Mediators that Matter: Psychological Distress, Developmental Assets, and Educational Outcomes among Black Youth

Abstract

Persistent inequities in the educational success of Black adolescents are a critical social justice concern. Though psychological distress has been associated with worse educational outcomes, less is known about the mechanisms that may influence this association. This study used nationally representative cross-sectional data from the National Survey of American Life-Adolescent (NSAL-A; 2001–2004) to explore how developmental assets (i.e., self-esteem, mastery, school bonding, educational aspirations, educational expectations) mediate associations between psychological distress (i.e., perceived stress, depressive symptoms) and educational outcomes (i.e., grades, grade repetition, suspensions, expulsions) among 1,170 Black adolescents ages 13–17 (52% female; mean age 15). The study found that educational expectations was a statistically significant mediator; lower psychological distress was associated with greater expectations which, in turn, was linked to better grades, lower grade repetition, and fewer expulsions. Additionally, school bonding was a statistically significant mediator such that lower perceived stress and depressive symptoms were associated with better school bonding. In turn, school bonding was associated with higher grades and fewer school expulsions. Self-esteem also significantly mediated the association between psychological distress and grade repetition. Study findings can contribute to precision in identifying culturally relevant targets of interventions among Black adolescents and help to address racial disparities in adolescent educational outcomes.

Persistent inequities in the educational success of Black American adolescents are a major social justice issue (Newman et al., 2021; Yearby et al., 2020). Due to structural and institutional racism (e.g., Yearby et al., 2020), Black adolescents are still overrepresented among adolescents experiencing underachievement, higher grade retention, and greater exclusionary discipline practices (e.g., U.S. Department of Education, 2021) despite improvements in educational achievement (e.g., Hanushek et al., 2009). Adolescence occasions both positive social emotional development and the potential emergence of psychological distress (National Academies of Sciences, Engineering, and Medicine, 2019; Rapee et al., 2022). Though research evidence points to adverse associations between psychological distress and educational outcomes (Clayborne et al., 2019; Wickersham et al., 2021), less is known about the mechanisms through which psychological distress influences adolescents’ educational trajectories (e.g., Pate et al., 2017), particularly for Black adolescents (e.g., Joe, Joe, et al., 2009).

Guided by positive youth development theory (e.g., Benson et al., 2007), this study explores the mediating role of developmental assets on associations between psychological distress and educational outcomes among Black American adolescents (i.e., African American and Caribbean Black adolescents). The prevailing educational research among Black students has given limited attention to factors that support Black students’ success, instead perpetuating deficit-based narratives, steeped in cultural stereotypes holding Black students as the inferior minority and centering primarily on risk factors for poor educational outcomes (Whaley & Noel, 2013; Williams et al., 2020). Our asset-based approach challenges that dominant lens. It is consistent with trends in the scientific community to engage in research around positive development among children and adolescents (Furlong et al., 2014) and empirical evidence demonstrating effectiveness of strengths-based interventions versus those that are problem-based (Weare & Nind, 2011). Through this research, we hope to identify a set of developmental assets that hold promise as culturally relevant targets of interventions designed to promote better educational experiences for Black adolescents.

Educational Outcomes among Black Youth

Structural inequities rooted in social systems precipitate disadvantage among people of color and can result in school policies and practices that are detrimental to Black adolescents’ educational experiences (Noguera & Alicea, 2020). Research on the disproportionate representation of Black youth across suspension and expulsion data suggests an intrinsic link between systemic racism and implicit bias in school discipline practices (Gregory et al., 2021; U.S. Department of Education Office of Civil Rights, 2021). Higher grade retention was found for Black elementary and secondary students across all grade levels (de Brey et al., 2019), and Black students have been suspended and expelled at rates more than twice their total enrollment (U.S. Department of Education, 2021). These longstanding and pervasive disparities are compounded by the disproportionate representation of Black adolescents in low-resourced settings. Black adolescents may experience higher exposure to violence, indigent living, family and community poverty, and other systemic inequities in the school setting, such as lower funded schools, which potentially contribute to negative educational trajectories (Busby et al., 2013; de Brey 2019; Francois et al., 2012; Sheats et al., 2018; U.S. Commission on Civil Rights, 2018).

Prior research has demonstrated that poor educational experiences are related to worse behavioral health, shorter lives, and lower economic well-being (Musu-Gillette et al., 2017; Olshansky et al., 2012; Schiller et al., 2012; U.S. Bureau of Labor Statistics, 2018). School success, at minimum high school graduation, is critical to an adolescent’s economic potential, life chances, and ability to become a productive member of society (Richman et al., 2004; UNICEF, 2012). Recent National Assessment of Educational Progress data revealed that Black students had lower math and reading scores (4^th, 8^th, 12^th grades) relative to their White, Hispanic, Asian, and multi-racial counterparts (U.S. Department of Education, 2023). Sufficiently addressing these longstanding educational disparities requires a clear delineation of factors that support Black adolescent educational success. For this study, educational outcomes include both academic (e.g., grades) and discipline outcomes (e.g., suspensions, expulsions).

Psychological Distress and Educational Outcomes

Psychological distress is used in this study as an overarching term encapsulating both perceived stress and depressive symptoms (e.g., Hirshberg et al., 2022). Levy et al. (2016) proposed a biopsychosocial stress response model suggesting that varying stressors (e.g., perceived discrimination) that challenge resources (e.g., family/community supports) may lead to perceived stress, which in turn influences other psychological responses (e.g., educational aspirations) and subsequent academic performance. This model suggests that other intervening factors may help explain the association between perception of stress and education outcomes.

Existing studies have shown that situations denoted as stressful (e.g., discrimination, negative police interactions) adversely influence educational outcomes among Black American adolescents (Anderson et al., 2019; Legewie & Fagan, 2019; Morsy & Rothstein, 2019). Though limited, prior studies on perceived stress, or appraising a situation as stressful, have reported that lower stress has predicted better educational outcomes, such as GPA, among Black adolescents (e.g., Schmeelk-Cone & Zimmerman, 2003). Other studies among adolescents and young adults have reported that higher perceived stress is associated with lower academic performance (Pascoe et al., 2020; Stoliker & Lafreniere, 2015).

Research indicates poorer educational trajectories for adolescents experiencing depression both concurrently (e.g., Wickersham et al., 2021) and longitudinally (Clayborne et al., 2019; Pate et al., 2017; Riglin et al., 2014). Depressive symptoms in African American adolescents were also associated with lower grades (Busby et al., 2013), lower future GPA (Pate et al., 2017), and lower educational expectations (Turcios-Cotto & Milan, 2013). Additionally, depressive symptoms were associated with higher odds of grade retention (Shinde et al., 2022). Further, discipline severity (e.g., suspensions) was associated with depressive symptoms among diverse youth (Eyllon et al., 2022). Convesely, better mental health has been associated with fewer suspensions and lower grade retention among Black adolescents (Rose et al., 2017).

