Racial Discrimination, School Climate, and Black Adolescents' Internalizing and Externalizing Behaviors: A Longitudinal Moderation Authors: E Cole Green, Jia Yan, Jasmine N Coleman, Elizabeth I Johnson Published: 2026 ABSTRACT ABSTRACTUsing data from Black adolescents in the Adolescent Brain Cognitive Development (ABCD) study (n = 1,316), we examined the moderating effect of school climate on the relationship between racial discrimination and later internalizing and externalizing behaviors during adolescence (M age = 12.00; SD = 0.65). Longitudinal regression analyses showed that positive perceptions of school climate were linked to a weaker association between racial discrimination and subsequent internalizing and externalizing behaviors. Findings highlight the need to reduce discrimination that Black youth face while simultaneously building safe and equitable school climates. Background Racial discrimination is a threat to the well‐being of Black adolescents. Although prevalence estimates vary by geography and age (Umaña‐Taylor 2016), several studies have indicated that Black youth experience high rates of discrimination. Prevalence estimates range from 64% to 73% in racially diverse samples (Niwa et al. 2014; Tobler et al. 2013) to 89% in samples of Black youth only (e.g., Roberts et al. 2012). Discrimination experiences can have many harmful effects on Black adolescents, contributing directly and indirectly to behavioral and mental health difficulties. Black adolescents' experiences of discrimination have been linked to increased internalizing symptoms, such as depressive symptoms and psychological distress (Gale and Dorsey 2020; Hughes et al. 2016; Smith‐Bynum et al. 2014). General strain theory posits that negative events may contribute to behavioral or mental health difficulties via negative emotions (Agnew 1992). Racial discrimination may serve as a negative event or strain for Black adolescents, potentially contributing to them developing negative emotions. This, in turn, may increase the risk for negative health outcomes, such as aggression and delinquency (Burt et al. 2012; Smith‐Bynum et al. 2014). Taken together, prior literature highlights the harmful effects of racial discrimination on Black adolescents and underscores the importance of working to eliminate discrimination while also identifying factors that serve protective functions. One promising avenue of research has highlighted the potential buffering effects of social connectedness on the relation between racial discrimination and health outcomes. Ransome et al. (2023), for example, put forth the “social connectedness buffering hypothesis,” which suggests that strong social connections may attenuate some of the harmful effects of racism (p. 22). They define social connectedness in terms of three core features: an individual's connection to others, their perception of belonging, and the resources made available to them through the networks to which they belong. Despite providing an overview of prior literature, this systematic review largely focused on adults, highlighting a need to continue investigating this phenomenon in adolescence. It is also worth noting that while this hypothesis was developed using research on various minoritized racial groups, over one‐third of the articles in the review were focused on the experiences of individuals racialized as Black. Although much of the empirical evidence for the social connectedness buffering hypothesis comes from adult samples, some work has been conducted with adolescents. Findings have been inconsistent, however, with some studies reporting protective effects (e.g., Liao et al. 2016; Park et al. 2018; Wright and Wachs 2019) and others reporting null or even amplification effects (Lorenzo‐Blanco et al. 2019). Furthermore, there is a dearth of studies focusing on Black adolescents. Given the pervasive and multidimensional racial discrimination experienced by Black youth (Benner et al. 2018; English et al. 2020), understanding the buffering potential of social connectedness remains an important gap. From a developmental lens, social connection is expected to be particularly important during adolescence. Adolescence is a period when social ties, such as peer and school connectedness, become more salient for critical developmental processes, such as seeking independence from caregivers (Birrell et al. 2025). These social ties play an important role in adolescent well‐being (Lamblin et al. 2017) through their stress‐buffering effects (Davidson and Demaray 2007; Mackin et al. 2017). Research on school‐based forms of social connectedness further demonstrates their associations with adolescent adjustment. For instance, greater school connectedness is prospectively associated with lower depressive and anxiety symptoms (Raniti et al. 2022), while more positive teacher–student relationships have been linked to greater school engagement and academic achievement (Roorda et al. 2011). Although these studies do not specifically test whether aspects of school climate buffer the effects of stressors on youth outcomes, they do suggest that school climate