Dr. YongJei Lee is Assistant Professor in the Department of Criminology at the University of South Florida. His researches focused on Crime Pattern Analysis, Crime Hot Spots, and Policing Effectiveness have appeared in Journal of Quantitative Criminology, Journal of Experimental Criminology, and in Translational Criminology. Phone: 941-359-4732 Address: Department of Criminology University of South Florida 8350 N. Tamiami Trail C243 | Sarasota, FL 34243
This is a video abstract of an article entitled, "Why Your Bar Has Crime but Not Mine: Resolving ... more This is a video abstract of an article entitled, "Why Your Bar Has Crime but Not Mine: Resolving the Land Use and Crime – Risky Facility Conflict." For the visually impaired, this video abstract is enabled with an audio transcript by the author.
Correctional authorities require accurate, unbiased, and interpretable tools to predict individua... more Correctional authorities require accurate, unbiased, and interpretable tools to predict individuals’ chances of recidivating if released into the community. However, existing prediction models have serious limitations meeting these requirements. We overcome these limitations by applying an established medical diagnostic approach: a relaxed naïve Bayes classifier. Using logistic regression in the form of a naïve Bayes classifier, we estimate the weights of observed features of offenders on recidivism. We apply these weights in a relaxed naïve Bayes classifier to predict the probability of recidivism. Results show that acquired features are stronger predictors of recidivism than innate features. Relaxed naïve Bayes classifier produces far less racial disparity than most alternatives. Critically, it is easier for users to interpret than its alternatives.
Supplemental material, PQX887809 Supplementary material for A Theory-Driven Algorithm for Real-Ti... more Supplemental material, PQX887809 Supplementary material for A Theory-Driven Algorithm for Real-Time Crime Hot Spot Forecasting by YongJei Lee, SooHyun O and John E. Eck in Police Quarterly
This study examines deviant identity in relation to youth offending by combining items tapping bo... more This study examines deviant identity in relation to youth offending by combining items tapping both self-appraisal and reflected appraisal. In particular, using survey data from 3,446 Korean youth across five waves of the Korea Youth Panel Survey (KYPS), findings from group-based trajectory modeling (GBTM) present four distinct offending groups—a high-rate chronic group, stable non-offending group, adolescence-limited group, and declining group. Then, findings from the multinomial logit model reveal that deviant identity is a robust predictor of offending for subgroups of adolescents involved in offending at any level in comparison to stable non-offenders. Accordingly, this study supports the idea that deviant identity should be considered as a prominent predictor of a variety of types of youth offending.
Supplemental online material for "Why Your Bar Has Crime but Not Mine: Resolving the Land Us... more Supplemental online material for "Why Your Bar Has Crime but Not Mine: Resolving the Land Use and Crime – Risky Facility Conflict" published in Justice Quarterly
Interpretations of two bodies of crime-place research conflict. Land use and crime studies claim ... more Interpretations of two bodies of crime-place research conflict. Land use and crime studies claim particular facilities increase crime. Risky facilities studies show most places of a single type hav...
This study examines the effect that individuals’ perceptions of police have on their adoption of ... more This study examines the effect that individuals’ perceptions of police have on their adoption of crime prevention measures. Unlike past research that conceptualized police perceptions as inversely ...
Real-time crime hot spot forecasting presents challenges to policing. There is a high volume of h... more Real-time crime hot spot forecasting presents challenges to policing. There is a high volume of hot spot misclassifications and a lack of theoretical support for forecasting algorithms, especially in disciplines outside the fields of criminology and criminal justice. Transparency is particularly important as most hot spot forecasting models do not provide their underlying mechanisms. To address these challenges, we operationalize two different theories in our algorithm to forecast crime hot spots over Portland and Cincinnati. First, we use a population heterogeneity framework to find places that are consistent hot spots. Second, we use a state dependence model of the number of crimes in the time periods prior to the predicted month. This algorithm is implemented in Excel, making it extremely simple to apply and completely transparent. Our forecasting models show high accuracy and high efficiency in hot spot forecasting in both Portland and Cincinnati context. We suggest previously d...
