as racism, are significant risk factors that contribute to their everyday stress. However, youth who achieve a positive racial-ethnic identity are able to use more effective coping like self-appraisal and social support, which, in turn, is linked to adaptive adjustment (Spencer et al., 2002; Spencer et al., 2003; Rivas-Drake et al., 2014a, 2014b).

Recent research supports these premises. Smith, Witherspoon, & Osgood (2017) found that quality settings where the adults were observed to be supportive, and personally engaged with children led to more PYD (competence, caring, and connection) and increased cultural values of respect for adults, particularly from African American children. These findings counter-indicate critical and negative approaches presumed to prepare youth for the often dangerous contexts of urban settings lending support to the axiom, “you have to give respect to get respect.” In another study identifying socio-cultural factors of PYD, Williams et al. (2014) found that PYD in urban African American and Latino adolescents could be seen as a bi-factorial model, including both racial-ethnic identity and a more general PYD component. PYD measures were linked to more pro-social activities and fewer criminal and externalizing behaviors, and an achieved racial-ethnic identity was linked to fewer internalizing problems. PYD and racial-ethnic factors are also connected in the way that racial-ethnic socialization is linked to the development of PYD (Evans et al., 2012). Parents may embed the promotion of PYD characteristics in their racial-ethnic socialization practices and foster PYD and racial-ethnic identity development at the same time (Hughes et al., 2006). In the following section, we further describe research exploring the role of racial-ethnic pride and perceived racial-ethnic barriers in positive youth development and adjustment.

Racial-Ethnic Pride and Youth Adjustment

In this article, the term, racial-ethnic identity, is used to represent related but distinct concepts. Racial identity refers to perceptions developed in response to presumed phenotypical characteristics and race-based social stratification. Ethnic identity refers to individuals’ feelings, involvement, and belongingness with their cultural background (Cokley, 2007). Aspects of ethnicity are informed by some shared but widely varied physiology, just as aspects of race are more likely to be more socially defined given the wide within-group variations (Murry, Smith, & Hill, 2001). Although racial and ethnic identities are distinct, these two concepts are interrelated and often difficult to separate, regarding both measurement and individuals’ lived experiences (Quintana, 2007; Rivas-Drake et al., 2014a, 2014b), so the hybrid term racial-ethnic identity is used in this study.

Identity development is one of the most salient tasks in adolescence (Erikson, 1968) though it is likely to begin even earlier in childhood, with racial-ethnic identity in 3^rd grade being related to children’s self-perceptions and academic performance (Aboud, 1988; Quintana, 2008; Quintana & Vera, 1999; Smith, Levine, Smith, Dumas, & Prinz, 2009). Attaining a positive racial-ethnic identity is an especially complex developmental task for racial-ethnic minority youth, given broader attitudes and perceptions of their over-arching racial-ethnic groups that may be negative, stereotypical, and/or disparaging (Spencer et al., 2002). A positive racial-ethnic identity is a unique strength in African American and Latino youth’s development, which has been found in previous research to be related to high self-esteem, academic achievement, and desirable adjustment (Rivas-Drake et al., 2014a, 2014b; C. Smith et al., 2009; Witherspoon, Daniels, Mason, & Smith, 2016).

A positive racial-ethnic identity refers overall to the degree to which youth feel good, happy, and proud about their ethnicity or race (Phinney & Ong, 2007; Rivas-Drake et al., 2014a, 2014b; Sellers, Rowley, Chavous, Shelton, & Smith, 1997; Umaña-Taylor, Yazedjian, & Bámaca-Gómez, 2004). A positive racial-ethnic identity has been repeatedly found to be associated with African American and Latino youth adjustment, and racial-ethnic pride is often used as an indicator of a positive racial-ethnic identity. Multiple studies with African American and Latino youth have linked higher racial-ethnic pride to positive youth outcomes, including more positive self-concept (Booth, Abercrombie, & Frey, 2017; Hughes et al., 2009; Thomas, Townsend, & Belgrave, 2003; C. Smith et al., 2009), better socio-emotional competence (Grills et al., 2016; Jagers, Sydnor, Mouttapa, & Flay, 2007), and social connectedness (Rew, Arheart, Johnson, & Spoden, 2015; Thomas et al., 2003). Similarly, racial-ethnic pride is linked to fewer negative outcomes, such as fewer problem behaviors (Davis, Smith-Bynum, Saleem, Francois, & Lambert, 2017; Rivas-Drake et al., 2014b; C. Smith et al., 2009), delinquent behaviors (e.g., substance use; Grindal, & Nieri, 2016), and depressive symptoms (Neblett, Banks, Cooper, & Smalls-Glover, 2013; Rivas-Drake et al., 2014b; Seaton, & Carter, 2017). Moreover, recent studies also find that positive racial-ethnic identity can buffer the negative impact from racial-ethnic discrimination and peer pressure, and can lead to desirable adjustment (Butler-Barnes et al., 2017; Derlan, & Umaña-Taylor, 2015; Douglass, & Umaña‐Taylor, 2017; Grindal, & Nieri, 2016).

Though racial-ethnic pride has exhibited positive associations with several socio-emotional and behavioral outcomes, the association between racial-ethnic pride and academic achievement evidences mixed results. Some studies find that higher racial-ethnic pride is related to better academic motivation and performance (Butler-Barnes et al., 2017; Butler-Barnes, Williams, & Chavous, 2012; Hurd, Sanchez, Zimmerman, & Caldwell, 2012). However, the role of racial-ethnic pride in academic achievement becomes complicated in different contexts. Byrd and Chavous (2009) found that among adolescents, higher pride was linked to better academic achievement in neighborhoods low in economic opportunity, but the association was the opposite in more economically advantaged neighborhood. This finding suggests that racial-ethnic pride can serve as a significant resource that helps African American and Latino adolescents, particularly when they are living with more contextual risks, such as lower socio-economic opportunities.

