Efficacy of the Responsive Classroom Approach: Results From a 3-Year, Longitudinal Randomized Controlled Trial

24 schools randomised; 2,904 children followed from the end of grade 2 to grade 5

United States; elementary schools.

A three-year cluster-randomised efficacy trial of the Responsive Classroom approach. Independence is marked unclear - the evaluating laboratory has a long research relationship with the programme's publisher, and a financial relationship could not be confirmed either way. The distinction that matters for reading this study is between its intention-to-treat result and its mediation result.

The intention-to-treat result is null - students at schools assigned to Responsive Classroom did not outperform control students in mathematics or reading achievement. The positive findings are MEDIATION estimates conditional on how much teachers actually used the practices - effect sizes of 0.26 for mathematics and 0.30 for reading on the practice-use-to-achievement relation, and 0.89 for lower-performing students in mathematics. Those are correlational estimates within a randomised trial, not causal intention-to-treat effects, and teachers who use a programme more are not a randomly selected group.

Low for the randomised contrast; the mediation analysis is not protected by randomisation.

Unreplicated at this scale. Its structure - null main effect, positive implementation-conditional effect - is the standard shape of a defended null in this literature.