Systematic literature reviews have supported the effectiveness of behavioral Tier 2 interventions, with research being particularly plentiful in relation to Check-In/Check-Out (CICO) and social skills training (SST). Though findings indicate either approach is effective in isolation, a recent study suggested these approaches might be particularly effective when integrated. The purpose of the current study was to build upon this initial single-case design study with a larger sample within a randomized controlled trial. Participants included 91 elementary students who had been identified as being at risk of social-emotional and behavioral difficulties. Students were randomized into three intervention conditions: CICO only, SST only, and CICO + SST. Prior to intervention, information was collected regarding the function of student problem behavior and the extent of social skill deficits. Systematic direct observation data were then completed at pre- and post-test regarding student positive and negative social engagement. Multivariate general linear models were then conducted, with pre-test scores serving as covariates and intervention group and behavioral function serving as fixed factors. Results indicated that implementation of CICO, SST, and a combined CICO + SST was functionally related to a reduction in negative social engagement. Follow-up post hoc tests indicated that after adjusting for pre-test responding, the difference between conditions in effectiveness was moderated by behavioral function. Specifically, SST was less effective for students whose behavior functioned to escape social and academic situations. No such differences were noted between students whose behavior functioned to attain adult or peer attention. Implications for practice, methodological limitations, and directions for future research are reviewed.