Abstract
Objectives
This study examines the relationship between homelessness and recidivism among people on probation with and without behavioral health problems. The study also illustrates a new way to summarize the effect of an exposure on an outcome, the Incremental Propensity Score (IPS) effect, which avoids pitfalls of other approaches commonly used in criminology.
Methods
We assessed the impact of homelessness at probation start on rearrest within one year among a cohort of people on probation (n = 2453). We estimated IPS effects, considering general and crime-specific recidivism if subjects were more or less likely to be unhoused, and assessed effect variation by behavioral health problem status. We used a doubly robust machine learning estimator to flexibly but efficiently estimate effects.
Results
A substantial intervention—reducing homelessness by roughly 65%—corresponded to a 9% reduction in the estimated average rate of recidivism (p < 0.05). Milder interventions showed smaller, non-significant effect sizes. Stratifying by behavioral health problem and rearrest type led to similar results without statistical significance.
Conclusions
Minding limitations related to observational data and generalizability, this study suggests large reductions in homelessness lead to significant reductions in rearrest rates. Efforts to reduce recidivism should include interventions that make homelessness less likely, but notable differences in recidivism will require these interventions be sizable. Meanwhile, efforts to establish recidivism risk factors should consider alternative effects, like IPS effects, to maximize validity and reduce bias.
Similar content being viewed by others
Notes
Authors acknowledge that the term “homelessness” may not be the most accurate descriptor for the phenomenon of interest, as advocates have indicated with their turn toward the term “houselessness.”. However, as “homeless” is the federally recognized term and remains a common term, we use”homelessness” in lieu of “houselessness.”.
Although IPS effects describe a curve of counterfactual average outcomes rather than a contrast between counterfactual average outcomes under different interventions, we refer to the values along this curve as effects to maintain consistency with the prevailing literature on the topic. A comprehensive review of IPS effects is provided in Bonvini et al. (2023).
Standard people on probation are defined as those who would typically be sentenced to supervision by county probation. Standard probationers excludes those who are supervised by probation under California’s Public Safety Realignment law, who have committed more serious felony offenses and would traditionally be supervised under state parole.
Comparison of those included and excluded indicated that those included were comparable to those excluded in terms of age and gender, but those included were more likely to be Black (\({x}^{2}\)= 39.54, p < .001) and to recidivate (\({x}^{2}\)= 265.94, p < .001).
References
Andrews DA, Bonta J, Wormith JS (2006) The recent past and near future of risk and/or need assessment. Crime Delinq 52(1):7–27. https://doi.org/10.1177/0011128705281756
Aurand A, Emmanuel D, Clarke M, Rafi I, Yentel D (2022) The gap: a shortage of affordable homes. National Low Income Housing Coalition, p 32. https://nlihc.org/gap
Baillargeon J, Binswanger IA, Penn JV, Williams BA, Murray OJ (2009) Psychiatric disorders and repeat incarcerations: the revolving prison door. Am J Psychiatry 166(1):103–109. https://doi.org/10.1176/appi.ajp.2008.08030416
Balyakina E, Mann C, Ellison M, Sivernell R, Fulda KG, Sarai SK, Cardarelli R (2014) Risk of future offense among probationers with co-occurring substance use and mental health disorders. Community Ment Health J 50(3):288–295. https://doi.org/10.1007/s10597-013-9624-4