International Journal of Police Science & Management
By operationalizing two theoretical frameworks, we forecast crime hot spots in Colorado Springs. ... more By operationalizing two theoretical frameworks, we forecast crime hot spots in Colorado Springs. First, we use a population heterogeneity (flag) framework to find places where the hot spot forecasting is consistently successful over months. Second, we use a state dependence (boost) framework of the number of crimes in the periods prior to the forecasted month. This algorithm is implemented in Microsoft Excel®, making it simple to apply and completely transparent. Results shows high accuracy and high efficiency in hot spot forecasting, even if the data set and the type of crime we used in this study were different from what the original algorithm was based on. Results imply that the underlying mechanisms of serious and non-serious crime for forecasting are different from each other. We also find that the spatial patterns of forecasted hot spots are different between calls for service and crime event. Future research should consider both flag and boost theories in hot spot forecasting.
The primary objectives of this research are (1) to introduce summary measures of concentration th... more The primary objectives of this research are (1) to introduce summary measures of concentration that are relatively new to our field; (2) compare four concentration measures to determine whether there are reasons to use one in favor of the others; and (3) apply the measures to a real-case data to further understand the concentration phenomenon. Using the crime data of Cincinnati, we compare four commonly used social science measures of concentration: Gini, Simpson, Shannon, and Decile indices. For some purposes, the measures are interchangeable, while for other purposes the measures may suggest different interpretations for the same set of data. This paper is the first quantitative comparison of multiple measures of crime place concentration. We describe the benefits and limitations of each index and the circumstances for which each is most useful. We also answer the question: is crime within street segments spread along the segment or is it concentrated at a few addresses, as most place-based crime studies have overlooked the interior variability of crime on street segments.
We examine the distribution of crime across street segments in Cincinnati (proximal places), and ... more We examine the distribution of crime across street segments in Cincinnati (proximal places), and then proceed to look at the distribution of crime at addresses (proprietary places) within the segments. We find that for a substantial proportion of the proximal places, crime is concentrated at a few proprietary places. This indicates that the address is potentially implicated in the cause of crime events. Theoretical and practical implications follow.
Recent legislation directs the Pennsylvania Commission on Sentencing to develop parole guidelines... more Recent legislation directs the Pennsylvania Commission on Sentencing to develop parole guidelines that can guide counties across Pennsylvania in responding to supervision violations with uniformity. This paper analyzes violations by criminal offenders sentenced to Allegheny County jail in 2008 as a basis for proposing sanctioning guidelines in the county that can serve as an example of analyses and guidelines that other counties in the state might undertake in the future. Specifically, structured decision tools utilized in three states‘ violation sanctioning process are examined, and a similar tool is devised for technical parole violations in Allegheny County that takes account of legal and jurisdictional particularities, as well as input and feedback from members of the Allegheny County Probation Office and the judiciary. The result consists primarily of two decision grids: the first (Stage 1) uses information about the violation severity, number of violations, and violator risk t...
Objectives: Design and estimate the impacts of a prevention program for part 1 violent crimes in ... more Objectives: Design and estimate the impacts of a prevention program for part 1 violent crimes in micro-place crime hot spots. Methods: A longitudinal study of crime hot spots using 21 years of crime offense report data on part 1 violent crimes from Pittsburgh, Pennsylvania. Based on kernel density smoothing for a definition of micro-place crime hot spots, we replicate past work on the existence of “chronic” hot spots, but then with such hot spots accounted for introduce “temporary” hot spots. Results: Chronic hot spots are good targets for prevention. They are easily identified and they tend to persist. Temporary hot spots, however, predominantly last only one month. Thus the common practice of identifying hot spots using a short time window of crime data and assuming that the resulting hot spots will persist is ineffective for temporary hot spots. Instead it is necessary to forecast the emergence of temporary hot spots to prevent their crimes. Over time chronic hot spots, while sti...
Objectives: Design and estimate the impacts of a prevention program for part 1 violent crimes in ... more Objectives: Design and estimate the impacts of a prevention program for part 1 violent crimes in micro-place crime hot spots. Methods: A longitudinal study of crime hot spots using 21 years of crime offense report data on part 1 violent crimes from Pittsburgh, Pennsylvania. Based on kernel density smoothing for a definition of micro-place crime hot spots, we replicate past work on the existence of “chronic” hot spots, but then with such hot spots accounted for introduce “temporary” hot spots. Results: Chronic hot spots are good targets for prevention. They are easily identified and they tend to persist. Temporary hot spots, however, predominantly last only one month. Thus the common practice of identifying hot spots using a short time window of crime data and assuming that the resulting hot spots will persist is ineffective for temporary hot spots. Instead it is necessary to forecast the emergence of temporary hot spots to prevent their crimes. Over time chronic hot spots, while sti...