may represent an important context for understanding adolescents' internalizing and externalizing problems and their broader school experiences. School climate is a general term that captures the overall inclusivity of a school's environment, including how safe it feels (Cohen et al. 2009). The five dimensions of school climate, safety, teaching and learning, institutional environment, school improvement process, and relationships, have all been linked to beneficial youth outcomes (Thapa et al. 2013). The National School Climate Council (2007) posits that school climate is “based on patterns of school life experiences and reflects norms, goals, values, interpersonal relationships, teaching, learning and leadership practices, and organizational structures” (p.5). This construct, particularly with its emphasis on relationships, complements Ransome and colleagues' (2023) three components of social connectedness: connections to others, belongingness, and the availability of resources within the social network. Because school climate encompasses the degree to which youth feel safe, valued, and supported in their school environment, it captures the same relational and resource‐based mechanisms theorized to buffer stress (Cohen and Wills 1985; Thoits 2011). Thus, when adolescents perceive their school climate positively, they may interpret stressors as less threatening, have more adaptive coping resources, and draw on supportive peer and adult relationships to mitigate stress. Indeed, school climate and the overlapping construct of school connectedness have been shown to moderate the relation between various stressors and negative adolescent mental health outcomes through the same mechanisms proposed by the social connectedness buffering hypothesis (Brookmeyer et al. 2006; Diggs et al. 2025; Goetschius et al. 2023; Zeng et al. 2025). However, in the context of racial discrimination, moderation findings have been inconsistent. Civitillo et al. (2021) illustrate the mixed nature of these findings: although specific aspects of school climate (i.e., teacher‐student relationship quality) attenuated some negative effects of discrimination on self‐esteem and school engagement, broader perceptions of school climate did not demonstrate this buffering effect. Other studies have found little evidence of buffering altogether. Chen et al. (2024), for instance, found that school climate did not moderate associations between personal experiences of racial discrimination and emotional or sleep health. Similarly, King et al. (2025) found no significant moderating effect of school connectedness on the association between discrimination and depressive symptoms among LGBQ Black youth. Together, these findings suggest that the protective role of school climate is inconsistent, underscoring the need for further investigation. Racial discrimination is uniquely and inversely associated with school climate. This pattern in part emerges from the highly racialized nature of exclusionary school discipline (Fix et al. 2021), which disproportionately penalizes Black students and contradicts principles of inclusive school climates (Wang et al. 2023). The racial disproportionality in suspensions suggests such practices are congruent with racial discrimination, a connection that Black adolescents, equipped with racial awareness, are positioned to recognize and interpret (Quintana 1998). The negative impact of exclusionary school discipline is well‐documented. It has been linked to a plethora of harmful outcomes, including poorer academic achievement, higher dropout rates, and more delinquent behaviors (Gerlinger et al. 2021; Noltemeyer et al. 2015). Such negative experiences (e.g., suspensions) are associated with poor perceptions of school climate not only amongst those suspended (Del Toro and Wang 2022) but also amongst those of their classmates (Wang et al. 2024). In contrast, the incorporation of anti‐discriminatory measures in school settings (e.g., school racial socialization) is associated with bolstering student perceptions of a positive school climate, regardless of race (Wang et al. 2023). Particularly, among girls, a reduction in both in‐school and general racial discrimination is associated with more positive perceptions of school climate (Gale and Dorsey 2020). The inverse association between racial discrimination and school climate underscores the importance of considering school climate as not only a correlate of racial discrimination but also as a potential protective factor that may alter the impact of discriminatory experiences on adolescents. In schools with negative climates, discrimination may be normalized and less likely to be addressed, amplifying negative outcomes. Alternatively, schools with positive climates, characterized by safety, equity, and relationality, may afford more resources to adolescents, thereby mitigating the impact of racial discrimination. The established relation between racial discrimination and school climate thus directly informs the rationale for testing it as a potential moderator of the relation between racial discrimination and mental health