The Evidence on Teaching

Examining the Effects of Schoolwide Positive Behavioral Interventions and Supports on Student Outcomes

Bradshaw CP, Mitchell MM, Leaf PJ · 2009

grade Brctdeveloper-involvedreplicated
Sample
37 elementary schools, group-randomised, 5-year longitudinal effectiveness trial
Population
Maryland elementary schools, US.
Design
The largest randomised effectiveness trial of school-wide PBIS and the source of its strongest behavioural claim. School-level randomisation, 5 years of follow-up, outcomes taken from administrative records (suspensions, office discipline referrals) rather than surveys — which is the reason this source carries weight in a literature dominated by teacher report. The evaluation team is closely tied to the SWPBIS dissemination network, hence developer-involved.
Key findings
Schools trained in SWPBIS implemented the model with high fidelity and showed significant reductions in student suspensions and office discipline referrals at school level. Academic achievement was examined and did not show the same clear benefit; the behavioural result is the durable one.
Genetic confound
Low — school-level randomisation.
Replication notes
Reproduced in the same trial's child-level analysis (Bradshaw, Waasdorp & Leaf 2012, Pediatrics) and broadly consistent with the pooled discipline reductions in Lee & Gage 2020. The achievement side of the claim is where the literature is thin and inconsistent.
DOI / URL
10.1177/1098300709334798

Effects

OutcomeMetricValueMeasureTimingVsHorizonClass
Student suspensions (administrative records), school levelsignificancesignificant reduction in SWPBIS schoolsstandardizedacross 5 school yearsbusiness-as-usualover-2yrbehaviour
Office discipline referrals (administrative records)significancesignificant reductionstandardizedacross 5 school yearsbusiness-as-usualover-2yrbehaviour
Academic achievementsignificanceexamined; not the clear positive that the discipline outcomes werestandardizedacross 5 school yearsbusiness-as-usualover-2yrdomain-skill

Cited by