Precinct or prejudice? Understanding racial disparities in New York City’s stop-and-frisk policy
Open accessThe DOI matches its Crossref record; 42 of the 44 references it cites were found in a registry or an index.
Abstract
Recent studies have examined racial disparities in stop-and-frisk, a widely employed but controversial policing tactic. The statistical evidence, however, has been limited and contradictory. We investigate by analyzing three million stops in New York City over five years, focusing on cases where officers suspected the stopped individual of criminal possession of a weapon (CPW). For each CPW stop, we estimate the ex ante probability that the detained suspect has a weapon. We find that in more than 40% of cases, the likelihood of finding a weapon (typically a knife) was less than 1%, raising concerns that the legal requirement of “reasonable suspicion” was often not met. We further find that blacks and Hispanics were disproportionately stopped in these low hit rate contexts, a phenomenon that we trace to two factors: (1) lower thresholds for stopping individuals—regardless of race—in high-crime, predominately minority areas, particularly public housing; and (2) lower thresholds for stopping minorities relative to similarly situated whites. Finally, we demonstrate that by conducting only the 6% of stops that are statistically most likely to result in weapons seizure, one can both recover the majority of weapons and mitigate racial disparities in who is stopped. We show that this statistically informed stopping strategy can be approximated by simple, easily implemented heuristics with little loss in efficiency.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
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
“The corresponding values of r from these two studies are 0.32 and a lower bound of 0.34, respectively, when considering black civilians.”
From Administrative Records Mask Racially Biased Policing (2020) · cites this paper as Goel, Rao, and Shroff (2016)
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
“We consider location and race as our domains because previous work has shown that they can produce substantial disparities in policing practices and in algorithmic performance (Goel et al., 2016).”
From WILDS: A Benchmark of in-the-Wild Distribution Shifts (2020) · cites this paper as Goel et al., 2016
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
“Criminal justice researchers have also established that racial bias permeates arrest records, as Black Americans and other minority groups are disproportionately stopped, searched, and arrested by police for the same behaviors exhibited by White Americans [71][72][73].”
From Efficiency vs. Equity: A Structured Interdisciplinary Review of AI in Criminal Justice Risk Assessments (2026) · cites this paper as [73]
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
“Criminal justice researchers have also established that racial bias permeates arrest records, as Black Americans and other minority groups are disproportionately stopped, searched, and arrested by police for the same behaviors exhibited by White Americans [71][72][73].”
From Efficiency vs. Equity: A Structured Interdisciplinary Review of AI in Criminal Justice Risk Assessments (2026) · cites this paper as [73]
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“Other researchers developed a simple scoring system to address unnecessary use of stop and frisk by the New York City Police Department (66).”
From Interpretable algorithmic forensics (2023) · cites this paper as [66]
3 more claims about this paper
Create a free account to see every claim other papers make about this one.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Scikit-learn: Machine Learning in Python
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Müller, A., Nothman, J., Louppe, G., et al. (2012) · arXiv · doi:10.48550/arxiv.1201.0490 · Cited by 71,016
Reasoning the fast and frugal way: Models of bounded rationality.
Gigerenzer, G., & Goldstein, D. G. (1996) · Psychological Review · doi:10.1037/0033-295x.103.4.650 · Cited by 3,504 · cited 2×
“…these difficulties, we draw on a large body of work which has found that simple, transparent and interpretable heuristics often work as well as complex statistical models [Czerlinski, Gigerenzer and Goldstein (1999), Gigerenzer and Goldstein (1996), Lovie and Lovie (1986), Ustun and Rudin (2014)].”
BackgroundThe Economics of Discrimination
Leonard, R. C., & Becker, G. S. (1957) · The American Catholic Sociological Review · doi:10.2307/3708738 · Cited by 2,818
Arrow, K. J. (2015) · Discrimination in Labor Markets · doi:10.1515/9781400867066-003 · Cited by 1,804 · cited 12×
“Arrow, K. (1973). The theory of discrimination. Discrimination in labor markets 3. Ayres, I. (2002). Outcome tests of racial disparities in police practices. Justice research and Policy 4 131–142. Becker, G. S. (1993). Nobel lecture: The economic way of looking at behavior. Journal of political economy 385–409. Bottou…”
BackgroundInfluentialNobel Lecture: The Economic Way of Looking at Behavior
Becker, G. S. (1993) · Journal of Political Economy · doi:10.1086/261880 · Cited by 1,609 · cited 7×
“Coviello and Persico (2013) fit an economic model of behavior to the stop-and-frisk data, and found no evidence of racial bias. Finally, investigating the ramifications of local events on policing, Legewie (2014) showed that in the days following fatal shootings of two NYPD officers by black suspects, there was an inc…”
BackgroundInfluentialSeeing Black: Race, Crime, and Visual Processing.