outcomes, specifically when considering the numerous adolescent outcomes already linked to school climate. Positive school climates are closely associated with academic and socioemotional outcomes. For instance, students who perceive equitable classrooms and school environments and feel that they belong report higher academic competence and school engagement (Griffin et al. 2020; Morris et al. 2020; Wang et al. 2020). Similarly, having culturally responsive teachers has also been linked to higher school engagement and fewer feelings of disconnection (Bottiani et al. 2020). These findings are well laid out in a conceptual model by Huguley et al. (2022), which argues that a school that integrates restorative practices into its climate will see a decrease in disciplinary outcomes and an increase in academic outcomes at the individual and school levels. Socioemotional outcomes are less explored; however, what research does exist seems promising. For instance, Wang et al. (2020) found that, across various dimensions, positive classroom climates were linked to positive academic outcomes and negatively associated with externalizing behaviors and socioemotional distress. Furthermore, some of these benefits were strengthened in more racially diverse classrooms. Still, despite these promising findings, some research has identified racial gaps in perceptions of school climate, with Black students having less favorable experiences across multiple domains (Voight and Nation 2016). As such, further research is needed to understand Black adolescents' experiences of school climate and their protective potential. Current Study Prior research suggests that positive school climates may help protect youth from adverse experiences, including discrimination (Morris et al. 2020). The current study seeks to build on this literature and to expand the social connectedness buffering hypothesis (Ransome et al. 2023) by focusing on perceptions of school climate as an aspect of social connectedness. Using data from the ongoing Adolescent Brain and Cognitive Development (ABCD) study, we address two research questions: 1) Do perceptions of school climate moderate the associations between racial discrimination and internalizing behaviors in adolescents racialized as Black? and 2) Do perceptions of school climate moderate the associations between racial discrimination and externalizing behaviors in adolescents racialized as Black? Methods Sample Participants are part of the ongoing Adolescent Brain Cognitive Development (ABCD) study, which is a longitudinal project observing the growth of U.S. children throughout their adolescent years (https://abcdstudy.org). This representative sample participates in frequent evaluations exploring various aspects of their lives and environments. Youth were initially recruited between the ages of 9 and 10 from 21 sites across 17 states and were, on average, 12 years old (SD = 7.85 months) at the time of the predictor measure (i.e., the 2‐year follow‐up). Early adolescents were targeted for study due to relevant ongoing identity developmental processes, particularly in relation to race (Quintana 1998). The current study uses a subsample of 1,316 Black adolescents (663 boys; 653 girls), comprising all Black youth in the ABCD who were retained through the 3‐year follow‐up. Measures Outcomes Variables Internalizing/Externalizing Problems The Brief Problem Monitor (BPM‐Y; Achenbach 2009) was used to measure both outcome variables. This youth‐reported measure consists of 19 items and asks participants to read various statements and rate them as 0 (not true) to 2 (very true) for themselves over the past 6 months. The internalizing subscale consisted of six items (α = 0.75) that were summed on a scale of 0–12 (M = 1.68; SD = 2.12). Sample items included: “I worry a lot” and “I feel worthless or inferior.” The externalizing subscale consisted of seven items (α = 0.70) that were summed on a scale of 0–14 (M = 2.12; SD = 2.22). Sample items included: “I argue a lot” and “I threaten to hurt people.” Higher scores indicated greater internalizing and externalizing problems. Scores were taken from the 3‐year follow‐up. Focal Predictor Variable Racial Discrimination Phinney's Perceived Discrimination Scale (Phinney et al. 1998) was used to capture adolescents' experiences of racial discrimination. This youth‐reported scale was first incorporated into the ABCD at the 1‐year follow‐up wave and consists of seven items (α = .75; ω t = 0.86; ω h = 0.59) that gauge the frequency of racial discrimination and feelings of nonacceptance, ranging from 1 (almost never) to 5 (very often). McDonald's omega was also reported due to prior validation work, also using the ABCD, which has shown this scale to have a bi‐factor structural model, with a strong unidimensional component (Xu et al. 2025). Mean scores (M = 1.29; SD = 0.48) were utilized if youth responded to at least four of the seven items. Sample items included: “How often do [teachers] treat you unfairly or negatively because of your ethnic background?” and “I feel that others behave in an unfair or negative way toward my ethnic