Skip to main content
Log in

Predictive Policing in a Developing Country: Evidence from Two Randomized Controlled Trials

  • Original Paper
  • Published:
Journal of Quantitative Criminology Aims and scope Submit manuscript

Abstract

Objectives

The paper studies the impact of predictive policing on crime in a developing country. It also assesses the impact of different police trainings.

Method

We analyze a randomized controlled trial conducted in Montevideo, Uruguay to assess the implementation of a predictive policing software developed in the United States. Half of the precincts were randomly assigned to the software and half to the local crime analysts (status quo). The second experiment allocated randomly a specially trained police force to targeted patrol areas per shift and day.

Results

No statistically significant differences were found in crime outcomes between the precincts assigned to the foreign predictive software and those assigned to local crime analysts. On the second experiment, given determined targeted places, the specially trained task force showed more compliance with the assigned patrol sites (20% more patrol time) and a greater potential for reducing crime (reduction of 30% in robberies only during high crime shifts in comparison to the control group (no special training). There is also evidence of a diffusion of benefits to adjacent areas.

Conclusions

The implementation of an international predictive policing software did not outperform local crime analysts in terms of crime reduction. Local crime analysts are more cost-effective. Given determined targeted places, a modest increase in police dosage of a specially trained police force could reduce crime in high-crime times. In developing countries new policing technologies and training require a deep understanding of the context to channel limited resources in the most efficient way.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Map 1

Source Fraiman (2016)

Fig. 1

Similar content being viewed by others

Notes

  1. Assuming that the intra-cluster correlation coefficient ranges between 0.05 and 0.10, and that the desired power is 0.8 and the significance level is 0.1, Experiment 1 can detect a minimum effect size between 0.25 and 0.3 standard deviations for the crime rate. As a rule of thumb, this effect is considered intermediate in the RCT literature, while up to 0.2 standard deviations is small and above 0.5 standard deviations is seen as being large (see, among others, Glennerster and Takavarasha 2013).

  2. We replicated the estimation of Eqs. (1) and (2) disaggregating every week by shift, thus using 2046 observations for (1) and 6864 for (2). The results are the same as using the complete weeks: there are no statistically significant differences on crime -including on robberies and thefts-.

  3. Assuming a power of 0.8 and a significance level of 0.1, Experiment 2 can detect a minimum effect size of 0.125 standard deviations when all shifts are considered and one of 0.15 standard deviations when only shifts 1 and 2 are considered. As a rule of thumb, these effects are considered to be small in the RCT literature.

  4. The results are robust to measuring the presence of vehicles and foot patrols based on the signals sent by the officers’ radios.

  5. Given that this experiment took place only in one police precinct, this was the only way that the government could try to test the impact of the different trainings in the predicted sites, had this been conducted in a broader area, another design that could address potential contamination could have been implemented.

  6. In Ariel et al. (2016), treatment involved increases in foot patrols by uniformed but unarmed Police Community Support Officers.

  7. In this paper, the authors test Koper’s (1995) finding that 15 min is the optimal time for an officer to patrol within a long-term hot spot.

References

  • Angrist J (2004) Treatment effect heterogeneity in theory and practice. Econ J 114(494):C52-83

    Article  Google Scholar 

  • Angrist JD, Imbens GW (1995) Two-stage least squares estimation of average causal effects in models with variable treatment intensity. J Am Stat Assoc 90(430):431–442

    Article  Google Scholar 

  • Ariel B, Weinborn C, Sherman LW (2016) “Soft” policing at hot spots—do police community support officers work? A randomized controlled trial. J Exp Criminol 12(3):277–317

    Article  Google Scholar 

  • Banerjee A, Chattopadhyay R, Duflo E, Keniston D, Singh N (2012) Improving police performance in Rajasthan, India: experimental evidence on incentives. Managerial autonomy and training, National Bureau of Economic Research, Cambridge, MA

    Google Scholar 

  • Bell B, Jaitman L, Machin S (2014) Crime deterrence: evidence from the London 2011 Riots. Econ J 124(576):480–506