This is a video abstract of an article entitled, "Why Your Bar Has Crime but Not Mine: Resolving ... more This is a video abstract of an article entitled, "Why Your Bar Has Crime but Not Mine: Resolving the Land Use and Crime – Risky Facility Conflict." For the visually impaired, this video abstract is enabled with an audio transcript by the author.
Correctional authorities require accurate, unbiased, and interpretable tools to predict individua... more Correctional authorities require accurate, unbiased, and interpretable tools to predict individuals’ chances of recidivating if released into the community. However, existing prediction models have serious limitations meeting these requirements. We overcome these limitations by applying an established medical diagnostic approach: a relaxed naïve Bayes classifier. Using logistic regression in the form of a naïve Bayes classifier, we estimate the weights of observed features of offenders on recidivism. We apply these weights in a relaxed naïve Bayes classifier to predict the probability of recidivism. Results show that acquired features are stronger predictors of recidivism than innate features. Relaxed naïve Bayes classifier produces far less racial disparity than most alternatives. Critically, it is easier for users to interpret than its alternatives.
Supplemental material, PQX887809 Supplementary material for A Theory-Driven Algorithm for Real-Ti... more Supplemental material, PQX887809 Supplementary material for A Theory-Driven Algorithm for Real-Time Crime Hot Spot Forecasting by YongJei Lee, SooHyun O and John E. Eck in Police Quarterly
This study examines deviant identity in relation to youth offending by combining items tapping bo... more This study examines deviant identity in relation to youth offending by combining items tapping both self-appraisal and reflected appraisal. In particular, using survey data from 3,446 Korean youth across five waves of the Korea Youth Panel Survey (KYPS), findings from group-based trajectory modeling (GBTM) present four distinct offending groups—a high-rate chronic group, stable non-offending group, adolescence-limited group, and declining group. Then, findings from the multinomial logit model reveal that deviant identity is a robust predictor of offending for subgroups of adolescents involved in offending at any level in comparison to stable non-offenders. Accordingly, this study supports the idea that deviant identity should be considered as a prominent predictor of a variety of types of youth offending.
Supplemental online material for "Why Your Bar Has Crime but Not Mine: Resolving the Land Us... more Supplemental online material for "Why Your Bar Has Crime but Not Mine: Resolving the Land Use and Crime – Risky Facility Conflict" published in Justice Quarterly
Interpretations of two bodies of crime-place research conflict. Land use and crime studies claim ... more Interpretations of two bodies of crime-place research conflict. Land use and crime studies claim particular facilities increase crime. Risky facilities studies show most places of a single type hav...
This study examines the effect that individuals’ perceptions of police have on their adoption of ... more This study examines the effect that individuals’ perceptions of police have on their adoption of crime prevention measures. Unlike past research that conceptualized police perceptions as inversely ...
Real-time crime hot spot forecasting presents challenges to policing. There is a high volume of h... more Real-time crime hot spot forecasting presents challenges to policing. There is a high volume of hot spot misclassifications and a lack of theoretical support for forecasting algorithms, especially in disciplines outside the fields of criminology and criminal justice. Transparency is particularly important as most hot spot forecasting models do not provide their underlying mechanisms. To address these challenges, we operationalize two different theories in our algorithm to forecast crime hot spots over Portland and Cincinnati. First, we use a population heterogeneity framework to find places that are consistent hot spots. Second, we use a state dependence model of the number of crimes in the time periods prior to the predicted month. This algorithm is implemented in Excel, making it extremely simple to apply and completely transparent. Our forecasting models show high accuracy and high efficiency in hot spot forecasting in both Portland and Cincinnati context. We suggest previously d...
International Journal of Police Science & Management
By operationalizing two theoretical frameworks, we forecast crime hot spots in Colorado Springs. ... more By operationalizing two theoretical frameworks, we forecast crime hot spots in Colorado Springs. First, we use a population heterogeneity (flag) framework to find places where the hot spot forecasting is consistently successful over months. Second, we use a state dependence (boost) framework of the number of crimes in the periods prior to the forecasted month. This algorithm is implemented in Microsoft Excel®, making it simple to apply and completely transparent. Results shows high accuracy and high efficiency in hot spot forecasting, even if the data set and the type of crime we used in this study were different from what the original algorithm was based on. Results imply that the underlying mechanisms of serious and non-serious crime for forecasting are different from each other. We also find that the spatial patterns of forecasted hot spots are different between calls for service and crime event. Future research should consider both flag and boost theories in hot spot forecasting.
