Praise vs. Feedback: What Stanford Research Says Actually Motivates Kids Long-Term

A parent watches her daughter ace a third-grade math test and says, with genuine love and pride: “You’re so smart!” The daughter beams. Over the next several weeks, this parent notices that her daughter has begun refusing math problems she’s not sure she can solve. When a harder chapter arrives, she becomes anxious and avoidant in ways she never was before. The parent cannot figure out what changed. The answer, backed by three decades of systematic research from Carol Dweck’s lab at Stanford University, is that the word “smart” changed it. The phrase meant as pure encouragement did something the parent did not intend: it gave the child an identity to protect.

Key Takeaways

• Carol Dweck’s research distinguishes between “fixed mindset” — believing intelligence and talent are innate and unchangeable — and “growth mindset” — believing ability develops through effort, strategy, and learning from failure.
• Praising children’s intelligence (“you’re so smart”) consistently produces fixed mindset responses: avoidance of challenge, helplessness under difficulty, and dishonesty about performance.
• Praising children’s effort, process, and strategy produces growth mindset responses: greater persistence, challenge-seeking, and resilience in the face of setbacks.
• The language that “hurts” is not criticism or high expectations — it is person-level praise that implies ability is a fixed trait. The language that “helps” is process-level feedback that links outcomes to specific actions.
• Mindset interventions — even brief ones delivered in school settings — produce measurable improvements in academic motivation and achievement, with stronger effects for students from underrepresented groups.

The Research Foundation: Dweck’s Studies

The Columbia Praise Study

The research that established the praise-versus-process distinction was conducted by Claudia Mueller and Carol Dweck and published in the Journal of Personality and Social Psychology in 1998. The study involved 412 fifth-graders across six schools. Children were given a set of moderately challenging nonverbal IQ problems and, regardless of actual performance, were told they did well. Half were then praised for their intelligence (“You must be smart at these problems”). Half were praised for their effort (“You must have worked really hard”).

What happened next across multiple experimental rounds was stark:

• When given the choice between a challenging problem set and an easy one for round two, 90% of the effort-praised children chose the challenging set. Most of the intelligence-praised children chose the easy set — protecting their “smart” label.
• When given a genuinely hard problem set (designed to be unsolvable), intelligence-praised children showed significantly more helplessness, negative affect, and attribution of failure to lack of ability. Effort-praised children showed more persistence and framed difficulty as something to work through.
• When asked to report their scores to a peer, 40% of intelligence-praised children lied and reported higher scores than they had received. Among effort-praised children, the lying rate was negligible.
• On a final round of problems at the original difficulty level, effort-praised children showed improved performance compared to their baseline. Intelligence-praised children showed decreased performance — their scores actually went down after experiencing the hard problems.

A single episode of praise for intelligence produced measurable avoidance, helplessness, dishonesty, and performance decline across the sample. The mechanism, Dweck argued, was the implicit message: “You are smart” creates a fixed label to be protected. “You worked hard” creates a process to be continued.

The Replication and Extension Record

The Mueller and Dweck (1998) finding has been replicated and extended in multiple subsequent studies. A 2007 study by Good, Aronson, and Inzlicht showed that brief growth mindset interventions for seventh-graders — even a single 25-minute session reframing intelligence as malleable — produced significant improvements in math and reading achievement over a full academic year. A 2012 study by Blackwell, Trzesniewski, and Dweck followed seventh-graders over two years and found that those who endorsed growth mindset beliefs showed consistently improving math trajectories while fixed mindset students showed flat or declining trajectories.

A large-scale replication across 65 countries published in Nature in 2019 by Yeager and colleagues, involving over 12,000 students, confirmed that a brief growth mindset intervention produced higher GPAs for lower-achieving students and reduced the achievement gap — particularly for first-generation college students and students from underrepresented groups.

The Language That Hurts

Person-Level Praise

The category of praise most consistently associated with fixed mindset outcomes is what researchers call person-level or trait-level praise: statements that attribute outcomes to fixed characteristics of the person rather than their actions.

Examples:

• “You’re so smart.”
• “You’re a natural at this.”
• “You’re so talented.”
• “You’re a math person.”
• “You’re just gifted.”

These statements are problematic not because they are unkind — they are clearly intended as positive — but because they create an identity that becomes a liability under difficulty. A child who believes they are “a math person” faces an existential threat when they encounter math difficulty: difficulty becomes evidence against the identity. The rational response is avoidance.

