Student or educator? Sign up with your school email and get 6 months of Pro free.

The DOI matches its Crossref record; 103 of the 104 references it cites were found in a registry or an index; none of the 37 sources its claims cite is checked for support yet.

Prejudice Reduction: Progress and Challenges

Open access

Abstract

The past decade has seen rapid growth in research that evaluates methods for reducing prejudice. This essay reviews 418 experiments reported in 309 manuscripts from 2007 to 2019 to assess which approaches work best and why. Our quantitative assessment uses meta-analysis to estimate average effects. Our qualitative assessment calls attention to landmark studies that are noteworthy for sustained interventions, imaginative measurement, and transparency. However, 76% of all studies evaluate light touch interventions, the long-term impact of which remains unclear. The modal intervention uses mentalizing as a salve for prejudice. Although these studies report optimistic conclusions, we identify troubling indications of publication bias that may exaggerate effects. Furthermore, landmark studies often find limited effects, which suggests the need for further theoretical innovation or synergies with other kinds of psychological or structural interventions. We conclude that much research effort is theoretically and empirically ill-suited to provide actionable, evidence-based recommendations for reducing prejudice.

Overview

CoChat Summary

TL;DR

This review synthesizes 418 experimental studies in 309 manuscripts (2007–2019) on prejudice reduction and finds an overall meta-analytic effect of d = 0.357, but also strong signs that small studies and publication bias inflate apparent effectiveness. The paper argues that many interventions—especially light-touch and implicit-bias approaches—look promising in small studies but shrink markedly in larger, more rigorous studies.

Key findings

  • The database includes 418 studies, 1,292 distinct point estimates, and a random-effects meta-analysis yields an overall effect size of d = 0.357 (SE = 0.02), with p < 0.0001.
  • Effects are much smaller in larger studies: the top quintile of treatment sample size yields d = 0.187 (SE = 0.022), compared with d = 0.606 (SE = 0.052) in the smallest quintile; the paper says the average effect size drops 48% to 0.187.
  • Light-touch interventions make up 76% of studies; a meta-analysis of light-touch interventions gives d = 0.35 (SE = 0.02), but the estimate falls to d = 0.16 (SE = 0.02) in the highest sample-size quintile. Only 8% of light-touch studies measure outcomes at least one day later, and 1% at least a month later.
  • Specific intervention categories show moderate average effects in the full sample, but smaller effects in larger studies: cognitive/emotional training d = 0.35 overall and d = 0.22 for treatment N ≥ 78; social categorization d = 0.37 overall and d = 0.31 for N ≥ 78; face-to-face contact d = 0.28 overall and d = 0.25 for outcomes measured at least one day later; extended/imagined contact d = 0.37 overall but d = 0.12 when restricted to studies with treatment samples of 78 or more.
  • The paper reports that 34 studies measured both implicit and explicit attitudes, with no correlation between estimated effects on the two measures (r = 0.02), and that only 7% of outcomes were behavioral.

Methods

  • Narrative review plus quantitative meta-analysis of experimental prejudice-reduction studies published from 2007 to 2019, following biomedical meta-analytic standards and PRISMA-style screening.
  • Search across five databases plus a text-based search in Articles+, using 19 keywords; the exhaustive search yielded over 16,000 manuscripts, winnowed to 1,835 full-text-reviewed manuscripts and 309 qualifying manuscripts.
  • The authors coded intervention type, outcomes, sample characteristics, timing, openness practices, and effect sizes; they averaged Cohen’s d within studies, then ran random-effects meta-analysis with standard errors clustered by article.

Limitations

  • The paper explicitly states that many studies are underpowered, with widespread publication bias and selective reporting; it notes that small studies report much stronger effects than large studies, suggesting inflated estimates.
  • The authors state that few studies make data and code public, many do not report key statistics such as standard errors, and some selectively report only subsets of results, which limited their ability to calculate statistics themselves.
  • Evident from the text: the review is restricted to experimental studies and excludes nonexperimental work, several related literatures, and most gender-based prejudice, so it does not cover all prejudice-reduction research.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1146/annurev-psych-071620-030619

Cites

103 of 103 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Verification

Claims the paper makes

0 Valid · 5 Not valid · 34 Not checked · 191 Own results and interpretations

162 more claims

Create a free account to see all 230 claims and to verify the 37 references still open.

See all claims — free account
  • It may state two separate claims as one. 30
  • Probably not a substantial claim. 36
  • The quote only partly states this claim. 3

Create a free account to see these claims.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

1 of 1

Citations

Cites

Show all 104 references

Research works printed but not linked: 6

  1. Allport G., 1954, The Nature of Prejudice
  2. Castillo JLÁ, 2014, Rev. Latinoam. Psicol., 46, 137
  3. Chaney KE., 2016, Ends: the endurance, depth, and scope of confronting as a prejudice reduction strategy
  4. Fiske S, 1999, Dual Process Theories in Social Psychology, 703
  5. Jost JT, 2009, Res. Organ. Behav., 29, 39
  6. Patel SL., 2013, Examining the influence of perceived social consensus information on weight prejudice across development

Cited by

See all 704

Metadata

Authors
Paluck, E. L., Porat, R., Clark, C. S., & Green, D. P.
Affiliations
  • Princeton University (US)
  • Hebrew University of Jerusalem (IL)
  • Columbia University (US)
Corresponding author
Paluck, Elizabeth Levy
Venue
Annual Review of Psychology
Published
2021-01-04
Volume
72
Issue
1
Pages
533-560
ISSN
  • 0066-4308 (ISSN-L)
  • 1545-2085
OpenAlex
W3085414015
Cited by
686
References
104
Citations by year
2026: 114 · 2025: 150 · 2024: 124 · 2023: 128 · 2022: 104 · 2021: 62
FWCI
12.94
Type
journal article
Language
en
Open access
Locations
  • Annual Review of Psychology (publishedVersion)
  • PubMed (publishedVersion)
Topics
  • Social Sciences › Sociology and Political Science › Social and Intergroup Psychology
  • Social Sciences › Psychology › Social Psychology › Death Anxiety and Social Exclusion
  • Social Sciences › Psychology › Social Psychology › Psychology of Social Influence
Keywords
prejudice reduction, intergroup bias, mentalization, publication bias, intervention effectiveness, evidence-based recommendations
MeSH terms
Female, Humans, Male, Prejudice

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

A language model drew the claims from the paper on 2026-10-11, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-11 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Prejudice Reduction: Progress and Challenges. https://app.cochat.ai/papers/10.1146/annurev-psych-071620-030619
The paper
Paluck, E. L., Porat, R., Clark, C. S., & Green, D. P. (2021). Prejudice Reduction: Progress and Challenges. Annual Review of Psychology. https://doi.org/10.1146/annurev-psych-071620-030619

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.