group.” Higher scores indicated greater racial discrimination. Scores were taken from the 2‐year follow‐up. Moderator School Climate The school environment subscale from the School Risk and Protective Factors protocol (SRPF; Arthur et al. 2007) was used to assess school climate, which was operationally defined as feeling valued and safe in one's school environment. This subscale assessed participants' perceptions of their school experiences. Participants indicated their level of agreement to various statements, ranging from 1 (definitely not true) to 4 (definitely true). Items were summed on a scale of 5‐24 (M = 19.69; SD = 3.02) if the youth responded to at least five of the six items. Sample items included: “My teacher(s) notices when I am doing a good job and lets me know about it” and “I feel safe at my school.” Higher scores indicated more positive school climates. Scores were taken from the 2‐year follow‐up. Perhaps due to the complexity and breadth of school climate as a construct (Cohen et al. 2009) as well as the brief nature of this six‐item subscale, reliability (α = 0.68) was slightly below the ideal threshold of α = 0.70. Sociodemographic Controls The present analyses employed a theoretically guided, parsimonious approach to the inclusion of control variables that involved selecting three sociodemographic covariates that have been consistently associated with discrimination experiences and adolescent adjustment (Benner et al. 2018; Seaton et al. 2010; Visser et al. 2021; Witherspoon et al. 2016). This approach reflects recommendations to avoid overcontrol bias, which can diminish meaningful associations when covariates are included without theoretical justification (Rohrer 2018; Spector and Brannick 2011). Age Participants' ages (months; M = 143.95; SD = 7.85) were reported at the time of each assessment. Ages from the 2‐year follow‐up were controlled for. Gender Participants reported their gender identity at baseline. It was coded as 0 (male) and 1 (female). Youth who identified as transgender were excluded from this subsample due to their potentially unique experiences with discrimination and mental health. Area Deprivation The Area Deprivation Index (ADI; Kind et al. 2014) was utilized to capture numerous potentially influential factors and to simplify models. This census‐block‐level indicator is weighted and incorporates data regarding income, education, employment, and housing quality. The subsample's range was 6.47‐125.75 (M = 108.86; SD = 13.46). Higher values indicated greater deprivation. This score was taken from baseline. Analytic Plan Data were cleaned in SPSS 27 and R, and the predictor and moderator variables were centered. Patterns of missing data were inspected and deemed to be not missing completely at random using Little (1988) criteria, and 19 patterns emerged (p < 0.001). Overall rates of missingness were low, with the highest being the ADI with 10.6% and the remaining variables all below 10%. Missingness was addressed using R's mice package to impute missing values (van Buuren and Groothuis‐Oudshoorn 2011). The imputation process consisted of 50 imputations, each with 10 iterations. Prior to model estimation, regression diagnostics were examined for linearity, homoscedasticity, normality of residuals, multicollinearity, and influential observations. Residuals showed some departure from normality, a known issue when modeling these outcomes (e.g., de Vries et al. 2020; Pedersen et al. 2021). To address this, we compared our findings with models estimated using structural equation modeling with robust maximum likelihood (MLR; see Supplement Table 1). These SEM models handled missing data using full‐information maximum likelihood (FIML). This approach is more robust to violations of normality, particularly with larger sample sizes like in the current study (Enders 2001; Lai 2018). Results were consistent across analytic approaches, suggesting the observed non‐normality did not significantly bias results. No other assumptions were violated. Four linear regression models were run, using pooled estimates, to answer the research questions: two without the interaction term to test the main effects and two with the interaction term to test potential moderation. Moderation was assessed via the interaction term. Interactions were probed using simple slopes and Johnson–Neyman plots to visualize the regions of significance. Findings Descriptives As seen in Table 1, adolescents' frequency of racial discrimination was low, and their perceptions of school climate were generally positive. Self‐reported internalizing and externalizing symptoms were also low. Area deprivation scores were high but with some variance (SD = 13.46). Table 1 also highlights the correlational relations between variables. Racial discrimination was significantly correlated with negative school climate and higher levels of internalizing and externalizing symptoms 1 year later. In contrast, positive school climate was significantly correlated with lower levels of internalizing and externalizing symptoms at the 3‐year follow‐up. Internalizing and externalizing symptoms