Eberhardt, J. L., Goff, P. A., Purdie, V. J., & Davies, P. G. (2004) · Journal of Personality and Social Psychology · doi:10.1037/0022-3514.87.6.876 · Cited by 1,208
Gelman, A., Fagan, J., & Kiss, A. (2007) · Journal of the American Statistical Association · doi:10.1198/016214506000001040 · Cited by 873
L1Penalized Estimation in the Cox Proportional Hazards Model
Goeman, J. J. (2010) · Biometrical Journal · doi:10.1002/bimj.200900028 · Cited by 816 · cited 1×
“Model coefficients were computed with the penalized package in R (Goeman, 2010).”
MethodPulled Over: How Police Stops Define Race and Citizenship
Epp, C. R., Maynard-Moody, S., & Haider-Markel, D. P. (2014) · doi:10.7208/chicago/9780226114040.001.0001 · Cited by 687
Racial Bias in Motor Vehicle Searches: Theory and Evidence
Knowles, J., Persico, N., & Todd, P. (2001) · Journal of Political Economy · doi:10.1086/318603 · Cited by 608
A Kernel Method for Smoothing Point Process Data
Diggle, P. (1985) · Applied Statistics · doi:10.2307/2347366 · Cited by 601
Penalized regression, standard errors, and Bayesian lassos
Casella, G., Ghosh, M., Gill, J., & Kyung, M. (2010) · Bayesian Analysis · doi:10.1214/10-ba607 · Cited by 567
On-line Learning and Stochastic Approximations
Bottou, L. (1999) · On-Line Learning in Neural Networks · doi:10.1017/cbo9780511569920.003 · Cited by 564
An Alternative Test of Racial Prejudice in Motor Vehicle Searches: Theory and Evidence
Anwar, S., & Fang, H. (2006) · American Economic Review · doi:10.1257/000282806776157579 · Cited by 293
A New Look at Racial Profiling: Evidence from the Boston Police Department
Antonovics, K., & Knight, B. G. (2009) · Review of Economics and Statistics · doi:10.1162/rest.91.1.163 · Cited by 292 · cited 1×
“Examining traffic stops by the Boston Police Department, however, Antonovics and Knight (2009) show that officers are more likely to conduct a search if the race of the officer differs from the race of the driver, consistent with taste-based racial discrimination.”
BackgroundHow good are simple heuristics
Czerlinski, J., Gigerenzer, G., & Goldstein, D. G. (1999) · Max Planck Institute for Plasma Physics · Cited by 280
Testing for Racial Profiling in Traffic Stops From Behind a Veil of Darkness
Grogger, J., & Ridgeway, G. (2006) · Journal of the American Statistical Association · doi:10.1198/016214506000000168 · Cited by 260 · cited 1×
“In a novel design, Grogger and Ridgeway (2006) analyzed traffic stops in Oakland and showed that the racial distribution of stopped individuals during the day, when a suspect’s race is readily apparent, matches the distribution at night, when the “veil of darkness” masks race, and thus concluded…”
BackgroundStatistical Discrimination or Prejudice? A Large Sample Field Experiment
Ewens, M., Tomlin, B., & Wang, L. C. (2014) · The Review of Economics and Statistics · doi:10.1162/rest_a_00365 · Cited by 254 · cited 1×
“Also, there is evidence that officers follow “scripts of suspicion” when filling out forms to justify stops [Fagan and Geller (2014)].”
