This study used crime count data from the Pittsburgh, Pennsylvania, Bureau of Police offense reports and 911 computer-aided dispatch (CAD) calls to determine the best univariate forecast method for crime and to evaluate the value of leading indicator crime forecast models. The researchers used the rolling-horizon experimental design, a design that maximizes the number of forecasts for a given time series at different times and under different conditions. Under this design, several forecast models are used to make alternative forecasts in parallel. For each forecast model included in an experiment, the researchers estimated models on training data, forecasted one month ahead to new data not previously seen by the model, and calculated and saved the forecast error. Then they added the observed value of the previously forecasted data point to the next month's training data, dropped the oldest historical data point, and forecasted the following month's data point. This process continued over a number of months. A total of 15 statistical datasets and 3 geographic information systems (GIS) shapefiles resulted from this study. The statistical datasets consist of Univariate Forecast Data by Police Precinct (Dataset 1) with 3,240 cases Output Data from the Univariate Forecasting Program: Sectors and Forecast Errors (Dataset 2) with 17,892 cases Multivariate, Leading Indicator Forecast Data by Grid Cell (Dataset 3) with 5,940 cases Output Data from the 911 Drug Calls Forecast Program (Dataset 4) with 5,112 cases Output Data from the Part One Property Crimes Forecast Program (Dataset 5) with 5,112 cases Output Data from the Part One Violent Crimes Forecast Program (Dataset 6) with 5,112 cases Input Data for the Regression Forecast Program for 911 Drug Calls (Dataset 7) with 10,011 cases Input Data for the Regression Forecast Program for Part One Property Crimes (Dataset 8) with 10,011 cases Input Data for the Regression Forecast Program for Part One Violent Crimes (Dataset 9) with 10,011 cases Output Data from Regression Forecast Program for 911 Drug Calls: Estimated Coefficients for Leading Indicator Models (Dataset 10) with 36 cases Output Data from Regression Forecast Program for Part One Property Crimes: Estimated Coefficients for Leading Indicator Models (Dataset 11) with 36 cases Output Data from Regression Forecast Program for Part One Violent Crimes: Estimated Coefficients for Leading Indicator Models (Dataset 12) with 36 cases Output Data from Regression Forecast Program for 911 Drug Calls: Forecast Errors (Dataset 13) with 4,936 cases Output Data from Regression Forecast Program for Part One Property Crimes: Forecast Errors (Dataset 14) with 4,936 cases Output Data from Regression Forecast Program for Part One Violent Crimes: Forecast Errors (Dataset 15) with 4,936 cases. The GIS Shapefiles (Dataset 16) are provided with the study in a single zip file: Included are polygon data for the 4,000 foot, square, uniform grid system used for much of the Pittsburgh crime data (grid400); polygon data for the 6 police precincts, alternatively called districts or zones, of Pittsburgh(policedist); and polygon data for the 3 major rivers in Pittsburgh the Allegheny, Monongahela, and Ohio (rivers).

This study was designed to develop crime forecasting as an application area for police in support of tactical deployment of resources. Data on crime offense reports and computer aided dispatch (CAD) drug calls and shots fired calls were collected from the Pittsburgh, Pennsylvania Bureau of Police for the years 1990 through 2001. Data on crime offense reports were collected from the Rochester, New York Police Department from January 1991 through December 2001. The Rochester CAD drug calls and shots fired calls were collected from January 1993 through May 2001. A total of 1,643,828 records (769,293 crime offense and 874,535 CAD) were collected from Pittsburgh, while 538,893 records (530,050 crime offense and 8,843 CAD) were collected from Rochester. ArcView 3.3 and GDT Dynamap 2000 Street centerline maps were used to address match the data, with some of the Pittsburgh data being cleaned to fix obvious errors and increase address match percentages. A SAS program was used to eliminate duplicate CAD calls based on time and location of the calls. For the 1990 through 1999 Pittsburgh crime offense data, the address match rate was 91 percent. The match rate for the 2000 through 2001 Pittsburgh crime offense data was 72 percent. The Pittsburgh CAD data address match rate for 1990 through 1999 was 85 percent, while for 2000 through 2001 the match rate was 100 percent because the new CAD system supplied incident coordinates. The address match rates for the Rochester crime offenses data was 96 percent, and 95 percent for the CAD data. Spatial overlay in ArcView was used to add geographic area identifiers for each data point: precinct, car beat, car beat plus, and 1990 Census tract. The crimes included for both Pittsburgh and Rochester were aggravated assault, arson, burglary, criminal mischief, misconduct, family violence, gambling, larceny, liquor law violations, motor vehicle theft, murder/manslaughter, prostitution, public drunkenness, rape, robbery, simple assaults, trespassing, vandalism, weapons, CAD drugs, and CAD shots fired.