Despite abundant studies linking racial-ethnic pride to adjustment, more literature on racial-ethnic pride and youth outcomes is focused on adolescence with far less attention on middle childhood (e.g., Butler‐Barnes et al., 2017; Derlan, & Umaña-Taylor, 2015; Douglass, & Umaña‐Taylor, 2017; Grindal, & Nieri, 2016; Rew et al. 2015). However, the process of identity development is suggested to begin earlier in childhood (Quintana, 2007; Quintana, & Segura-Herrera, 2003). Swanson and colleagues (2009) claimed that the foundations of identity development (i.e., cognitive and social competence) often emerge during middle childhood, thus leading to identity formation during adolescence. Furthermore, Umaña-Taylor and colleagues (2004) noted that during early to middle childhood, racial-ethnic labeling, knowledge, and constancy emerge (i.e., ethnic identification). Moreover, by third grade or fourth grade, a positive racial-ethnic identity was found to be associated with higher academic self-efficacy, better cognitive and academic performance, and fewer problem behaviors (Smith, Atkins, & Connell, 2003; C. Smith et al., 2009; Witherspoon, Daniels, Mason, & Smith, 2016). These studies suggest the importance of investigating the role of racial-ethnic pride in middle childhood.

Perceived Racial-Ethnic Barriers and Youth Adjustment

In addition to pride, perceived racial-ethnic barriers may be another significant component in African American and Latino children’s development. Perceived racial-ethnic barriers are defined as beliefs that one is more likely to be challenged because of his/her racial-ethnic background. It is potentially a unique element for African American and Latino youth who are challenged disproportionately by discrimination in both institutional settings and everyday life (Darity, 2003; Darity & Hamilton, 2018; Pachter et al., 2010; Skiba et al., 2011; Umaña-Taylor, 2016). Studies find that racial-ethnic discrimination experience is linked to less academic engagement and achievement (Alliman-Brissett, & Turner, 2010; O’Hara, Gibbons, Weng, Gerrard, & Simons, 2012) and more psychological distress, including more self-reported internalizing symptoms, lower self-concept, more depressive symptoms, more suicidal thoughts and higher levels of hopelessness (Ahmed, Kia-Keating, & Tsai, 2011; Nyborg & Curry, 2003; Saleem, & Lambert, 2016; Seaton, & Carter, 2017; Sellers, Copeland-Linder, Martin, & Lewis, 2006; Walker et al., 2017). However, perceived racial-ethnic barriers are not identical to discrimination experiences, as perceived racial-ethnic barriers involve perceptions of the structural challenges independent from personal discrimination experiences.

Perceived barriers have been linked to both desirable and undesirable adjustment (Hughes et al., 2006). A realistic understanding of potential structural barriers is often viewed as essential for African American and Latino youth adjustment. Denial of racism in the face of discrimination is found to be related to more depressive symptoms for African American populations (Kwate & Goodman, 2015; Seaton, Upton, Gilbert, & Volpe, 2014). Studies also point out that promoting awareness of structural barriers can prepare African American and Latino youth to be ready to deal with bias and racial-ethnic barriers (Cooper & McLoyd, 2011; Hughes et al., 2006; Neblett, Rivas-Drake, & Umaña-Taylor, 2012; Saleem,& Lambert, 2016). Conveying messages to promote awareness for racial-ethnic barriers together with messages promoting racial-ethnic pride is linked to fewer behavioral problems and higher self-esteem (Murry et al., 2005; Stevenson et al., 2002). In sum, perceived barriers can be a sign of readiness to deal with the potential discrimination that racial-ethnic minority youth might encounter in society, yet some findings exhibit deleterious effects of perceived racial barriers. It is a delicate task to prepare youth, particularly elementary-age children, for bias while also fostering a positive sense of identity and adjustment.

Overemphasis on the challenges of racial-ethnic barriers and mistrust of others may result in pessimism in the prospects for opportunity and achievement (Hughes et al., 2006; Harris-Britt et al., 2007). Ogbu (1991, 2003) suggests that when minority youth perceive barriers and negative stereotypes from the dominant culture, some of them develop cultural mistrust and cope with this perception with resistance or opposition to the success defined in the dominant culture. As a result, more desirable developmental outcomes are diminished. Perceived racial-ethnic barriers are associated with worse academic achievement and more depressive symptoms in African American youth (Rumbaut, 1994; Smith et al., 2003). Moreover, mistrust of others is positively related to deviant behaviors and inversely related to academic engagement for African American and Latino adolescents (Biafora, Taylor, Warheit, Zimmerman, & Vega, 1993; Cooper, & Sanchez, 2016; Irving, & Hudley, 2008).

There are limitations in the current literature on perceived racial-ethnic barriers and youth adjustment. Similar to the paucity of research on racial-ethnic pride, there is a lack of research on perceived barriers conducted with school-aged children. The majority of the research on perceived barriers has been conducted with adolescents (e.g., Ahmed et al., 2011; Cooper, & Sanchez, 2016; Saleem, & Lambert, 2016; Walker et al., 2017). However, as early as age 6, African American children showed understanding of racial-ethnic discrimination and how it can impact people’s occupation (Bigler, Averhart, & Liben, 2003). Thus, researchers should attend to perceived barriers in middle childhood. Moreover, an abundance of studies focus on the message of racial-ethnic barriers conveyed to youth (for a review, see Hughes et al., 2006), but less attention has been paid to youth perceptions of racial-ethnic barriers. However, given the transactional process of racial-ethnic socialization between adults and children, perceptions of barriers can differ from the messages they received, potentially impacted by other components in their ecological system (Spencer et al., 2003). Therefore, there is a need to directly measure school-aged children’s perception of racial-ethnic barriers in studies.

