Updating and expanding systematic reviews and meta-analyses on the effects of school-wide positive behavior interventions and supports

29 studies (7 randomised controlled trials, 22 quasi-experimental group designs) from the United States and Europe

School-level outcomes; K-12.

Robust variance estimation meta-analysis of all available school-LEVEL outcomes, including unpublished studies — a real strength against publication bias. Graded C because three quarters of the included evidence is quasi-experimental, because school-level aggregate outcomes cannot separate changed behaviour from changed recording practice, and because the discipline outcome that SWPBIS moves is also the outcome SWPBIS trains staff to record differently.

Statistically significant reductions in school discipline and statistically significant increases in academic achievement, with effect sizes described as small to medium. Not all individual studies reported improved academic achievement — the achievement literature is mixed, and the pooled positive rests on aggregation across designs of unequal quality.

Low for the randomised subset; the quasi-experimental majority is subject to school selection into adoption, which is a selection problem rather than a heredity problem.

Pools the Bradshaw and Horner trials with 22 non-randomised studies, so it does not supply independent confirmation so much as it summarises a field whose randomised core is small and developer-adjacent.