Positive Youth Development Theory (PYDT) and Developmental Assets

Positive youth development theory (PYDT; e.g., Benson et al., 2007) is grounded in developmental and ecological models and encompasses multiple theoretical strands (Benson et al., 2007) and provides the guiding theoretical framework for this study. According to this theory, adolescents are seen as “co-producers” of their development, as they are influenced by and influence their contexts (Benson et al., 2007). PYDT helps to elucidate the intrinsic ability, assets, and potential that youth have for positive growth and development. It also looks at the critical role that context plays in effecting positive change, including how to maximize the constructive interplay between person and context (Benson et al., 2007).

The Search Institute’s developmental asset framework (Search Institute, n.d.) is grounded in PYDT, aligning with a focus on the capacity and strengths of adolescents. The assets are described as essential nutrients needed by all youth for healthy development (Sesma et al., 2005). They can be conceptualized as “opportunities, experiences, and supports” (p. 197) that serve protective or promotive functions, contributing to better developmental outcomes and effective functioning despite adversity (Benson et al., 2011). Research suggests that these assets should be actively explored by young people and actively nurtured through intervention (Benson et al., 1999). For this study, assets are conceptualized as promoting better educational outcomes as well as intervening in the psychological distress and educational outcomes link. For example, lower distress has been associated with higher levels of assets (e.g., Manna et al., 2016), and higher levels of assets have been associated with better educational outcomes (e.g., Uka et al., 2021).

This study focused on specific internal assets which are “social-emotional strengths, commitments, and values youth need to make good choices, take responsibility for their own lives, and be independent and fulfilled” (Search Institute, n.d.). Two sets of internal developmental assets, positive identity, and commitment to learning, were explored. These selected assets have been extensively documented in adolescent research as associated with both psychological distress and educational outcomes (e.g., Trzesniewski et al., 2006; Uka et al., 2021), making them relevant as potential mediators. Notably, these assets are malleable, making them ideal targets for individual- or group-level interventions, particularly at the school-level.

Positive Identity Assets

Identity development is a key task during adolescence (Erikson, 1968). During this stage, adolescents seek to develop and internalize a sense of who they are, including a sense of self-esteem and power to influence outcomes in their lives (Benson et al., 1999). Self-esteem and personal power are positive identity assets in the Search Institute’s framework of developmental assets (Benson et al., 1999; Sesma et al., 2005). Self-esteem is described as a person’s beliefs about their value, aptitude, ability to achieve success, importance, and, in general, their appreciation or lack of appreciation of self (Modrcin-Talbott et al., 1998). Depression (Manna et al., 2016; Millings et al., 2012; Trzesniewski et al., 2006) and stress (Tuominen-Soini & Salmela-Aro, 2014) have been negatively associated with self-esteem. Higher self-esteem has also been associated with better school performance (e.g., Zhao et al., 2021).

Personal power refers to an individuals’ perception about what they can control in their lives and includes constructs such as mastery (Benson et al., 1999). Theoretically, youth’s perception of their circumstances as due mainly to personal effort leads to potential maximization of that effort (Luthar & Zigler, 1992). This may, in turn, result in greater school success. Though less explored in the adolescent research literature, mastery has been associated with academic success (Józsa et al., 2019) and lower depression (e.g., Assari & Caldwell, 2017) among youth. Though adolescent research is lacking on self-esteem and mastery as mediators between psychological distress and educational outcomes, both constructs have been mediators in other adolescent development studies (e.g., Bong, 2008; Ybrandt & Armelius, 2010).

Commitment to Learning Assets

School bonding and achievement motivation are commitment to learning assets, significant facets of adolescent development (e.g., Sesma et al., 2005) associated with better academic achievement (Uka et al., 2021). Though terminology overlaps (e.g., school connectedness, school belonging), school bonding refers to the close connections students have with school (including school personnel, e.g., teachers) and the investment they make in school (Catalano et al., 2004). These ties have been important to positive educational experiences and outcomes among adolescents (Libbey, 2004). Notably, some studies have found that Black students had lower school bonding compared with White and Asian students (Yang & Anyon, 2016), suggesting the need to better understand this construct among Black adolescents. Generally, depressive symptoms (e.g., Klinck et al., 2020; Pate et al., 2017) and perceived stress (Maiya et al., 2021) have been negatively associated with school bonding, while positive connections to school have been associated with better achievement (Dotterer & Wehrspann, 2016; Scales et al., 2020). Pate et al. (2017) also reported school connectedness significantly mediated the association between depressive symptoms and grade in later years.

Achievement motivation is described as the desire to do well in school, including a young person’s aspirations or expectation to achieve (Sesma et al., 2005). Expectations and aspirations have been negatively correlated with depressive symptoms (e.g., Chen & Hesketh, 2021) and positively correlated with better GPA and overall educational success (Chen & Hesketh, 2021; Cunningham et al., 2009; Rutherford, 2015; Sesma & Roehlkepartain, 2003). Higher aspirations have also been associated with lower suspensions and expulsions (Mizel et al., 2016). Like self-esteem and mastery, research is lacking on aspirations and expectations as mediators of psychological distress and educational outcomes among adolescents; however, educational expectations have been used as a mediator in prior adolescent research (e.g., Suizzo et al., 2012).

Contextual and Demographic Influences

Given the significance of youth–environment transactions, contextual factors (i.e., parent and peer support, and neighborhood quality) potentially underlie the development of assets and have been associated with psychological distress and educational outcomes. For example, parent and peer support have been related to better self-esteem (e.g., Scott et al., 2016; Smokowski et al., 2014), mastery (Song et al., 2015; Surjadi et al., 2011), greater school connections (Estell & Perdue, 2013; Pan et al., 2017), educational aspirations (Berzin, 2010), and lower depressive symptoms (Ringdal et al., 2021). Similarly, neighborhood cohesion and quality have been related to better self-esteem (DiClemente et al., 2018) and higher educational expectations (Mello & Swanson, 2007), respectively. Better neighborhood quality (e.g., neighborhood advantage) is also associated with better academic achievement (e.g., Wodtke & Parbst, 2017).

Studies highlight the potential association of common demographic variables with assets, psychological distress, and educational outcomes. Previous studies suggest that strength-based assets may vary based on ethnicity (Butler-Barnes et al., 2018). Additionally, sex differences have been noted in mental health among adolescents, with girls experiencing greater prevalence of depression compared to boys (Kessler et al., 2012). Sex differences have also been noted in school discipline outcomes, with boys reporting greater school suspensions (Camacho & Krezmien, 2019). Higher income has also been associated with better self-esteem (Bai et al., 2021), better school bonding (e.g., Assari, 2019), less grade retention (Giano et al., 2022), while lower income has been associated with greater suspensions (Camacho & Krezmien, 2019) and higher depression (Santiago et al., 2011). Finally, identity formation (e.g., self-esteem) has been found to change or increase across adolescent age (Meeus et al., 2010), and school connections and mental health have been found to worsen with age (e.g., Cavioni et al., 2021). Given the relevance of these contextual and demographic associations to our study variables, they are included as covariates in the present study.