Phenomenological Variant of Ecological System Theory (PVEST) and Person-Centered Approaches

Based on the review above, we consider multiple aspects of PYD, racial-ethnic pride, and perceived barriers to be all associated with youth adjustment. Understanding the complex processes of PYD, racial-ethnic identity, perceived racial-ethnic barriers, and the mixed research findings, positing a unidirectional association between these variables that fit all racial-ethnic minority youth might not be well-informed. Multifinality suggests that any one factor may function differently depending on the context in which it operates (Cicchetti, & Rogosch, 1996). Spencer et al.’s Phenomenological Variant of Ecological System Theory (PVEST, 1990, 1995, 1997, 2002) points out that it is important to recognize African American and Latino youth’s perception of different experiences, including the challenges they may face, their strengths and coping resources in their contexts (such as a positive racial-ethnic identity, supportive adults, neighborhood contexts, and social support networks), as a whole instead of isolating different aspects of individuals’ experiences in order to understand the overall patterns of developmental adjustment. Therefore, it is important to examine the role of PYD in conjunction with racial-ethnic pride and perceived barriers. In order to explore the multiple patterns of PYD, pride, and barriers, a person-centered approach is potentially enlightening. A person-centered approach can help classify potential subgroups of youth based on their overall personal and contextual experiences. This approach also allows us to see the within-group variation among African American and Latino youth. Certain demographic factors may contribute to the within-group variation.

The Roles of Gender and Age

Some studies find different patterns of racial-ethnic identity between boys and girls (Fhagen-Smith, Vandiver, Worrell, & Cross, 2010; Lott, 2011), whereas others find that boys and girls show similar levels of centrality, affirmation and belonging (Chavous, Rivas-Drake, Smalls, Griffin, & Cogburn, 2008; Cokley, 2001; Pahl & Way, 2006). In regard to potential developmental phases, sense of racial-ethnic identity increases with age (Rew et al., 2015; C. Smith et al., 2009). As youth enter adolescence, they report greater levels of barriers and discrimination, which may be attributed to increased salience of racial-ethnic identity and growing capacity to recognize racial-ethnic barriers (Hughes, Del Toro, Harding, Way, & Rarick, 2016). Moreover, girls and older children might demonstrate more PYD characteristics, such as pro-social behavior, than boys and younger children (Furrow, King, & White, 2004; Li, Lynch, Kalvin, Liu, & Lerner, 2011; Van der Graaff, Carlo, Crocetti, Koot, & Branje, 2018), and PYD is more likely to be linked to adjustment for girls than for boys (Årdal, Holsen, Diseth, & Larsen, 2018). In sum, gender and age can be potentially linked to the typology of different patterns of PYD, racial-ethnic pride and perceived racial-ethnic barriers, questions we explored in the current research.

Method

PYD is linked to youth adjustment, and for African American and Latino youth, racial-ethnic pride and perceived racial-ethnic barriers are also significant components in their cultural-ecological, developmental contexts (Lerner et al., 2017; Ogbu, 1981; Spencer et al., 2002, 2003). The present study aims at empirically identifying latent patterns among African American and Latino children’s set of experiences and perceptions, including PYD, racial-ethnic pride, and perceived racial-ethnic barriers, and comparing adjustments among the subgroups. In addition, the present study investigates whether gender and age add to the prediction of different subgroup memberships.

Participants and Procedure

Data were collected from 76 community-based afterschool programs in local school districts of a mid-Atlantic state across urban, suburban, and rural areas. Afterschool programs serving public elementary school children in grades kindergarten to fifth were identified via multiple approaches: (1) contacting and searching the websites of local school districts for their afterschool care providers; and (2) systematically searching for local community-based agencies such as the YM/YWCA, Boys and Girls Club, and local Parks and Recreation Commissions that provided afterschool programming. Agencies offering programs for this age group most days of the week throughout the academic school year were included, and no programs meeting these criteria were excluded. The team successfully recruited 12 of 14 (86%) program providers who operated a range of 2–12 program sites. At the site level, we retained 76 of 83 program sites recruited (92%). Program sites were matched on geographic locale, racial-ethnic composition, and socio-economic status and randomly assigned to intervention and control groups. Two sites whose match dropped were excluded from the analyses. Furthermore, one site that failed to collect sufficient child data across time was not included, and its matching site was excluded as well resulting in 72 total sites (Smith, Osgood, & Witherspoon, 2018). The number of children served by each program site ranged from 15-63 with an average of 40 children. The programs varied in the proportion of free/reduced lunch eligible children, ranging from 17-98% with a mean of 45 percent.

Consent forms were sent to the parents of participating children in grades 2-5 to which they could refuse participation; children provided assent before completing the survey. Participation rates for the child survey varied across the sites ranging from 72 to 90 percent (kindergarten and first-grade children were not included due to lower literacy levels at these stages of development). Of the total 800 participating children, 48% were White/European-American, 28.7% African American, 6.7% Latino/a, and 16.4% mixed race or other racial-ethnic groups. The participants ranged from 7-11 years old when they first attended this study.

In this study, we focused on the children of color, 306 children, who identified as African American or Latino. We then excluded 72 children (23.5%) who had missing data on the five variables of focus for the current study. The final sample consisted of 234 African American and Latino children. The demographic information of the sample is shown in Table 1. The children who were excluded did not differ from the children who were included in age, gender, race, and other study variables.

Table 1.

Demographic Information for Study Participants

	African American (N= 181, 77.4%) N(%) / M(SD)	Latino (N=53, 22.6%) N(%) / M(SD)	F/χ2
Age	8.42 (1.10)	8.66 (1.24)	3.21
Gender			3.06
Boy	93 (51.4%)	20 (37.7%)
Girl	88 (48.6%)	33 (62.3%)
Grade			9.34
1	3 (1.7%)	0 (0%)
2	61 (33.7%)	16 (30.2%)
3	53 (29.3%)	12 (22.6%)
4	45 (24.9%)	11 (20.8%)
5	19 (10.5%)	14 (26.4)

^*p < .05,

^**p < .01. None of the above differences were statistically significant at the .01-.05 levels.