The primary objectives of this research are (1) to introduce summary measures of concentration th... more The primary objectives of this research are (1) to introduce summary measures of concentration that are relatively new to our field; (2) compare four concentration measures to determine whether there are reasons to use one in favor of the others; and (3) apply the measures to a real-case data to further understand the concentration phenomenon. Using the crime data of Cincinnati, we compare four commonly used social science measures of concentration: Gini, Simpson, Shannon, and Decile indices. For some purposes, the measures are interchangeable, while for other purposes the measures may suggest different interpretations for the same set of data. This paper is the first quantitative comparison of multiple measures of crime place concentration. We describe the benefits and limitations of each index and the circumstances for which each is most useful. We also answer the question: is crime within street segments spread along the segment or is it concentrated at a few addresses, as most place-based crime studies have overlooked the interior variability of crime on street segments.
We examine the distribution of crime across street segments in Cincinnati (proximal places), and ... more We examine the distribution of crime across street segments in Cincinnati (proximal places), and then proceed to look at the distribution of crime at addresses (proprietary places) within the segments. We find that for a substantial proportion of the proximal places, crime is concentrated at a few proprietary places. This indicates that the address is potentially implicated in the cause of crime events. Theoretical and practical implications follow.
Recent legislation directs the Pennsylvania Commission on Sentencing to develop parole guidelines... more Recent legislation directs the Pennsylvania Commission on Sentencing to develop parole guidelines that can guide counties across Pennsylvania in responding to supervision violations with uniformity. This paper analyzes violations by criminal offenders sentenced to Allegheny County jail in 2008 as a basis for proposing sanctioning guidelines in the county that can serve as an example of analyses and guidelines that other counties in the state might undertake in the future. Specifically, structured decision tools utilized in three states‘ violation sanctioning process are examined, and a similar tool is devised for technical parole violations in Allegheny County that takes account of legal and jurisdictional particularities, as well as input and feedback from members of the Allegheny County Probation Office and the judiciary. The result consists primarily of two decision grids: the first (Stage 1) uses information about the violation severity, number of violations, and violator risk t...
Objectives: Design and estimate the impacts of a prevention program for part 1 violent crimes in ... more Objectives: Design and estimate the impacts of a prevention program for part 1 violent crimes in micro-place crime hot spots. Methods: A longitudinal study of crime hot spots using 21 years of crime offense report data on part 1 violent crimes from Pittsburgh, Pennsylvania. Based on kernel density smoothing for a definition of micro-place crime hot spots, we replicate past work on the existence of “chronic” hot spots, but then with such hot spots accounted for introduce “temporary” hot spots. Results: Chronic hot spots are good targets for prevention. They are easily identified and they tend to persist. Temporary hot spots, however, predominantly last only one month. Thus the common practice of identifying hot spots using a short time window of crime data and assuming that the resulting hot spots will persist is ineffective for temporary hot spots. Instead it is necessary to forecast the emergence of temporary hot spots to prevent their crimes. Over time chronic hot spots, while sti...
Objectives: Design and estimate the impacts of a prevention program for part 1 violent crimes in ... more Objectives: Design and estimate the impacts of a prevention program for part 1 violent crimes in micro-place crime hot spots. Methods: A longitudinal study of crime hot spots using 21 years of crime offense report data on part 1 violent crimes from Pittsburgh, Pennsylvania. Based on kernel density smoothing for a definition of micro-place crime hot spots, we replicate past work on the existence of “chronic” hot spots, but then with such hot spots accounted for introduce “temporary” hot spots. Results: Chronic hot spots are good targets for prevention. They are easily identified and they tend to persist. Temporary hot spots, however, predominantly last only one month. Thus the common practice of identifying hot spots using a short time window of crime data and assuming that the resulting hot spots will persist is ineffective for temporary hot spots. Instead it is necessary to forecast the emergence of temporary hot spots to prevent their crimes. Over time chronic hot spots, while sti...
