Purpose: To examine effects of a 10-week after-school physical activity (PA) program on academic performance of 6- to 12-year-old African American children with behavior problems. Methods: Participants were randomized to PA (n = 19) or sedentary attention control (n = 16) programs. Academic records, curriculum-based measures, and classroom observations were obtained at baseline, postintervention, and/or follow-up. Mixed models tested group × time interactions on academic records and curriculum-based measures. One-way analysis of variance or Kruskal–Wallis tested for differences in postintervention classroom observations. Results: Intent-to-treat analyses demonstrated a moderate effect within groups from baseline to postintervention on disciplinary referrals (PA: d = −0.47; attention control: d = −0.36) and a null moderate effect on academic assessments (PA: d = 0.11 to 0.36; attention control: d = 0.05 to 0.40). No significant group × time interactions emerged on direct academic assessments (all Ps ≥ .05, d = −0.23 to 0.26) or academic records (all Ps ≥ .05, d = −0.28 to 0.16). Classroom observations revealed that intervention participants were off-task due to moving at twice the rate of comparative classmates (F = 15.74, P < .001) and were off-task due to talking 33% more often (F = 1.39, P = .257). Conclusion: Academic outcome improvements were small within and between groups and did not sustain at follow-up. Academic benefits of after-school PA programs for children with attention-deficit hyperactivity disorder and/or disruptive behavior disorders were smaller than neurobiological, behavioral, and cognitive outcomes as previously reported.