The data at Time 1 were collected in the fall semester of the academic school year, and data at Time 2 was collected in the following spring semester. Children completed the surveys during the afterschool program using personal digital assistants that they read on their own or with assistance from research staff. The survey ranged in time from 45-60 minutes with short cartoon and joke breaks programmed between 15-minute sections on the personal digital assistants. Given the study was focused on using group-based approaches to behavioral co-regulation , and had to be limited to 30-45 minutes in order to maintain children’s attention and maximum participation, the survey included 3 measures of PYD (i.e., competence, caring & connection) highly relevant to the overall study goals, racial-ethnic pride, perceived racial-ethnic barriers, and adjustment problems. Math and reading proficiency was assessed by obtaining standardized test scores on the Pennsylvania System of School Assessment (PSSA) in the spring semester offered to students from grade 3 to 8; children from Grades 1 and 2 did not have PSSA scores.

Measures

Positive Youth Development (PYD).

Three dimensions of PYD, competence, caring, and connection, were measured at Time 1. Given we were interested in using group-based approaches to self- and co-regulation, we were particularly interested in behavioral competence. Competence was measured using a 12-item subscale of collective efficacy focused on measuring behavioral influence upon others (Sampson, Raudenbush, & Earls, 1997; Smith et al., 2013). This subscale examined how competent children felt collectively in the afterschool program as a whole, encouraging positive behaviors and discouraging negative behaviors among their peers (e.g., if children in this program are misbehaving, other children remind them to act their best). Children responded on a 3-point Likert scale, ranging from 1 (not true) to 3 (very true). The score was calculated as the mean of all the items. Ranging from 1 to 3, higher scores indicated higher levels of collective efficacy. Internal consistency was good overall and across African American and Latino youth (Cronbach’s alpha = .89). Our group-based approaches were designed to foster a sense of care and encouragement of positive behavior among the children. Caring was measured by the prosocial behaviors subscale of the Strengths and Difficulties Questionnaire (SDQ; Goodman, Meltzer, & Bailey, 2003; Mellor, 2004). Children responded on a 3-point Likert scale, ranging from 1 (not true) to 3 (very true), to 6 items (e.g., I am helpful if someone is hurt, upset, or feeling ill). The score was calculated as the mean of all the items. Ranging from 1 to 3, higher scores indicated more use of prosocial behaviors. The internal consistency was good overall and across African American and Latino youth (Cronbach’s alpha = .77). Furthermore, we thought that working together as a group to praise and encourage each other and behave well would foster a sense of connection, not only among children but staff as well. Connection was assessed using the 8-item Afterschool Connectedness scale adapted from measures of school connectedness (Resnick et al., 1997) to assess the degree to which children felt they were a part of their afterschool program group (e.g., I feel that my afterschool program staff cares about me). Children responded on a 3-point Likert scale ranging from 1 (not true) to 3 (very true), and the score was calculated as the mean of all the items. Ranging from 1 to 3, higher scores indicated a greater attachment and belongingness to the afterschool program. The internal consistency was good overall and across African American and Latino youth (Cronbach’s alpha = .78). Confidence and character are the other “2Cs” in Lerner et al.’s (2005, 2016, 2017) model of PYD. While one might hypothesize that confidence and character would also be conceptually relevant, behavioral competence, caring, and connection were included as original aims in the broader study, beyond which we were limited by time and measurement constraints with our sample.

Racial-ethnic pride & perceived racial-ethnic barriers.

Children’s racial-ethnic pride and perceived racial-ethnic barriers were assessed using a measure for young school-aged children (Smith et al., 2003). There were three items to measure racial-ethnic pride (e.g., you like being the color you are) and two items for racial-ethnic barriers (e.g., No matter how hard a person tries, they might not make it in life because of color). Participants responded to statements using a 3-point scale ranging from 1 (Not True) to 3 (Very True). The scores were the mean of three items for racial-ethnic pride and the mean of two items for perceived racial-ethnic barriers. Racial-ethnic pride’s internal consistency for African American youth was acceptable (Cronbach’s alpha = .75), but low for Latino youth (Cronbach’s alpha = .46). However, the internal consistency of the 2-item measure of perceived racial-ethnic barriers was adequate and similar across groups (range of .66 to .73). Overall internal consistencies were acceptable for racial-ethnic pride and 2-item barriers at Time 1 (range of .68 to .70).

Adjustment problems.

Children’s adjustment problems were assessed using child reports of the Strengths and Difficulties Questionnaire (SDQ) at both Time 1 and 2 (Goodman, Meltzer, & Bailey, 2003; Mellor, 2004). The SDQ in this study was comprised of 16 items to which participants responded on a 3-point scale indicating the degree to which each item was “not true, sometimes true, or very true.” Several specific externalizing and internalizing problems were also measured using SDQ subscales, including hyperactivity (5 items), emotional symptoms (6 items) and conduct problems (4 items). The SDQ items were used to calculate a mean score of all the items on overall problem behaviors. Ranging from 1 to 3, higher scores indicated more adjustment problems. The internal consistency was .86 at both time points. The internal consistency for African American and Latino youth was acceptable (Reliabilities for all the measures are shown in Table 2).

Table 2.

Internal Consistency of Measures for African American and Latino Children

	Overall	African American	Latino
Pride	.70	.75	.46
Barriers	.68	.66	.73
Competence (collective efficacy)	.89	.89	.89
Caring (pro-social)	.77	.77	.77
Connection (afterschool connectedness)	.78	.78	.73
Problem behavior (Time 1)	.86	.86	.86
Problem behavior (Time 2)	.87	.86	.91
Hyperactivity (Time 1)	.80	.80	.87
Hyperactivity (Time 2)	.81	.78	.82
Emotional symptoms (Time 1)	.78	.79	.74
Emotional symptoms (Time 1)	.74	.74	.74
Conduct problems (Time 1)	.69	.69	.72
Conduct problems (Time 2)	.74	.71	.81

Math and reading proficiency.

