Syntes: Gazmuri (2025) — in the most trustworthy school trials, growth-mindset interventions moved whole-sample achievement by about zero

Källa: Gazmuri, C. (2025). Can growth mindset interventions improve academic achievement? A structured review of the existing evidence. Review of Education 13(2), e70066. https://doi.org/10.1002/rev3.70066

Verified content

• [abstract] All included studies were RCTs of growth-mindset-of-intelligence interventions for school-age children with an academic outcome; the strongest studies (large samples, minimal missing data, high data quality) show effects of Cohen's d = −0.01 to +0.065.
• [full text] 29 papers met the criteria; 3 were excluded because they reported results only for a subgroup (at-risk or lower-achieving students), not the whole randomised sample: Paunesku et al. 2015, Yeager et al. 2016 and Yeager et al. 2019. 26 papers with 62 outcomes remained (the abstract says 24 studies). 14 outcomes rated 4, 16 rated 3.
• [full text] The five 4* studies: Outes-León et al. 2020, Peru (800 secondary schools; ITT maths d = 0.054, reading 0.04 at two months; 0.038 and 0.044 at 14 months); Rege et al. 2021, Norway (6,541 students; d = 0.025 on advanced-maths enrolment and passing); Zhou et al. 2023, rural China (1,680 Year 5 pupils; 0.03–0.06 on official exams); Foliano et al. 2019, England (101 primary schools, 5,018 Year 6 pupils; KS2 maths −0.01, reading 0.00); Ganimian 2020, Argentina (202 schools; maths 0.015, reading −0.008).
• [full text] Among 3* studies: Wanzek et al. 2021 (240 Grade 4 pupils with reading difficulties; reading plus mindset vs reading only: comprehension d = 0.03, oral reading fluency −0.15); Burnette et al. 2018 (222 rural low-income Grade 10 girls; GPA d = 0.07); Binning et al. 2019 GPA −0.09; Chao et al. 2017 (India) −0.13.
• [full text] The author flags studies with an apparent conflict of interest (for example Rege et al. 2021, Porter et al. 2022); among 3/4 outcomes these were not larger than the others.

Limits

• [full text] Whole-sample effects only; studies reporting only subgroup effects were excluded, so the review does not estimate the lower-achiever effect that C13 is about.
• [full text] Not preregistered; English-language studies only; the 400-student threshold for 4* is the author's judgement.
• [assessment] Mixes primary and secondary schools and very different interventions (one session to eight-week teacher programmes).

Why it is in the library

A recent, independent quality-graded review that puts the best school trials side by side. It agrees with C13's 'does little on average' clause and shows the lower-achiever clause rests on subgroup reports this review could not include.

Added 2026-10-18 by the project (weekly source sweep). [abstract] = from the published abstract; [full text] = read in the full text; [assessment] = the project's own assessment.