Evidence audit
Mindset, Rebuilt on the Research
All 8 chapters kept as scaffolding — every claim checked against twenty years of replication attempts. Growth mindset is the most famous idea in modern self-help, and one of the most shrunken.
Published 8 July 2026 · Last checked 14 July 2026 · 8 chapters audited
The quick verdict
Growth mindset is the most famous idea in modern self-help, and one of the most shrunken: across 273 studies of 365,000+ people, mindset explains about 1% of the differences in achievement, and mindset-training programmes average an effect close to zero. What survives is smaller and still useful — a real benefit for struggling students, and the genuinely solid finding that explaining a setback as temporary and fixable predicts trying again.
1 chapter holds up strongly · 7 mixed or overstated
Source book: Mindset · Carol Dweck (2006)
The 8-chapter structure of Carol Dweck's Mindset, kept as scaffolding — but every claim underneath checked against what two decades of replication attempts actually found. This is the hardest audit in this section, because the idea is genuinely appealing and its author is a serious scientist. The finding is nonetheless clear: across 273 studies and more than 365,000 people, mindset accounts for roughly 1% of the variation in academic achievement, and growth-mindset interventions average an effect of d = 0.08 — close to nothing. What survives is a small, real, targeted benefit for struggling and low-income students. That is a much smaller claim than the book makes, and it is still worth having. A research-grounded companion to the book, not a replacement for reading it.
How we audited this book
We kept the book's own chapters and checked every claim underneath against the original studies. Each chapter gets one of two verdicts:
- Holds up
- The core idea is backed by findings that different research teams have confirmed independently. Safe to build habits on.
- Mixed
- A real effect exists, but it's smaller, more conditional, or more contested than the popular version suggests.
- ⚑ Where the book overstates it
- Marks the specific place the book's language runs ahead of the science — with the correction right next to it.
The research terms, in plain language
- Meta-analysis:
- One study that pools all previous studies on a question — the closest thing science has to a final answer.
- Replication:
- Re-running a study to see if the result appears again. Many famous findings failed this test in the 2010s.
- Correlation (r or ρ):
- How tightly two things move together, from 0 (no link) to 1 (lockstep). In psychology, .18 is weak, .50 is strong, and .84 means two questionnaires are nearly measuring the same thing.
- Effect size (d):
- How big a difference an intervention makes. Rough guide: 0.2 is small, 0.5 medium, 0.8 large — so d = 0.08 is close to nothing.
- Variance explained:
- The share of the differences between people that one factor accounts for. “1%” means the other 99% is something else.
- Pre-registered:
- The analysis plan was locked in before the data was collected, so researchers can't quietly pick the flattering result.
Chapter by chapter
- 01The Two Mindsets Are Real — and Explain About 1% of Anything
- 02How Mindsets Change the Meaning of Failure — the Best Part of the Book
- 03Praise the Process — a Real Effect That Shrank on Replication
- 04Sports: Anecdote, Selected After the Fact
- 05Business: The Weakest Chapter, and It Has Aged Badly
- 06Relationships: Interesting, and Well Outside the Evidence
- 07Where Mindsets Come From — and the Bias in the Literature
- 08Changing Mindsets: What Actually Survives
holds up · mixed or overstated
The Two Mindsets Are Real — and Explain About 1% of Anything
In the book: “The Mindsets”
Dweck's distinction — do you believe ability is fixed, or can it grow? — is a genuine psychological difference that can be measured reliably. The question is how much it predicts. The answer, from the largest synthesis available, is: very little.
What the research says
Sisk, Burgoyne, Sun, Butler & Macnamara, 2018 (Psychological Science) — meta-analysis 1: 273 studies, 365,000+ participants
⚠️ The central correction. The relationship between a person's mindset and their academic achievement is weak: mindset accounts for approximately **1% of the difference** in achievement between students. It is a real association. It is a tiny one.
Sisk et al., 2018 — meta-analysis 2: 43 studies, 57,000+ participants
Growth-mindset *interventions* produce an average effect on academic achievement of **d = 0.08** — statistically detectable at that sample size, and practically almost nothing.
⚑ Where the book overstates it
The book presents mindset as the hidden variable behind success, failure, sport, business and love. The measured effect on the one outcome studied most heavily — school achievement — is about 1% of the variance. Almost everything else in your life is doing the other 99%.
Do this
Believe ability can grow — it's true, and it costs nothing. Just don't expect the belief itself to change your results. What changes results is the scheduled, specific action from the habits research.
How Mindsets Change the Meaning of Failure — the Best Part of the Book
In the book: “Inside the Mindsets”
Strip away the achievement claims and something sturdier remains: how you interpret a setback genuinely predicts whether you try again. This chapter's psychology is better supported than its headline promise.