were significantly positively correlated with one another. Adolescent girls reported higher levels of internalizing symptoms. Higher levels of externalizing symptoms were correlated with higher area deprivation, which itself was associated with younger ages. Finally, cross‐item correlations were calculated to assess potential content overlap between the racial discrimination and school climate measures. However, no evidence of redundancy emerged (see Supplement Table 2). Table 1Means, standard deviations, and correlations. M SDRDSCINTEXTAgea GenderADIRD1.290.48—SC19.693.02−0.21c —INT1.682.120.19c −0.10c —EXT2.122.220.18c −0.19c 0.46c —Agea 143.957.850.02−0.120.010.03—Gender——−0.02−0.020.12c 0.03−0.02—ADI108.8613.460.06−0.040.050.07b −0.09c 0.00—Abbreviations: ADI, area deprivation index; EXT, externalizing; INT, internalizing; RD, racial discrimination; SC, school climate. a Age in months. b p < 0.05. c p < 0.01. Means, standard deviations, and correlations. Abbreviations: ADI, area deprivation index; EXT, externalizing; INT, internalizing; RD, racial discrimination; SC, school climate. Age in months. p < 0.05. p < 0.01. Main and Moderating Effects Four multiple regression models were run to test main and moderating effects on internalizing and externalizing behaviors (Table 2; Supplement Table 3). These models utilized pooled data from the imputed mice sets. Model 1, which examined the main effects of racial discrimination and school climate on later internalizing behaviors, was statistically significant, F(5, 20047) = 13.29; p < 0.001, and explained 6% of the variance (R 2 = 0.06). Racial discrimination (β = 0.18, p < 0.001) and school climate (β = −0.06, p = 0.05) both significantly predicted later internalizing behaviors, as did gender (β = 0.26, p < 0.001). Model 2, which examined the main effects of racial discrimination and school climate on later externalizing behaviors, was also statistically significant, F(5, 19731) = 10.89; p < 0.001), and explained 5.7% of the variance (R 2 = 0.06). Racial discrimination (β =0.13, p < 0.001) and school climate (β = −0.16, p < 0.001) both significantly predicted later externalizing behaviors. Table 2Moderation models.Models 3 and 4InternalizingExternalizingPredictorEstimateSEβEstimateSEβ(Intercept)−0.271.26−0.140.451.32−0.04Racial discrimination0.72§ 0.130.160.51§ 0.140.11School climate−0.04* 0.02−0.05−0.11§ 0.02−0.15Interaction−0.09† 0.04−0.06−0.11‡ 0.04−0.07Age0.010.010.040.010.010.02Gender: female0.55§ 0.120.260.110.120.05ADI0.000.010.000.01* 0.010.05* p < 0.1. † p < 0.05. ‡ p < 0.01. § p < 0.001. Moderation models. p < 0.1. p < 0.05. p < 0.01. p < 0.001. Model 3 introduced an interaction term between racial discrimination and school climate on later internalizing behaviors. The model was statistically significant, F(6,22640) = 10.36; p < 0.001, and explained 6.5% of the variance (R 2 = 0.07). School climate moderated the impact of racial discrimination on later internalizing behaviors (β = −0.06, p = 0.02). The effect size of the interaction was modest but indicated that the association between racial discrimination and internalizing symptoms was weaker in more positive school climates. Simple slopes analysis (Table 3) indicated that the association between racial discrimination and later internalizing behaviors was strongest at low levels of school climate (B = 0.90, SE = 0.14, p < 0.01) and weakest at high levels (B = 0.40, SE = 0.18, p = 0.02). Gender continued to significantly predict later internalizing behaviors (β = 0.26, p < 0.001). No other controls were significant. The interaction is plotted in Figure 1a using a randomly selected imputation. The pattern reaffirmed that high levels of positive school climate attenuated the relation between racial discrimination and later internalizing behaviors. Further analysis via a Johnson‐Neyman interval (Figure 1b) revealed that when school climate values were outside the interval [3.44, 37.58], the slope of racial discrimination was statistically significant. The observed range of values for school climate was −13.69, 4.31. Table 3Simple slopes.Moderator LevelInternalizingExternalizingEstimateSEEstimateSEM − 1 SD0.90*** 0.140.80*** 0.14M0.65*** 0.120.46*** 0.13M + 1 SD0.40* 0.180.120.18* p < 0.05**p < 0.01*** p < 0.001. Simple slopes. p < 0.05 **p < 0.01 p < 0.001. (a) Moderation plot (internalizing). (b) Johnson–Neyman plot (internalizing). Model 4 introduced an interaction term between racial discrimination and school climate on later externalizing behaviors. Model 4 was statistically significant, F(6,16991) = 8.47; p < 0.001, and explained 6.5% of the variance (R 2 = 0.07). School climate moderated the impact of racial discrimination on later externalizing behaviors (β = −0.07, p = 0.01). The effect size of the interaction was modest but indicated that the association between racial discrimination and internalizing symptoms was weaker in more positive school climates. Simple slopes analysis (Table 3) indicated that the association between racial discrimination and later externalizing behaviors was strongest at low levels of school climate (B = 0.80, SE = 