Since place-based crime has been studied, scholars have employed a variety of ways to describe t... more Since place-based crime has been studied, scholars have employed a variety of ways to describe the concentration of crime at places. Most usefully, they sometimes provide a full distribution of crime across street segments, or among addresses, or other small geographic areas of interest. This is feasible if the researcher is showing the distribution of crime at places throughout one or two larger areas, such as a city. In such circumstances, a few tables or graphs will be sufficient. But once researchers started looking at spatial areas numbering in the hundreds and thousands, like street segments, then describing the internal distribution of crime within each becomes cumbersome. We need summary measures of crime concentration. The mean, median, and mode are not appropriate for this task: the first two because of the highly skewed nature of crime distributions, and while the mode is better, it does not provide enough information.
If there is a standard summary measure, it is the Decile index, showing the proportion of crime located at the 10 percent most crime ridden spatial units. There are closely related measures: the five percent, 20 percent, or other n percent most crime ridden spatial units. The choice of 10 percent (or other percent) is in some sense arbitrary. And this measure ignores much of the distribution. Fortunately, other disciplines have developed measures of concentration, or its opposite, dispersion. So, in this dissertation, I have compared three well established measures to the Decile: the Gini coefficient, Simpson index, and Shannon index.
In addition to crime concentration at places, scholars have become interested in the stability of high crime locations. Here too, a number of measures have been used, with the same problem arising. If the task is to show the stability of hot crime places in a city, we can create a table or graph to show the proportion of places that stay hot, stay cold, or shift in temperature. But how does one show this for thousands of street segments? The same measures that can provide useful summary information about concentration may be able to be used to provide summary information about stability.
The primary objectives of this dissertation are 1) to compare for concentration measures and determine if there were reasons to use one in favor of the others; and 2) to apply the same measures to stability and determine if some measures were superior to others. I show that for spatial concentration, the Shannon index is often preferred, and that for temporal stability, the Simpson is usually superior to other indices. The more commonly used Decile index is best used as supplement to the other indices, rather than by itself. I also demonstrate how these indices can be used together to explore the interior characteristics of spatial units. I conclude by showing how my research can be expanded to help us understand crime patterns and to prevent crime.
Keywords: Crime and Place, Concentration Measures, Stability Measures, Concentration of Crime, Stability of Crime, Systematic Review
The Encyclopedia of Research Methods and Statistical Techniques in Criminology and Criminal Justice, 2019
A bettor is often looking for new tools to refine the process to accurately specify the probabili... more A bettor is often looking for new tools to refine the process to accurately specify the probability of an uncertain event occurring. Once new information is introduced, a bettor uses the new information to update his prior information. This repetitive refining process is the key to understand the Bayesian analysis. This article introduces how Bayesian analysis, a theory devised by 18th-century English Presbyterian pastor Thomas Bayes, can help measure the outcome of events in a crime.
Weisburd (2015) has proposed the law of crime concentration at places, which states that urban cr... more Weisburd (2015) has proposed the law of crime concentration at places, which states that urban crimes concentrate in micro geospatial units, hot spots, within a narrow band of high percentages. For example, in large cities around 50% of crime occurs in only 5% of a city’s block‐long street segments. Should hot spot programs for crime prevention by police be constant for a year or more over the range from 0% to 5% of a city’s street network, or should they be dynamic? Of course, in the trivial case of a hot spot being permanently eliminated, police would no longer allocate special prevention. This chapter is concerned with non‐trivial cases. While the Koper curve (Koper 1995) addresses optimal duration of directed patrols on a daily basis in crime hot spots, the literature has been largely silent on longer time‐scales of weeks, months, or years for optimal duration of hot spot programs. For example, the Telep, Mitchel, and Weisburd (2014) hot spot field trials in Sacramento called for 15 minute patrols (following the Koper curve) every two hours from 9:00 AM until 1:00 AM, for a total of 8 patrols per hot spot and day. For implementation by the Sacramento police, however, should some hot spots have treatment for a year or longer but others for shorter durations of time—weeks or months? If so, for how long? To address these questions, this chapter further decomposes hot spots into two primary types by duration, chronic and temporary. Chronic hot spots are fixed with durations of at least a year, but often last much longer, even for decades. Temporary hot spots are dynamic with an on‐and‐off behavior for weeks or months at a time. As such, chronic hot spots merit constant crime prevention programs by police while temporary hot spots should have dynamic allocations of police resources. Needed for dynamic allocations are accurate models for predicting temporary hot spots, or early detection with accurate predictions of persistence. Criminological theories suggest crime patterns that are the basis for accurate‐enough crime prediction models. Notable in our paper (Gorr and Lee 2015) and in this chapter is the introduction of distributional equity of police crime prevention services as an important performance criterion for hot spot program design, in addition to the commonly‐used effectiveness criterion. This chapter makes the case that a combination chronic and temporary hot spot program is the most effective and equitable program. In doing so the chapter reviews related criminological theories, prediction models, police interventions for crime prevention, performance measures, and police policy.
