The DOI matches its Crossref record; 18 of the 18 references it cites were found in a registry or an index; none of the 9 sources its claims cite is checked for support yet.
The accuracy, fairness, and limits of predicting recidivism
Open accessAbstract
Should we trust computers to make life-altering decisions in the criminal justice system?
Overview
CoChat Summary
TL;DR
In a Broward County sample, nonexpert Amazon Mechanical Turk participants predicted recidivism about as accurately and with similar racial disparities as COMPAS, and COMPAS’s proprietary 137-feature score could be matched by simple linear classifiers using as few as two features. The authors conclude that algorithmic recidivism prediction, at least as represented by COMPAS, is no better than untrained human judgment and reaches a performance ceiling around 65% accuracy.
Key findings
- A crowd of nonexperts was statistically indistinguishable from COMPAS: crowd accuracy was 67.0% versus COMPAS’s 65.2% on the 1000-defendant subset (P = 0.85), and AUC-ROC was 0.71 ± 0.03 for participants versus 0.70 ± 0.04 for COMPAS.
- Both humans and COMPAS showed similar racial disparities in error rates: participants’ false-positive rate was 37.1% for black defendants versus 27.2% for white defendants (P = 0.027), while COMPAS’s was 40.4% versus 25.4% (P = 0.002); participants’ false-negative rate was 29.2% for black defendants versus 40.3% for white defendants (P = 0.034), while COMPAS’s was 30.9% versus 47.9% (P = 0.003).
- COMPAS’s accuracy was reproducible with simpler models: logistic regression with 7 features achieved 66.6% accuracy [64.4, 68.9], logistic regression with only 2 features—age and total number of previous convictions—achieved 66.8% [64.3, 69.2], and a nonlinear SVM achieved 65.2% [63.0, 67.2], all comparable to COMPAS’s 65.4% [64.3, 66.5].
- Including race in the human prediction task did not significantly change overall accuracy or fairness: crowd accuracy was 66.5% with race versus 67.0% without race (P = 0.66), while false-positive rates for black versus white defendants remained significantly different (40.0% vs. 26.2%, P = 0.001).
Methods
- Data came from 7214 pretrial defendants in Broward County, Florida, from 2013–2014, including demographics, criminal history, COMPAS recidivism risk scores, and 2-year rearrest records; human assessment used a fixed random subset of 1000 defendants.
- Human participants were recruited via Amazon’s Mechanical Turk: 462 in the no-race condition and 449 in the race condition, with catch trials removing inattentive respondents; each participant judged one of 20 subsets of 50 defendants, and crowd predictions used a majority-rules criterion.
- Algorithmic assessment trained logistic regression and nonlinear SVM classifiers on 7 features over 1000 random 80%/20% training/testing splits, and also tested all subsets of the 7 features, including a 2-feature logistic regression.
Limitations
- Stated in the paper: recidivism is operationalized as rearrest, not direct reoffending, and differences in arrest rates across race complicate direct comparison of false-positive and false-negative rates.
- Stated in the paper: participants were not instructed how to weight features; it remains unknown whether guidelines, dynamic risk factors, larger crowds, or criminal justice expertise would improve accuracy.
- Evident from the text: the human sample was drawn from Mechanical Turk and assumed to have little to no criminal justice expertise, there were not enough nonwhite participants to reliably measure differences across participant race, and COMPAS’s proprietary inner workings were not revealed, limiting independent verification.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Figures
Verification
Claims the paper makes
0 Valid · 1 Not valid · 8 Not checked · 47 Own results and interpretations
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“This predictor yields similar results to COMPAS in terms of predictive fairness [Table 2, (A) and (D) columns].”
§ The accuracy, fairness, and limits of predicting recidivism > RESULTS > Algorithmic assessment
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The COMPAS accuracy for this subset of 1000 defendants was 65.2%.”