Current Study

Minimal research has examined the specific pathways among psychological distress, developmental assets, and educational outcomes, particularly for Black adolescents. Adolescent research (e.g., Pate et al., 2017; Yang & Anyon, 2016) supports the potential mediating role of our developmental assets. Developmental assets may fully or partially reduce negative sequelae and are included in this study as mediators. However, research is scant on the indirect processes through which psychological distress influences educational outcomes (e.g., Pate et al., 2017). We acknowledge the potential for bidirectionality between psychological distress and educational outcomes. Some literature supports the influence of school discipline on depressive symptoms (e.g., Perryman et al., 2022) as well as lower grades on mental health problems (e.g., depression; Wallin et al., 2019). However, other studies, including longitudinal studies, highlight the influence of mental health problems on later educational experiences (e.g., Clayborne et al., 2019; Pate et al., 2017). Consistent with this latter literature base, we explore the influence of psychological distress on educational outcomes.

Our overarching research question is: Among Black adolescents, are associations between psychological distress (i.e., depression symptoms, perceived stress) and educational outcomes (i.e., grades, grade repetition, suspensions, expulsions) mediated by developmental assets (i.e., self-esteem, mastery, school bonding, educational aspirations, educational expectations), controlling for contextual and demographic influences? We hypothesize that all developmental assets will significantly mediate the associations between psychological distress and educational outcomes. Specifically, and as noted in Figure 1, we expect that adolescents with lower depressive symptoms and perceived stress will report higher self-esteem, mastery, school bonding, educational aspirations, and educational expectations. We also expect that higher self-esteem, mastery, school bonding, educational aspirations, and educational expectations will be associated with better grades, less grade repetition, suspensions, and expulsions.

Figure 1. Conceptual Model of Psychological Distress, Developmental Assets, and Educational Outcomes.

Note: Demographic (ethnicity, sex, age, income) and contextual (parent support, peer support, neighborhood quality) covariates are not shown.

Method

Study Population

The following research is a secondary analysis of the 2001–2004 National Survey of American Life adolescent supplement (NSAL-A; Jackson et al., 2004), administered by researchers at the Program for Research on Black Americans (PRBA) and approved through the University of Michigan’s IRB. The NSAL-A is based on the NSAL (see Jackson et al., 2004 for more detailed information about the NSAL). To generate the NSAL-A, all adult NSAL participant households including African American (AA) and Caribbean Black (CB) adolescents were screened for eligible adolescents living in those households; subsequently, adolescents were selected utilizing a randomized procedure. If a household had multiple eligible adolescents, up to two were selected to participate with the aim of having the second adolescent being a different gender (e.g., Joe, Baser, et al., 2009). The NSAL-A weight was created to adjust for variation in probabilities of selection within specific households and as non-response rates for both adolescents and households. The weighted data were post-stratified to estimate the national population distributions for both sex (male and female) as well as age (13–17 years old) subgroups for Black adolescents (e.g., Joe, Baser, et al., 2009), allowing research using these data to make more precise inferences on the population of Black adolescents at a national level.

Prior to being interviewed, assent was gathered from the adolescent and informed consent was obtained from the adolescent’s legal guardian. Interviews were conducted by trained interviewers who utilized a computer-assisted instrument and most interviews occurred in the adolescents’ homes. Eighteen percent of interviews were conducted either partially or fully by telephone. Adolescents were given $50 for participating and the total study response rate was 80.6% (83.5% for CB adolescents and 80.4% for AA adolescents). While the original sample included 1,193 cases, 23 cases over age 18 at the time of the interview were excluded. As a result, the final analysis sample was composed of 1,170 AA (n = 810) and CB (n = 360) adolescents, 52% female, who were between the ages of 13–17. The majority of CB adolescents were U.S.-born (60.0%) and had one or both parents born in the Caribbean (72.2%).

Measures

Psychological Distress

Perceived Stress.

This construct was measured using Cohen’s perceived stress scale (Cohen et al., 1983), which measures an individual’s evaluation or appraisal of situations in their life as stressful. A sample item is “In the last month, how often have you dealt successfully with daily hassles?” and includes 14 questions with frequency responses ranging from 1 (never) to 5 (very often). Positively worded items were reverse coded, and higher scores indicated higher levels of perceived stress (α = .77). Previous research has generated α reliability estimates for the scale scores ranging from .74 to .80 for Black youth (Schmeelk-Cone & Zimmerman, 2003; McGlumphy et al., 2019). Concurrent and predictive validity has also been demonstrated generally (Cohen et al., 1983), with predictive validity established for Black youth for a 13-item version of the scale (Ahmed, 2023).

Depressive Symptoms.

The Center for Epidemiologic Studies Depression (CES-D) 12-item scale (Radloff, 1977) was used to measure depressive symptoms that adolescents experienced within the past week. One sample item from the scale is “I felt depressed” and responses ranged from 0 (rarely or none of the time/less than one day) to 3 (most or all the time/5 to 7 days). Positively worded items were reverse scored, and high mean CES-D scores were indicative of greater depressive symptoms (α = .68). Internal consistency estimates among Black adolescents and young adults from prior studies ranged from .64 to .79 (Prelow et al., 2006; Roberts & Sobhan, 1992). Though validity studies of the 12-item version of the scale among youth are limited, the CES-D 12 has demonstrated acceptable discriminant validity in a previous study conducted with adolescents (Poulin et al., 2005). Convergent validity of the 20-item version of the scale has been established among Black youth (Lu et al., 2017).

Developmental Assets

Positive Identity.

Self-esteem.

This variable was measured using the 10-item Rosenberg Self-esteem scale (Rosenberg, 1965). A sample item from this scale is: “On the whole, I am satisfied with myself.” It used a 4-point response set ranging from 1 (strongly agree) to 4 (strongly disagree). Positively worded items were reverse coded, with higher mean scores indicating higher levels of self-esteem (α = .72). Reliability estimates from prior studies among Black adolescents ranged from .79 to .85 (Lockett & Harrell, 2003; Salazar et al., 2005; Seaton, 2010). Construct (Bagley & Mallick, 2001) and concurrent (Hagborg, 1993) validity has been established among youth. Though no validation studies have been conducted exclusively among Black youth, factorial validity has been reported among African American college students (Chao et al., 2016).

Mastery.

Mastery was measured using Pearlin’s 7-item scale (Pearlin & Schooler, 1978), which assesses an individual’s perceived sense of control of their life. A sample scale item is “I can do just about anything I set my mind to,” and a 4-item response set was used ranging from 1 (strongly agree) to 4 (strongly disagree). Positive items were reverse coded, with higher mean scores meaning higher levels of mastery (α = .68). Reliability estimates from prior studies among Black adolescents ranged from .69 to .77 (Blash & Unger, 1995; Nebbitt et al., 2008). Construct validity has been reported for adults (Pearlin et al., 1981).

Commitment to Learning.

School Bonding.