Beckett K, Herbert S (2009) Banished: the new social control in urban America. Oxford University Press, London
Belloni A, Chernozhukov V, Chetverikov D, Wei Y (2018) Uniformly valid post-regularization confidence regions for many functional parameters in Z-estimation framewok. Ann Stat 46(6B):3643–3675. https://doi.org/10.1214/17-AOS1671
Bickel PJ, Klaassen CA, Ritov YA, Wellner JA (1993) Efficient and adaptive estimation for semiparametric models, vol 4. Johns Hopkins University Press, Baltimore
Bonvini M, McClean A, Branson Z, Kennedy EH (2023) Incremental causal effects: an introduction and review. In: Zubizarreta JR, Stuart EA, Small DS, Rosenbaum PR (eds) Handbook of matching and weighting adjustments for causal inference. Chapman and Hall/CRC, pp 349–372
Brennan T, Dieterich W, Ehret B (2009) Evaluating the predictive validity of the COMPAS risk and needs assessment system. Crim Justice Behav 36(1):21–40. https://doi.org/10.1177/0093854808326545
Carneiro P, Heckman JJ, Vytlacil EJ (2011) Estimating marginal returns to education. Am Econ Rev 101(6):2754–2781. https://doi.org/10.1257/aer.101.6.2754
Chernozhukov V, Chetverikov D, Demirer M, Duflo E, Hansen C, Newey W, Robins J (2018) Double/debiased machine learning for treatment and structural parameters. Economet J 21(1):C1–C68. https://doi.org/10.1111/ectj.12097
Clark VA (2016) Predicting two types of recidivism among newly released prisoners. Crime Delinq 62(10):1364–1400. https://doi.org/10.1177/0011128714555760
Couloute L (2018) Nowhere to go: homelessness among formerly incarcerated people. Prison Policy Initiative. https://www.prisonpolicy.org/reports/housing.html
Desmarais SL, Johnson KL, Singh JP (2016) Performance of recidivism risk assessment instruments in U.S. correctional settings. Psychol Serv 13(3):206–222. https://doi.org/10.1037/ser0000075
Farabee D, Zhang S, Roberts REL, Yang J (2010) COMPAS validation study: final report. Semel Institute for Neuroscience and Human Behavior
Gendreau P, Little T, Goggin C (1996) A meta-analysis of the predictors of adult offender recidivism: What works! Criminology 34(4):575–608. https://doi.org/10.1111/j.1745-9125.1996.tb01220.x
Giné E, Zinn J (1984) Some limit theorems for empirical processes. Ann Probab 12(4):929–989. https://doi.org/10.1214/aop/1176993138
Haneuse S, Rotnitzky A (2013) Estimation of the effect of interventions that modify the received treatment. Stat Med 32(30):5260–5277. https://doi.org/10.1002/sim.5907
Harding DJ, Morenoff JD, Dobson CC, Lane EB, Opatovsky K, Williams E-DG, Wyse J (2016) Families, prisoner reentry, and reintegration. In: Burton LM, Burton D, McHale SM, King V, Van Hook J (eds) Boys and men in African American families. Springer, pp 105–160. https://doi.org/10.1007/978-3-319-43847-4_8
Herbert CW, Morenoff JD, Harding DJ (2015) Homelessness and housing insecurity among former prisoners. Russell Sage Founda J Soc Sci 1(2):44–79. https://doi.org/10.7758/rsf.2015.1.2.04
Hines O, Dukes O, Diaz-Ordaz K, Vansteelandt S (2022) Demystifying statistical learning based on efficient influence functions. Am Stat 76(3):292–304. https://doi.org/10.1080/00031305.2021.2021984
Jacobs LA, Gottlieb A (2020) The effect of housing circumstances on recidivism: evidence from a sample of people on probation in San Francisco. Crim Justice Behav 47(9):1097–1115. https://doi.org/10.1177/0093854820942285
Jacobs LA, Fixler A, Labrum T, Givens A, Newhill C (2021) Risk factors for criminal recidivism among persons with serious psychiatric diagnoses: disentangling what matters for whom. Front Psych 12:778399. https://doi.org/10.3389/fpsyt.2021.778399
Kennedy EH (2019) Nonparametric causal effects based on incremental propensity score interventions. J Am Stat Assoc 114(526):645–656. https://doi.org/10.1080/01621459.2017.1422737
