Papers
Topics
Authors
Recent
Search
2000 character limit reached

Rethinking recidivism through a causal lens

Published 19 Nov 2020 in cs.LG, cs.CY, and stat.AP | (2011.11483v4)

Abstract: Predictive modeling of criminal recidivism, or whether people will re-offend in the future, has a long and contentious history. Modern causal inference methods allow us to move beyond prediction and target the "treatment effect" of a specific intervention on an outcome in an observational dataset. In this paper, we look specifically at the effect of incarceration (prison time) on recidivism, using a well-known dataset from North Carolina. Two popular causal methods for addressing confounding bias are explained and demonstrated: directed acyclic graph (DAG) adjustment and double machine learning (DML), including a sensitivity analysis for unobserved confounders. We find that incarceration has a detrimental effect on recidivism, i.e., longer prison sentences make it more likely that individuals will re-offend after release, although this conclusion should not be generalized beyond the scope of our data. We hope that this case study can inform future applications of causal inference to criminal justice analysis.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (46)
  1. Machine bias. ProPublica., 2016. URL https://www.propublica.org/article/machine-bias-risk-assessments-in-criminal-sentencing.
  2. DoubleML: An Object-Oriented Implementation of Double Machine Learning in R. Journal of Statistical Software, 108(3), 2024. ISSN 1548-7660. URL https://www.jstatsoft.org/v108/i03/.
  3. Classifying adult probationers by forecasting future offending. National Institute of Justice., 4:2020, 2012. URL https://www.ojp.gov/pdffiles1/nij/grants/238082.pdf.
  4. Improving Fairness in Criminal Justice Algorithmic Risk Assessments Using Optimal Transport and Conformal Prediction Sets. Sociological Methods & Research, Mar. 2023. ISSN 0049-1241, 1552-8294. URL http://journals.sagepub.com/doi/10.1177/00491241231155883.
  5. K. Biza. MXM: Feature Selection (Including Multiple Solutions) and Bayesian Networks, 2022.
  6. R. A. Bradshaw. Cross-Validation of the Iowa Offender Risk Assessment Model in Michigan, 1980-1982: Version 1, 1989. URL http://www.icpsr.umich.edu/icpsrweb/NACJD/studies/9236/version/1.
  7. L. Breiman. Random forests. Machine Learning, 45(1):5–32, 2001a. ISSN 08856125. URL http://link.springer.com/10.1023/A:1010933404324.
  8. L. Breiman. Statistical Modeling: The Two Cultures (with comments and a rejoinder by the author). Statistical Science, 16(3), Aug. 2001b. ISSN 0883-4237. doi: 10.1214/ss/1009213726. URL https://projecteuclid.org/journals/statistical-science/volume-16/issue-3/Statistical-Modeling--The-Two-Cultures-with-comments-and-a/10.1214/ss/1009213726.full.
  9. Predicting criminal recidivism: A comparison of neural network models with statistical methods. Journal of Criminal Justice, 24(3):227–240, Jan. 1996. ISSN 0047-2352. URL https://www.sciencedirect.com/science/article/pii/0047235296000128.
  10. Double/debiased machine learning for treatment and structural parameters. The Econometrics Journal, 21(1):C1–C68, Feb. 2018. ISSN 1368-4221, 1368-423X. URL https://academic.oup.com/ectj/article/21/1/C1/5056401.
  11. Long Story Short: Omitted Variable Bias in Causal Machine Learning. Technical Report w30302, National Bureau of Economic Research, Cambridge, MA, July 2022. URL http://www.nber.org/papers/w30302.pdf.
  12. A. Chouldechova. Fair prediction with disparate impact: A study of bias in recidivism prediction instruments. Oct. 2016. URL http://arxiv.org/abs/1610.07524.
  13. Survival Analysis: A Survey. Journal of Quantitative Criminology, 7(1):59–98, 1991. ISSN 0748-4518. URL https://www.jstor.org/stable/23365824.