This variable was assessed using a 9-item school bonding scale measuring adolescent affective connection and attitudes towards school. The measure was derived from the National Comorbidity Survey: Adolescent Supplement (NCS-A; Kessler, 2011), from Hawkins et al.’s (1992) conceptualizations of experiences in school as protective mechanisms, and from Zimmerman and Arunkumar’s (1994) resiliency factors. A sample scale item is “most of my teachers treat/ed me fairly,” measured on a 4-point scale ranging from 1 (very true) to 4 (not at all true). Positively worded items were reverse scored such that higher scores represented greater levels of school bonding (α = .71).

Educational Aspirations.

Adolescents’ educational aspirations were measured by a single question asking: “Please tell me how far you would like to go in school?” The response set ranged from not graduate high school (1) to graduate/professional degree (6).

Educational Expectations.

Educational expectations were measured by a single question asking: “As things stand now, how far in school do you think you will get?” The response set ranged from not graduate high school (1) to graduate/professional degree (6).

Educational Outcomes

Grades.

Grades were measured by asking adolescents about the grades they most often receive, including mostly As, mostly Bs, mostly Cs, mostly Ds, or mostly failing grades. The letter grades were converted into a 5-point scale (A = 5, B = 4, C = 3, D = 2, Failing = 1).

Grade Repetition.

Repeating a grade was measured by asking adolescents: “Did you ever stay back or repeat a grade in school?” The question was derived from the National Longitudinal Study on Adolescents (Add Health; Harris et al., 2013). Responses were coded as 1 (yes) and 0 (no).

School Suspension.

School suspension was measured by asking adolescents: “Were you ever suspended from school for a day or longer?” The question was derived from Add Health (Harris et al., 2013). Responses were dichotomized as yes (1) or no (0).

School Expulsion.

School expulsion was measured by asking adolescents: “Were you ever expelled from school?” The question was derived from Add Health (Harris et al., 2013). Responses were dichotomized as yes (1) or no (0).

Contextual and Control Variables

Parent Support.

This variable was measured with five items asking how often adolescents receive emotional and tangible support from family members (e.g., “How often do your family members make you feel loved and cared for?”). Response options ranged from 1 (very often) to 4 (never), and all items were reverse coded such that higher mean scores represented greater family support (α = .72). Parent support questions were adapted from Fetzer Institute/National Institute of Aging Working Group (1999) for use in the NSAL-A.

Peer Support.

This variable was measured using five items asking how often adolescents receive emotional and tangible support from peers (e.g., “How often do friends listen to you talk about your private problems and concerns?”). Responses ranged from 1 (very often) to 4 (never). All items were reverse coded so that higher mean scores represented greater levels of support (α = .72). Peer support questions were adapted from Fetzer Institute/National Institute of Aging Working Group (1999) for use in the NSAL-A.

Neighborhood Quality.

This variable was measured using seven items asking adolescents about perceptions of their neighborhood (e.g., “I have neighbors who would help me if I had an emergency”). Response items ranged from 1 (very true) to 4 (not true at all). Positive items were reverse coded such that higher mean scores represented better perceptions of neighborhood quality (α = .79). Items were adapted from the National Survey on Black Americans (Jackson & Neighbors, 2005) and Add Health (Harris et al., 2013) for NSAL-A use.

Control Variables.

Control variables included sex, ethnicity, age, and family income. Sex was a binary variable (male, female) as was ethnicity (AA, CB). Age was used as a continuous variable measured in years. Income was a continuous variable. Ethnicity and income were based on the adult in each reported household.

Transparency and Openness

We used the complete NSAL-A data sample (n = 1170) for this study. We described all measures and discussed our selection of measures in the background section. The study is consistent with Journal Article Reporting Standards (JARS; Kazak, 2018). The NSAL-A public dataset and supporting materials are available through the Resource Center for Minority Data, https://www.icpsr.umich.edu/web/RCMD/studies/36380. Analysis code for the current study is available upon request from the corresponding author. As noted below, data were analyzed using Mplus 8.6 (Muthén & Muthén, 2021). This study was not pre-registered.

Analysis Plan

Following from Figure 1, a measured variable path analysis was conducted separately for each of the four key outcomes (i.e., grades, grade repetition, suspension, and expulsion) using Mplus 8.6 (Muthén & Muthén, 2021), accommodating the complex study design’s sampling weights, stratification, and clustering using design-based corrections (see, e.g., Stapleton, 2013) and accommodating the binary nature of the latter three outcomes using categorical estimation (see, e.g., Finney & DiStefano, 2013). Given the potential connections hypothesized among the variables, all models were just-identified (and hence had perfect fit indices by default). Model direct effects were assessed using robust maximum likelihood, while indirect and total effects were assessed using bootstrapping (Preacher & Hayes, 2008). The structure of each model included the psychological distress variables (perceived stress, depressive symptoms), contextual variables (parent support, peer support, neighborhood quality), and control variables (sex, ethnicity, age, family income) all as potential determinants of the developmental assets (self-esteem, mastery, school bonding, educational aspirations, educational expectations), and with all four blocks of variables as potential determinants of the outcome being modeled. Of primary importance was the influence of psychological distress on each outcome, and the extent to which it was mediated by the developmental assets.

Results

Descriptive Statistics of Demographic, Contextual, and Main Study Variables

Sample descriptives are reported in Table 1. The sample included slightly more female (n = 607; 52%) and AA (n = 810; 69%) adolescents. Average age was about 15 years old (SD = 1.43) and mean income was $37,687 (SD = 39,695). Mean scores and standard deviations of parent support (M = 3.41, SD = .52), peer support (M = 2.91, SD = .62), neighborhood quality (M = 3.14, SD = .63), perceived stress (M = 35.57, SD = 7.80), depressive symptoms (M = 8.92, SD = 5.26), self-esteem (M = 3.56, SD = .42), mastery (M = 3.14, SD = .55), school bonding (M = 3.39, SD = .45), educational aspirations (M = 5.01, SD = 1.20), educational expectations (M = 4.52, SD = 1.44), and grades (M = 3.69, SD = .82) were reported. Twenty-seven percent of the sample had been retained a grade, 52.6 % had been suspended, and 7% had been expelled.

Table 1.

Descriptive Statistics of Demographic, Contextual, and Main Study Variables (N =1170)

Demographics	n (%)	M (SD)
Ethnicity
African American	810 (69.2)
Caribbean Black	360 (30.8)
Sex
Female	607 (51.9)
Male	563 (48.1)
Age		15.03 (1.43)
Family income		$37,687 (39,695)
Contextual Variables
Parent support1		3.41 (.52)
Peer support1		2.91 (.62)
Neighborhood quality1		3.14 (.63)
Psychological Distress
Perceived stress2		35.57 (7.8)
Depressive symptoms3		8.92 (5.26)
Developmental Assets
Self-esteem1		3.56 (.42)
Mastery1		3.14 (.55)
School bonding1		3.39 (.45)
Educational aspirations4		5.01 (1.20)
Educational expectations4		4.52 (1.44)
Educational Outcomes
Grades5		3.69 (.82)
Grade retention (1 = yes)	327 (27.9)
School suspension (1 = yes)	615 (52.6)
School expulsion (1 = yes)	82 (7.0)

Note. Range of scores:

¹1–4

²14–62

³0–30

⁴1–6

⁵1–5

N, percentage, means (M), and standard deviation (SD) are unweighted.