BackgroundStaying out of Sight? Concentrated Policing and Local Political Action
Lerman, A. E., & Weaver, V. (2014) · The ANNALS of the American Academy of Political and Social Science · doi:10.1177/0002716213503085 · Cited by 226
Criminal Justice Forecasts of Risk: A Machine Learning Approach
Berk, R. A. (2012) · SpringerBriefs in computer science · doi:10.1007/978-1-4614-3085-8 · Cited by 192
Legewie, J. (2016) · American Journal of Sociology · doi:10.1086/687518 · Cited by 187
Assessing the Effect of Race Bias in Post-traffic Stop Outcomes Using Propensity Scores
Ridgeway, G. (2006) · Journal of Quantitative Criminology · doi:10.1007/s10940-005-9000-9 · Cited by 176 · cited 2×
“In a novel design, Grogger and Ridgeway (2006) analyzed traffic stops in Oakland and showed that the racial distribution of stopped individuals during the day, when a suspect’s race is readily apparent, matches the distribution at night, when the “veil of darkness” masks race, and thus concluded…”
BackgroundOutcome Tests of Racial Disparities in Police Practices
Ayres, I. (2002) · Justice Research and Policy · doi:10.3818/jrp.4.1.2002.131 · Cited by 156 · cited 3×
“Hit rates are regularly used to assess the level of proof applied when stopping a suspect, with lower hit rates corresponding to less stringent standards (Ayres, 2002; Becker, 1993).”
BackgroundRacial Bias in Motor Vehicle Searches: Theory and Evidence
Knowles, J., Persico, N., & Todd, P. (1999) · National Bureau of Economic Research · doi:10.3386/w7449 · Cited by 132
Ridgeway, G. K. (2008) · Cited by 127
Considering Cost Asymmetry in Learning Classifiers
Bach, F., Heckerman, D. E., & Horvitz, E. (2005) · Journal of Machine Learning Research · doi:10.5555/1248547.1248610 · Cited by 120
The flat maximum effect and linear scoring models for prediction
Lovie, A. D., & Lovie, P. (1986) · Journal of Forecasting · doi:10.1002/for.3980050303 · Cited by 106 · cited 2×
“…these difficulties, we draw on a large body of work which has found that simple, transparent and interpretable heuristics often work as well as complex statistical models [Czerlinski, Gigerenzer and Goldstein (1999), Gigerenzer and Goldstein (1996), Lovie and Lovie (1986), Ustun and Rudin (2014)].”
BackgroundA New Look at Racial Profiling: Evidence from the Boston Police Department
Antonovics, K., & Knight, B. (2004) · National Bureau of Economic Research · doi:10.3386/w10634 · Cited by 98
An Alternative Test of Racial Prejudice in Motor Vehicle Searches: Theory and Evidence
Anwar, S., & Fang, H. (2005) · National Bureau of Economic Research · doi:10.3386/w11264 · Cited by 92
Ridgeway, G., & MacDonald, J. M. (2009) · Journal of the American Statistical Association · doi:10.1198/jasa.2009.0034 · Cited by 89
Racial Profiling? Detecting Bias Using Statistical Evidence
Persico, N. (2009) · Annual Review of Economics · doi:10.1146/annurev.economics.050708.143307 · Cited by 58
Statistical Discrimination or Prejudice? A Large Sample Field ExperimentPreprint
Ewens, M., Tomlin, B., & Wang, L. C. (2012) · SSRN Electronic Journal · doi:10.2139/ssrn.1816702 · Cited by 53
Following the Script: Narratives of Suspicion in Terry Stops in Street Policing
Fagan, J., & Geller, A. (2014) · The University of Chicago Law Review · doi:10.7916/d8jw8dzh · Cited by 50 · cited 1×
“Finally, we repeat our primary analysis with a random forest classifier, considered to be one of the best statistical methods for large-scale classification [Fernández-Delgado et al. (2014)].”
MethodMethods and Models for Interpretable Linear Classification
Ustun, B., & Rudin, C. D. (2014) · arXiv (Cornell University) · doi:10.48550/arxiv.1405.4047 · Cited by 28 · cited 1×
“…these difficulties, we draw on a large body of work which has found that simple, transparent and interpretable heuristics often work as well as complex statistical models [Czerlinski, Gigerenzer and Goldstein (1999), Gigerenzer and Goldstein (1996), Lovie and Lovie (1986), Ustun and Rudin (2014)].”