This study assessed the implementation and impact of the One Vision One Life (OVOL) violence-prevention strategy in Pittsburgh, Pennsylvania. In 2003, the rise in violence in Pittsburgh prompted community leaders to form the Allegheny County Violence Prevention Imitative, which became the OVOL program. The OVOL program sought to prevent violence using a problem-solving, data-driven model to inform how community organizations and outreach teams respond to homicide incidents. The research team examined the impact of the OVOL program on violence using a quasi-experimental design to compare violence trends in the program's target areas before and after implementation to (1) trends in Pittsburgh neighborhoods where One Vision was not implemented, and (2) trends in specific nontarget neighborhoods whose violence and neighborhood dynamics One Vision staff contended were most similar to those of target neighborhoods. The Pittsburgh Bureau of Police provided the violent-crime data, which the research team aggregated into monthly counts. The Pittsburgh Department of City Planning provided neighborhood characteristics data, which were extracted from the 2000 Census. Monthly data were collected on 90 neighborhoods in Pittsburgh, Pennsylvania from 1996 to 2007, resulting in 12,960 neighborhood-by-month observations.

Tree canopy and crime in Pittsburgh, PA. Tree canopy data were from the 2010 Tree Canopy Assessment carried out by the University of Vermont Spatial Analysis Laboratory in collaboration with Tree Pittsburgh. Crime data was obtained from 2010 police reporting compiled by the Pittsburgh Neighborhood & Community Information System (PNCIS).

Highland Park. Crime = 3%, White = 66%, Poverty = 9%, Tree Canopy = 49% Larimer. Crime = 13%, White = 9%, Poverty = 22%, Tree Canopy = 22%

