Papers
Topics
Authors
Recent
Search
2000 character limit reached

Estimating the Likelihood of Arrest from Police Records in Presence of Unreported Crimes

Published 11 Oct 2023 in stat.ME and stat.AP | (2310.07935v1)

Abstract: Many important policy decisions concerning policing hinge on our understanding of how likely various criminal offenses are to result in arrests. Since many crimes are never reported to law enforcement, estimates based on police records alone must be adjusted to account for the likelihood that each crime would have been reported to the police. In this paper, we present a methodological framework for estimating the likelihood of arrest from police data that incorporates estimates of crime reporting rates computed from a victimization survey. We propose a parametric regression-based two-step estimator that (i) estimates the likelihood of crime reporting using logistic regression with survey weights; and then (ii) applies a second regression step to model the likelihood of arrest. Our empirical analysis focuses on racial disparities in arrests for violent crimes (sex offenses, robbery, aggravated and simple assaults) from 2006--2015 police records from the National Incident Based Reporting System (NIBRS), with estimates of crime reporting obtained using 2003--2020 data from the National Crime Victimization Survey (NCVS). We find that, after adjusting for unreported crimes, the likelihood of arrest computed from police records decreases significantly. We also find that, while incidents with white offenders on average result in arrests more often than those with black offenders, the disparities tend to be small after accounting for crime characteristics and unreported crimes.

Citations (2)

Summary

  • The paper introduces a novel two-step regression estimator that adjusts arrest likelihoods by integrating crime reporting rates.
  • It utilizes NCVS and NIBRS data to empirically show that accounting for unreported crimes lowers estimated arrest probabilities, especially for violent offenses.
  • The study highlights how unreported crimes distort arrest data, urging a reexamination of racial disparities and policy reforms in criminal justice.

Estimating the Likelihood of Arrest from Police Records in the Presence of Unreported Crimes

The paper "Estimating the likelihood of arrest from police records in presence of unreported crimes" addresses a critical issue in criminology: the estimation of arrest likelihood for crimes, given the anticipated dark figure of unreported offenses. Traditional analysis of crime data often overlooks the discrepancy between crimes committed and those reported to law enforcement, potentially skewing arrest likelihood calculations that rely merely on officially recorded data.

Methodological Framework

The authors present a robust statistical framework to estimate the arrest likelihood using police data while accounting for crimes that remain unreported. This approach employs a parametric regression-based two-step estimator. The first step involves estimating crime reporting likelihood using logistic regression models and data from victimization surveys like the National Crime Victimization Survey (NCVS). In the second step, information from the National Incident Based Reporting System (NIBRS) is employed to estimate arrest likelihoods, adjusted using the crime reporting rates derived in the first step.

Empirical Analysis

The empirical focus of the paper lies in exploring racial disparities in arrest rates for violent crimes, such as sex offenses, robbery, and assault, over the period from 2006 to 2015 with data from NIBRS and crime reporting estimates from NCVS surveys spanning 2003 to 2020. Findings reveal a significant decrease in estimated arrest likelihood after adjustments for unreported crimes are accounted for. Notably, it is observed that offenses involving white offenders result in arrests at slightly higher rates than those involving black offenders, particularly after adjusting for crime specifics and unreported incidents.

Implications and Future Directions

The paper's methodology contributes to a more nuanced understanding of racial disparities in arrests, highlighting the necessity for accounting for the non-reporting bias in crime data analysis. By proposing a statistical framework that can withstand the complexities of non-random missing data (such as through capture-recapture methods and regression with inverse probability weighting), this study bridges the gap between officially reported figures and actual crime occurrences.

The results have both theoretical and practical implications, suggesting the need for revisiting previous criminological analyses and policies relying heavily on official crime data. This could lead to better-informed decisions in law enforcement and criminal justice systems aiming for equity.

Future work could enhance this framework by integrating nonparametric models, such as those handling mixed data types, to improve flexibility and application scope. Furthermore, the extension of this methodology could encompass additional socio-demographic variables, which might further elucidate the complex interactions influencing arrest outcomes. This might involve incorporating real-time or more granular data that could capture geospatial crime pattern variations more effectively.

Ultimately, this research underscores the critical importance of addressing data issues in criminal justice research and develops a promising pathway for improving arrest likelihood estimates amidst pervasive data limitations.

Paper to Video (Beta)

No one has generated a video about this paper yet.

Whiteboard

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

Open Problems

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

Continue Learning

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

Collections

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

Tweets

Sign up for free to view the 14 tweets with 338 likes about this paper.