Academic achievement was measured by the annual Pennsylvania System School Assessment (PSSA), which is a standards-based, criterion-referenced assessment. Every Pennsylvania student in grades 3 through 8 was assessed in English Language Arts and Math. Math and reading achievement was assessed by PSSA at Time 2.

Analytic Plan

Due to the small size of Latino participants (N = 53), we decided to combine African American and Latino youth in statistical analyses. Previous research demonstrates that both racial-ethnic identity and perceived barriers are salient for these groups (Rivas-Drake et al., 2014b). In this sample, African American and Latino children did not have different distributions of grade or gender, and their age was not significantly different across the groups. Internal consistency reliabilities were good across groups except for the measure of racial-ethnic pride, which was more focused on color and low for the Latino group. Moreover, t-tests showed that none of the study variables varied significantly across groups. Bivariate correlation between variables demonstrated similar patterns for African American and Latino youth. Given that there were no statistically significant differences between the African American and Latino children regarding study variables in this sample, and the small sample of Latino youth, it is reasonable to combine the African American and Latino youth in the analyses. In case of potential racial-ethnic difference, we examined whether race-ethnicity was predictive of class memberships.

All analyses were conducted in Mplus 8.2 (Muthén & Muthén, 1998–2017). Full information maximal likelihood was used to deal with missing data at random and to utilize all the data available (Little & Rubin, 1987). Latent profile analysis (LPA; Lanza, Flaherty, & Collins, 2003) was used to obtain a typology of African American and Latino PYD attuned to their racial-ethnic pride and perceived racial-ethnic barriers. LPA simultaneously estimated youth profiles and individuals’ probability in each profile membership, and classified individuals into the most appropriate class based on maximal probability (Hill, Degnan, Calkins, & Keane, 2006). The determination of optimal class solution was based on conceptual considerations and a range of model fit criteria provided in the literature (Lubke & Muthén, 2005; Nylund, Asparouhov, & Muthén, 2007; Stanley, Kellermanns, & Zellweger, 2017). We started by testing a 2-class solution and hierarchically added one profile at a time through an iterative process until we tested a 6-class solution. Akaike’s Information Criterion (AIC; Akaike, 1974), Bayesian Information Criterion (BIC; Schwarz, 1978), Sample-Size adjusted BIC (SABIC; Sclove, 1987), Lo-Mendell-Rubin adjusted test (LMRT; Lo, Mendell, & Rubin, 2001), entropy (Celeux & Soromenho, 1996) and the smallest latent profile (Collins & Lanza, 2010) were used to evaluate model fit. Smaller AIC, BIC, and SABIC indicate better model fit. Significant LMRT indicates that the target model has a significantly better model fit than the model with one profile less. Entropy reflects the distinctiveness among profiles, and values approaching 1 indicate clear delineation of profiles.

A 3-step approach to modeling was utilized to compare covariates and distal outcomes between profiles and identify predictors for profile membership (Asparouhov & Muthén, 2014). This approach is recommended because profiles remain fixed when covariates are entered into the model, and it accounts for uncertainty associated with the profile solution (Asparouhov & Muthén, 2014; Vermunt, 2010). After we chose the profile solution, we compared adjustment problems, math and reading achievement across profiles and tested whether race-ethnicity, age, and gender predicted individual profile membership.

Results

Descriptive Results

The descriptive results of study measures and bivariate correlation matrix are shown in Table 2. Racial-ethnic pride was positively related to three PYD indicators, while perceived racial-ethnic barriers were not related to PYD indicators. Competence was not related to adjustment problems and academic achievement. Caring and connection were negatively associated with overall adjustment problems and all the specific problems at one or both time points, but not related to academic achievement. Pride was negatively associated with overall adjustment problems at Time 1, and emotional symptoms at both time points, whereas barriers were positively related to overall adjustment problems and all the specific problems at both time points. Child-reported pride was also positively related to standardized reading test scores, whereas child-reported perceived barriers were negatively related to reading scores. Children’s self-reported hyperactivity and conduct problems were inversely related to standardized reading test scores; hyperactivity, and emotional problems were inversely related to reading scores.

Latent Profile Analysis (LPA) of PYD, Pride, and Barriers

We examined models with one through six latent profiles and then compared model fit indices (see Table 3). The AIC, BIC, and SABIC dropped and model LMRT improved significantly when the classes increased from 1 to 2 and from 2 to 3, indicating that the 3-class solution fit the data better compared to the 2- and 1-class solutions. The AIC, BIC, and SABIC kept dropping, and model LMRT improved significantly when the classes increased from 3 to 4, and these two solutions had similar entropy. However, the smallest class in the 4-class solution only contained 5.6% of the sample (N = 13), whereas the smallest class in the 3-class solution contained 10.7 % of the sample (N = 25). Thus, the 3-class solution had a more conceptually meaningful smallest profile compared to the 4-class solution. The LMRT did not improve significantly for the 5-class solution or the 6-class solution indicating no significant change in model fit. Based on all the model fit indices as well as conceptual meaning, a 3-class solution was chosen.

Table 3.