The following sections explicate 1) the standardized direct paths from contextual variables to developmental assets; 2) the standardized direct paths from psychological distress to developmental assets; 3) the standardized direct paths from developmental assets to educational outcomes; and 4) the mediation analysis - standardized direct and indirect path associations for the psychological distress, developmental asset, and educational outcomes variables, controlling for the demographic and contextual variables (see Table 2).

Table 2.

Decomposition of Standardized Effects of Psychological Distress on Educational Outcomes, as Mediated by Developmental Assets

	Grades	95% CI	Grade Repetition	95% CI	Suspensions	95% CI	Expulsions	95% CI
Perceived Stress
Total effect	0.010	[−0.100, 0.131]	0.116	[0.015, 0.222]	0.029	[−0.098, 0.151]	0.120	[−0.052, 0.294]
Total indirect effect	−0.058	[−0.117, −0.007]	0.093	[0.042, 0.145]	−0.016	[−0.077, 0.039]	0.075	[0.003, 0.163]
Indirect via self-esteem	−0.005	[−0.038, 0.027]	0.029	[0.004, 0.061]	−0.020	[−0.059, 0.010]	−0.023	[−0.079, 0.019]
Indirect via mastery	−0.013	[−0.055, 0.026]	0.041	[−0.008, 0.083]	−0.017	[−0.075, 0.048]	0.066	[−0.016, 0.163]
Indirect via school bonding	−0.019	[−0.040, −0.004]	0.002	[−0.010, 0.017]	0.005	[−0.003, 0.022]	0.014	[0.001, 0.039]
Indirect via educational aspirations	−0.003	[−0.014, 0.001]	0.005	[−0.001, 0.025]	0.004	[−0.002, 0.027]	−0.001	[−0.014, 0.003]
Indirect via educational expectations	−0.017	[−0.043, −0.004]	0.017	[0.003, 0.042]	0.012	[−0.001, 0.035]	0.019	[0.005, 0.041]
Direct effect	0.067	[−0.047, 0.182]	0.023	[−0.074, 0.129]	0.045	[−0.096, 0.173]	0.046	[−0.156, 0.249]
Depressive Symptoms
Total effect	−0.155	[−0.257, −0.062]	0.115	[0.004, 0.225]	0.122	[−0.010, 0.250]	0.085	[−0.041, 0.209]
Total indirect effect	−0.066	[−0.121, −0.020]	0.080	[0.036, 0.136]	0.001	[−0.047, 0.043]	0.050	[0.000, 0.105]
Indirect via self-esteem	−0.006	[−0.044, 0.025]	0.030	[0.006, 0.064]	−0.021	[−0.062, 0.009]	−0.024	[−0.083, 0.018]
Indirect via mastery	−0.005	[−0.023, 0.011]	0.017	[−0.002, 0.043]	−0.007	[−0.036, 0.015]	0.027	[−0.005, 0.059]
Indirect via school bonding	−0.026	[−0.056, −0.007]	0.002	[−0.015, 0.019]	0.007	[−0.004, 0.018]	0.019	[0.003, 0.050]
Indirect via educational aspirations	−0.003	[−0.017, 0.001]	0.006	[−0.002, 0.021]	0.004	[−0.002, 0.022]	−0.002	[−0.012, 0.003]
Indirect via educational expectations	−0.026	[−0.051, −0.010]	0.025	[0.009, 0.046]	0.018	[−0.003, 0.045]	0.029	[0.004, 0.060]
Direct effect	−0.089	[−0.100, 0.131]	0.035	[−0.092, 0.170]	0.121	[−0.007, 0.235]	0.035	[−0.113, 0.173]

Note. Controlling for ethnicity, sex, age, income, parent support, peer support, and neighborhood quality. Significant effects are bolded.

CI = Confidence Interval

Bivariate Associations of Contextual Variables and Developmental Assets

Parent support was significantly associated with higher school bonding (β = 0.25, SE = 0.05) and educational expectations (β = 0.08, SE = 0.03), while peer support was associated with better self-esteem (β = 0.10, SE = 0.03) and higher educational aspirations (β = 0.09, SE = 0.04). Neighborhood quality was only associated with better self-esteem (β = 0.07, SE = 0.03).

Bivariate Associations of Psychological Distress and Developmental Assets

Perceived stress was significantly associated with lower self-esteem (β = −0.30, SE = 0.03, p < 0.001), mastery (β = −0.48, SE = 0.03, p < 0.001), school bonding (β = −0.09, SE = 0.04, p = 0.019), and educational expectations (β = −0.10, SE = 0.04, p = 0.012), but not associated with educational aspirations. Depressive symptoms were significantly associated with lower self-esteem (β = −0.31, SE = 0.05, p < 0.001), mastery (β = −0.20, SE = 0.03, p < 0.001), school bonding (β = −0.12, SE = 0.05, p = 0.009), and educational expectations (β = −0.16, SE = 0.04, p < 0.001), but not associated with educational aspirations.

Bivariate Associations of Developmental Assets and Educational Outcomes

Self-esteem was significantly associated with less grade repetition (β = −0.10, SE = 0.05, p = 0.032) but not associated with grades, suspensions, or expulsions. School bonding was associated with better grades (β = 0.21, SE = 0.04, p < 0.001) and fewer expulsions (β = −0.15, SE = 0.07, p = 0.032) but not with grade repetition or suspensions. Educational expectations were associated with better grades (β = 0.17, SE = 0.05, p = 0.001), less grade repetition (β = −0.16, SE = 0.05, p = 0.002), and fewer expulsions (β = −0.19, SE = 0.07, p = 0.007). Neither mastery nor educational aspirations were associated with any of the educational outcomes.

Mediation Results: Psychological Distress, Developmental Assets, Educational Outcomes

Grades

Regarding grades, the effect of each key predictor (perceived stress and depressive symptoms) will be taken in turn. For perceived stress, the bootstrap confidence interval revealed that the total effect on grades was not statistically significant. That said, because a total effect is an aggregate of individual mediated relations (i.e., through developmental asset variables) as well as any additional direct effect, a decomposition of the effect of perceived stress on grades can be revealing. In this case, the sum of all indirect effects through developmental assets was itself statistically significant (as indicated by a 95% bootstrap confidence interval not containing zero), with a standardized estimate of −.058. Further, the bootstrap confidence intervals revealed statistically significant indirect mediated relations involving school bonding and educational expectations (with estimated standardized effect sizes of −.019 and −.017, respectively). None of the remaining developmental assets were detected as significant mediators. Finally, the remaining direct effect of perceived stress on grades, above and beyond all mediating pathways, was likewise not statistically significant (see Table 2).

