What can be learned from growth mindset controversies?
Yeager DS, Dweck CS · 2020
grade Dcritiquedeveloper-lednot-applicable
Sample
Not applicable - a reply article
Population
Not applicable.
Design
An invited article in an award issue rather than a competitively submitted paper, which matters for how much independent scrutiny it received. Graded D as expert commentary, and retained because the archive's job includes recording what the strongest advocates say when pressed.
Key findings
Concedes Sisk et al.'s r = .09 overall correlation but argues it is uninterpretable given I-squared of 96%, and counters with large-sample correlations of r = .24 to .34 with grades and with achievement tests among students facing adversity. Disputes the effect-size benchmark, noting that Hattie's own long-horizon effects average about .10, and invokes the argument that 0.10-0.15 SD is large for a brief cheap scalable intervention. On the EEF pilot it argues a non-significant +2 months is not evidence against the theory. It does not engage with the EEF effectiveness trial, which reported exact zeros on national tests with 4,584 pupils. The most important concession is structural - heterogeneity is granted, and the retreat position is that mindset works in some contexts, not that it works.
Genetic confound
Not applicable.
Replication notes
Not applicable. Recorded as the developer-side position statement against which the independent evidence is read.
DOI / URL
10.1037/amp0000794
Effects
| Outcome | Metric | Value | Measure | Timing | Vs | Horizon | Class |
|---|---|---|---|---|---|---|---|
| Mindset-achievement correlation as defended by the authors | r | 0.24 with high-school grades pooled; up to 0.34 among students facing adversity | mixed | cross-sectional | none | not-applicable | domain-skill |
| Position on the deflated intervention effect | argument | 0.10-0.15 SD is large for a brief cheap scalable intervention | mixed | not applicable | unclear | not-applicable | domain-skill |
Cited by
- Does teaching a growth mindset raise achievement?mixedconf: highgc: low