Growth Mindset: Does It Actually Work? What the Meta-Analyses Show

Few ideas in educational psychology have achieved the cultural penetration of Carol Dweck’s growth mindset theory. The concept — that believing intelligence is malleable rather than fixed leads to greater academic achievement — has been adopted by school districts, corporate training programs, and parenting guides worldwide. But as the idea has scaled from laboratory to classroom to boardroom, a growing body of rigorous research has raised uncomfortable questions about how large the effect actually is and when it works. Here is what the evidence shows.

What Is Growth Mindset Theory?

Dweck’s theory, developed over decades of research, proposes that people hold implicit beliefs about the nature of intelligence that fall on a continuum:

• Fixed mindset: The belief that intelligence is a stable trait — you’re either smart or you’re not. People with fixed mindsets tend to avoid challenges (which might reveal low ability), give up easily when faced with difficulty, and interpret failure as evidence of permanent limitations.
• Growth mindset: The belief that intelligence can be developed through effort, learning, and persistence. People with growth mindsets tend to embrace challenges, persist through setbacks, and view failure as an opportunity to learn.

The theory predicts that students with growth mindsets will achieve more because they engage more productively with the learning process — seeking feedback, choosing harder tasks, and sustaining effort in the face of difficulty.

What Did the Original Research Show?

Dweck’s early laboratory studies produced striking results. In controlled experiments, students who were praised for effort (“You worked really hard”) rather than ability (“You’re so smart”) were more likely to choose challenging tasks, persist longer, and perform better on subsequent tests. These findings were replicated across age groups and cultures, generating enormous enthusiasm.

Research on the impact of growth mindset interventions on academic achievement examined whether these laboratory effects translate to real-world educational outcomes when mindset interventions are delivered at scale.

What Do the Large-Scale Studies Show?

As growth mindset moved from lab to classroom, the effect sizes shrank dramatically. Several landmark studies illustrate this pattern:

The National Study of Learning Mindsets (2019): The largest randomized trial of a growth mindset intervention, conducted with over 12,000 ninth-graders across 65 U.S. schools. The intervention was a brief online module (~50 minutes total). Key findings:

• The average effect on GPA was 0.03 grade points — statistically significant but practically tiny
• Effects were concentrated among lower-achieving students (0.10 grade points) and were essentially zero for average and high-achieving students
• School context mattered: effects were larger in schools with supportive peer norms (where challenge-seeking behavior was socially accepted)

Meta-analyses: Multiple meta-analyses have now synthesized the growth mindset literature:

To put these numbers in context: an effect size of d = 0.05–0.08 means that a growth mindset intervention explains less than 1% of the variance in academic achievement. Compare this to the effect of IQ on academic performance (r = 0.50–0.60, explaining 25–36% of variance) or even the effect of self-control on test performance, which is substantially larger.

Why Are the Effects So Small?

Several explanations have been proposed:

• Beliefs alone don’t change behavior: Believing that intelligence is malleable is necessary but not sufficient for improved learning. Students also need effective study strategies, adequate prior knowledge, access to quality instruction, and environmental support. Mindset interventions that deliver only the belief component without the behavioral scaffolding may fail to produce meaningful change.
• The dosage problem: Most scalable interventions are brief (30–90 minutes of online content). Changing deeply held beliefs about intelligence through such minimal exposure may be unrealistic. Laboratory studies that produced larger effects typically involved more intensive, sustained interventions.
• Ceiling effects in mindset: Many students already hold growth mindsets to some degree. In populations where baseline growth mindset levels are high, there is less room for intervention-driven change.
• Cognitive ability constraints: Effort and persistence cannot fully compensate for differences in cognitive ability. A student with lower fluid reasoning capacity who works harder will improve but may not close the gap with a higher-ability peer who works less. Research on genetic and environmental origins of cognitive abilities demonstrates the substantial heritability of intelligence, setting biological limits on how much mindset can compensate.
• Publication bias: The Macnamara & Burgoyne (2023) meta-analysis found clear evidence that studies with positive results were more likely to be published, inflating earlier meta-analytic estimates. When statistical corrections for publication bias are applied, the average effect approaches zero.

When Does Growth Mindset Work?

