This paper addresses two major and potentially conflicting movements: the importance of diversity as both a conceptual and political issue and the rise of the evidence-based practice movement in education. This tension is particularly important when evaluating and reporting universal interventions because of their intended applicability across diverse groups of children and adolescents. This study contributes to this discussion through an analysis of published school-based universal social and emotional learning (SEL) intervention evaluations in terms of their theoretical and empirical attention to student diversity characteristics. We defined student diversity in terms of five characteristics: gender, race/ethnicity, SES, disability status, and sexual orientation/gender identity. We assessed how and when demographic characteristics were reported, how these characteristics were analyzed as moderators of program outcomes, and how differential effects based on diversity were incorporated into reported intervention generalizability discussions. Results showed that diversity characteristics were inconsistently reported across articles. Most studies did not test for moderating effects, but those that did found inconsistent effects across diversity characteristics. Further, conceptual and/or empirical support for conducting the moderation analyses was often not provided or sufficiently supported by previous literature or a hypothesis. This research highlights the need for a more nuanced understanding of how SEL program effects may be moderated by student demographic characteristics and suggests caution about the generalizability of the reviewed SEL programs across diverse groups of children and adolescents.