This project will: 1) evaluate the efficacy of a comprehensive, multi-grade, multi-year program, the Positive Action program (PA) designed to reduce student problem behaviors (including drug use and violence), disciplinary referrals and suspensions, and academic underachievement; 2) evaluate the effects of the program on the presumed/theoretical mediator variables; and 3) assess which of these changes in mediator variables actually mediate the effects of the program on behavior and school performance. The Hawaiian school district has agreed to implement a randomized trial of the PA program. We randomly assigned schools to PA or control conditions from strata matched on multiple indicators of risk. PA schools will receive the full PA program for four years (the 2002-03 through 2005-06 school years). This includes all materials for all students (at all grades), families, teachers, and principal, plus community kits; training for teachers and other interested staff; and continuous technical support and monitoring by a PA staff person. Data on behavior and academic achievement will be obtained from school archival records for all students for all years and, where appropriate for several years prior to the study. Students in grades 1 and 3 in the 2001-02 school year and their parents and teachers will also be followed through the project with surveys every year. The subject population will contain equal portions of males and females, and proportionate numbers of the major ethnic groups on Oahu (Hawaiian or part-Hawaiian, white, Asian, Other). Analyses will test for program effects on all hypothesized mediating variables and outcomes, and the hypothesized mediational pathways.