Outcome Praise Without Process

Outcome praise — celebrating results without connecting them to process — is a milder form of the same problem: “You got an A! Great job!” The A is the outcome; the praise attaches to the outcome rather than to the strategies, effort, and choices that produced it. Over time, this trains children to value the letter grade rather than the learning — a pattern that predicts grade-driven motivation, academic dishonesty, and fragility when grades are harder to obtain.

The Uninstructive “Good Job”

Uninstructive positive feedback — “good job,” “great work,” “awesome,” applied generically — provides approval without information. Research by Wulf, Lewthwaite, and colleagues suggests that generic positive feedback does not improve performance and, for children who are accustomed to it, may actually impair performance on new tasks because it creates unrealistic calibration of baseline difficulty.

The Language That Helps

Process-Level Praise and Feedback

Process-level praise connects outcomes to specific actions, choices, and strategies that the child can repeat or modify:

• “You tried a different strategy when the first one didn’t work — that’s what good problem-solvers do.”
• “You spent a long time on that paragraph and changed it three times. It shows.”
• “That problem was hard and you stayed with it. That’s how mathematicians actually work.”
• “I noticed you asked for help when you got stuck instead of giving up. That was a smart move.”

The key features: specific (not generic), behavioral (connected to what the child did), and replicable (the child can do it again). Process-level feedback teaches a theory of performance: what produces outcomes is not fixed ability but specific, learnable actions.

Accurate and Specific Critical Feedback

One of the most important findings in Dweck’s research is that appropriate critical feedback — when framed in a growth context — is not demotivating. The famous formulation from Dweck’s Mindset (2006): “Not yet.” Rather than a grade of F, a grade of “Not Yet” communicates that the student is on a learning trajectory, not at a permanent stopping point.

Specific feedback on what went wrong and what to try differently is more motivating under growth mindset conditions than generic positive feedback. “Your argument in the second paragraph doesn’t follow from the first — try connecting them with a transition that shows how one leads to the other” gives the student a specific action and implies they are capable of executing it.

Normalizing Difficulty and Mistake

Growth mindset parenting involves explicitly and repeatedly communicating that difficulty and error are the conditions under which learning happens — not signs of failure. This is not a platitude; it is specific information about how the brain and learning work that children can internalize as a theory of their own performance.

Research by Haimovitz and Dweck (2016) found that whether children develop growth or fixed mindsets is influenced not only by the praise they receive but by their parents’ theories of failure: parents who believe failure is a positive, informative event raise children with stronger growth mindsets than parents who experience failure — even their children’s failure — as threatening.

Comparing Praise and Feedback Approaches

Type	Example Language	Effect on Challenge-Seeking	Effect on Resilience Under Failure	Effect on Honesty
Intelligence praise	”You’re so smart.”	Decreases (protects label)	Decreases (failure threatens identity)	Decreases (protects image)
Talent praise	”You’re a natural.”	Decreases	Decreases	Mixed
Effort praise	”You worked really hard on this.”	Increases	Moderate increase	Increases
Strategy/process praise	”You tried a new approach when the first didn’t work.”	Strong increase	Strong increase	Increases
Outcome praise (generic)	“Great job! You got an A.”	No change	No change	Neutral
Critical feedback (process)	“That paragraph needs a clearer argument — here’s what to try.”	Increases with growth mindset framing	Increases	Increases
Normalizing difficulty	”This is supposed to be hard — hard is how learning happens.”	Strong increase	Strong increase	Increases

Complications and Nuances

The Sincerity Problem

A concern that emerged in the replication literature is the risk of formulaic process praise — using the “right” language without authenticity. Children, particularly school-age children, detect inauthentic or inflated praise and discount it. Research by Kamins and Dweck (1999) showed that children who received inflated positive feedback developed more fragile self-assessments than children who received accurate feedback. Process-level praise needs to be accurate and genuine, not a scripted formula.

High-Ability Children and Fixed Mindset Risk

Ironically, the children at greatest risk for fixed mindset formation are those who have been told most repeatedly that they are smart — high-achieving children from families and schools that prize academic achievement and routinely label high performance as evidence of intelligence. Research by Haimovitz, Wormington, and Corpus found that students at highly selective schools showed higher rates of fixed mindset beliefs than national averages, likely as a result of chronic ability-focused feedback. Gifted programs that emphasize “smartness” as identity create the exact conditions Dweck’s research identifies as harmful.

