Harvard Education Literature Review Example: Growth Mindset Interventions

5 education-specific scoring dimensions (combined with Harvard)

Dimension Poor signal Strong signal

1 Theory specification "Growth mindset raises grades" Names the causal chain: implicit theory → response to challenge/failure → sustained effort → grade outcome

2 Effect-size + moderator "d = .08, so it works" Reports d̄ = .08 [.02, .14] and the moderators: low SES, academically at-risk (Sisk 2018) + racial-minority, low-achieving, US schools (Yeager 2019)

3 Outcome fidelity Self-reported mindset only Achievement via GPA / standardised test / course grades; intervention delivered with fidelity

4 Design ladder One quasi-experiment RCT (Blackwell 2007) → meta-analysis (Sisk 2018) → national experiment (Yeager 2019) → longitudinal (Golke 2020)

5 Nonlinearity Linear "more malleability = better" Curvilinear account (Golke 2020): moderate malleability associated with better trajectories; extreme fixed and extreme growth not

Strong draft (Harvard / Cite Them Right 12th, annotated)

[Thesis / scope] Growth-mindset interventions do change achievement, but the effect is small on average, substantively conditional, and curvilinear rather than linear: the pupils who gain most are those already underperforming or from low-SES and minority backgrounds, while a simplistic "more malleability is always better" message may miss the pupils who need a calibrated ("work smarter, not just harder") message. [Narrative, theory] Dweck (2006) proposed that implicit theories of intelligence—whether ability is fixed or malleable—shape how pupils interpret challenge and respond to failure; a malleable theory sustains effort where a fixed theory produces helplessness. [RCT] Blackwell, Trzesniewski and Dweck (2007) tested this chain in an RCT with approximately 400 seventh-graders followed across eight months: pupils randomly assigned to an intervention teaching that intelligence is malleable showed greater increases in mathematics grades and more positive response-to-challenge beliefs than control pupils, controlling for prior achievement. [Meta-analysis] Sisk, Burgoyne, Sun, Butler and Macnamara (2018) pooled 43 studies and 57,155 participants and estimated a mean effect of d̄ = .08, 95% CI [.02, .14]—small overall—but with meaningful moderation: effects were larger for pupils with lower socioeconomic status and those academically at-risk. [National experiment — compensatory profile] The compensatory pattern is corroborated by a national US experiment: Yeager, Hanselman, Walton, Murray, Chang, et al. (2019) tested a brief growth-mindset intervention across 65 US high schools (approximately 12,500 ninth-graders) and found that lower-achieving pupils, racial-minority pupils, and those in low-SES schools showed the largest GPA gains—confirming that mindset effects are concentrated among the pupils who need them most, not universal. [Longitudinal + curvilinearity] Golke, Dörfler and Artelt (2020) supplied the nonlinearity evidence: a German longitudinal study found that pupils with moderate (malleable-but-not-too-malleable) implicit-theory beliefs showed more favourable achievement trajectories than those with extreme fixed or extreme growth mindsets. [Synthesis] Across the evidence, the defensible conclusion is a curvilinear, context-bound mechanism: moderate malleability beliefs, paired with a strategy-rich environment, support achievement; an uncalibrated growth-mindset slogan does not. [Precise gap] What remains untested is whether the US/low-SES/low-achiever advantage generalises to high-achieving and high-SES pupils in non-US, non-English-speaking school systems—a gap that matters because the Sisk et al. (2018) moderator profile and Yeager et al. (2019) national findings are pronounced enough to invalidate a universal effect-size claim.

[Evidence-quality subsection] Design quality strengthens the curvilinear account. [Narrative, RCT] Blackwell, Trzesniewski and Dweck (2007) supplied the causal prototype: random assignment, longitudinal follow-up, and control for prior achievement make its estimate credible for the mechanism, if not for a population effect size. [Parenthetical, meta] Sisk et al. (2018) broadened the base to 43 studies and 57,155 participants, but its d̄ = .08 [.02, .14] means any single-school claim must be modest. [Narrative, national experiment] Yeager et al. (2019) is the largest real-world test (65 schools, ~12,500 pupils), confirming the compensatory profile at scale. [Narrative, 3 authors] Golke, Dörfler and Artelt (2020) adds the longitudinal nonlinearity: moderate malleability, not extreme growth, tracks with achievement gains. [Synthesis] A GRADE-style reading yields moderate certainty for the causal chain in underperforming/low-SES/minority samples, low certainty for a universal effect size, and moderate certainty for curvilinearity. [Gap restatement] Future reviews should separate fidelity of delivery, pupil baseline achievement, and national/linguistic context before pooling.