For depressive symptoms, a similar story emerges. The bootstrap confidence interval revealed that the total effect on grades was statistically significant, with an estimated standardized effect of −.155. Disaggregating this, to start, the sum of all indirect effects was also statistically significant, with a standardized estimate of −.066. Unpacking these further, bootstrap confidence intervals revealed statistically significant indirect mediated relations involving school bonding and educational expectations (with estimated standardized effect sizes of −.026 and −.026, respectively). None of the remaining developmental assets were identified as significant mediators, nor was the remaining direct effect of depressive symptoms on grades statistically significant (Table 2).

Grade Repetition

For the first predictor, perceived stress, the bootstrap confidence interval revealed that the total effect on grade repetition was statistically significant, with a standardized estimate of .116. The sum of all indirect effects through developmental assets was also statistically significant (standardized estimate of .093). Further, the bootstrap confidence intervals revealed statistically significant indirect mediated relations involving self-esteem and educational expectations (with estimated standardized effect sizes of .029 and .017, respectively). None of the remaining developmental assets were detected as significant mediators. Finally, the remaining direct effect of perceived stress on grade repetition, above and beyond all mediating pathways, was likewise not statistically significant (Table 2).

A comparable story emerges for depressive symptoms (see Table 2). The bootstrap confidence interval revealed that the total effect on grade repetition was statistically significant, with an estimated standardized effect of .115. Disaggregating this, the sum of all indirect effects was also statistically significant, with a standardized estimate of .080. Further unpacking this, bootstrap confidence intervals revealed statistically significant indirect mediated relations involving self-esteem and educational expectations (with estimated standardized effect sizes of .030 and .025, respectively). None of the remaining developmental assets were detected as significant mediators, nor was the remaining direct effect of depressive symptoms on grade repetition statistically significant.

Suspensions

Regarding suspensions, for perceived stress the bootstrap confidence interval revealed that neither the total effect nor the sum of all indirect effects on suspensions were statistically significant (Table 2). Further, none of the developmental assets were detected as significant mediators. Finally, the remaining direct effect of perceived stress on suspensions, above and beyond all mediating pathways, was similarly not statistically significant.

The same findings emerged for depressive symptoms. The bootstrap confidence interval revealed that neither the total effect nor the sum of all indirect effects on suspensions were statistically significant. Further, none of the developmental assets were detected as significant mediators. Finally, the remaining direct effect of depressive symptoms on suspensions was similarly not statistically significant (see Table 2).

Expulsions

Considering expulsions, the effect of each key predictor will be taken in turn. For perceived stress, the bootstrap confidence interval revealed that the total effect on expulsions was not statistically significant. However, a decomposition of the effect of perceived stress on expulsions revealed that the sum of all indirect effects through developmental assets was itself statistically significant, with a standardized estimate of .075. Further, the bootstrap confidence intervals revealed statistically significant indirect mediated relations involving school bonding and educational expectations (with estimated standardized effect sizes of .014 and .019, respectively). None of the remaining developmental assets were detected as significant mediators. Finally, the remaining direct effect of perceived stress on expulsions, above and beyond all mediating pathways, was likewise not statistically significant.

Again, a related picture arises for depressive symptoms. The bootstrap confidence interval revealed that the total effect on grades was not statistically significant, with an estimated standardized effect of .085 (Table 2). Disaggregating this, the sum of all indirect effects was statistically significant, with a standardized estimate of .050. Unpacking these further, bootstrap confidence intervals revealed statistically significant indirect mediated relations involving school bonding and educational expectations (with estimated standardized effect sizes of .019 and .029, respectively). None of the remaining developmental assets were significant mediators, nor was the direct effect of depressive symptoms on expulsions statistically significant.

Discussion

Educational disparities, grounded in structural racism, challenge socially just educational experiences for Black adolescents. As such, Black adolescents remain disproportionately represented among adolescents with negative educational trajectories, warranting continued attention to factors that may support their educational success. Stressors (e.g., perceived discrimination) can precipitate psychological distress and the literature confirms that psychological distress is a key influence on education outcomes. Despite this, minimal research has examined the psychological distress–asset–education link among Black adolescents. The current study helps to fill this gap. PYDT provides a context for understanding the significance of developmental assets for promoting healthy growth and development and disrupting negative outcomes. As hypothesized, school bonding and educational expectations emerged as consistent mediators of psychological distress and educational outcomes, with similar patterns of influence observed. Self-esteem also emerged as a significant mediator of psychological distress and likelihood of repeating a grade. Contrary to expectations, neither educational aspirations nor mastery were significant mediators in this study. Thus, our hypotheses were partially supported.

Mediation findings suggest further exploration of self-esteem, school bonding, and educational expectations as potential intervention targets for programs and services designed to foster better educational experiences among Black adolescents. These findings also fill a gap in the literature on potential mechanisms through which psychological distress is associated with educational outcomes among Black adolescents.

Self-Esteem Remains an Important Positive Identity Asset

Self-esteem is a key part of identity development during adolescence (Erikson & Erikson, 1998) as well as critical to fostering adolescents’ ability to manage typical developmental tasks (Mruk, 2006). Given the majority of adolescents spend a significant portion of their day in school, their self-esteem might be meaningfully shaped by interactions with and support from non-familial adults and peers, particularly as they navigate new or different standards and values inherent to school settings (Catalano et al., 2004). In fact, consistent with prior research (e.g., Smokowski et al., 2014), higher peer support was associated with better self-esteem in this study. Subsequently, school may be one relevant context for fostering adolescent self-esteem, especially as positive identity assets have been noted as significant to the promotion of better academic achievement (Uka et al., 2021). This is aligned with the importance of context for positive growth and development highlighted by PYDT (Benson et al., 2007). Given that neighborhood quality, which includes supportive neighbors, was also associated with better self-esteem in this study, the neighborhood context also warrants further exploration in efforts to promote Black adolescent self-esteem (e.g., DiClemente et al., 2018).

For this study, self-esteem significantly influenced the link between perceived stress and repeating a grade. Similarly, self-esteem significantly mediated the associations between depression symptoms and repeating a grade. Our findings were consistent with prior literature reporting a negative association between psychological distress and self-esteem (e.g., Millings et al., 2012; Tuominen-Soini & Salmela-Aro, 2014) as well as studies highlighting higher self-esteem as promotive of better academic outcomes (e.g., Zhao et al., 2021).

Similar to our study, some research has highlighted a lack of connection between self-esteem and discipline outcomes, like suspensions (e.g., Graves Jr., 2023). It could be that the development of one’s identity, through promoting self-esteem, may be relevant consideration for some aspects of Black adolescents’ school experiences (e.g., academic), while macro-level changes in school climate need to be considered given the systemic and discriminatory nature of school discipline practices. Alternatively, it is possible that school discipline practices may predict lower self-esteem (e.g., Omulema et al., 2015). Even so, the pathway to less grade repetition for youth experiencing depressive symptoms or perceived stress might be through a positive view of self, suggesting further examination of self-esteem in efforts to address grade repetition among Black youth.