Descriptive Statistics and Bivariate Correlation for Study Variables

		M(SD)	1	2	3	4	5	6	7	8	9	10	11	12	13	14
1	Competence (T1)	2.38 (.48)
2	Caring (T1)	2.45 (.50)	.25**
3	Connection (T1)	2.46 (.47)	.49**	.42**
4	Pride (T1)	2.72 (.49)	.28**	.26**	.36**
5	Barriers (T1)	1.66 (.75)	.05	.10	−.01	.06
6	Adjustment problems (T1)	1.55 (.44)	−.07	−.15*	−.24**	−.21**	.31**
7	Adjustment problems (T2)	1.57 (.45)	−.16	−.28**	−.27**	−.17	.26**	.58**
8	Hyperactivity (T1)	1.53 (.57)	−.11	−.15*	−.25**	−.13	.26**	.84**	.44**
9	Hyperactivity (T2)	1.64 (.59)	−.15	−.16	−.21*	−.12	.19*	.52**	.86**	.44**
10	Emotional symptoms (T1)	1.65 (.54)	.05	−.02	−.07	−.18**	.23**	.85**	.48**	.51**	.34**
11	Emotional symptoms (T2)	1.57 (.49)	−.11	−.21*	−.11	−.18*	.21*	.51**	.82**	.30**	.49**	.57**
12	Conduct problems (T1)	1.51 (.54)	−.05	−.13*	−.21**	−.14*	.28**	.81**	.55**	.57**	.48**	.53**	.42**
13	Conduct problems (T2)	1.52 (.56)	−.14	−.31**	−.24**	−.07	.27**	.41**	.83**	.31**	.63**	.28**	.52**	.45**
14	Math (T2)	1312.37 (172.37)	−.03	.11	.15	.22	−.20	−.28*	−.22	−.27*	−.30*	−.21	−.15	−.23*	−.08
15	Reading (T2)	1259.81 (172.99)	.01	.20	.06	.32**	−.25*	−.31**	−.37**	−.30**	−.40**	−.23*	−.32*	−.14	−.17	.69**

^*p < .05,

^**p < .01. Note. T1 = Time 1, T2 = Time 2.

The description of the class characteristics is shown in Table 4 and Figure 1. The majority (56.4%, N=132) of the children fell in a class featured by high competence, caring, connection and racial-ethnic pride, but low perceived barriers. This group of children subjectively did not appear to anticipate the experience of racial-ethnic barriers. Thus, we named this class the High PYD, Proud & Optimistic. The second class consisted of 77 (33.9%) children in this sample and was featured by high competence, caring, connection, racial-ethnic pride, and perceived barriers. This group of children appeared to be aware of the structural challenges that were related to their race-ethnicity and anticipated experiencing some racial-ethnic barriers. Thus, this class was named the High PYD, Proud & Aware. The smallest class contained 25 (10.6%) children and was featured by low competence, caring, connection, pride, and perceived barriers. This group of children did not report a strong sense of racial-ethnic pride in their group nor did they anticipate being challenged by racial-ethnic barriers. Race-ethnicity may not be a salient part of their life, and they appeared to be disconnected from their racial-ethnic group. Thus, this class was named the Low PYD & Disconnected.

Table 4.

Model fit indices for LCA

	Number of Classes
	1 Class	2 Classes	3 Classes	4 Classes	5 Classes	6 Classes
AIC	1713.06	1554.05	1492.47	1414.64	1386.48	1361.78
BIC	1747.61	1609.34	1568.49	1511.39	1503.96	1499.99
Adjusted BIC	1715.91	1558.62	1498.76	1422.65	1396.20	1373.21
LMRT	-	165.94*	71.40**	54.97*	38.97	−32.49
Entropy	-	.92	.81	.84	.80	.81
Smallest class (%)	-	11.1%	10.7%	5.6%	5.1%	5.1%

^*p < .05,

^**p < .01.

Figure 1.

Youth Adjustment Profiles

Note. The results were standardized to make the interpretation of the profiles easier.

Demographic Predictors for Class Memberships

In order to examine whether age and gender predicted LPA memberships, a 3-step approach was used. The result is shown in table 4. Children’s race-ethnicity, age or gender did not significantly predict their class memberships.

Comparison of Youth Adjustment among Latent Profiles

The chi-square test comparing adjustment across profiles showed that the three profiles had different levels of adjustment, with the High PYD, Proud & Optimistic outperforming the other two profiles at both Time 1 and Time 2 (see Table 5). At Time 1, the High PYD, Proud & Optimistic had less overall adjustment problems than both the Aware (χ² = 18.64, p < .01) and Disconnected (χ² = 12.06, p < .01) profiles. Regarding the specific problems, the High PYD, Proud & Optimistic profile exhibited less hyperactivity, emotional symptoms and conduct problems than both the Aware and the Disconnected profiles. The Aware and the Disconnected did not differ from each other in overall adjustment problems or specific problems.

Table 5.

Latent profile analysis classes descriptions

	High PYD, Proud & Optimistic		High PYD, Proud & Aware		Low PYD & Disconnected
	(n=132, 56.4%)		(n= 77, 33.9%)		(n=25, 10.6%)
	M	SE	M	SE	M	SE
Competence	2.42	.05	2.44	.06	2.03	.11
Caring	2.48	.05	2.52	.06	2.05	.18
Connection	2.53	.05	2.49	.06	2.00	.16
Pride	2.87	.03	2.84	.04	1.57	.15
Barriers	1.14	.06	2.55	.09	1.49	.11

At Time 2, the High PYD, Proud & Optimistic still had fewer overall adjustment problems than both the Aware (χ² = 16.23, p < .01) and the Disconnected (χ² = 12.10, p < .01). Regarding the specific problems, the Optimistic still had fewer conduct problems than both the Aware (χ² = 11.53, p < .01) and the Disconnected (χ² = 5.15, p < .05), but did not show differences in hyperactivity or emotional symptoms. Just like Time 1, the Aware and the Disconnected did not show differences in overall adjustment problems or specific problems.