    Article  Google Scholar 

  • Bennett Moses L, Chan J (2016) Algorithmic prediction in policing: assumptions, evaluation, and accountability. Polic Soc 28(7):1–17

    Google Scholar 

  • Birzer ML (2003) The theory of andragogy applied to police training. Policing 26(1):29–42

    Article  Google Scholar 

  • Bo H, Galiani S (2020) Assessing external validity. NBER Working Paper No. 26422

  • Chaszczewski M (2015) Community crime mapping. Increasing predictive policing with dynamic symbol sets. Rochester Institute of Technology. Thesis. Available at: http://scholarworks.rit.edu/theses/8693/

  • Eck J, Weisburd D (1995) Crime and place, crime prevention studies. Criminal Justice Press, Monsey, NY

    Google Scholar 

  • Evans D, Herbert D (1989) The geography of crime. Routledge, London, UK

    Google Scholar 

  • Felson M (1987) Routine activities and crime prevention in the developing metropolis. Criminology 25(4):911–932

    Article  Google Scholar 

  • Ferguson A (2012) Predictive policing and reasonable suspicion. Emory Law J 62(2):259–313

    Google Scholar 

  • Fraiman R (2016) Patrullaje inteligente basado en la evidencia: la estrategia de patrullaje policial en la ciudad de montevideo. presentation at America latina crime and policy network (AL-CAPONE). Universidad Torcuato Di Tella

  • Gill C, Wooditch A, Weisburd D (2016) Testing the “Law of Crime at Place” in a Suburban Setting: implications for research and practice. Unpublished

  • Glennerster R, Takavarasha K (2013) Running randomized evaluations: a practical guide. Princeton University Press, Princeton, NJ

    Book  Google Scholar 

  • Haberman CP, Ratcliffe JH (2012) The predictive policing challenges of near repeat armed street robberies. Policing 6(2):151–166

    Article  Google Scholar 

  • Hunt P, Saunders J, Hollywood JS (2014) Evaluation of the shreveport predictive policing experiment. Rand Corporation. Available at: http://www.rand.org/pubs/research_reports/RR531.html

  • Jaitman L (ed) (2017) The Costs of crime and violence: new evidence and insights in Latin America and the Caribbean. Washington, DC: Inter-American Development Bank (IDB). Available at: https://publications.iadb.org/handle/11319/8133

  • Jaitman L, Ajzenman N (2016) Crime concentration and hot spots dynamics in Latin America. Washington, DC: Inter-American Development Bank (IDB). Available at: https://publications.iadb.org/handle/11319/7702

  • Johnson SD, Bernasco W, Bowers KJ, Elffers H, Ratcliffe JH, Rengert G, Townsley M (2007) Space-time patterns of risk: a cross-national assessment of residential burglary victimization. J Quant Criminol 23:201–219

    Article  Google Scholar 

  • Kennedy LW, Caplan JM, Piza E (2011) Risk clusters, hotspots, and spatial intelligence: risk terrain modeling as an algorithm for police resource allocation strategies. J Quant Criminol 27(3):339–362

    Article  Google Scholar 

  • Koper CS, Lum C, Willis J (2014) Optimizing the use of technology in policing: results and implications from a multi-site study of the social, organizational, and behavioural aspects of implementing police technologies. Policing 8(2):212–221

    Article  Google Scholar 

  • Koper CS (1995) Just enough police presence: reducing crime and disorderly behavior by optimizing patrol time in crime hot spots. Justice Q 12(4):649–672

    Article  Google Scholar 

  • Lum C, Koper CS, Willis J (2016) Understanding the limits of technology’s impact on police effectiveness. Police Q. https://doi.org/10.1177/1098611116667279

    Article  Google Scholar 

  • Mohler GO, Short MB, Malinowski S, Johnson M, Tita GE, Bertozzi AL, Brantingham PJ (2015) Randomized controlled field trials of predictive policing. J Am Stat Assoc 110(512):1399–1411

