Background/Context: Well-documented achievement gaps persist between members of disadvantaged social groups and their more advantaged peers (Domina et al., 2018; Harackiewicz et al., 2014; Musu-Gillette et al., 2017; Stoet & Geary, 2013). Certainly, systemic and structural factors explain most of these gaps; yet, there is another contributor "down on the ground" that plays out every day in the classroom: social identity threat (Steele, 1997). Social identity threat can impair students' performance due to the distraction and stress they cause in high-stakes, evaluative environments (Cohen & Sherman, 2014). Because social identity threat takes a psychological form, it is possible to alleviate it with strategies that address how students think and feel in the classroom. One such strategy is known as "values affirmation," a brief writing activity in which students write about cherished personal values that transcend the evaluative situation (Cohen et al., 2006, 2009). Purpose/Objective/Research Question: Despite the breadth of empirical evidence, no meta-analysis has assessed the overall effectiveness of values affirmation in improving the real-world academic achievement of identity-threatened students in educational settings. The primary aim of the present meta-analytical review is to estimate this average treatment effect. We also assess whether, as predicted, the effect of values affirmation is apparent for members of identity-threatened groups and not for members of identity-nonthreatened groups. The second research aim is to explore whether values affirmation is heterogeneous or homogeneous in its effects, and, if heterogeneous, the variables that predict when, where, and for whom it yields the most benefit. Criteria for inclusion of studies: Design. To be eligible for inclusion, a study had to randomly allocate students or groups of students either to a values affirmation treatment group or to a control group. Participants. The student participants could be drawn from any grade level ranging from kindergarten to 12th grade (primary, presecondary, and secondary schools) to colleges, universities, and graduate schools (tertiary education) and other educational institutions. Intervention. The study had to test a values affirmation intervention, which required three key elements. First, students had to be presented with a list of personal values (value menu). Next, they had to be given a chance to write a brief essay (writing task) about their most important value, typically a value in a domain unrelated to school (value domain) (Harris & Napper, 2005; McQueen & Klein, 2006). Control. The study had to have a control condition, specifically one that was neutral, typically accomplished through a non-affirming writing activity. Outcomes: The primary outcome was grade point average, or GPA. Secondary outcomes encompassed other official measures of academic achievement, such as standardized test scores, exam performance, classroom grade, and percentile rank in class. Data Collection: Studies were identified through a database search of published articles using wildcard search terms. Data analysis: We calculated two types of effect sizes. The first was the raw effect size unadjusted for students' baseline performance or other preintervention variables. The second was the covariate-adjusted effect size adjusted for students' baseline performance and other preintervention variables. The two types of affirmation effect sizes were calculated for both identity-threatened students and identity-nonthreatened students. We built two sets of models to estimate the average treatment effect. First, we estimated the weighted average affirmation effect separately for threatened students and nonthreatened students at the level of project cluster. We also conducted multilevel meta-analyses to model the nested organization of effect sizes by project clusters with effect sizes (Level 1) nested in project clusters (Level 2). We identified four sets of moderators from past studies: student characteristics, implementation fidelity, social context, procedural characteristics, and tested how they predicted the average treatment effect. Findings/Results: After a systematic search yielded 58 relevant studies, multilevel analyses identified a moderate to large overall affirmation effect for identity-threatened students (Hedges' g = 0.15; see Figure 1) against the empirical benchmark (Bakker et al., 2019; Kraft, 2020), but not for identity-nonthreatened students (Hedges' g = 0.01; see Figure 2). Heterogeneity in the affirmation effect was moderate to high for identity-threatened students, with larger effect sizes associated with (1) a larger covariate-controlled achievement gap between nonthreatened and threatened students in the control condition, suggestive of psychological underperformance (b = 0.28, SE = 0.07, 95% CI [0.13, 0.43], p < 0.001), (2) the availability of financial resources in school (b = 0.59, SE = 0.34, 95% CI [-0.08, 1.27], p = 0.083), (3) more distal performance outcomes (b = 0.01, SE = 0.002, 95% CI [0.003, 0.01], p = 0.001), and (4) the presentation of values affirmation as a normal classroom activity rather than a research study or a nonnormal classroom activity (b = 0.16, SE = 0.08, 95% CI [0.01, 0.32], p = 0.038). Affirmation also appears to work best when it is delivered as a normal classroom activity and where identity threat co-occurs with resources for improvement and time to await cumulative benefits (see Figure 3). Conclusions: Our results suggest that values affirmation intervention is one cost-effective tool to close achievement gaps in educational contexts. However, values affirmation, like other social-psychological interventions, should not replace larger structural and systemic efforts to close achievement gaps. Rather, they should complement them. Aspiration, hope, a sense of belonging, and other precious psychological states do not bear fruit by themselves. They require structural pathways and material resources for their benefits to come to fruition.