In regard to academic achievement at Time 2, both the Optimistic and the Aware evidenced trends of higher scores in reading and math than the Disconnected. However, the Optimistic demonstrated statistically significantly higher math achievement than the Aware (χ² = 5.49, p < .05). We did not detect statistically significant differences between the Optimistic and the Disconnected or between the Aware and the Disconnected. The Optimistic also had higher levels of reading achievement than the Aware (χ² = 4.08, p < .05) and the Disconnected (χ² = 5.04, p < .05). However, there were no statistically significant differences between the Aware and the Disconnected.

Discussion

This study sought to identify African American and Latino youth profiles of PYD integrating models of racial-ethnic pride and perceived racial-ethnic barriers. Using PVEST as the theoretical framework that considers youth strengths and competencies in the face of challenges, a person-centered approach helped to capture a set of individual perceptions of experience. Previous studies have found that PYD is linked to better youth adjustment, but the concept of PYD might be better informed by including socio-cultural dimensions for African American and Latino youth (Lerner et al., 2017; Williams et al., 2014). Racial-ethnic pride has been found to be associated with better youth adjustment in most cases, but the role of pride is less clear for children living in economically advantaged neighborhood (Byrd & Chavous, 2009). Perceived racial-ethnic barriers have exhibited mixed findings, associated with both better and worse youth adjustment (Hughes et al., 2006; Quintana, 2007; Rivas-Drake et al., 2014a, 2014b; Smith et al., 2003: Stevenson et al., 2002).

Using a person-centered approach, we found three profiles of African American and Latino youth. The High PYD, Proud & Optimistic comprised the majority of the sample while the High PYD, Proud & Aware and the Low PYD & Disconnected were proportionally smaller profiles. The Optimistic outperformed the other two groups regarding adjustment problems and academic achievement. Interestingly, the Aware and the Disconnected did not demonstrate statistically significant differences in adjustment problems or academic achievement. However, the Aware tended to have higher achievement scores than the Disconnected, but the smaller sample sizes might be attenuating the findings.

These findings suggest some detrimental impact from perceived racial-ethnic barriers on adjustment even when these elementary-school African American and Latino children are equipped with PYD and pride in their racial-ethnic group. These findings were also consistent with variable-centered literature suggesting that perceived barriers is negatively related to desirable youth outcomes (French, Kim, & Pillado, 2006; Irving & Hudley, 2005; McMahon & Watts, 2002; Nyborg & Curry, 2010; Smith et al., 2003; Wei, Wang, Heppner, & Du, 2012). From these findings using a person-centered approach, we found that elementary school children who perceived racial-ethnic barriers evidenced lower achievement scores than children who are more optimistic about racial-ethnic barriers but, a trend of higher scores academically than children without the resources afforded by PYD, that is feeling competent, connected, and caring. Young people with both a positive sense of their racial-ethnic group, and PYD (e.g., competent, caring, and connected) evidence the more optimal patterns of adjustment and achievement.

The smallest profile was characterized by low PYD, low pride and low perceived barriers. The High PYD, Proud & Aware had a trend of higher means of both math and reading scores than the Low PYD & Disconnected. It is possible that we did not have enough statistical power to test the difference because of the small size of the Low PYD & Disconnected group. However, there is a rich body of research suggesting that disconnection and a lack of interpersonal resources do not bode well for child adjustment.

Both profiles who perceived high racial-ethnic barriers and who lacked PYD did not exhibit optimal adjustment. Neither having higher levels of PYD nor perceiving low levels of racial-ethnic barriers was enough for optimal behavioral outcomes and academic achievement. The different pathways to youth adjustment suggest that importance of comprehensive evaluation of different components in the youth of color’s ecological context when exploring the pathway to adjustment. Compared to the variable-centered literature regarding PYD, pride, and barriers, this person-centered study showed that it is essential to evaluate youth experience as a whole (Spencer et al., 2002, 2003). Moreover, this study contributed to the current literature on racial-ethnic identity and adjustment in school-aged children. As early as middle childhood, racial-ethnic pride and perceived racial-ethnic barriers can shape youth adjustment in conjunction with other developmental characteristics like PYD.

Gender, Age, and Race-Ethnicity as Predictors.

In this study, we found that gender and age did not predict profile membership. It is consistent with some studies that did not find gender differences in racial-ethnic identity development (Chavous et al., 2008; Pahl & Way, 2006). Even though age is related to a more comprehensive understanding and more significance of racial-ethnic identity (Aboud, 2005; Byrd, 2012; Hughes et al., 2016; Quintana, 2008; Umaña-Taylor et al., 2004), it was not related to profile membership in this study. The participants were mainly from Grade 2 to Grade 5 (see Table 1), and it was possible that they were at similar developmental stages with a lack of significant variation. Future study is needed to further examine whether age is related to different PYD and racial-ethnic patterns, across childhood and adolescence, to help inform families and communities regarding the timing and effects of socializing children about racial-ethnic barriers. We also did not find a link between race-ethnicity and specific profiles. This finding is not surprising as previous research demonstrates that both racial-ethnic identity and perceived barriers are salient for African American and Latino groups (Rivas-Drake et al., 2014b).

Strengths and Limitations

Unlike a variable-centered approach, the person-centered study enabled us to capture individuals’ experiences and empirically suggested subgroups. This study used an ample sample size of 234 participants, which provided sufficient statistical power for the LPA. In addition, we used SDQ, which was an adequate measure for different aspects of youth adjustment, including both problem behaviors and internalized problems (e.g., emotional symptoms). The strong relations between child-reported PYD, racial-ethnic identity and barriers demonstrate the internal validity of this child survey data to their self-reported adjustment, and between children’s self-report of their PYD, identity, and adjustment to standardized test scores conducted by their state schools. Furthermore, this research included two time points with some of the relations being even stronger across time. Yet, future research using more longitudinal data could help identify critical time points at which identity and indicators of positive development might emerge, pointing to critical periods for parental and school practitioners’ socialization approaches regarding both positive identity and preparing children for potential barriers, a source of current concern for parents and youth in the U.S. where race and ethnicity still matter in such a young age (Smith et al., 2009). Another limitation is the low reliability of the racial-ethnic pride measure with Latino youth. The low reliability may be due to the focus on “the color of your skin” in the items, which may not reflect the variation of race and skin color within the Latino population. Future research is needed to develop a more inclusive measure of racial-ethnic pride for both African American and Latino youth. Last but not least, driven by the primary study aims and the limited time to survey youth in afterschool programs, we only utilized “3 Cs” out of “5 Cs”. We did not have data on confidence and character, though some might say caring is an aspect of character. Future studies could include all 5 “Cs” and possibly even the “6^th C”-contribution to have a more comprehensive evaluation of PYD and sociocultural dimensions of youth adjustment and achievement.

