UC Riverside

Using a sample of sixth graders in 11 public schools in a large Southwestern city, this longitudinal study examined how a model substance use prevention program, keepin’ it REAL, that was implemented in 7^th grade, influenced three other problem behaviors (fighting, weapon use, stealing), measured in 8^th grade. Using a non-equivalent control group design, we compared 259 students in the intervention to 322 students in a treatment-as-usual condition. At baseline, 37% of the sample reported fighting in the last 30 days; 31% reported stealing in the last 30 days, and 16% reported using a weapon in the last 30 days. Regression analyses adjusted for students nested in schools through multi-level modeling and for missing data through multiple imputation. We found that at posttest the rates of all three behaviors were lower in the intervention group than the control group at posttest: 35% versus 37% got into a fight in the last 30 days; 24% versus 31% stole something in the last 30 days; and 16% versus 25% used a weapon in the last 30 days. The program impact for fighting and stealing was not statistically significant and involved minimal effect sizes. The program impact for weapon use was not statistically significant but had an effect size comparable to that for other problem behavior interventions. Promoting positive development via life skills may be a key to broadening program impact.