[Implementation subsection] Translation is the second gap. [Narrative, 2 authors] Yeager and Walton (2011) argued that successful mindset interventions must change the meaning of difficulty—framing struggle as evidence of learning rather than lack of ability—and must be embedded in supportive contexts that supply the strategies pupils need. [Parenthetical] This explains why the Golke et al. (2020) middle band, not the extreme growth group, improved: an extreme growth message without strategy support may simply demand more effort in the absence of tools. [Precise gap] A literature review that recommends a universal growth-mindset programme without modelling pupil baseline, intervention fidelity, and strategy support will systematically overstate transferability to well-resourced, high-achieving schools.

In-text citation audit (Harvard / CTR 12th, convention locked from 11a)

• 1 author: Dweck (2006) / (Dweck, 2006)

• 2 authors: Yeager and Walton (2011) / (Yeager and Walton, 2011) — and

• 3 authors, FIRST citation: (Blackwell, Trzesniewski and Dweck, 2007) — all three listed

• 3 authors, SUBSEQUENT: (Blackwell et al., 2007) — et al. roman

• 5 authors, FIRST citation: (Sisk, Burgoyne, Sun, Butler and Macnamara, 2018) — all five listed

• 5 authors, SUBSEQUENT: (Sisk et al., 2018)

• 18 authors, FIRST citation: (Yeager, Hanselman, Walton, Murray, Chang et al., 2019) — first six + et al. per CTR 12th

• 18 authors, SUBSEQUENT: (Yeager et al., 2019)

• Multi-source: (Blackwell et al., 2007; Golke, Dörfler and Artelt, 2020; Sisk et al., 2018; Yeager et al., 2019) — alphabetical

• Page number: (Sisk et al., 2018, p. 560)

Reference list (Harvard / Cite Them Right 12th)

Reference list

Blackwell, L. S., Trzesniewski, K. H. and Dweck, C. S. (2007) 'Implicit theories of intelligence predict achievement across an adolescent transition: a longitudinal study and an intervention', Child Development, 78(1), pp. 246–263. Available at: https://doi.org/10.1111/j.1467-8624.2007 .00995.x

Dweck, C. S. (2006) Mindset: the new psychology of success. New York: Random House.

Golke, S., Dörfler, T. and Artelt, C. (2020) 'The development of implicit theories of intelligence across the transition to secondary school: a longitudinal intervention study', AERA Open, 6(4). Available at: https://doi.org/10.1177/2332858420976034

Sisk, V. F., Burgoyne, A. P., Sun, J., Butler, J. L. and Macnamara, B. N. (2018) 'To what extent and under which circumstances are growth mindset interventions effective? Two meta-analyses', Psychological Science, 29(4), pp. 549–571. Available at: https://doi.org/10.1177/0956797617739704

Yeager, D. S., Hanselman, P., Walton, G. M., Murray, J. S., Chang, R., et al. (2019) 'A national experiment reveals where a growth mindset improves achievement', Nature, 573, pp. 364–369. Available at: https://doi.org/10.1038/s41586-019-1466-y

Yeager, D. S. and Walton, G. M. (2011) 'Social-psychological interventions in education: they're not "magic bullets"', Psychological Science in the Public Interest, 15(3), pp. 159–196. Available at: https://doi.org/10.1177/1529100614524592

Format checks (CTR 12th): • "Reference list", alphabetized B → D → G → S → Y → Y

• Surname, Initial(s). — Blackwell, L. S. not Blackwell, LS

• Single quotes on article titles, sentence case; journals italicized title case

• 29(4), pp. 549–571 (en dash); 6(4) with no page range acceptable for AERA Open

• et al. roman; all authors listed in list (Sisk = 5 fully listed; Yeager 2019 = 18 fully listed)

• Available at: https://doi.org/... — no access date

• Book: Dweck, C. S. (2006) Mindset: the new psychology of success. New York: Random House. — place: publisher retained

What changed: poor draft → strong Harvard draft

Element Poor draft Strong Harvard draft (above)

Opening "Growth mindset raises grades." Curvilinear, compensatory, context-bound thesis

Theory Named Dweck only Full causal chain: implicit theory → response to challenge → effort → grade

Effect size None or overstated d̄ = .08 [.02, .14] (Sisk 2018) + moderators (low SES / at-risk)

National corroboration Absent Yeager et al. 2019 Nature 573:364–369 (65 schools, ~12,500 pupils, compensatory profile)

