A systematic review and meta-analysis of growth mindset interventions: For whom, how, and why might such interventions work?

53 independent samples testing distinct interventions, published 2002-2020, randomised designs only

Mixed ages; academic, mental-health and social-functioning outcomes.

The favourable meta-analysis, and the reason it is favourable is the conditioning. It reports no unconditional average treatment effect on achievement; the headline number is estimated jointly on two moderators - a targeted or at-risk subsample AND high implementation fidelity. No publication-bias-corrected estimate is reported. Graded C because a conditional estimate with no bias correction is a weaker object than Macnamara and Burgoyne's quality ladder. Independence marked unclear - the authors are not Dweck or Yeager but several have authored their own mindset interventions and earlier mindset meta-analyses.

Academic achievement in targeted subsamples with high fidelity d = 0.14 (95% CI 0.06 to 0.22), with a 95% PREDICTION interval of -0.08 to 0.35 - meaning a new intervention of this kind is not predicted to do anything reliably. Mental health under the same conditioning d = 0.32 (0.10 to 0.54, PI 0.07 to 0.57). Social functioning, unmoderated because too few studies, d = 0.36 (0.03 to 0.68, PI -0.50 to 1.22).

Low - randomised trials only.

Same literature as Macnamara and Burgoyne 2023, opposite headline, driven by analytic choices rather than by different data. The two appeared in the same issue of Psychological Bulletin, which is the clearest signal available that the field's disagreement is about method, not evidence.