What Is a Mindset? How Your Beliefs Shape What You Do

A mindset is a belief about how something basic works, such as whether your abilities can grow or whether stress helps or harms you, and it shapes what you notice, how you explain what happens and how you respond.¹ Psychologists call the belief that intelligence can be developed a growth mindset, and the belief that it cannot a fixed mindset.²

The question what is a mindset has a settled answer; the question of what changing one achieves does not. The idea is heavily studied, but its measured payoff is smaller than its fame. In a 2019 randomized experiment with 12,490 US ninth graders from a nationally representative sample of schools, a growth mindset program lifted grades slightly, and only among lower-achieving students.³ A large independent trial in English primary schools found no gain at all.⁴

What is a mindset in psychology: a frame that filters experience

In psychology, a mindset is a mental frame that selectively organizes information, steering a person toward one way of understanding an experience and toward matching actions. That definition comes from a 2013 paper by Alia Crum, based on her doctoral research at Yale University with Peter Salovey, and co-written with Shawn Achor. The authors adapted it from the Stanford psychologist Carol Dweck. Mindsets are a normal part of how people think, they argued, but they still shape judgment, health and behavior.¹

Dweck’s own framing is narrower. People can believe that an attribute such as intelligence or personality is simply fixed, or that it can be shaped and developed, she and David Yeager of the University of Texas at Austin wrote in a 2019 review of the field. As they define it, a growth mindset about intelligence is the belief that it can be developed through effort, good learning strategies and plenty of mentoring and support from others; a fixed mindset is the belief that it cannot.²

Dweck first called these beliefs implicit theories, because people are often unaware of holding them, and named the two versions entity and incremental theories. She switched to the friendlier labels fixed and growth mindset in a 2006 book.² The same idea has since been applied to beliefs about personality and prejudice, about stress, and about what a company thinks of talent, which is why the word now covers much more than intelligence.²¹⁵

Where fixed and growth mindsets came from

Fixed and growth mindsets came out of Carol Dweck’s research on why children of similar ability react so differently to failure. In studies from the 1970s, Dweck and colleagues found that how children explained a failure predicted whether they gave up, a helpless response, or treated it as a challenge, a mastery-oriented one, even among children equally able at the task.²

1. 1970sChildren's explanations of failure predict helpless or mastery-oriented responses
2. 1998Praise for intelligence, rather than effort, is followed by less persistence after failure in six studies
3. 2006Dweck's book renames entity and incremental theories as fixed and growth mindsets
4. 2019A national randomized experiment in US high schools is published in Nature
5. 2023Two meta-analyses of growth mindset programs reach different verdicts

Two early studies are still widely cited. In six studies published in 1998, Claudia Mueller and Dweck found that fifth graders praised for their intelligence after a success showed less persistence, less enjoyment and worse performance after a later failure than children praised for effort, and were more likely to describe intelligence as a fixed trait.⁶ In a 2007 study, Lisa Blackwell and colleagues followed 373 seventh graders through two years of junior high: a belief that intelligence is malleable predicted rising math grades, while a belief that it is fixed predicted a flat line.⁷ In a small follow-up intervention study, 48 students taught the malleable view reversed a decline in grades that continued among 43 controls.⁷

Not every classic result has held up. In 2019, Yue Li and Timothy Bates of the University of Edinburgh closely repeated the praise experiment with Chinese children aged 9 to 13, 624 children across their studies. The growth mindset manipulation was linked to scores on one post-failure test but to none of the original’s 8 motivation and attribution measures, and two follow-up studies with an active control condition found no effect; children’s mindsets were also unrelated to their grades.⁸

How a belief about ability turns into behavior

Mindset theory holds that a belief about ability changes what effort and failure mean. In Dweck and Yeager’s account, when ability looks fixed, proving it matters more, hard effort can look like a sign of low ability and setbacks are easily blamed on low ability, so persistence can drop. When ability looks improvable, effort becomes a tool and a setback becomes information about the learning process.²