Implications and Conclusions

This study furthered burgeoning research exploring socio-cultural elements of positive youth development among African American and Latino school-aged children (Cabrera & the SRCD Ethnic-Racial Issues Committee, 2013; Smith et al., 2018; Williams et al., 2014). The results point to the importance of taking into account multiple aspects of PYD, that is, competence, connection, and competence integrated with the dimensions of racial-ethnic pride and perceived racial-ethnic barriers into consideration when exploring the pathway to desirable adjustment. For children who may be overcoming structural challenges due to their race-ethnicity, we did not find that perceived racial-ethnic barriers helped them cope with the challenges. On the other hand, racial-ethnic pride, in concert with PYD fosters more optimal adjustment. Moreover, the findings of the significance of racial-ethnic pride and perceived racial-ethnic barriers suggest that racial-ethnic factors may start contributing to youth adjustment as early as elementary school. It is also possible that for older adolescent populations, these profiles may vary and the perceptions of racial barriers might evidence different results. Taken together, these findings suggest that promoting racial-ethnic pride could start with school-aged children in order to promote optimal adjustment among African American and Latino youth.

This study provides further support for relational developmental systems (RDS, Lerner et al., 2016, 2017) perspective as well as the PVEST model. Consistent with the RDS perspective, our findings suggest that both individual (competence) and contextual developmental assets (connection to adults and youth in their communities) contribute much to youth adjustment. Moreover, besides the “Cs” of PYD, the additional socio-cultural dimensions are also helpful. The PVEST conceptual framework focuses on coping strategies particularly relevant to race and ethnicity, the value of resisting stereotypes and possessing a sense of group pride is found to be helpful both in regard to adjustment and academic performance. In some ways, this framework is consistent with the ideas of stereotype threat that find when stereotypes regarding one’s race are invoked, it suppresses the performance of racial-ethnic minority students (Steele, 1997). Steele (2011) points to the value of supportive mentors and group-based teamwork in building a sense of competence and connection, such as would be conceptualized in the PYD framework.

Although we recognize that parents, teachers, and others in the community want desperately to prepare these youth for racial-ethnic barriers and discrimination, we found that fostering awareness of racial-ethnic barriers in these young elementary-age children did not help them adapt well behaviorally or academically. On the contrary, helping them to feel like competent, caring, and connected members of their community, and proud of their racial-ethnic background continues to demonstrate benefits to this sample of African American and Latino children of diverse geographic and socio-economic backgrounds. These children are able to judge themselves by the “content of their character” and define themselves as proud, important and valuable members of society in the face of negative appraisals from elsewhere.

In conclusion, fostering children’s connection to others in their community, caring for others, feeling competent as a group as well as a sense of affirmation, happiness, and belonging to their racial-ethnic group proves again to be helpful, particularly given the challenges, stereotypes, and discrimination these youth are likely to experience. Moreover, this study was done in an afterschool setting, implying that feeling competent as a whole, caring for others, and being connected with adults and peers within the afterschool programs mattered and contributed to youth adjustment. The findings suggest the potential of using afterschool programs to promote competence, caring, and connection as well as racial-ethnic pride in order to foster optimal adjustment among African American and Latino youth.

Table 6.

Chi-square test of youth adjustment among three classes

	High PYD, Proud & Optimistic		High PYD, Proud & Aware		Low PYD & Disconnected
Outcomes	M	SE	M	SE	M	SE
Time 1
Adjustment problems	1.38a,b	.04	1.76a	.07	1.78b	.10
Hyperactivity	1.26a,b	.09	1.85a	.09	1.74b	.16
Emotion symptoms	1.49a,b	.06	1.85a	.08	1.84b	.13
Conduct problems	1.24a,b	.05	1.81a	.08	1.81b	.13
Time 2
Adjustment problems	1.37a,b	.05	1.80a	.08	1.83b	.12
Hyperactivity	1.36	.22	1.93	.19	2.01	.19
Emotion symptoms	1.35	.13	1.83	.13	1.78	.12
Conduct problems	1.31a,b	.07	1.80a	.10	1.75b	.17
PSSA math	1358.90a	27.37	1270.74a	24.02	1175.66	96.50
PSSA reading	1306.73a,b	25.09	1217.35a	33.98	1110.14b	81.86

Note. When two classes have the same alphabetic letter, it means the two measures were significantly different between classes.

Table 7.

Age and Gender as Predictors for Class Memberships

	Age		Gender		Race-Ethnicity
	Logit	OR	Logit	OR	Logit	OR
High PYD, Proud & Optimistic to High PYD, Proud & Aware	.17	1.19	.29	1.34	.66	1.93
High PYD, Proud & Optimistic to Low PYD & Disconnected	.28	1.32	.56	1.75	.65	1.92
High PYD, Proud & Aware to Low PYD & Disconnected	.11	1.12	.27	1.31	−.01	.99

^*p < .05,

^**p < .01. Note. Boy was coded 1 and girl was coded 2. African American was coded 1 and Latino was coded 0. OR = Odds ratio.