§ The accuracy, fairness, and limits of predicting recidivism > MATERIALS AND METHODS
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“With respect to fairness, participant accuracy is not significantly different for black defendants (66.2%) compared with white defendants (67.6%; P = 0.65, unpaired t test).”
§ The accuracy, fairness, and limits of predicting recidivism > RESULTS > Human assessment
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“This crowd-based approach yields a prediction accuracy of 67.0%.”
§ The accuracy, fairness, and limits of predicting recidivism > RESULTS > Human assessment
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The average COMPAS accuracy on 10,000 random subsets of size 1000 each was 65.4% with a 95% confidence interval of (62.6, 68.1).”
§ The accuracy, fairness, and limits of predicting recidivism > MATERIALS AND METHODS
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
20 more claims
Create a free account to see all 56 claims and to verify the 6 references still open.
- It may state two separate claims as one. 28
- Probably not a substantial claim. 6
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?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“In fact, quite a lot of effort has been made by scientists to understand how COMPAS depends on important variables like race and age, without a lot of success (59, 60).”
From Interpretable algorithmic forensics (2023) · cites this paper as [59]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“For instance, this work has demonstrated that simple interpretable models (with two and respectively three features) can perform as well as the widely used criminal justice COMPAS algorithm, which instead draws on 137 features.”
From Accountable Artificial Intelligence: Holding Algorithms to Account (2021) · cites this paper as Dressel and Farid 2018
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Another example is [399], where the authors propose a fair-by-design approach in order to develop ML models that jointly have less bias and include as explanations human comprehensible rules.”
From Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI (2020) · cites this paper as [400]
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“Dressel and Farid (2018) achieved this result by using a linear predictor-logistic regressor that made use of only two variables (age and total number of previous convictions of the subject).”
From Ethical principles in machine learning and artificial intelligence: cases from the field and possible ways forward (2020) · cites this paper as Dressel and Farid (2018)
The cited source says what the claim attributes to it.
In the citing paper
“Recent research for predicting social outcomes has shown that even with complex models and large datasets, there are strong limits to predictive performance.49,83”
From Leakage and the reproducibility crisis in machine-learning-based science (2023) · cites this paper as [83]
2 more claims about this paper
Create a free account to see every claim other papers make about this one.
Citations
Cites
Cortes, C., & Vapnik, V. (1995) · Machine Learning · doi:10.1007/bf00994018 · Cited by 59,421
The Regression Analysis of Binary Sequences
Cox, D. R. (1958) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/j.2517-6161.1958.tb00292.x · Cited by 2,813
Predicting relapse: A meta-analysis of sexual offender recidivism studies.
Hanson, R. K., & Bussière, M. T. (1998) · Journal of Consulting and Clinical Psychology · doi:10.1037/0022-006x.66.2.348 · Cited by 2,375
A META‐ANALYSIS OF THE PREDICTORS OF ADULT OFFENDER RECIDIVISM: WHAT WORKS!*
GENDREAU, P., LITTLE, T., & GOGGIN, C. (1996) · Criminology · doi:10.1111/j.1745-9125.1996.tb01220.x · Cited by 1,244
Hanson, R. K., & Morton-Bourgon, K. E. (2009) · Psychological Assessment · doi:10.1037/a0014421 · Cited by 1,050
Inherent Trade-Offs in the Fair Determination of Risk Scores
Kleinberg, J., Mullainathan, S., & Raghavan, M. (2016) · arXiv · doi:10.48550/arxiv.1609.05807 · Cited by 941
Where Should We Intervene?: Dynamic Predictors of Sexual Offense Recidivism
HANSON, R. K., & HARRIS, A. J. R. (2000) · Criminal Justice and Behavior · doi:10.1177/0093854800027001002 · Cited by 793
Predictive Policing: The Role of Crime Forecasting in Law Enforcement Operations
Perry, W., McInnis, B., Price, C., Smith, S., & Hollywood, J. (2013) · doi:10.7249/rr233 · Cited by 693
The efficacy of violence prediction: a meta-analytic comparison of nine risk assessment tools.