Kennedy EH (2021) npcausal: nonparametric causal inference methods in R (0.1.0) [Computer software]
Listwan SJ, Hartman JL, LaCourse A (2018) Impact of the MeckFUSE Pilot Project: Recidivism among the chronically homeless. Justice Eval J 1(1):96–108. https://doi.org/10.1080/24751979.2018.1478236
Lutze FE, Rosky JW, Hamilton ZK (2014) Homelessness and reentry: a multisite outcome evaluation of Washington state’s reentry housing program for high risk offenders. Crim Justice Behav 41(4):471–491. https://doi.org/10.1177/0093854813510164
Monahan J, Skeem JL (2014) Risk redux: the resurgence of risk assessment in criminal sanctioning. Fed Sentenc Rep 26(3):158–166. https://doi.org/10.1525/fsr.2014.26.3.158
Monahan J, Skeem J, Lowenkamp C (2017) Age, risk assessment, and sanctioning: Overestimating the old, underestimating the young. Law Hum Behav 41(2):191–201. https://doi.org/10.1037/lhb0000233
Mulvey EP, Schubert CA (2017) Mentally ill individuals in jails and prisons. Crime Justice 46(1):231–277. https://doi.org/10.1086/688461
Naimi AI, Rudolph JE, Kennedy EH, Cartus A, Kirkpatrick SI, Haas DM, Bodnar LM (2021) Incremental propensity score effects for time-fixed exposures. Epidemiology 32(2):202–208
Nyamathi AM, Zhang S, Salem BE, Farabee D, Hall B, Marlow E, Faucette M, Bond D, Yadav K (2016) A randomized clinical trial of tailored interventions for health promotion and recidivism reduction among homeless parolees: outcomes and cost analysis. J Exp Criminol. https://doi.org/10.1007/s11292-015-9236-9
Ostermann M, Salerno LM, Hyatt JM (2015) How different operationalizations of recidivism impact conclusions of effectiveness of parole supervision. J Res Crime Delinq 52(6):771–796. https://doi.org/10.1177/0022427815580626
Polley E, LeDell E, Kennedy C, Lendle S, van der Laan M (2021) SuperLearner (Version 2.0–28). https://cran.r-project.org/web/packages/SuperLearner/index.html
American Psychiatric Association (2000) Diagnostic and statistical manual of mental disorders, text revision (DSM-IV-TR). American Psychiatric Association
Rice D (2016) Chart book: cuts in federal assistance have exacerbated families’ struggles to afford housing. Center on Budget and Policy Priorities. https://www.cbpp.org/research/housing/cuts-in-federal-assistance-have-exacerbated-families-struggles-to-afford-housing
Robinson T (2019) No right to rest: police enforcement patterns and quality of life consequences of the criminalization of homelessness. Urban Aff Rev 55(1):41–73. https://doi.org/10.1177/1078087417690833
Skeem JL, Manchak S, Peterson JK (2011) Correctional policy for offenders with mental illness: creating a new paradigm for recidivism reduction. Law Hum Behav 35(2):110–126. https://doi.org/10.1007/s10979-010-9223-7
Steiner B, Makarios MD, Travis LF (2015) Examining the effects of residential situtions and residential mobility on offender recidivism. Crime Delinq 61(3):375–401. https://doi.org/10.1177/0011128711399409
Tsiatis AA (2006) Semiparametric theory and missing data. Springer
U.S. Census Bureau & U.S. Department of Housing and Urban Development (1963) Average sales price of houses sold for the United States. FRED, Federal Reserve Bank of St. Louis; FRED, Federal Reserve Bank of St. Louis. https://fred.stlouisfed.org/series/ASPUS
van der Laan MJ, Robins JM (2003) Unified methods for censored longitudinal data and causality. Springer
Vansteelandt S, Bekaert M, Claeskens G (2012) On model selection and model misspecification in causal inference. Stat Methods Med Res 21(1):7–30. https://doi.org/10.1177/0962280210387717
Westreich D, Cole SR (2010) Invited commentary: positivity in practice. Am J Epidemiol 171(6):674–677. https://doi.org/10.1093/aje/kwp436