Commitment to Learning Assets are Important to School Success

Schools are a significant environmental context for youth, next to family. Most adolescents spend a substantial part of their day in school settings, which provides opportunities for educational, personal, and social growth and development. Within these contexts, youth develop their educational selves (Kirk et al., 2012), and consequent desire and anticipation of future educational achievement. As well, school environments that foster positive relationships with peers, teachers, other adults; connections to school itself; and high expectations may, in turn, support better educational outcomes (Eccles & Roeser, 2011). Aligned with PYDT, commitment to learning assets show promise in interrupting negative pathways and reflect the potential of school-related internal assets to promote positive youth developmental outcomes in a school context.

School Bonding Matters

School bonding, or the significant connections and investments students have to school, is a consistent asset in the context of the school environment (Catalano, et al., 2004; Yang & Anyon, 2016). Students bonding to school can take the form of positive connections to and support from teachers and other adults in the school setting, as well as a favorable perspective of the school environment and meaningfulness of classes and school (Yang & Anyon, 2016). Parents also play a role in their youth’s engagement in and relationship to school, such that students with more supportive parents demonstrate greater classroom engagement (Estell & Perdue, 2013). Our finding that parent support was positively associated with school bonding aligns with this supposition.

Extensive research supports the benefits of school bonding to positive educational outcomes (e.g., Dotterer & Wehrspann, 2016), particularly as it relates to the important role teachers may play in regard to promoting student engagement. Indeed, some research has highlighted the significance of positive teacher relationships to support Black youth as they negotiate their varied environments (school, home) and manage discriminatory concerns at school (Gay, 2002). The positive influence of school bonding for Black adolescents was observed in our study, which was consistent with literature that noted positive associations between depressive symptoms and perceived stress and school bonding (e.g., Klinck et al., 2020; Maiya et al., 2021) as well as school bonding and academic achievement (e.g., Scales et al., 2020). Our findings also align with studies reporting a significant mediating role of school connections on the association between depressive symptoms and academic outcomes (Pate et al., 2017). Results showed perceived stress and depressive symptoms were negatively associated with school bonding. Concurrently, school bonding was associated with better grades and fewer school expulsions.

These findings support the importance of school bonding given its significant link between psychological distress experienced by Black youth and their educational success. However, structural inequities, along with racism, low student expectations, and under-resourced schools may contribute to a school climate that hinders the development of school bonding in youth of color (Bottiani et al., 2016; Yang & Anyon, 2016). Accordingly, intervention at the school level to promote school bonding should consider a multi-level approach.

Educational Expectations Matter

Greater educational expectation can motivate someone to pursue actions towards a particular goal (Rutherford, 2015) and have been associated with better school outcomes (Cunningham et al., 2009; Rutherford, 2015) and later adult attainment (Beal & Crockett, 2010). Educational expectations, also referred to as educational probable selves, are related to anticipated outcomes (Kirk et al., 2012), or youth’s comprehension of how their abilities and opportunities relate to their educational achievement (Greenaway et al., 2015). Expectations, then, are grounded in some sense of concrete, versus abstract (i.e., aspirations), assessment of the future. For Black adolescents, expectations can be shaped by external expectations (e.g., teachers, parents), discrimination, parental educational achievement (i.e., modeling), financial resources of the family, and whether the school is lower resourced (e.g., Bingham & Okagaki, 2012; Demie, 2022; Glogowski & Rakoff, 2019). Expectations may also be influenced by greater parent support, as observed in our study, as parents provide a nurturing environment (e.g., Pan et al, 2017) for an adolescent’s educational probable self.

Some of these factors are structural or contextual, while others might be addressable at micro-levels (e.g., norms, social capital, social networks). Given our study results, educational expectations are potentially important to Black youth’s educational outcomes and thus warrant further attention. Educational expectations significantly influenced the link between perceived stress and grades, repeating a grade, and expulsions. Similarly, educational expectations significantly mediated the associations between depression symptoms and grades, repeating a grade, and expulsions. Our findings were consistent with prior literature reporting that lower psychological distress was associated with greater expectations (e.g., Chen & Hesketh, 2021). In turn, greater expectations were associated with better grades, lower grade repetition, and fewer expulsions (e.g., Rutherford, 2015). The finding that expectations are associated with fewer expulsions is interesting, given that Black adolescents are disproportionately subjected to harsher disciplinary practices in school settings (e.g., Newman et al., 2021). Findings suggest that the future expectations students have for their education deserve closer examination in the development of interventions at both the micro- and contextual- levels to help address the influence of psychological distress on Black adolescents’ educational success.

Strengths, Limitations, and Future Research

A primary strength of the current study was the intentional focus on the mechanisms through which psychological distress influences educational outcomes. Much prior research has established the link between psychological distress and educational outcomes without fully considering the process through which those constructs are associated. Understanding that pathway gives a more complete picture of how psychological distress affects educational outcomes and provides additional and more precise targets of intervention to improve educational outcomes among Black adolescents. Furthermore, we utilized the only existing national probability sample of Black adolescents to date, which enhances our ability to generalize our findings to similar adolescents across the United States.

Although these strengths are significant and expand the literature on Black adolescent development, we recognize some limitations and propose some options for future research. The study used cross-sectional survey data, and consequently causality, directionality, and variation in the psychological distress-asset-educational outcomes link over time could not be fully evaluated. Future research could benefit from longitudinal data collection and prospective research designs to document the assets that may be relevant at varying developmental stages as well as confirm directionality. For example, lower grades could predict depressive symptoms. As well, greater suspensions could be associated with lower self-esteem, which in turn, could be associated with greater depressive symptoms.

Further, the use of secondary data limited a full exploration of some variables. For example, sex was a binary variable. Future studies should consider a more inclusive gender construct. Self-report data could also portend social desirability bias in reporting, thus other sources of information about our constructs may be helpful to validate adolescent responses.

The study also explored a limited set of developmental assets, some of varied reliability. Future studies should explore a broader set of assets, possibly gathered to be modeled as more reliable latent variables, aligned with the Search Institute’s developmental asset framework (Search Institute, n.d.), particularly those considered commitment to learning (e.g., school engagement) or positive identity assets (e.g., sense of purpose).

Additionally, given lower reliability of some of the available scales, identifying other scales that may be more reflective of the population’s cultural or ethnic background should be considered. Validation studies of the instruments among Black youth are also warranted, for example with the CES-D scale (e.g., Lu et a., 2017), as symptom expression may be unique for Black adolescents. Likewise, some studies have highlighted the educational aspiration–expectation discrepancy, with larger discrepancies associated with worse outcomes. Thus, future research should explore this discrepancy especially among Black adolescents and particularly as aspirations did not emerge as a significant mediator in the current study.