Uploads
Videos by YongJei Lee
Article DOI: https://doi.org/10.1080/07418825.2021.1903068
Papers by YongJei Lee
Article DOI: https://doi.org/10.1080/07418825.2021.1903068
If there is a standard summary measure, it is the Decile index, showing the proportion of crime located at the 10 percent most crime ridden spatial units. There are closely related measures: the five percent, 20 percent, or other n percent most crime ridden spatial units. The choice of 10 percent (or other percent) is in some sense arbitrary. And this measure ignores much of the distribution. Fortunately, other disciplines have developed measures of concentration, or its opposite, dispersion. So, in this dissertation, I have compared three well established measures to the Decile: the Gini coefficient, Simpson index, and Shannon index.
In addition to crime concentration at places, scholars have become interested in the stability of high crime locations. Here too, a number of measures have been used, with the same problem arising. If the task is to show the stability of hot crime places in a city, we can create a table or graph to show the proportion of places that stay hot, stay cold, or shift in temperature. But how does one show this for thousands of street segments? The same measures that can provide useful summary information about concentration may be able to be used to provide summary information about stability.
The primary objectives of this dissertation are 1) to compare for concentration measures and determine if there were reasons to use one in favor of the others; and 2) to apply the same measures to stability and determine if some measures were superior to others. I show that for spatial concentration, the Shannon index is often preferred, and that for temporal stability, the Simpson is usually superior to other indices. The more commonly used Decile index is best used as supplement to the other indices, rather than by itself. I also demonstrate how these indices can be used together to explore the interior characteristics of spatial units. I conclude by showing how my research can be expanded to help us understand crime patterns and to prevent crime.
Keywords: Crime and Place, Concentration Measures, Stability Measures, Concentration of Crime, Stability of Crime, Systematic Review
concentrate in micro geospatial units, hot spots, within a narrow band of high percentages. For example,
in large cities around 50% of crime occurs in only 5% of a city’s block‐long street segments. Should hot
spot programs for crime prevention by police be constant for a year or more over the range from 0% to
5% of a city’s street network, or should they be dynamic? Of course, in the trivial case of a hot spot
being permanently eliminated, police would no longer allocate special prevention. This chapter is
concerned with non‐trivial cases.
While the Koper curve (Koper 1995) addresses optimal duration of directed patrols on a daily basis in
crime hot spots, the literature has been largely silent on longer time‐scales of weeks, months, or years
for optimal duration of hot spot programs. For example, the Telep, Mitchel, and Weisburd (2014) hot
spot field trials in Sacramento called for 15 minute patrols (following the Koper curve) every two hours
from 9:00 AM until 1:00 AM, for a total of 8 patrols per hot spot and day. For implementation by the
Sacramento police, however, should some hot spots have treatment for a year or longer but others for
shorter durations of time—weeks or months? If so, for how long?
To address these questions, this chapter further decomposes hot spots into two primary types by
duration, chronic and temporary. Chronic hot spots are fixed with durations of at least a year, but often
last much longer, even for decades. Temporary hot spots are dynamic with an on‐and‐off behavior for
weeks or months at a time. As such, chronic hot spots merit constant crime prevention programs by
police while temporary hot spots should have dynamic allocations of police resources. Needed for
dynamic allocations are accurate models for predicting temporary hot spots, or early detection with
accurate predictions of persistence. Criminological theories suggest crime patterns that are the basis for
accurate‐enough crime prediction models.
Notable in our paper (Gorr and Lee 2015) and in this chapter is the introduction of distributional equity
of police crime prevention services as an important performance criterion for hot spot program design,
in addition to the commonly‐used effectiveness criterion. This chapter makes the case that a
combination chronic and temporary hot spot program is the most effective and equitable program. In
doing so the chapter reviews related criminological theories, prediction models, police interventions for
crime prevention, performance measures, and police policy.