The present study evaluated the effects of the On-Task in a Box intervention on student on-task behavior when used as a class-wide intervention. The intervention package includes self-monitoring, video modeling, and reinforcement contingency components. A multiple baseline design across three elementary classrooms was used to determine the effects of the intervention on student behavior. Results indicated immediate increases in on-task behavior in all classrooms. Calculation of nonoverlap of all pairs' effect sizes indicated strong intervention and maintenance effects on on-task behavior of students in all classrooms. Social validity data were also collected, which indicated adequate validity of the intervention. Results of the study suggest that the intervention package may be effective for improving class-wide levels of on-task behavior.