Yukhnenko D, Blackwood N, Fazel S (2020a) Risk factors for recidivism in individuals receiving community sentences: a systematic review and meta-analysis. CNS Spectr 25(2):252–263. https://doi.org/10.1017/S1092852919001056
Yukhnenko D, Sridhar S, Fazel S (2020b) A systematic review of criminal recidivism rates worldwide: 3-year update. Wellcome Open Res 4:28. https://doi.org/10.12688/wellcomeopenres.14970.3
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendix 1: Balance Table
Appendix 1: Balance Table
Variable | Type | Unadjusted standardized mean difference | Adjusted mean housed | Adjusted mean homeless | Adjusted standardized mean difference |
---|---|---|---|---|---|
Age | Cont | − 0.2915 | 36.1 | 36.4 | − 0.0195 |
Sex | Bin | 0.0389 | 0.1519 | 0.1609 | − 0.0090 |
Race/ethnicity | |||||
Black | Bin | 0.0234 | 0.4945 | 0.4815 | 0.0130 |
Latino | Bin | 0.0296 | 0.1439 | 0.1518 | − 0.0079 |
Another race/ethnicity | Bin | 0.0388 | 0.1165 | 0.1271 | − 0.0106 |
White | Bin | − 0.0919 | 0.2452 | 0.2397 | 0.0055 |
Recidivism Risk | Cont | − 0.5371 | − 0.0057 | 0.0662 | − 0.0775 |
Financial Insecurity | Cont | − 0.6336 | − 0.0018 | 0.0515 | − 0.0567 |
Diagnosis | |||||
Co-occurring | Bin | − 0.0599 | 0.0393 | 0.0439 | − 0.0046 |
No Diagnosis | Bin | 0.1508 | 0.8114 | 0.8019 | 0.0095 |
SMI Only | Bin | − 0.0290 | 0.0639 | 0.0643 | − 0.0004 |
SUD Only | Bin | − 0.0609 | 0.0854 | 0.0898 | − 0.0045 |
Effective sample size | |||||
Homeless | 648 | 383.3 | |||
Housed | 1805 | 1661.7 |
Standardized mean differences are housed minus homeless, divided by pooled standard deviation across both groups; SMI = Serious mental illness; SUD = Substance use disorder; Cont. = Continuous; Bin. = Binary.
The table shows traditional balance diagnostics. The first column shows the variable in the data while the second column shows the type of the variable (“cont.” equals continuous and “bin.” equals binary). The third column shows the unadjusted mean differences between the housed and homeless groups. For binary variables, this is the raw difference in proportions, while for continuous variables it is the standardized mean difference, where the standardized mean difference equals the mean among housed individuals minus the mean for homeless individuals, divided by the pooled standard deviation across both groups. The fourth and fifth columns show the adjusted means among housed and homeless individuals, respectively. Here, the data is re-weighted according to the propensity score estimates. Specifically, the weights for individual \(i\), which we denote by \(W_{i}\), are:
where \(A_{i} = 1\) if individual \(i\) was housed at the started of probation and \(0\) otherwise, and \(\hat{\pi }\left( {X_{i} } \right)\) are the estimated propensity score values for individual \(i\). Finally, the sixth column shows the adjusted standardized mean differences. For binary variables, this is again the raw difference in proportions; i.e., the adjusted mean for housed individuals minus the adjusted mean for homeless individuals. For continuous variables, this is the standardized mean difference. The final two rows of the table show the true and weighted sizes for the housed and homeless groups.
The table demonstrates that balance is improved by weighting according to the estimated propensity scores, indicating that the SuperLearner estimator for the propensity score is appropriately capturing confounding in the observed data.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Jacobs, L.A., McClean, A., Branson, Z. et al. Incremental Propensity Score Effects for Criminology: An Application Assessing the Relationship Between Homelessness, Behavioral Health Problems, and Recidivism. J Quant Criminol (2024). https://doi.org/10.1007/s10940-024-09582-7
Accepted:
Published:
DOI: https://doi.org/10.1007/s10940-024-09582-7