0.14, p < 0.001) and weakest at high levels (B = 0.12, SE = 0.18, p = 0.50). No other controls significantly predicted later externalizing behaviors. The interaction is plotted in Figure 2a using a randomly selected imputation, and simple slopes are provided in Table 3. The pattern reaffirmed that high levels of positive school climate attenuated the relation between racial discrimination and later externalizing behaviors. Further analysis via a Johnson‐Neyman interval (Figure 2b) revealed that when school climate values were outside the interval [1.50, 12.60], the slope of racial discrimination was statistically significant. The observed range of values for school climate was −13.69, 4.31. (a) Moderation plot (externalizing). (b) Johnson–Neyman plot (externalizing). Discussion Racial discrimination is an adverse childhood experience that increases risk for internalizing and externalizing behaviors among youth (Bernard et al. 2022). Identifying factors that may protect youth from the harmful effects of discrimination is critical, especially for Black youth in the U.S. School climate is one such promising factor, given the amount of time Black adolescents spend in a school setting and the numerous benefits associated with it. Using data from the Adolescent Brain Cognitive Development (ABCD) study, the current study examined whether positive school climates mitigate the effect of racial discrimination on later internalizing and externalizing behaviors among Black early adolescents. Descriptive data revealed relatively low levels of racial discrimination, which contrasts with some previous research (Umaña‐Taylor 2016) and may reflect the developmental moment during which youth were assessed. Rates of racial discrimination often increase during the middle school years (Hughes et al. 2016), suggesting that we might expect higher levels of discrimination as the participants enter middle adolescence. These low levels could also indicate measurement constraints, as early adolescents may underreport or fail to recognize subtle forms of discrimination (Wray‐Lake et al. 2023). Meta‐analytic work has highlighted shortcomings of adolescent discrimination measures, indicating that fewer items, shorter response scales, and longer retrospective reference periods could all contribute to underreported discrimination experiences (Benner et al. 2022). Although Phinney's Perceived Discrimination Scale seems to have an adequate number of items and a sizable Likert range, its ambiguous reference period may have led to lower mean scores (Benner et al. 2018; Goosby et al. 2018). When discrimination did occur, it was associated with greater internalizing and externalizing behaviors. This pattern of results aligns with past research linking discrimination to poorer mental health outcomes and delinquency (e.g., Burt et al. 2012; Hughes et al. 2016), continuing to highlight a need to reduce the frequency of racial discrimination experienced by Black youth. Consistent with prior research (Morris et al. 2020), positive school climate appeared to serve a protective function. We found that as perceptions of school climate grew more positive, the impact of racial discrimination on Black adolescents' internalizing and externalizing behaviors weakened, albeit with a small effect size. Indeed, the overall variance explained in both moderation models as well as the individual moderator effect sizes were modest; however, this should not discount their meaningfulness. Even small effect sizes, such as those seen here, can produce consequential changes as they accumulate over time, particularly in the context of a dynamic construct like school climate (Funder and Ozer 2019), and meta‐analytic work on other moderators of this relation, such as ethnic/racial identity, has found similar effect sizes (Yip et al. 2019). Our findings, therefore, support and extend Ransome and colleagues' (2023) social connectedness buffering hypothesis. The direction of the interactions indicates that more positive school climates may attenuate the associations between racial discrimination and both internalizing and externalizing behaviors. Specifically, associations between racial discrimination and both outcomes were stronger among adolescents reporting less positive school climates and weakened when perceptions were more positive. The Johnson‐Neyman analyses further indicated that these associations lost their statistical significance at sufficiently high levels of positive school climates. Thus, by providing students with a nurturing environment in which they feel safe, schools can interrupt the adverse effects of racial discrimination. These findings add to a rich body of literature on school climate (e.g., Bottiani et al. 2020; Griffin et al. 2020; Morris et al. 2020; Wang et al. 2020) and highlight the importance of fair and equitable learning environments, co‐constructed by teachers and peers alike, that foster a sense of belonging in students. This is particularly critical considering the unique developmental context of adolescence, particularly for Black youth. The potential for schools to offer a protective