  14. dml.sensemakr: Sensitivity Analysis for Debiased Machine Learning, 2023. R package version 0.1.0.
  15. D. R. Cox. Regression Models and Life-Tables. Journal of the Royal Statistical Society. Series B (Methodological), 34(2):187–220, 1972. ISSN 0035-9246. URL https://www.jstor.org/stable/2985181.
  16. Performance of recidivism risk assessment instruments in U.S. correctional settings. Psychological Services, 13(3):206–222, Aug. 2016. ISSN 1939-148X, 1541-1559. URL http://doi.apa.org/getdoi.cfm?doi=10.1037/ser0000075.
  17. J. Dressel and H. Farid. The accuracy, fairness, and limits of predicting recidivism. Science Advances, 4(1), Jan. 2018. ISSN 2375-2548. URL http://advances.sciencemag.org/lookup/doi/10.1126/sciadv.aao5580.
  18. Review of Causal Discovery Methods Based on Graphical Models. Frontiers in Genetics, 10:524, June 2019. ISSN 1664-8021. URL https://www.frontiersin.org/article/10.3389/fgene.2019.00524/full.
  19. S. D. Gottfredson. Prediction: An Overview of Selected Methodological Issues. Crime and Justice, 9:21–51, 1987. ISSN 0192-3234. URL https://www.jstor.org/stable/1147388.
  20. Criminal Violence and Incapacitation in California, 1962-1988: Version 1, 1993. URL http://www.icpsr.umich.edu/icpsrweb/NACJD/studies/9922/version/1.
  21. A. Khademi and V. Honavar. Algorithmic Bias in Recidivism Prediction: A Causal Perspective, Nov. 2019. URL http://arxiv.org/abs/1911.10640.
  22. Fairness in Algorithmic Decision Making: An Excursion Through the Lens of Causality. In The World Wide Web Conference, pages 2907–2914, May 2019. URL http://arxiv.org/abs/1903.11719.
  23. A Comparison of Logistic Regression, Classification and Regression Tree, and Neural Networks Models in Predicting Violent Re-Offending. Journal of Quantitative Criminology, 27(4):547–573, Dec. 2011. ISSN 1573-7799. URL https://doi.org/10.1007/s10940-011-9137-7.
  24. The Impact of Incarceration on Recidivism. Annual Review of Criminology, 5(1):133–152, Jan. 2022. ISSN 2572-4568, 2572-4568. URL https://www.annualreviews.org/doi/10.1146/annurev-criminol-030920-112506.
  25. M. D. Maltz and R. McCleary. The Mathematics of Behavioral Change: Recidivism and Construct Validity. Evaluation Quarterly, 1(3):421–438, Aug. 1977. ISSN 0145-4692. URL http://journals.sagepub.com/doi/10.1177/0193841X7700100304.
  26. B. Monnery. The Determinants of Recidivism among Ex-Prisoners: A Survival Analysis on French Data, May 2013. URL https://papers.ssrn.com/abstract=2265349.
  27. Predicting criminal recidivism using neural networks. Socio-Economic Planning Sciences, 34(4):271–284, Dec. 2000. ISSN 0038-0121. URL https://www.sciencedirect.com/science/article/pii/S0038012100000033.
  28. J. Pearl. Causality: models, reasoning, and inference. Cambridge University Press, Cambridge, U.K. ; New York, 2000. ISBN 9780521895606 9780521773621. URL http://bayes.cs.ucla.edu/BOOK-2K/.
  29. Parolee Recidivism and Successful Treatment Completion: Comparing Hazard Models Across Propensity Methods. Journal of Quantitative Criminology, 31(1):149–181, Mar. 2015. ISSN 0748-4518, 1573-7799. URL http://link.springer.com/10.1007/s10940-014-9229-2.
  30. E. K. Rose and Y. Shem-Tov. How Does Incarceration Affect Reoffending? Estimating the Dose-Response Function. Journal of Political Economy, 129(12):3302–3356, Dec. 2021. ISSN 0022-3808, 1537-534X. URL https://www.journals.uchicago.edu/doi/10.1086/716561.
  31. D. B. Rubin. Estimating causal effects of treatments in randomized and nonrandomized studies. Journal of Educational Psychology, 66(5):688–701, 1974. ISSN 0022-0663. URL http://content.apa.org/journals/edu/66/5/688.
  32. The Age of Secrecy and Unfairness in Recidivism Prediction. Harvard Data Science Review, 2(1), Jan. 2020. URL https://hdsr.mitpress.mit.edu/pub/7z10o269.