    Article  Google Scholar 

  • Perry WL, McInnis B, Price CC, Smith SC, Hollywood JS (2013) Predictive Policing: The Role of Crime Forecasting in Law Enforcement Operations. Santa Monica, CA: RAND Corporation. Available at: http://www.rand.org/pubs/research_reports/RR233.html

  • Ratcliffe JH, Taniguchi T, Groff ER, Wood JD (2011) The philadelphia foot patrol experiment: a randomized controlled trial of police patrol effectiveness in violent crime hotspots. Criminology 49(3):795–831

    Article  Google Scholar 

  • Ratcliffe JH, Taylor RB, Askey AP, Thomas K, Grasso J, Bethel K, Fisher R, Koehnlein J (2020) The philadelphia predictive policing experiment. J Exp Criminol. https://doi.org/10.1007/s11292-019-09400-2

    Article  Google Scholar 

  • Santos RB, Santos RG (2015) Examination of police dosage in residential burglary and residential theft from vehicle micro time hot spots. Crime Sci 4(1):27. https://doi.org/10.1186/s40163-015-0041-6

    Article  Google Scholar 

  • Sherman LW, Weisburd D (1995) General deterrent effects of police patrol in crime “Hot Spots”: a randomized. Controll Trial Justice Q 12(4):625–648

    Google Scholar 

  • Sherman L, Gartin P, Buerger M (1989) Hot spots of predatory crime: routine activities and the criminology of place. Criminology 27(1):27–56

    Article  Google Scholar 

  • Telep C, Mitchell R, Weisburd D (2014) How much time should the police spend at crime hot spots? Answers from a police agency directed randomized field trial in sacramento, California. Justice Q 31:905–933

    Article  Google Scholar 

  • Uchida CD (2009) A national discussion on predictive policing: defining our terms and mapping successful implementation strategies. National Institute of Justice Working Document No. 230404. National Criminal Justice Reference Service. Available at: https://www.ncjrs.gov/App/Publications/abstract.aspx?ID=252437

  • Weisburd D (2015) The law of crime concentration and the criminology of place. Criminology 53(2):133–157

    Article  Google Scholar 

  • Weisburd D, Braga A, Groff E, Alese W (2017) Can hot spots policing reduce crime in Urban areas? An agent-based simulation. Criminology 55(1):137–173

    Article  Google Scholar 

  • Weisburd D, Eck JE (2004) What can police do to reduce crime, disorder, and fear? Ann Am Acad Pol Soc Sci 593(1):42–65

    Article  Google Scholar 

  • Weisburd D, Green L (1994) Defining the street-level drug market drugs and crime evaluating public policy initiatives. In: MacKenzie DL, Uchida CD (eds) Drugs and crime: evaluating public policy initiatives. Sage, Thousand Oaks, CA

    Google Scholar 

  • Weisburd D, Maher L, Sherman LW (1992) Contrasting crime general and crime specific theory the case of hot spots of crime. In: Adler F, Laufer WS (eds) Advances in criminological theory. New Brunswick, NJ, Transaction

    Google Scholar 

  • Weisburd D, Majmundar MK (eds) (2018) Proactive policing: effects on crime and communities. National academies of sciences, engineering, and medicine. National Academic Press, Washington DC

Download references

Acknowledgements

This paper relies on experiments conducted by the Montevideo Police Department under the leadership of the Chief of the Montevideo Police Department, M. Layera; and the participation of members of his team: the Director of the Dirección de Información Táctica, A. Sosa; and the Chief of the Grupo de Respuesta Táctica, A. Clavijo. R. Boba Santos and R. Santos provided valuable technical assistance and E. Maguire participated in early stages of the design. The opinions expressed in this publication are those of the authors and do not necessarily reflect the views of the Government of Uruguay, or the institutions they are affiliated to.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Sebastian Galiani.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Galiani, S., Jaitman, L. Predictive Policing in a Developing Country: Evidence from Two Randomized Controlled Trials. J Quant Criminol 39, 805–831 (2023). https://doi.org/10.1007/s10940-022-09551-y

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10940-022-09551-y

Keywords

JEL Classification

Navigation