1. The belief: ability is fixed (padlock), or it can be developed through effort, good strategies and help from others (seedling)
2. How the setback is read: as a sign of low ability, or as information about how the learning is going
3. What matters most: proving ability and not looking stupid, or developing ability by taking on challenges
4. What happens next: hard effort can look like a sign of low ability and persistence can drop, or effort and asking for help become tools and persistence can hold

A 1999 study at the University of Hong Kong, where all teaching and exams were in English, gives a concrete case. Students with a fixed mindset showed little interest in a remedial English course even when their English was weak, while weak students with a growth mindset were much keener, Dweck and Yeager report.² The national growth mindset program described below targeted the same chain: the belief that having to try hard or ask for help means you lack ability, the habit of blaming failure on low ability, and the goal of never looking stupid.³

The link between belief and results is weaker than the story suggests. A 2018 meta-analysis led by Victoria Sisk of Case Western Reserve University, covering 129 studies with 365,915 people, found that the overall relationship between mindset and academic achievement was weak.⁹¹⁰

The national US trial: small grade gains for lower achievers only

The best-designed single test of a growth mindset program we found, the National Study of Learning Mindsets, raised grades slightly for lower-achieving students and made no clear difference for the rest. It was a randomized experiment in a nationally representative sample of 65 US public high schools, published in Nature in 2019 by David Yeager and 24 colleagues, including Dweck. The data were collected and processed by independent research companies and the analysis plan was registered in advance.³

The study

Moderate evidence

Two online lessons, 65 US high schools: the National Study of Learning Mindsets

Students were randomly assigned to a growth mindset program or a control program about brain functions that did not address beliefs about intelligence, each delivered in two self-guided online sessions about three weeks apart. Among lower-achieving students, those at or below their school’s median, the program raised core-course grades at the end of ninth grade by 0.10 grade points, a standardized effect of 0.11. Higher-achieving students showed no clear difference. In an exploratory analysis of 41 schools, 36 percent of the program group took advanced math the next year, against 33 percent of controls.³

For a program lasting less than an hour, with no teacher training, that is a real but modest result, and it depended on the setting: grades rose more in schools where peer norms supported taking on challenges.³ Critics Brooke Macnamara and Alexander Burgoyne point out that the grade result covers only the lower-achieving half of the students.¹⁰ The authors say that focus was set in advance, because higher achievers may already have the habits and support that the program aims to build.³

The study was funded by foundations, US government grants and a personal gift from one of the authors. Several authors disclosed paid or unpaid talks and consulting on mindset research, and the paper states that none had a financial relationship with a company selling growth mindset products.³

Two 2023 meta-analyses, one body of research, two verdicts

Whether growth mindset programs work in general is contested. Two meta-analyses published together in Psychological Bulletin in 2023, one by Macnamara and Burgoyne and one by Jeni Burnette and six colleagues, examined the same body of intervention studies with different methods and reached different conclusions.¹¹

Macnamara and Burgoyne pooled 63 studies with 97,672 students and found an average effect on academic achievement of d = 0.05, which was not statistically significant after correcting for possible publication bias. In the 6 highest-quality studies, with 13,571 students, the effect was d = 0.02, again not significant. Authors with a financial incentive, such as income from mindset books or talks, published larger effects than authors without one, and the pair concluded that apparent benefits are likely due to weak study design, reporting flaws and bias.¹⁰

Burnette’s team included 53 independent samples from randomized studies published from 2002 to 2020 and asked for whom the programs work. In the groups expected to benefit most, with programs delivered as designed, the effect on academic achievement was d = 0.14. Their prediction interval, the range of effects to expect from a new program, ran from −0.08 to 0.35: anything from slightly negative to modest.¹²

The gap is largely about method. A 2023 commentary by Elizabeth Tipton, David Yeager and four colleagues argued that Macnamara and Burgoyne averaged effects within each study and tested explanations one at a time, which hides real variation, and that re-analyzing their dataset with Burnette’s methods showed a meaningful effect in at-risk groups. Four of the commentary’s six authors also wrote the national experiment, so it is not a neutral referee. The commentary also disputed how Macnamara and Burgoyne defined a financial conflict of interest.¹¹

An earlier 2018 meta-analysis led by Sisk, of 38 independent samples with 57,155 students, had also found weak overall effects, with some support for benefits among students with low socioeconomic status or at academic risk.⁹¹⁰

Myth

A growth mindset program reliably raises grades.