  33. Retrospective Causal Inference with Machine Learning Ensembles: An Application to Anti-Recidivism Policies in Colombia, July 2016. URL http://arxiv.org/abs/1607.03026.
  34. P. Schmidt and A. D. Witte. Predicting Recidivism Using Survival Models. Research in Criminology. Springer, New York, NY, 1988. ISBN 9781461283430 9781461237723. URL http://link.springer.com/10.1007/978-1-4612-3772-3.
  35. P. Schmidt and A. D. Witte. Predicting Recidivism in North Carolina, 1978 and 1980: Version 1, 1989a. URL https://www.icpsr.umich.edu/web/NACJD/studies/8987/versions/V1.
  36. P. Schmidt and A. D. Witte. Predicting criminal recidivism using ‘split population’ survival time models. Journal of Econometrics, 40(1):141–159, Jan. 1989b. ISSN 0304-4076. URL https://www.sciencedirect.com/science/article/pii/0304407689900341.
  37. P. Schmidt and A. D. Witte. Some Thoughts on How and When to Predict in Criminal Justice Settings. In M. J. Lerner, editor, New Directions in the Study of Justice, Law, and Social Control, pages 247–270. Springer US, Boston, MA, 1990. ISBN 9781489936103 9781489936080. URL http://link.springer.com/10.1007/978-1-4899-3608-0_11.
  38. A Linear Non-Gaussian Acyclic Model for Causal Discovery. Journal of Machine Learning Research, 7(72):2003–2030, 2006. ISSN 1533-7928. URL http://jmlr.org/papers/v7/shimizu06a.html.
  39. Constructing Bayesian Network Models of Gene Expression Networks from Microarray Data. 2000. URL https://www.researchgate.net/publication/2412831_Constructing_Bayesian_Network_Models_of_Gene_Expression_Networks_from_Microarray_Data.
  40. Conviction, Incarceration, and Recidivism: Understanding the Revolving Door. 2023. ISSN 1556-5068. URL https://www.ssrn.com/abstract=4507597.
  41. Which method predicts recidivism best?: a comparison of statistical, machine learning and data mining predictive models. Journal of the Royal Statistical Society: Series A (Statistics in Society), 176(2):565–584, Feb. 2013. ISSN 09641998. URL http://doi.wiley.com/10.1111/j.1467-985X.2012.01056.x.
  42. Optimizing predictive performance of criminal recidivism models using registration data with binary and survival outcomes. PLOS ONE, 14(3):e0213245, Mar. 2019. ISSN 1932-6203. URL https://dx.plos.org/10.1371/journal.pone.0213245.
  43. Constraint-based causal discovery with mixed data. International Journal of Data Science and Analytics, 6(1):19–30, Aug. 2018. ISSN 2364-415X, 2364-4168. URL http://link.springer.com/10.1007/s41060-018-0097-y.
  44. In Pursuit of Interpretable, Fair and Accurate Machine Learning for Criminal Recidivism Prediction. Journal of Quantitative Criminology, 39(2):519–581, June 2023. ISSN 1573-7799. URL https://doi.org/10.1007/s10940-022-09545-w.
  45. Interpretable Classification Models for Recidivism Prediction. Journal of the Royal Statistical Society: Series A (Statistics in Society), 180(3):689–722, June 2017. ISSN 09641998. URL http://arxiv.org/abs/1503.07810.
  46. DAGs with NO TEARS: Continuous Optimization for Structure Learning. 2018. URL https://arxiv.org/abs/1803.01422.

Summary

No one has generated a summary of this paper yet.

Sign Up to Summarize

Paper to Video (Beta)

No one has generated a video about this paper yet.

Sign Up to Generate All Videos Subscribe on YouTube

Whiteboard

No one has generated a whiteboard explanation for this paper yet.

Sign Up to Generate

Open Problems

We haven't generated a list of open problems mentioned in this paper yet.

Generate Now

Continue Learning

We haven't generated follow-up questions for this paper yet.

Generate Now

Collections

Sign up for free to add this paper to one or more collections.

Sign Up

Tweets

Sign up for free to view the 1 tweet with 0 likes about this paper.

Sign Up for Free