What the research says
Attribution theory / learned helplessness (Seligman, Weiner and successors)
Explaining failure as permanent and about you ("I'm not smart") predicts giving up and predicts depression risk; explaining it as temporary and specific ("I didn't prepare for that") predicts trying again. A large, old, well-replicated literature — and the real engine under this chapter.
Dweck's own early work on helpless vs mastery responses
Children who respond to failure by protecting their self-image, rather than by changing strategy, do perform worse on subsequent tasks. The observation is real, even if the intervention built on it turned out to be weak.
Do this
After a rejection, write the reason down as something specific and fixable — "I didn't tailor the CV to the job description" — not as a verdict on you. This is the piece of Dweck's work with the deepest evidence behind it.
Praise the Process — a Real Effect That Shrank on Replication
In the book: “The Truth About Ability and Accomplishment”
The most actionable, most repeated advice in the book: praise effort and strategy, not intelligence. Tell a child "you're so clever" and you make them fragile. This finding is genuinely important — and its evidence base is thinner than its fame.
What the research says
Mueller & Dweck, 1998 — the praise studies
Children praised for intelligence after success subsequently chose easier tasks, showed less persistence and performed worse than children praised for effort. Highly influential — but the classic studies ran on small samples typical of 1990s social psychology, an era whose findings have replicated poorly as a class.
Li & Bates, 2019 (Journal of Experimental Psychology: General) — attempted replication, 600+ Chinese children
⚠️ Contested, and we are flagging the contest rather than picking a side. Li & Bates ran the Mueller & Dweck paradigm on more than 600 children in China. The mindset manipulation shifted beliefs about intelligence — but showed no effect on the eight motivation and attribution measures Mueller & Dweck had used (mean p = .48), and no effect on post-failure performance in two further studies. **However**, a published re-analysis argues the 'failure to replicate' does not hold up and that the original finding is confirmed. This is an unresolved dispute, not a clean debunking.
⚑ Where the book overstates it
The advice is harmless and probably still worth following — process praise is more informative than trait praise regardless. But it is presented as settled behavioural fact, and the replication record is genuinely disputed: one large attempted replication found nothing, and a re-analysis of that same data disagrees. Confidence is not warranted in either direction.
Do this
Praise what someone did, not what they are — "you kept revising that until it worked" beats "you're brilliant." Costs nothing, likely helps a little, and don't be surprised when it isn't transformative.
Sports: Anecdote, Selected After the Fact
In the book: “Sports: The Mindset of a Champion”
Michael Jordan cut from his school team; Muhammad Ali without the physique. Great stories, and a textbook example of the reasoning error this section exists to catch.
What the research says
The champion narratives
❌ Not evidence. These are champions selected *because* they succeeded, then examined for a common trait. Nobody counts the thousands with identical determination who never made it — and there is no comparison group, so no causal claim can be drawn.
What actually predicts elite sport
Deliberate practice explains around 18% of the difference between athletes (Macnamara et al., 2014) — meaningful, and far from everything. Physical attributes, coaching, timing and access matter enormously, and no mindset overcomes them.
⚑ Where the book overstates it
Survivorship bias, presented as proof. If mindset made champions, we would find it by studying people *before* they succeeded and seeing who rose — not by reading it backwards off the winners.
Do this
Skip this chapter. If you want the sport lesson that survives: practise deliberately, and be honest that access and circumstance are doing a lot of the work.
Business: The Weakest Chapter, and It Has Aged Badly
In the book: “Business: Mindset and Leadership”
Dweck reads Enron's collapse, Lee Iacocca's decline and Jack Welch's success through the mindset lens. This is where the framework stops being a psychological theory and becomes a way of narrating anything.
What the research says
The corporate case studies
❌ Not evidence. Post-hoc explanation of famous business outcomes by a trait the author has already decided matters. Any outcome can be fitted: a failure becomes fixed mindset, a success becomes growth mindset, and no result could ever have contradicted the theory.
The 'unfalsifiable' problem
A framework that explains every outcome equally well after the fact predicts none of them in advance. That is the definition of an untestable claim, and much of this chapter has that shape.
⚑ Where the book overstates it
Notice what would have happened had Enron succeeded: the same culture would have been narrated as bold and growth-oriented. When a theory can absorb any result, it has stopped doing scientific work.
Do this
Nothing here. Judge a workplace by whether it actually trains and promotes people, not by whether its language sounds growth-minded.
Relationships: Interesting, and Well Outside the Evidence
In the book: “Relationships: Mindsets in Love (or Not)”
Do you believe relationships either work or don't (fixed), or that they're built (growth)? A nice idea, borrowed into a domain where the mindset literature has done very little work.
What the research says
Implicit theories of relationships (Knee and others)
There is a small literature finding that believing relationships require work, rather than destiny, correlates with better relationship persistence. Real, but correlational, modest, and not the same construct as academic mindset.