Design ladder Quasi-experiment RCT → meta-analysis → national experiment → longitudinal

Curvilinearity Omitted Core mechanism: moderate malleability associated with better trajectories; extremes do not (Golke 2020)

Citation mechanics APA &/double-quote/et al. italic and/single-quote/roman et al./place retained

Multi-source One per sentence Alphabetical, semicolon-separated

Gap "More research needed." High-achiever × high-SES × non-US/non-English context

Reference list APA "References" "Reference list" CTR 12th

5 fatal signals (Harvard × Education)

1. Treating d̄ = .08 as either "it works" or "it's useless"—Sisk et al. (2018) shows it is small on average but larger for the pupils who need it most (low SES, academically at-risk).

2. Citing Yeager and Walton (2011) as a primary study—it is a review/position piece (and in Psychological Science in the Public Interest, not the Psychological Science main journal).

3. Using & in citations ((Yeager & Walton, 2011)) — Harvard uses and.

4. Italicizing et al. or double-quoting article titles.

5. Recommending a universal programme without fidelity, baseline achievement, and strategy-support moderators—overclaims a conditional intervention into a slogan.

Rewrite Checklist (Harvard × Education)

1. Open with a mechanism-level, conditional thesis (curvilinear + compensatory), not "does mindset work?"

2. Name the causal chain from Dweck (2006): implicit theory → challenge interpretation → sustained effort → grade.

3. Anchor with Blackwell et al. (2007) for the RCT prototype; report Sisk et al.'s (2018) d̄ = .08 [.02, .14] with its moderators (low SES, academically at-risk).

4. Add Yeager et al. (2019) for the national-scale compensatory profile (racial-minority, low-achieving, low-SES schools).

5. Make Golke, Dörfler and Artelt (2020) the curvilinearity anchor: moderate malleability, not extreme growth, tracks with achievement.

6. Convert ≥40% parentheticals to narrative; et al. roman; 3+ authors full list on first citation.

7. Audit reference list: single quotes, italics journal, Place: Publisher for books, Available at: https://doi.org/..., no access date for DOI.

8. Alphabetize by surname; list all authors in the reference list.

9. Close with a gap naming pupil baseline × intervention fidelity × national context (e.g., high-achieving, high-SES, non-US schools).

10. Verify every DOI resolves before publishing.

FAQ

Why is the effect size so small (d̄ = .08)? Does that mean mindset interventions fail? No. Sisk et al. (2018) found d̄ = .08, 95% CI [.02, .14] overall—small, but with larger effects for low-SES and academically at-risk pupils. Yeager et al. (2019) then confirmed at national scale that lower-achieving, racial-minority, and low-SES-school pupils gained the most. A small average with a strong subgroup profile is a compensatory intervention, not a failure. Report the moderator profile, not just the mean.

What does "malleable-but-not-too-malleable" mean? Golke, Dörfler and Artelt (2020) found that pupils with moderate malleability beliefs showed more favourable achievement trajectories; those with extreme fixed mindsets and those with extreme growth mindsets did not. The interpretation: an extreme growth message without strategy support can become "just try harder," which is ineffective. The review-grade mechanism is therefore curvilinear, not linear.

Is Yeager and Walton 2011 in Psychological Science? No. It is in Psychological Science in the Public Interest (PSPI), volume 15(3), 159–196—a different Sage publication from the Psychological Science main journal. This matters because PSPI publishes extended reviews, not empirical reports. Cite it as a review/position piece, not primary evidence.

How does this page relate to APA × Education? /apa-education-literature-review-example covers the same literature cluster in APA 7th (US convention: & in citations, double quotes, italic et al., no place of publication). This page applies the Harvard (CTR 12th) format layer (and, single quotes, roman et al., place retained) to the same or adjacent sources. Read the two together to see how the format changes while the argument stays.

Can I reuse this review's citations? No. Use the structure and Harvard mechanics as a model; replace sources with your own verified search, and confirm each DOI resolves.

/harvard-literature-review-example · /literature-review-example-annotated-sample#by-citation-style · /apa-education-literature-review-example · /harvard-sociology-literature-review-example

Continue the research workflow with Acade

Acade can help you search for literature, organize evidence notes, build a synthesis outline, and revise citation-supported academic writing. It cannot guarantee source accuracy or replace your reading, methodological judgment, institutional rules, or responsibility for the submitted work.

Tell Acade your topic, discipline, citation style, evidence requirements, and assignment constraints, then ask it to help organize the next verifiable research step.