Practice Profile: Universal School-Based Prevention and Intervention Programs for Aggressive and Disruptive Behavior

Wilson and Lipsey (2007) analyzed 249 studies, published in 1960 or later, of school-based prevention programs and their effects on students' aggressive and disruptive behaviors. Eligible studies had to be reported in English; involve a school-based program for youth in pre–K through 12th grade; assess effects on at least one outcome related to aggressive, violent, or disruptive behaviors; and use at least one quasi-experimental design with either random assignment to conditions or reported initial equivalence.

Included studies came from journals (61 percent), dissertations or theses (30 percent), and other unpublished work (9 percent). Study participants were boys and girls predominantly ages 6–10 (43 percent), and ethnicities included white (33 percent), Black (25 percent), Hispanic (8 percent), other (2 percent), and mixed (4 percent). No information on the number of students in the treatment and control groups was reported. While various intervention formats were analyzed separately, the focus of this review was on 77 studies that examined universal programs.

Authors reported the effect size as a standardized mean difference with 95 percent confidence intervals. The inverse variance weights were computed using the subject-level sample size for each effect. Only one effect size from each subject sample was used in the analysis. Authors used a random effects analysis.

Park-Higgerson and colleagues (2008) analyzed 26 studies of school-based intervention programs on outcomes of aggressive and violent behaviors with students from grades 1–12. All of the studies used randomized controlled designs and were published between 1970 and 2004.

To be included, studies had to involve school-based programs designed to prevent aggression or violence in children and adolescents by providing education, improving social skills, or changing the environment; had to include students enrolled in grades 1–11; had to use a randomized assignment procedure for treatment and control groups; and had to examine aggression or violent behavior measures that were reported with necessary statistical information to measure effect. No demographic information was provided for students in the analyzed studies. Authors analyzed studies on five program characteristics related to program success: 1) the application of theory, 2) universal or selective programming, 3) single- or multiple-approach intervention, 4) characteristics of target population, and 5) type of instructor. For this review, the focus was on the seven studies that examined universal programs.

The overall effect size was calculated for the full sample. Pooled results were expressed as standardized mean differences with 95 percent confidence intervals. The analysis makes no mention of weighting.