This study examined spatial and temporal features of crime guns in Pittsburgh, Pennsylvania, in order to ascertain how gun availability affected criminal behavior among youth, whether the effects differed between young adults and juveniles, and whether that relationship changed over time. Rather than investigating the general prevalence of guns, this study focused only on those firearms used in the commission of crimes. Crime guns were defined specifically as those used in murders, assaults, robberies, weapons offenses, and drug offenses. The emphasis of the project was on the attributes of crime guns and those who possess them, the geographic sources of those guns, the distribution of crime guns over neighborhoods in a city, and the relationship between the prevalence of crime guns and the incidence of homicide. Data for Part 1, Traced Guns Data, came from the City of Pittsburgh Bureau of Police. Gun trace data provided a detailed view of crime guns recovered by police during a two-year period, from 1995 to 1997. These data identified the original source of each crime gun (first sale to a non-FFL, i.e., a person not holding a Federal Firearms License) as well as attributes of the gun and the person possessing the gun at the time of the precipitating crime, and the ZIP-code location where the gun was recovered. For Part 2, Crime Laboratory Data, data were gathered from the local county crime laboratory on guns submitted by Pittsburgh police for forensic testing. These data were from 1993 to 1998 and provided a longer time series for examining changes in crime guns over time than the data in Part 1. In Parts 3 and 4, Stolen Guns by ZIP-Code Data and Stolen Guns by Census Tract Data, data on stolen guns came from the local police. These data included the attributes of the guns and residential neighborhoods of owners. Part 3 contains data from 1987 to 1996 organized by ZIP code, whereas Part 4 contains data from 1993 to 1996 organized by census tract. Part 5, Shots Fired Data, contains the final indicator of crime gun prevalence for this study, which was 911 calls of incidents involving shots fired. These data provided vital information on both the geographic location and timing of these incidents. Shots-fired incidents not only captured varying levels of access to crime guns, but also variations in the willingness to actually use crime guns in a criminal manner. Part 6, Homicide Data, contains homicide data for the city of Pittsburgh from 1990 to 1995. These data were used to examine the relationship between varying levels of crime gun prevalence and levels of homicide, especially youth homicide, in the same city. Part 7, Pilot Mapping Application, is a pilot application illustrating the potential uses of mapping tools in police investigations of crime guns traced back to original point of sale. NTC. It consists of two ArcView 3.1 project files and 90 supporting data and mapping files. Variables in Part 1 include date of manufacture and sale of the crime gun, weapon type, gun model, caliber, firing mechanism, dealer location (ZIP code and state), recovery date and location (ZIP code and state), age and state of residence of purchaser and possessor, and possessor role. Part 2 also contains gun type and model, as well as gun make, precipitating offense, police zone submitting the gun, and year the gun was submitted to the crime lab. Variables in Parts 3 and 4 include month and year the gun was stolen, gun type, make, and caliber, and owner residence. Residence locations are limited to owner ZIP code in Part 3, and 1990 Census tract number and neighborhood name in Part 4. Part 5 contains the date, time, census tract and police zone of 911 calls relating to shots fired. Part 6 contains the date and census tract of the homicide incident, drug involvement, gang involvement, weapon, and victim and offender ages. Data in Part 7 include state, county, and ZIP code of traced guns, population figures, and counts of crime guns recovered at various geographic locations (states, counties, and ZIP codes) where the traced guns first originated in sales by an FFL to a non-FFL individual. Data for individual guns are not provided in Part 7.

Graph and download economic data for Combined Violent and Property Crime Offenses Known to Law Enforcement in Westmoreland County, PA (DISCONTINUED) (FBITC042129) from 2011 to 2018 about Westmoreland County, PA; Pittsburgh; crime; violent crime; property crime; PA; and USA.

This data ceased updating with the transition to a new records management system on 11/14/2023. Access to the updated data set has been added as of April 11, 2025 here: Crime Data Guide.

Graph and download economic data for Combined Violent and Property Crime Offenses Known to Law Enforcement in Washington County, PA (DISCONTINUED) (FBITC042125) from 2005 to 2020 about Washington County, PA; Pittsburgh; crime; violent crime; property crime; PA; and USA.

The 30 Day Police Blotter contains the most recent initial crime incident data, updated on a nightly basis. All data is reported at the block/intersection level, with the exception of sex crimes, which are reported at the police zone level. The information is "semi-refined" meaning a police report was taken, but it has not made its way through the court system. This data is subject to change once it is processed and republished using Uniform Crime Reporting (UCR) standards. The UCR coding process creates a necessary delay before processed data is available for publication. Therefore, the 30 Day blotter will provide information for users seeking the most current information available.

This dataset will be continually overwritten and any records older than thirty days will be removed. Validated incidents will be moved to the Police Blotter Archive Dataset. Data in the archived file is of a higher quality and is the file most appropriate for reporting crime statistics.

This dataset only contains information reported by City of Pittsburgh Police, and does not contain incidents that solely involve other police departments operating within the city (campus police, Port Authority, etc.)