Cultural Variation

Cross-cultural research has found that the relationship between effort praise and motivation varies by cultural context. In societies where effort is already valued as a core virtue (certain East Asian contexts, for example), effort praise may not produce the same differential effects as in societies where natural talent is more prominently celebrated. This does not invalidate the core findings but suggests that context shapes how praise messages land.

What to Watch For Over 3 Months

Over the next three months, track your child’s response to new challenges and difficulty — not just their reaction to success. A child developing a growth orientation will begin to show curiosity or interest at the beginning of difficult tasks (“this is hard — let me figure out how to start”) rather than immediate avoidance or distress. This shift does not happen after a single conversation; it develops over weeks and months of consistent process-focused feedback.

Also track your own language. Keep a loose mental note of how often you comment on who your child is versus what your child did. The research suggests that even parents who are intellectually committed to growth mindset messaging default to ability-focused language in moments of genuine pride — “you’re so brilliant” comes out naturally and automatically. The intervention is building the habit of adding a process frame: “You worked through that really carefully. That’s what brilliant looks like.”

Frequently Asked Questions

Is it ever okay to tell my child they are smart?

The research does not say that calling a child smart is catastrophically harmful in isolation. The concern is with chronic, primary identity-level praise that makes “being smart” the core of a child’s academic self-concept. Occasional positive comments about intelligence in the context of a consistently process-focused feedback environment are not the problem Dweck’s research identifies.

What if my child is genuinely talented? Won’t they figure out they’re special?

Yes — talented children do know they find certain things easier than peers. The question is what theory of performance they develop around that fact. A talented child who understands they can always improve, who values the effort and strategy that underlies performance, and who does not use their current ability as a reason to avoid challenge develops very differently from a talented child whose identity is built on “being the smart one.”

My child’s teacher gives a lot of ability praise. What can I do at home?

Research by Haimovitz and Dweck shows that parents’ own mindset language at home is a significant predictor of children’s mindsets independent of school culture. You cannot control what the teacher says, but you can provide a consistent counter-frame at home: normalizing difficulty, praising process, and discussing the purpose and meaning of challenge in ways the school environment may not.

Does growth mindset apply to non-academic contexts like sports or music?

Yes. The same mechanisms operate in athletic and artistic domains. Research on young musicians and athletes shows that those praised for talent develop more fragile motivation under difficulty than those whose coaches and parents praise the specific practice behaviors, strategies, and improvements that develop skill over time. The domain changes; the principles do not.

How young can growth mindset language start?

Evidence suggests that praise language affects motivation from preschool age. Research by Gunderson and colleagues (2013) found that the type of praise parents directed at toddlers (aged 1–3) predicted that child’s mindset orientation at ages 7–8, with process praise in the toddler years associated with stronger growth mindset five years later. Starting early is substantively meaningful.

---

About the author

Ricky Flores is the founder of HiWave Makers and an electrical engineer with 15+ years of experience building consumer technology at Apple, Samsung, and Texas Instruments. He writes about how kids learn to build, think, and create in a tech-saturated world. Read more at hiwavemakers.com.

Sources

1. Mueller, C. M., & Dweck, C. S. (1998). Praise for intelligence can undermine children’s motivation and performance. Journal of Personality and Social Psychology, 75(1), 33–52.
2. Dweck, C. S. (2006). Mindset: The New Psychology of Success. Random House.
3. Blackwell, L. S., Trzesniewski, K. H., & Dweck, C. S. (2007). Implicit theories of intelligence predict achievement across an adolescent transition. Child Development, 78(1), 246–263.
4. Yeager, D. S., Hanselman, P., Walton, G. M., Murray, J. S., Crosnoe, R., Muller, C., … & Dweck, C. S. (2019). A national experiment reveals where a growth mindset improves achievement. Nature, 573(7774), 364–369.
5. Haimovitz, K., & Dweck, C. S. (2016). What predicts children’s fixed and growth intelligence mind-sets? Not their parents’ views of intelligence but their parents’ views of failure. Psychological Science, 27(6), 859–869.
6. Gunderson, E. A., Gripshover, S. J., Romero, C., Dweck, C. S., Goldin-Meadow, S., & Levine, S. C. (2013). Parent praise to 1- to 3-year-olds predicts children’s motivational frameworks 5 years later. Child Development, 84(5), 1526–1541.
7. Kamins, M. L., & Dweck, C. S. (1999). Person versus process praise and criticism: Implications for contingent self-worth and coping. Developmental Psychology, 35(3), 835–847.