Yang, M., Wong, S. C. P., & Coid, J. (2010) · Psychological Bulletin · doi:10.1037/a0020473 · Cited by 525
The Robust Beauty of Majority Rules in Group Decisions.
Hastie, R., & Kameda, T. (2005) · Psychological Review · doi:10.1037/0033-295x.112.2.494 · Cited by 462
Cited by
Arrieta, A. B., Díaz-Rodríguez, N., Del Ser, J., Bennetot, A., Tabik, S., Barbado, A., García, S., Gil-López, S., Molina, D., Benjamins, R., et al. (2019) · arXiv · doi:10.48550/arxiv.1910.10045 · Cited by 11,321
Barredo Arrieta, A., Díaz-Rodríguez, N., Del Ser, J., Bennetot, A., Tabik, S., Barbado, A., Garcia, S., Gil-Lopez, S., Molina, D., Benjamins, R., et al. (2020) · Information Fusion · doi:10.1016/j.inffus.2019.12.012 · Cited by 10,423
A Survey on Bias and Fairness in Machine Learning
Mehrabi, N., Morstatter, F., Saxena, N., Lerman, K., & Galstyan, A. (2022) · ACM Computing Surveys · doi:10.1145/3457607 · Cited by 5,812
Leakage and the reproducibility crisis in machine-learning-based science
Kapoor, S., & Narayanan, A. (2023) · Patterns · doi:10.1016/j.patter.2023.100804 · Cited by 1,387
“Recent research for predicting social outcomes has shown that even with complex models and large datasets, there are strong limits to predictive performance.(49,83) However, results such as the better-than-human performance of ML models in perception tasks such as image classification(88,89) give the impression of ML…”
BackgroundResultRahwan, I., Cebrian, M., Obradovich, N., Bongard, J., Bonnefon, J.-F., Breazeal, C., Crandall, J. W., Christakis, N. A., Couzin, I. D., Jackson, M. O., et al. (2019) · Nature · doi:10.1038/s41586-019-1138-y · Cited by 1,364
Metadata
- Authors
- Dressel, J., & Farid, H.
- Affiliations
- Dartmouth College (US)
- Corresponding author
- Farid, Hany
- Venue
- Science Advances
- Published
- 2018-01-05
- Volume
- 4
- Issue
- 1
- Pages
- eaao5580-eaao5580
- ISSN
- 2375-2548 (ISSN-L)
- PMC
- PMC5777393
- OpenAlex
- W2785011159
- Cited by
- 1,071
- References
- 17
- Citations by year
- 2026: 138 · 2025: 124 · 2024: 149 · 2023: 153 · 2022: 126 · 2021: 127 · 2020: 148 · 2019: 76 · 2018: 29
- FWCI
- 188.76
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- cc-by-nc
- Locations
- Science Advances (publishedVersion, cc-by-nc)
- Europe PMC (PubMed Central) (publishedVersion, cc-by-nc)
- PubMed Central (publishedVersion, cc-by-nc)
- PubMed (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Ethics and Social Impacts of AI
- Social Sciences › Sociology and Political Science › Criminal Justice and Corrections Analysis
- Social Sciences › Law › Artificial Intelligence in Law
- Keywords
- recidivism prediction, COMPAS, risk assessment software, algorithmic fairness, algorithmic accuracy, criminal justice, parole decisions, sentencing, linear predictor, algorithmic bias
- MeSH terms
- Recidivism, Algorithms, Humans, ROC Curve, Area Under Curve
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 2026-11-09, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-11-09 (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: The accuracy, fairness, and limits of predicting recidivism. https://app.cochat.ai/papers/10.1126/sciadv.aao5580
- The paper
- Dressel, J., & Farid, H. (2018). The accuracy, fairness, and limits of predicting recidivism. Science Advances. https://doi.org/10.1126/sciadv.aao5580
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.