Crime Data Guide

The Department of Justice launched Operation Weed and Seed in 1991 as a means of mobilizing a large and varied array of resources in a comprehensive, coordinated effort to control crime and drug problems and improve the quality of life in targeted high-crime neighborhoods. In the long term, Weed and Seed programs are intended to reduce levels of crime, violence, drug trafficking, and fear of crime, and to create new jobs, improve housing, enhance the quality of neighborhood life, and reduce alcohol and drug use. This baseline data collection effort is the initial step toward assessing the achievement of the long-term objectives. The evaluation was conducted using a quasi-experimental design, matching households in comparison neighborhoods with the Weed and Seed target neighborhoods. Comparison neighborhoods were chosen to match Weed and Seed target neighborhoods on the basis of crime rates, population demographics, housing characteristics, and size and density. Neighborhoods in eight sites were selected: Akron, OH, Bradenton (North Manatee), FL, Hartford, CT, Las Vegas, NV, Pittsburgh, PA, Salt Lake City, UT, Seattle, WA, and Shreveport, LA. The "neighborhood" in Hartford, CT, was actually a public housing development, which is part of the reason for the smaller number of interviews at this site. Baseline data collection tasks included the completion of in-person surveys with residents in the target and matched comparison neighborhoods, and the provision of guidance to the sites in the collection of important process data on a routine uniform basis. The survey questions can be broadly divided into these areas: (1) respondent demographics, (2) household size and income, (3) perceptions of the neighborhood, and (4) perceptions of city services. Questions addressed in the course of gathering the baseline data include: Are the target and comparison areas sufficiently well-matched that analytic contrasts between the areas over time are valid? Is there evidence that the survey measures are accurate and valid measures of the dependent variables of interest -- fear of crime, victimization, etc.? Are the sample sizes and response rates sufficient to provide ample statistical power for later analyses? Variables cover respondents' perceptions of the neighborhood, safety and observed security measures, police effectiveness, and city services, as well as their ratings of neighborhood crime, disorder, and other problems. Other items included respondents' experiences with victimization, calls/contacts with police and satisfaction with police response, and involvement in community meetings and events. Demographic information on respondents includes year of birth, gender, ethnicity, household income, and employment status.

This compilation includes five historical datasets that are part of the University of Pittsburgh Library collection. The datasets were transcribed from The Pittsburgh Neighborhood Atlas, published in 1977. The atlas was prepared by the Pittsburgh Neighborhood Alliance. The information provides an insight into the neighborhoods conditions and the direction in which they were moving at the time of preparation. Much of the material describing neighborhood characteristics came from figures compiled for smaller areas: voting districts or census blocks. The five datasets in this collection provide data about overall neighborhood satisfaction and satisfaction with public services, based on a city-wide citizen survey. Also included are statistics about public assistance, the crime rate and the changes in real estate and mortgage loans transactions.

The Police Blotter Archive contains crime incident data after it has been validated and processed to meet Uniform Crime Reporting (UCR) standards, published on a nightly basis. This data validation process creates a data publishing delay of approximately thirty days. Users who require the most recent incident data should use the 30 Day Police Blotter. The 30 Day Police Blotter dataset contains more recent data, but has not yet been run through quality control and standardization procedures by the Police Bureau. All data is reported at the block/intersection level, with the exception of sex crimes, which are reported at the police zone level.

This dataset only contains information reported by City of Pittsburgh Police, and does not contain incidents that solely involve other police departments operating within the city (campus police, Port Authority, etc.)

