Andrew Palmer Wheeler

How Long After? A Natural Experiment Assessing the Impact of the Length of Aftercare Service Delivery on Recidivism:

Although aftercare programs have been gaining popularity as a mechanism for helping offenders readjust to society, evaluations of their success remain varied. This is most likely due to the diversity of programs labeled as aftercare and the inability of research to isolate specific program components. The current study capitalizes on a natural experiment to examine the impact of one particular component, length of service delivery, on recidivism. The study employs survival analysis techniques on a population of inmates graduating from a motivational boot camp who either received no aftercare, 30 days of aftercare, or 90 days of aftercare (depending on the existing policy on their graduation date). Findings show that those receiving 30 days of aftercare services are indistinguishable from those receiving no aftercare services in terms of recidivism. Also, we find that although those receiving 90 days of aftercare did recidivate substantially less than those receiving 0 or 30 days of aftercare, after accounting for sample attrition, however, these findings also lacked statistical significance.

Testing for Racial Profiling With the Veil-of-Darkness Method

The “veil-of-darkness” method is an innovative and low-cost approach that circumvents many of the benchmarking issues that arise in testing for racial profiling. Changes in natural lighting are used to establish a presumptively more race-neutral benchmark on the assumption that after dark, police suffer an impaired ability to detect motorists’ race. Applying the veil-of-darkness method to vehicle stops by the Syracuse (NY) police between 2006 and 2009 and examining differences among officers assigned to specialized traffic units and crime-suppression units, we found that African Americans were no more likely to be stopped during daylight than during darkness, indicating no racial bias.

The effect of police paramilitary unit raids on crime at micro-places in Buffalo, New York

This study examines the deterrent effect of the short-term application of police paramilitary units. The police in Buffalo conducted 39 police paramilitary unit raids during a two-day period in 2012. It is hypothesized that if police paramilitary units are used then there will be a reduction in calls for service, Part 1 crimes and drug arrests. The effect of the police paramilitary unit raids at micro-place street units was estimated by drawing a comparable sample of control units via propensity score modeling. A fixed effects negative binomial panel model was estimated for counts 35 weeks before and 35 weeks after the intervention. Calls for service and drug arrests increased in the treated areas compared with controls. There is slight evidence of a deterrent effect for Part 1 crimes in the two weeks after the intervention. The null findings suggest that, at best, police paramilitary unit raids have a short-term deterrent effect on crime.

Tables and Graphs for Monitoring Temporal Crime Patterns

This article is a practical review on how to construct tables and graphs to monitor temporal crime trends. Such advice is mostly applicable to crime analysts to improve the readability of their products, but is also useful to general consumers of crime statistics in trying to identify crime trends in reported data. First, the use of percent change to identify significant changes in crime trends is critiqued, and an alternative metric based on the Poisson distribution is provided. Second, visualization principles for constructing tables are provided, and a practical example of remaking a poor table using these guidelines is shown. Finally, the utility of using time series charts to easily identify short- and long-term increases, as well as outliers in seasonal data using examples with actual crime data is illustrated.

Visualization techniques for journey to crime flow data

Research applications using journey to crime data present unique challenges for modeling and visualization not encountered with typical data that are only georeferenced at one point in space. The article provides graphical examples on how to visualize flow lines directly, how to effectively aggregate and visualize statistical summaries of distributions of flow lines, and supplemental statistical graphics to visualize flow data. The article also provides supplementary material demonstrating and implementing several of the techniques discussed in the article.

Testing for Similarity in Area-Based Spatial Patterns: Alternative Methods to Andresen's Spatial Point Pattern Test

Andresen’s spatial point pattern test (SPPT) compares two spatial point patterns on defined areal units: it identifies areas where the spatial point patterns diverge and aggregates these local (dis)similarities to one global measure. We discuss the limitations of the SPPT and provide two alternative methods to calculate differences in the point patterns. In the first approach we use differences in proportions tests corrected for multiple comparisons. We show how the size of differences matter, as with large point patterns many areas will be identified by SPPT as statistically different, even if those differences are substantively trivial. The second approach uses multinomial logistic regression, which can be extended to identify differences in proportions over continuous time. We demonstrate these methods on identifying areas where pedestrian stops by the New York City Police Department are different from violent crimes from 2006 through 2016.

What Factors Influence an Officer’s Decision to Shoot? The Promise and Limitations of Using Public Data:

We analyze a set of 207 Dallas Police Department officer-involved shooting incidents in reference to 1,702 instances in which officers from the same agency drew their firearms but did not shoot at the suspect. We find that situational factors of whether the suspect was armed and whether an officer was injured were the best predictors of the decision to shoot. We also find that African Americans are less likely than Whites to be shot. It is important to collect data on encounters in which weapons are and are not discharged. Analyses examining only shootings is fundamentally limited in assessing racial bias.

Evaluating Community Prosecution Code Enforcement in Dallas, Texas

We evaluated a community prosecution program in Dallas, Texas. City attorneys, who in Dallas are the chief prosecutors for specified misdemeanors, were paired with code enforcement officers to improve property conditions in a number of proactive focus areas, or PFAs, throughout the city. We conducted a panel data analysis, focusing on the effects of PFA activity on crime in 19 PFAs over a six-year period (monthly observations from 2010 to 2015). Control areas with similar levels of pre-intervention crime were also included. Statistical analyses controlled for pre-existing crime trends, seasonality effects, and other law enforcement activities. With and without dosage data, the total crime rate decreased in PFA areas relative to control areas. City attorney/code enforcement teams, by seeking the voluntary or court-ordered abatement of code violations and criminal activity at residential and commercial properties, apparently improved public safety in targeted areas.

Exploring Bias in Police Shooting Decisions With Real Shoot/Don’t Shoot Cases:

The controversy surrounding recent high-profile police shootings (e.g., Michael Brown in Ferguson, Missouri; Laquan McDonald in Chicago) has prompted inquiry into the possible existence of bias in officers’ use-of-force decisions. Using a balanced mix of shoot/don’t shoot cases from a large municipal police department in the Southwestern United States, this study analyzed the effect of suspect race on officers’ decisions to shoot—while accounting for other theoretically relevant factors. Findings suggest that Black suspects were not disproportionately the target of police shootings; Black suspects were approximately one third as likely to be shot as other suspects. This finding challenges the current bias narrative and is consistent with the other race-related findings in recently published research.

Monitoring Volatile Homicide Trends Across U.S. Cities

The recent increase in the national homicide rate in the United States has generated much speculation about its causes among the media. In this article we show how two data visualization tools, funnel charts and time series fan charts, can show the typical volatility in homicide rates in different cities over time. Many of the recent increases are not out of the norm given historical patterns, and so one need not rely on various ex-ante hypotheses to explain recent homicide spikes occurring in some U.S. cities.