Fact

Two 2023 meta-analyses put the average effect on achievement at d = 0.05, not significant after bias correction, or d = 0.14 in targeted groups. The national US trial showed small gains for lower-achieving students only, and a large English trial found none.

Independent evaluations have found little. In England, a randomized trial funded by the Education Endowment Foundation and evaluated by the National Institute of Economic and Social Research, not the program’s developers, tested an eight-week, teacher-delivered program for Year 6 pupils, aged 10 to 11, in 101 schools with 5,018 pupils. Pupils made no additional progress in reading, grammar, punctuation and spelling, or math compared with the control group, a finding the evaluators rated high security.⁴ A 2025 review by Carolina Gazmuri of 24 randomized trials with school-age children found that the strongest studies showed effects between d = −0.01 and d = 0.065. It advised schools and governments not to commit significant time or money to such programs, though very low-cost ones might still be reasonable.¹³

Mindsets about stress and about the workplace

Mindsets are not only about ability. Alia Crum’s 2013 studies introduced the stress mindset, the extent to which you believe the effects of stress are enhancing or debilitating, with an 8-item questionnaire to measure it.¹

In one of those studies, Crum’s team randomly assigned 388 employees of a large financial institution going through downsizing to watch three short videos over a week presenting stress as enhancing, or as debilitating, or to watch none. According to the authors, the stress-is-enhancing group moved toward that mindset on the questionnaire and rated its own work performance higher afterward, while the other groups did not. These were self-reports by staff of a single company, followed for about two weeks, so this is limited evidence. A third study, with 63 undergraduates, linked a stress-is-enhancing mindset to more interest in feedback during a stressful speech task. The authors also describe more moderate responses of the stress hormone cortisol, but that test fell short of statistical significance. They call for research on for whom, and for which kinds of stress, such interventions work best.¹

A 2024 meta-analysis of randomized trials by Michel Bosshard and Patrick Gomez found that programs teaching people that stress can be useful slightly improved performance on stressful tasks overall (d = 0.23 across 44 effect sizes), but programs limited to the stress-is-enhancing mindset message made no clear difference, in just 6 effect sizes.¹⁴

Companies can also be seen as holding a mindset. In 2020, Elizabeth Canning, Mary Murphy, Dweck and colleagues reported that among more than 500 employees of seven Fortune 1000 companies, those who saw their employer as believing talent is fixed described a culture with less collaboration, innovation and integrity, and reported less trust in and commitment to the company.⁵ These are employees’ own ratings, so they show a link, not that the belief causes the culture.

Applying the idea at work: effort, strategy and support

Two leading mindset researchers, Carol Dweck and David Yeager, wrote in 2019 that it is not clear their school program would carry over easily to colleges or the workplace, and that doing so may require new rounds of research.² The idea still offers a more careful way of talking about effort, strategy and setbacks, for yourself and for a team.

Dweck and Yeager warn against what they call a false growth mindset: equating a growth mindset with effort alone, for example praising effort that is not working, which can leave people feeling incompetent. Their own definition puts good strategies and support from others next to effort.² The study techniques with the strongest evidence are one concrete place to start on the strategy part.

For managers, the Canning findings suggest that what a company signals about talent, down to the wording of its mission statement, is part of the picture, though the evidence from real companies is correlational.⁵

The bottom line

In trials, teaching students that ability can grow has changed their grades little or not at all: in the national US trial, grades rose slightly for lower-achieving students only, and a large English trial found no gain.³⁴ The belief itself is real and measurable, and it colors how people read setbacks.¹ Treat a growth mindset as one useful way to interpret setbacks, paired with good strategies and support from others, not as a fix on its own.²