Growth Mindset, Revisited: What the Evidence Actually Shows

Introduction

Few ideas have swept through education like growth mindset. Popularized by the Stanford psychologist Carol Dweck, the core claim is intuitive and appealing: students who believe their abilities can be developed through effort tend to do better than students who believe ability is fixed. ¹ Schools adopted it enthusiastically, sometimes reducing it to posters and slogans.

Then came the reckoning. As researchers ran larger and more rigorous studies, the picture got more complicated. The honest summary today is not that growth mindset is a myth, and not that it is a miracle. It is that the effect is real, small, and conditional, and that it works mainly for the students who need it most. ³

This article walks through what the evidence actually shows, because the nuance is the useful part. It also makes a case that matters for anyone serious about developing thinking: a belief that ability can grow is only valuable if it comes with a genuine way to grow it. That second half is where Arq.training lives.

The original claim

Dweck's distinction is between a fixed mindset, the belief that intelligence and ability are static, and a growth mindset, the belief that they can be developed through effort, strategy, and learning. ¹ The theory predicts that students with a growth mindset respond to difficulty by trying harder and adapting, while those with a fixed mindset are more likely to interpret struggle as evidence of low ability and give up.

There was real research behind the idea. A well-known longitudinal study by Lisa Blackwell, Kali Trzesniewski, and Carol Dweck followed students across the transition to junior high and found that those holding a growth theory of intelligence showed an upward trajectory in math grades, while a brief intervention helped reverse a decline in a small treatment group. ²

The trouble was not the original studies. It was what happened next: a compelling idea, often stripped of nuance, was scaled across thousands of schools faster than the evidence could keep up.

What the meta-analyses found

When researchers pooled the accumulating studies, the average effects turned out to be modest. A pair of meta-analyses by Victoria Sisk and colleagues found a weak correlation between mindset and academic achievement of about 0.10, and an average effect of mindset interventions of roughly 0.08 standard deviations, drawing on data from hundreds of thousands of students. ³

A later and more stringent review by Brooke Macnamara and Alexander Burgoyne reached an even more cautious conclusion: once study quality and potential bias were accounted for, the pooled effect of growth mindset interventions was very small and not statistically distinguishable from zero. The review noted that the large majority of intervention designs had methodological confounds. ⁴

This is exactly how science is supposed to work. An exciting early finding meets larger, better-controlled tests, and the estimate shrinks toward a more honest value. The lesson is not that belief is irrelevant. It is that a mindset intervention, on its own, is not the lever it was sold as.

Who it actually helps

The most useful finding in this literature is not the average. It is the pattern underneath it. The effects of growth mindset are not spread evenly; they concentrate among specific students.

Sisk and colleagues found that the benefits were largest for academically high-risk students and those from lower-income backgrounds. ³ The most rigorous single trial, the National Study of Learning Mindsets led by David Yeager and colleagues and published in Nature, makes the point precisely. Across more than 12,000 students in a nationally representative sample, a short online growth mindset intervention produced a small improvement in grades, concentrated among lower-achieving students and in schools with peer norms that supported it. ⁵

So the conditional story is the real story. A growth mindset matters most for students who are struggling, who doubt they can improve, and who are in an environment that makes the new belief safe to act on. For students who already believe they can grow and have the support to do so, telling them again does little.

Belief is the door, not the room

Here is the synthesis that makes all of this practically useful. A growth mindset is a belief about whether ability can change. It is genuinely helpful for getting a discouraged student to engage. But a belief is not a method. Believing you can get better at thinking does not, by itself, make you better at thinking.

This explains the modest average effects without dismissing the idea. Mindset interventions that change belief but provide no real path to develop the underlying skill are pushing on a door with no room behind it. The students who improve are often those for whom the new belief unlocks effort that then meets real instruction. Where the instruction is missing, the belief has little to act on.

The implication is clear. Pair the belief with the method. Help people believe their thinking can develop, and then give them the genuine developmental work that makes it true: problems just beyond their current level, paired with feedback on the structure of their reasoning. ⁶ That second part is the harder and more important part, and it is what Arq.training is built to deliver. The mindset opens the door. The deliberate development of cognitive complexity is the room.

Originally published on Arq.