<p>This study employed a cluster randomized controlled trial design to investigate the impact of BEST in CLASS-Elementary, a Tier 2 intervention that trains and coaches teachers to support students with unwanted problem behavior, on teachers’ implementation of evidence-based practices, self-efficacy, burnout, and attributions for problem behavior. A total of 114 teachers (97% female; 70% White, 22% Black) and 237 focal students (71% male; 44% Black, 32% White, 8% Hispanic/Latinx) in two southeastern states participated in the study. Results indicated that compared to the comparison group, BEST in CLASS-Elementary teachers increased Adherence (<i>d</i> = 0.68) and Competence (<i>d</i> = 0.43) from pretest to posttest. BEST in CLASS-Elementary participants reported higher efficacy for Student Engagement (<i>d</i> = 0.34), Instructional Practices (<i>d</i> = 0.34), and Classroom Management (<i>d</i> = 0.36). BEST in CLASS-Elementary did not produce statistically significant changes on any of the burnout or attribution scales apart from the Purposeful dimension on the attribution measure (<i>d</i> = − 0.33). Findings suggest that BEST in CLASS-Elementary is effective at increasing teacher use and quality of delivery of BEST in CLASS-Elementary practices with focal students, with positive effects for teachers’ sense of self-efficacy as well as to what they attribute the purpose of focal student problem behavior. Findings, implications, and future research are discussed.</p>