The evidence does not show that growth mindset is entirely ineffective. Rather, it shows that effects are moderated by context and population:

• Low-achieving and at-risk students: The most consistent positive effects appear among students who are struggling academically — precisely those for whom a fixed mindset (“I’m not smart enough”) may be most debilitating. For these students, shifting beliefs about the malleability of intelligence can reduce learned helplessness and re-engage them with learning.
• Transitional periods: Interventions show larger effects during challenging academic transitions (entering high school, starting college) when students face new demands and are forming beliefs about their capabilities in the new context.
• Supportive environments: Growth mindset effects are larger in schools and classrooms where the culture supports challenge-seeking — where asking for help is normalized, mistakes are treated as learning opportunities, and effort is valued. In environments that are indifferent or hostile to struggle, mindset interventions may be insufficient.
• When combined with strategy instruction: Telling students that intelligence is malleable is more effective when coupled with explicit instruction in how to learn — metacognitive strategies, study skills, and self-regulated learning techniques.

What Are the Criticisms?

The growth mindset literature has faced several serious criticisms:

• Replication failures: Several high-profile attempts to replicate Dweck’s original findings have failed or produced substantially smaller effects. This is part of the broader “replication crisis” in psychology but is particularly damaging for a theory that has been so widely implemented.
• Conceptual stretching: The concept of “growth mindset” has been expanded far beyond its original definition. In popular use, it often becomes indistinguishable from general optimism, resilience, or grit — constructs that have their own measurement traditions and evidence bases.
• Oversimplification of intelligence: The theory implies a simple dichotomy (intelligence is either fixed or growable) that does not reflect the nuanced reality. Intelligence has both heritable and environmental components, is stable in some respects and malleable in others, and changes differently across different abilities and life stages. The Flynn Effect demonstrates population-level malleability while individual-level stability remains high.
• Potential for blame: A troubling implication of growth mindset rhetoric is that students who fail despite effort may be blamed for having the “wrong” mindset. This ignores structural barriers (poverty, inadequate schools, discrimination) and cognitive ability differences that constrain outcomes regardless of belief.

How Does Mindset Compare to Other Interventions?

When judged against other educational interventions, growth mindset’s effect size is modest:

• High-quality tutoring: d = 0.40–0.80 (far larger than mindset)
• Formative assessment and feedback: d = 0.40–0.70
• Metacognitive strategy instruction: d = 0.50–0.60
• Reducing class size: d = 0.20–0.30
• Growth mindset intervention: d = 0.05–0.10

This does not mean mindset is worthless — its scalability (delivery via brief online modules to thousands of students at low cost) means even small effects can aggregate meaningfully at the population level. But it does mean that growth mindset should not be treated as a primary lever for improving educational outcomes. It is, at best, a modest supplement to the structural, pedagogical, and resource-based interventions that produce substantially larger effects.

What Should Educators Do?

Based on the current evidence:

1. Don’t abandon mindset — but don’t rely on it. Encouraging students to view challenges as growth opportunities is unlikely to cause harm and may help struggling students. But it should not replace investment in quality instruction, adequate resources, and structural support.
2. Combine mindset with strategy. Telling students “you can grow your brain” is less effective than teaching them how to learn — spaced practice, retrieval practice, elaborative encoding, and self-monitoring.
3. Avoid overpromising. Telling students that effort alone determines success sets them up for disillusionment when they encounter genuine ability constraints. A more honest message: effort matters, strategy matters, and ability sets a context within which effort and strategy operate.
4. Focus on high-impact interventions first. For limited educational budgets and attention, the evidence suggests prioritizing high-quality instruction, formative assessment, and targeted support for struggling students over mindset programming.

Conclusion

Growth mindset theory contains a kernel of truth: beliefs about intelligence influence motivation and learning behavior, and shifting those beliefs can help — particularly for struggling students in supportive environments. But the popular narrative — that mindset is a transformative force rivaling or exceeding the importance of cognitive ability — is not supported by the evidence. Meta-analyses consistently show very small average effects, with substantial evidence of publication bias inflating earlier estimates. Growth mindset is best understood as a modest, low-cost supplement to more impactful educational practices, not a replacement for them. The science of how education improves intelligence points to far more powerful levers — quality instruction, adequate time, and enriched learning environments — that deserve priority in educational investment.