I came across this new paper thanks to Dylan Wiliam. It will probably stir some debate, as always, when I discuss this theory. Carolina Gazmuri (2025) has just published a structured review on growth mindset interventions and their effects on academic achievement. The research seeks to answer the question: Do growth mindset interventions really improve student achievement? The idea, of course, is well known. If students believe their intelligence is malleable rather than fixed, they are more likely to learn. It sounds elegant, attractive, and—at least in theory—cheap. But does it really deliver?

What the strongest studies say

Gazmuri screened and evaluated 24 randomised controlled trials (RCTs). She focused strictly on interventions with school-age children and outcomes tied to academic performance. And here’s the sobering finding: the strongest studies—those with large samples, low attrition, and high-quality data—show effect sizes hovering between −0.01 and +0.065. In plain words, close to zero, as was demonstrated in earlier meta-analysis. This was also shown in a huge replication study.

When weak studies look impressive

It gets more interesting. Some of the studies with the biggest effects are precisely those with weaker designs, small samples, or high dropout rates. In contrast, when you zoom in on the most trustworthy research, the impact all but disappears. Add to this the issue of conflicts of interest. Several studies were conducted by researchers connected to organisations selling mindset programmes. Consequently, the picture becomes even murkier.

Implications for schools and policymakers

So what should schools or policymakers do? The review is clear: don’t invest heavily in mindset interventions with the expectation of raising test scores. At best, the evidence suggests very modest effects. That doesn’t mean growth mindset is meaningless—encouraging students to see ability as malleable may still be valuable for other reasons. If it can be done at negligible cost, it may be harmless or even slightly beneficial. But as a lever for boosting academic achievement, the promise looks overstated.

This reminds me of a broader pattern we often see in education research. An idea starts with a strong theoretical and intuitive appeal, gets supported by small or lab-based studies, and spreads quickly. This occurs through books, talks, and training packages, but then falters when rigorously tested at scale. Growth mindset, it seems, is following that script.

No magic bullet

For me, the key takeaway is not to abandon the concept altogether. Instead, approach it with caution. Beliefs matter, but they are not magic. As Gazmuri concludes, if schools want genuine gains in learning, they should think twice before betting on growth mindset interventions as their primary strategy.

Abstract of the review:

Encouraging the idea of a growth mindset in which students believe that they can improve their ability, as opposed to a fixed mindset, has been suggested as an effective and relatively cheap approach to improving student attainment at school. This paper offers a comprehensive review of the evidence from growth mindset interventions. After a rigorous search, screening and evaluation, the inclusion criteria led to 24 studies. All were randomised control trials (RCTs) focused on growth mindset of intelligence interventions for school-age children and included output measures for academic performance assessment. Their findings reveal that the strongest studies, characterised by larger sample sizes, minimal missing data and high data quality, exhibit null or very small effect sizes, ranging from Cohen’s d = −0.01 to +0.065. Additionally, certain findings raise concerns about a potential conflict-of-interest bias, suggesting that some negative or null results may remain unpublished. The review identifies four evaluations with a high degree of trustworthiness and non-conflict of interest. Among these, two studies indicate no discernible impact, while the other two show a very small impact. Given these findings, we found evidence that suggest that growth mindset interventions targeted for school-age students, do not have much or any relevant impact in academic achievement. It is therefore not advisable for schools, school districts or governments to allocate significant time or resources to the implementation of growth mindset interventions for school-age students, as the anticipated outcomes are likely to be either null or very modest. However, if there is an opportunity to implement such interventions at a minimal or negligible cost, or as part of another objective, it might be reasonable to proceed with them, considering the potential for a small positive impact.

Image: https://commons.wikimedia.org/wiki/File:Fixed_versus_growth_mindset.png