Content area
Abstract
Schools have long sought measures to improve student behavior. Positive Behavior Support (PBS) aims to provide students with clearly defined, carefully taught, and consistently rewarded behavioral expectations, as well as punishers consistent with the level and function of misbehavior. School-wide PBS addresses student behavior in the public areas of the school, such as the hallways, cafeteria, playground, and so on, through PBS practices and the use of office discipline referrals for evaluation and decision making. This study assessed the influence of school-wide PBS in middle schools using student surveys collected by the Oregon Healthy Teens project. The research hypotheses posited that (a) school-wide PBS would improve school context measures, such as teacher praise; (b) school-wide PBS would improve student target behaviors, such as delinquency and truancy; (c) school-wide PBS would not necessarily change measures of student competence, such as social skills; (d) student competence would moderate the effect of school-wide PBS on student target behaviors; and (e) if the target behaviors were associated with school-wide PBS, then school context measures would mediate those effects.
The analyses compared student reports of school context, target behaviors, and competence in a quasi-experimental design with 103 randomly selected middle schools across three years, among which 32 received training in school-wide PBS. Student survey data and office discipline referrals were also correlated with measures of school-wide PBS implementation collected in 19 schools. The longitudinal, multilevel statistical models accounted for the nested structure of the data, which included student, cohort, and school levels. The results failed to support the research hypotheses, as no tests of context, target behaviors, competence, or moderator effects obtained statistical significance. With no differences in target behaviors, mediational analyses were not applicable. A thorough appraisal of this investigation revealed that the lack of support for the research hypotheses likely arose from limitations of the quasi-experimental design as well as survey questions that inadequately measured the changes in behavior expected from the implementation of school-wide PBS.