Given our focus on establishing mediators that matter as a first step in better understanding the mediation chain, we did not explore demographic variation beyond their inclusion as covariates. Additional research should examine variation in the adolescent age, sex, and ethnicity using moderation approaches, as, for example, some mediators may be more relevant for younger versus older adolescents, or African American versus Caribbean Black adolescents. Also, the important contexts for the promotion of developmental assets (e.g., parent, peer support) warrant a fuller exploration given their significance in this study to some developmental assets. Similarly, parent and teacher expectations and parent educational attainment may be important constructs to include as a part of future research given their overall importance to assets such as educational expectations and youth educational outcomes.

Implications

Academic environments may be intrinsically shaped by discriminatory cultural ecologies (Bingham & Okagoki, 2012; Yang & Anyon, 2016), which can inhibit high expectations and positive connections to school for Black adolescents, particularly in schools that may be more diverse in ethnic composition (Bottiani et al., 2016). Building an anti-racist culture of expectation and a sense of connectedness in schools, where youth self-esteem can also be fostered, likely requires multi-tiered and multi-level approaches. These approaches are critical to efforts towards educational equity and addressing social injustice in Black adolescent educational experiences.

School-level policies can allocate funds to provide training for teachers and school staff on structural racism, discrimination, and implicit bias, where school staff reflect on and explore the role bias plays in shaping their perceptions and expectations of and connections to adolescents of color (e.g., Bottiani et al., 2016; Yang & Anyon, 2016). Accordingly, approaches can be generated to address racism-based hostility (Yang & Anyon, 2016) and acknowledge and celebrate cultural differences in an effort to improve school bonding and foster a climate of high educational expectations. The Double Check intervention (Bradshaw & Rosenberg, 2018) is one example of a coaching and professional development model design that promotes culturally responsive teacher practices and reduces disparity in exclusionary discipline practices among students of color. Bradshaw et al. (2018) conducted an efficacy study of components of Double Check that showed better classroom management and a significant reduction in discipline referrals for Black students. Reduction in discipline referral, in particular, is a step towards a more socially just educational experience for Black adolescents.

Additionally, schools could create access and opportunities for students to be integrated into decision-making processes, with specific attention to involving diverse student groups. Students could be a part of school-level rule development, culturally responsive discipline approaches, curriculum changes, as well as school-wide programs and events (e.g., Mahoney et al., 2021; Rose et al., 2017; Yang & Anyon, 2016). This level of involvement may foster a sense of belonging and a greater investment in the school community (e.g., Catalano et al., 2004), which, in turn, may enhance student–staff relationships and contribute to a safer, and more supportive and inclusive academic environment (e.g., Rose et al., 2017). Involvement in decision-making could also extend to the classroom level, with students co-developing classroom norms. As well, schools could facilitate classroom-level interventions that foster better student–teacher or student–staff relationships and enhance self-esteem and better academic performance.

Intrinsically aligned with PYDT, Taylor et al. (2017) highlighted SEL interventions as a type of asset development approach focused on both promotion of positive development and prevention of adverse outcomes. Systemic SEL emphasizes adolescents as collaborators in the SEL process, and the critical importance of their voice in school- and classroom-level problem-solving and decision-making (Mahoney et al., 2021). A meta-analysis of follow-up effects for SEL interventions reported better attitudes towards self (e.g., self-esteem, school bonding), improved academic performance, and lower emotional distress; these findings held across diverse subgroups of children and youth (Taylor et al., 2017). Importantly, equity-focused SEL research emphasizes the crucial awareness of contexts within which social and emotional competencies develop as well as the importance of addressing implicit bias and promoting cultural responsiveness in program implementation (Comer, 2009; Gregory & Fergus, 2019; Jagers et al., 2019). In fact, Jager et al. (2019) propose transformative SEL, which intrinsically considers the root causes of longstanding inequities (e.g., educational) that may affect optimal development of children and youth of color and contribute to further injustice.

Further, guidance and school counselors, school social workers, and mentors can increase awareness of career prospects and corresponding behavioral strategies, school and community scholarships, and college trips. Indeed, targeting Black youths’ perspectives about their probable selves (Kirk et al., 2012) might open up opportunities for them to broaden their own expectations of future possibilities. Alongside the interventions mentioned above, psychological interventions might help to improve students’ beliefs about their abilities and self-esteem within a school context (e.g., Yeager et al., 2013).

School climates that supported strict but fair discipline, alongside high student expectations, and strong teacher–student support were found to be associated with fewer suspensions and less suspension disparity between Black and White students (e.g., Heilbrun et al., 2018), and thus may warrant more exploration in the development of initiatives designed to foster positive school climate. One example of a multi-tiered and multi-level approach to enhancing school culture and climate is the Positive Schools Center (PSC; Positive Schools Center, n.d.), part of the Center for Restorative Change. The PSC, created in 2015, was developed to help cultivate positive and supportive school cultures and promotes engaging learning communities, thereby reducing punitive discipline. To accomplish this goal, the PSC uses five core principles including: Racial justice and equity; restorative approaches; social and emotional learning; trauma response strategies; and student, family, and community voice. In collaboration with school teams, PSC facilitates program planning with school leaders, staff training, leader and education coaching, school and classroom observations, and technical support of school teams and committees. Importantly, PSC utilizes a performance improvement model where strategies are implemented, analyzed, and modified with critical input from students, families, and communities. Support is provided to school leaders to implement the revised strategy to continually improve school culture. Thus, the PSC proactively and restoratively transforms school climates to foster connection between students and staff and encourages an environment where high expectations are fostered. Outcome data on the effects of the PSC are forthcoming.

Summarily, these proposed best practices and existing interventions are aligned with PYDT as they consider the adolescent’s potential for positive growth and development, the importance of adolescent input, and the significance of the youth–context exchange (Benson et al., 2007). Notably, given that parent, peer, and neighborhood contexts were associated with varying developmental assets in this study, schools could consider these contexts in the development and implementation of these interventions. Importantly, these approaches need to be considered within the context of the cultural underpinnings of each unique school environment (Durlak et al., 2011). Future studies should continue to explore these suggested practices and interventions in empirical studies, using diverse research methodologies, to build the evidence base for asset development and subsequent better educational experiences among Black adolescents.

Conclusion

For Black adolescents, educational success remains a key social justice priority due to disproportionately lower achievement and grade promotion, as well as higher suspension and expulsion rates. Psychological distress is a key predictor of educational outcomes among youth; however, the mechanisms through which psychological distress influences educational outcomes have been understudied. Using a nationally representative sample of Black adolescents, this study applied a strength-based approach to examine links between psychological distress and educational experiences, filling a gap in the adolescent development, psychology, and education literature. Overall, the study helps to elucidate the mediating roles of adolescent self-esteem, school bonding, and educational expectations that future interventionists and policy makers could further consider and explore in order to effectively address disparities in Black adolescent educational outcomes.

Educational Impact and Implications Statement.

This research used a strength-based approach to explore developmental assets that have the potential to promote better educational outcomes among Black adolescents, despite experiences of psychological distress. We found that self-esteem, positive connections to school, and higher educational expectations supported better grades, as well as less grade repetition and fewer expulsions. The findings warrant further exploration in the development of school-based efforts to address racial disparities in adolescent educational outcomes.