space for Black adolescents as they continue to develop their sense of identity is yet another beneficial effect of positive school climates. More broadly, findings support Huguley et al. (2022) argument for envisioning schools as places of restoration rather than punishment. Although our study leverages data from a large, geographically diverse sample of youth and provides important new information about the protective role of positive school climates in Black early adolescents' well‐being, several methodological constraints are worth noting. First, the measure available in the ABCD study for racial discrimination contained multiple items referencing schools specifically, thus overlapping with the school climate measures. Despite this content overlap, the constructs remain empirically and theoretically distinct. Cross‐item Spearman correlations indicated that associations between similar items were small, suggesting that redundancy was not a statistical concern. Conceptually, the racial discrimination items capture experiences of targeted unfair treatment and a sense of a lack of belonging due to one's race in general as well as in the school setting. In contrast, school climate reflects adolescents' broader perceptions of safety, belonging, and relational support within the school environment. Future research may also benefit from distinguishing discrimination occurring within schools from broader experiences of racial discrimination. For example, Gale and Dorsey (2020) examined in‐school and general racial discrimination separately and found unique contributions of the former, highlighting the value of considering the context in which discriminatory experiences occur. Second, the available school climate measure demonstrated lower‐than‐ideal reliability, aligning with prior work using this particular measure (Gonzalez et al. 2021). School climate is a multidimensional construct (Thapa et al. 2013), and the available six‐item subscale likely failed to fully capture its complexity. Future work should use more comprehensive, multidimensional measures of school climate that can distinguish among its various domains. Doing so may help identify which aspects of school climate are most protective and, in turn, provide more specific points for intervention. Finally, the ABCD study is rich with numerous variables, some of which may play a role in youths' internalizing or externalizing behaviors, such as school racial composition (Morris et al. 2020) or school suspensions (Wang et al. 2024). To maintain a parsimonious analytic approach, the current study included only a limited set of covariates selected based on prior literature. Although this approach reduced the risk of overadjustment and unnecessary model complexity, other contextual factors may also help shape the associations examined here. Future research should investigate these additional factors while also addressing the measurement limitations noted above. Beyond these methodological considerations, the longitudinal structure of the ABCD study provides important opportunities to examine how experiences of racial discrimination and school climates change over time. Relatedly, although the current investigation focused on the protective role of positive school climates, other aspects of school environments, such as school‐based racial discrimination and disciplinary inequities, may harm Black adolescents (e.g., Cooper et al. 2022) and warrant continued scrutiny. Implications Racial discrimination is a race‐based stressor that has lasting adverse effects on Black adolescents' behavioral and mental health (Benner et al. 2018; English et al. 2020), and positive school climate attenuates these adverse effects. It is important to appropriately assess for these experiences not only in clinical settings, but also in settings where adolescents spend most of their time, such as in schools. The 33‐item Expanded Family Adverse Childhood Experiences Scale (ACEs; Karatekin and Hill 2019), a commonly used tool to screen for stressful life experiences, has two items that assess discrimination (e.g., being hit or attacked because of your skin color, religion, or where your family comes from; feeling discriminated against because of these characteristics). Given the prevalence of racial discrimination for Black youth and the overlap between some racial discrimination and school climate items, adding a third item that assesses for school‐based, non‐physical racial discrimination may provide useful information about both constructs. The findings of the present study also point to the need to implement strategies to reduce Black adolescents' experiences of racial discrimination and to improve perceptions of school climates. Schools may benefit from implementing empathy‐based training and interventions to reduce racial discrimination. Programs that promote empathy reduce implicit biases (e.g., Whitford and Emerson 2019) and negative intergroup attitudes (e.g., Beelmann and Heinemann 2014), both of which contribute to discrimination. Consistent with the social connectedness buffering hypothesis, schools may also benefit from