BackgroundAn Economic Analysis of Black-White Disparities in NYPD's Stop and Frisk Program
Coviello, D., & Persico, N. (2013) · National Bureau of Economic Research · doi:10.3386/w18803 · Cited by 25 · cited 1×
“Coviello and Persico (2013) fit an economic model of behavior to the stop-and-frisk data and found no evidence of racial bias.”
BackgroundMilgram, A., Holsinger, A. M., Vannostrand, M., & Alsdorf, M. W. (2015) · Federal Sentencing Reporter · doi:10.1525/fsr.2015.27.4.216 · Cited by 21
(1911) · The Yale Law Journal · doi:10.2307/785460 · Cited by 7
Debate: The Constitutionality of Stop-and-Frisk in New York City
Rudovsky, D., & Rosenthal, L. (2013) · Penn Carey Law Legal Scholarship Repository (University of Pennsylvania) · Cited by 5
(2014) · De Gruyter eBooks · doi:10.1515/9781614512127.1 · Cited by 3
New York City to conduct urban airflow test
(2013) · Physics Today · doi:10.1063/pt.5.026959 · Cited by 1
Legewie, J. (2014) · SSRN Electronic Journal · doi:10.2139/ssrn.2388410 · Cited by 1
(2014) · doi:10.5555/2627435.2697065
Cited by
Algorithmic Decision Making and the Cost of Fairness
Corbett-Davies, S., Pierson, E., Feller, A., Goel, S., & Huq, A. (2017) · Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining · doi:10.1145/3097983.3098095 · Cited by 1,149
Self-citationZafar, M. B., Valera, I., Gomez Rodriguez, M., & Gummadi, K. P. (2017) · Proceedings of the 26th International Conference on World Wide Web · doi:10.1145/3038912.3052660 · Cited by 632
“Finally, we experiment with synthetic as well as real world datasets and show that our methodology can be effectively used to avoid disparate mistreatment.”
MethodA large-scale analysis of racial disparities in police stops across the United States
Pierson, E., Simoiu, C., Overgoor, J., Corbett-Davies, S., Jenson, D., Shoemaker, A., Ramachandran, V., Barghouty, P., Phillips, C., Shroff, R., et al. (2020) · Nature Human Behaviour · doi:10.1038/s41562-020-0858-1 · Cited by 552
Self-citationLanguage from police body camera footage shows racial disparities in officer respect
Voigt, R., Camp, N. P., Prabhakaran, V., Hamilton, W. L., Hetey, R. C., Griffiths, C. M., Jurgens, D., Jurafsky, D., & Eberhardt, J. L. (2017) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.1702413114 · Cited by 407
“This contrasts with the case of stop-and-frisk, where individual outlier officers account for a substantial proportion of racial disparities (27); the disparities we observe here cannot be explained by a small number of extreme officers.”
ResultPredict and Surveil: Data, Discretion, and the Future of Policing
Brayne, S. (2020) · doi:10.1093/oso/9780190684099.001.0001 · Cited by 334
Metadata
- Authors
- Goel, S., Rao, J. M., & Shroff, R.
- Affiliations
- Microsoft Research New York City (United States) (US)
- Venue
- The Annals of Applied Statistics
- Publisher
- Institute of Mathematical Statistics
- Published
- 2016-03-01
- Volume
- 10
- Issue
- 1
- Pages
- 365-394
- ISSN
- 1932-6157 (ISSN-L)
- 1941-7330
- OpenAlex
- W3121654114
- Cited by
- 285
- References
- 44
- Citations by year
- 2026: 11 · 2025: 17 · 2024: 20 · 2023: 34 · 2022: 25 · 2021: 34 · 2020: 44 · 2019: 34 · 2018: 30 · 2017: 25 · 2016: 10
- FWCI
- 142.55
- Type
- journal article
- Language
- en
- Open access
- green
- doi.org
- Locations
- The Annals of Applied Statistics (publishedVersion)
- SSRN Electronic Journal (publishedVersion)
- Project Euclid (Cornell University) (submittedVersion)
- Topics
- Social Sciences › Political Science and International Relations › Policing Practices and Perceptions
- Social Sciences › Sociology and Political Science › Criminal Justice and Corrections Analysis
- Social Sciences › Law › Criminal Law and Evidence
- Keywords
- stop and frisk, racial disparities in policing, reasonable suspicion, public housing, high-crime neighborhoods, racial bias in policing, New York City
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.
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
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.