The extrapolation
⚠️ The book generalises a school-achievement construct across sport, business and romance. Even if mindset worked as advertised in classrooms — and it explains ~1% there — nothing licenses assuming it transfers to marriage.
⚑ Where the book overstates it
The pattern of this book's back half is one idea stretched across every domain of life. That is what makes it a bestseller and what makes it weak science.
Do this
Read as reflection, not research.
Where Mindsets Come From — and the Bias in the Literature
In the book: “Parents, Teachers, and Coaches: Where Do Mindsets Come From?”
This chapter argues that adults transmit mindsets to children. It's also the right place to record what happened when researchers audited the quality of the studies behind the whole intervention industry.
What the research says
Macnamara & Burgoyne, 2023 (Psychological Bulletin) — 122 studies
⚠️ The most damning finding in this audit. Reviewing the growth-mindset intervention literature, they found **94% of studies contained confounds**; authors with a **financial incentive** (selling mindset programmes) were about **two and a half times more likely** to report positive effects; higher-quality studies were *less* likely to find a benefit; and among the studies that best followed methodological best practice, the overall effect was not significant (p = .666). Their conclusion: apparent effects are likely attributable to inadequate design, reporting flaws and bias.
Teacher and parent mindset transmission
The claim that adults' mindsets shape children's is intuitive but weakly evidenced; the effects reported are small and come from the same literature the audit above found compromised.
⚑ Where the book overstates it
A billion-dollar school-programme industry grew out of this book. When the literature behind it was audited, nearly all of it had confounds and the people selling it were far more likely to publish good news about it. That is a conflict-of-interest story as much as a psychology one.
Do this
If someone sells you or your school a growth-mindset programme, ask what the effect size is and who ran the study. The honest answer is: small, and often the vendor.
Changing Mindsets: What Actually Survives
In the book: “Changing Mindsets”
Here is the fair, final accounting — and it is not nothing. When the largest and most rigorous experiment ever run on growth mindset reported, it found a real effect. A small one, for a specific group.
What the research says
Yeager et al., 2019 (Nature) — the National Study of Learning Mindsets
✅ The best evidence in the book's favour, and it is Dweck's own team's. A pre-registered, nationally representative randomised trial of about **12,000 ninth-graders across 65 US high schools**: a short online growth-mindset intervention raised core-course GPA by **0.10 grade points** among lower-achieving students — rising to about 0.15 (and 0.17 in STEM) where the school's peer climate supported taking on challenge. No effect for higher-achieving students.
How to read that honestly
A 0.10 GPA bump, for struggling students only, from a cheap 50-minute online exercise, is a genuinely good return on cost — and it is a fraction of what the book promises. It is also conditional on the environment: the intervention did nothing where the school culture didn't back it. Mindset is a small lever that works when the surrounding conditions are already right.
⚑ Where the book overstates it
This is the honest shape of growth mindset after twenty years: not a hoax, and not a revolution. A cheap intervention with a small, real benefit for students who are already struggling — which is close to the *opposite* of the book's message that mindset is the master key to all achievement.
Do this
Take the version the evidence supports. Believe your skills can improve, especially when you're behind — that belief measurably helps people who are struggling. Then go and do the work, because the belief on its own moves a GPA by a tenth of a point, and the work moves everything else.
Questions people ask
Is growth mindset debunked?
Shrunken, not debunked. Mindset explains about 1% of achievement differences and interventions average close to zero — but the largest, most rigorous trial (about 12,000 students) found a real 0.10 GPA gain for lower-achieving students from one cheap online session. It's a small lever that helps people who are struggling, not the master key the book describes.
Does praising effort instead of intelligence really matter?
It's genuinely disputed. The original studies found children praised for intelligence chose easier tasks and crumbled after failure — but a large replication attempt in China found no effect on motivation or performance, and a re-analysis of that same data disagrees. Praise the process anyway: it's more informative and costs nothing. Just don't expect it to transform anyone.
Why did the growth mindset effect shrink so much?
The evidence base had problems. An audit of 122 intervention studies found 94% contained design flaws that muddy the result, researchers with a financial stake in mindset programmes were about two and a half times more likely to report positive effects, and the studies that best followed good methods found no significant benefit at all.
What part of Mindset should I actually use?
The setback part. How you explain a failure to yourself — permanent and about you, or temporary and specific — reliably predicts whether you try again. After a job rejection, write down one specific, fixable reason ("my CV didn't show the software they wanted") instead of a verdict on yourself. That habit rests on decades of solid research.
Compiled as a research-grounded companion to the book — not a replacement for reading it. Citations are author/year pointers to real, findable studies rather than formatted references; verify specifics before relying on them for high-stakes decisions.