implementing practices that improve Black adolescents' school connectedness. For instance, Afrocentric‐infused curriculum and student‐involved school reform efforts have improved students' school connectedness (Voight and Nation 2016). Black adolescents' exposure to Afrocentric‐infused curriculum may provide necessary opportunities for them to see parts of themselves represented in their class material. In addition, the importance of having their voices included in school reform efforts may be particularly salient as adolescents may otherwise feel that their voices are silenced. Although schools may provide an optimal setting for such practices, it is important to not put the impetus of interrupting racism solely on schools. Other valuable assets for youth are their families and the broader communities in which they are embedded. By working to create a united front of schools, families, and communities, we can work to further strengthen Black adolescents' well‐being. Conclusion Consistent with prior research, these findings suggest that whereas racial discrimination is a risk factor, positive school climate is a protective factor for Black adolescents' internalizing and externalizing behaviors. Importantly, this study extends the existing literature by providing evidence for the protective role of positive school climate in decreasing the impact of racial discrimination on later internalizing and externalizing behaviors. Given the risks associated with the discrimination that Black adolescents often experience, these findings emphasize a pressing need to both decrease adolescents' experiences of racial discrimination and to improve their perceptions of school climates. Addressing both by implementing school‐based practices may be advantageous, as adolescents spend a great deal of time in school. Future research should focus on appropriately screening for racial discrimination and assessing for different aspects of school climate, especially considering the potential overlap of items. Identifying other potential protective factors, such as those in the family and community contexts, is also necessary. Author Contributions E. Cole Green: conceptualization, formal analysis; writing – original draft preparation. Jia Yan: formal analysis; writing – review and editing. Jasmine N. Coleman: writing – review and editing. Elizabeth I. Johnson: writing – review and editing. Ethics Statement Our research was conducted in accordance with the American Psychological Association's ethical standards pertaining to research and publication. This manuscript is based on de‐identified data from the ABCD study that have been deposited in the National Institute of Mental Health Data Archive (NDA) and made available to the authors via a Data Use Certification (DUC) agreement. Work with the data has been certified by the lead author's Institutional Review Board as meeting the NIH criteria for Exemption 4: “study of data if publicly available or recorded such that subjects cannot be identified.” Conflicts of Interest The authors declare no conflicts of interest. Supporting information Supporting File Acknowledgements Data used in the preparation of this article were obtained from the Adolescent Brain Cognitive Development™ (ABCD) Study, held in the NIH Brain Development Cohorts Data Sharing Platform. This is a multisite, longitudinal study designed to recruit more than 10,000 children aged 9–10 and follow them over 10 years into early adulthood. The ABCD Study® is supported by the National Institutes of Health and additional federal partners under award numbers: U01DA041048, U01DA050989, U01DA051016, U01DA041022, U01DA051018, U01DA051037, U01DA050987, U01DA041174, U01DA041106, U01DA041117, U01DA041028, U01DA041134, U01DA050988, U01DA051039, U01DA041156, U01DA041025, U01DA041120, U01DA051038, U01DA041148, U01DA041093, U01DA041089, U24DA041123, U24DA041147. A full list of supporters is available at Federal Partners – ABCD Study. ABCD Consortium investigators designed and implemented the study and/or provided data but did not necessarily participate in the analysis or writing of this report. This manuscript reflects the views of the authors and may not reflect the opinions or views of the NIH or ABCD Consortium investigators. The ABCD data repository grows and changes over time. The ABCD data used in this report came from https://doi.org/10.82525/wjkw-bp63. DOIs can be found at https://www.nbdc-datahub.org/abcd-study. Additional support for this work was made possible from NIEHS R01‐ES032295 and R01‐ES031074. Data Availability Statement This manuscript is based on de‐identified data from the ABCD study that have been deposited in the National Institute of Mental Health Data Archive (NDA) and made available to the authors via a Data Use Certification (DUC) agreement. References Associated Data Supplementary Materials Supporting File Data Availability Statement This manuscript is based on de‐identified data from the ABCD study that have been deposited in the National Institute of Mental Health Data Archive (NDA) and made available to the authors via a Data Use Certification (DUC) agreement.