Crime Data Guide

The Pittsburgh Youth Study (PYS) is part of the larger "Program of Research on the Causes and Correlates of Delinquency" initiated by the Office of Juvenile Justice and Delinquency Prevention in 1986. PYS aims to document the development of antisocial and delinquent behavior from childhood to early adulthood, the risk factors that impinge on that development, and help seeking and service provision of boys' behavior problems. The study also focuses on boys' development of alcohol and drug use, and internalizing problems. PYS consists of three samples of boys who were in the first, fourth, and seventh grades in Pittsburgh, Pennsylvania public schools during the 1987-1988 academic year (called the youngest, middle, and oldest sample, respectively). Using a screening risk score that measured each boy's antisocial behavior, boys identified at the top 30 percent within each grade sample on the screening risk measure (n=~250), as well as an equal number of boys randomly selected from the remainder (n=~250), were selected for follow-up. Consequently, the final sample for the study consisted of 1,517 total students selected for follow-up. 506 of these students were in the oldest sample, 508 were in the middle sample, and 503 were in the youngest sample. Assessments were conducted semiannually and then annually using multiple informants (i.e., boys, parents, teachers) between 1987 and 2010. The youngest sample was assessed from ages 6-19 and again at ages 25 and 28. The middle sample was assessed from ages 9-13 and again at age 23. The oldest sample was assessed from ages 13-25, with an additional assessment at age 35. Information has been collected on a broad range of risk and protective factors across multiple domains (e.g., individual, family, peer, school, neighborhood). Measures of conduct problems, substance use/abuse, criminal behavior, mental health problems have been collected. This collection contains data and syntax files for delinquency constructs. The datasets include constructs on the frequency and level of criminal and delinquent activities, including theft, violence, weapons used, delinquency, drug-selling, white collar crime, as well as police contacts and past incarceration. Additionally, the collection includes data on delinquency risk (high vs. low) and the associated weight. The delinquency constructs were created by using the PYS raw data. The raw data are available at ICPSR in the following studies: Pittsburgh Youth Study Youngest Sample (1987 - 2001) [Pittsburgh, Pennsylvania], Pittsburgh Youth Study Middle Sample (1987 - 1991) [Pittsburgh, Pennsylvania] , and Pittsburgh Youth Study Oldest Sample (1987 - 2000) [Pittsburgh, Pennsylvania].

Non-traffic citations (NTCs, also known as "summary offenses") document low-level criminal offenses where a law enforcement officer or other authorized official issued a citation in lieu of arrest. These citations normally include a fine. In Pennsylvania, NTCs often include a notice to appear before a magistrate if the person does not provide a guilty plea. Offenses that normally result in a citation include disorderly conduct, loitering, harassment and retail theft. This dataset only contains information reported by City of Pittsburgh Police. It does not contain incidents that solely involve other police departments operating within the city (for example, campus police or Port Authority police). Latinos are not included in this data as a race and they will not be reflected in this data. More documentation is available in our Crime Data Guide.

The National Initiative for Building Community Trust and Justice (the National Initiative) is a joint project of the National Network for Safe Communities, the Center for Policing Equity, the Justice Collaboratory at Yale Law School, and the Urban Institute, designed to improve relationships and increase trust between communities and law enforcement. Funded by the Department of Justice, this mixed-methods evaluation aimed to assess outcomes and impacts in six cities that participated in the National Initiative, which include Birmingham, AL; Fort Worth, TX; Gary, IN; Minneapolis, MN; Pittsburgh, PA; and Stockton, CA. The data described herein represent two waves of surveys of residents living in the highest-crime, lowest-income residential street segments in the six National Initiative cities. The first wave was conducted between September 2015 and January 2016, and the second wave was conducted between July and October 2017. Survey items were designed to measure neighborhood residents' perceptions of their neighborhood conditions--with particular emphases on neighborhood safety, disorder, and victimization--and perceptions of the police as it relates to procedural justice, police legitimacy, officer trust, community-focused policing, police bias, willingness to partner with the police on solving crime, and the law. The data described herein are from pre- and post-training assessment surveys of officers who participated in three trainings: 1) procedural justice (PJ) conceptual training, which is the application of PJ in the context of law enforcement-civilian interactions, as well as its role in mitigating historical tensions between law enforcement and communities of color; 2) procedural justice tactical, which provided simulation and scenario-based exercises and techniques to operationalize PJ principles in officers' daily activities; and 3) implicit bias, which engaged officers in critical thought about racial bias, and prepared them to better identify and handle identity traps that enable implicit biases. Surveys for the procedural justice conceptual training were fielded between December 2015 and July 2016; procedural justice tactical between February 2016 and June 2017; and implicit bias between September 2016 and April 2018. Survey items were designed to measure officers' understanding of procedural justice and implicit bias concepts, as well as officers' levels of satisfaction with the trainings.