Predicting Crime through Incarceration: The Impact of Prison Cycling on Crime in Communities in Boston, Massachusetts, Newark, New Jersey, Trenton, New Jersey, and Rural New Jersey, 2000-2010 (ICPSR 35014)

Version Date: Mar 22, 2017

Cite this study |

Principal Investigator(s):
Todd R. Clear, Rutgers University; Natasha A. Frost, Northeastern University; Michael Carr, University of Massachusetts - Boston; Geert Dhondt, John Jay College of Criminal Justice; Anthony Braga, Rutgers University; Garrett A.R. Warfield, The Boston Foundation

https://doi.org/10.3886/ICPSR35014.v1

Version V1

Slide tabs to view more

Summary

These data are part of NACJD's Fast Track Release and are distributed as they were received from the data depositor. The files have been zipped by NACJD for release, but not checked or processed except for the removal of direct identifiers. Users should refer to the accompanying readme file for a brief description of the files available with this collection and consult the investigator(s) if further information is needed.

Researchers compiled datasets on prison admissions and releases that would be comparable across places and geocoded and mapped those data onto crime rates across those same places. The data used were panel data. The data were quarterly or annual data, depending on the location, from a mix of urban (Boston, Newark and Trenton) and rural communities in New Jersey covering various years between 2000 and 2010.

The crime, release, and admission data were individual level data that were then aggregated from the individual incident level to the census tract level by quarter (in Boston and Newark) or year (in Trenton). The analyses centered on the effects of rates of prison removals and returns on rates of crime in communities (defined as census tracts) in the cities of Boston, Massachusetts, Newark, New Jersey, and Trenton, New Jersey, and across rural municipalities in New Jersey.

There are 4 Stata data files. The Boston data file has 6,862 cases, and 44 variables. The Newark data file has 1,440 cases, and 45 variables. The Trenton data file has 66 cases, and 32 variables. The New Jersey Rural data file has 1,170 cases, and 32 variables.

Citation

Clear, Todd R., Frost, Natasha A., Carr, Michael, Dhondt, Geert, Braga, Anthony, and Warfield, Garrett A.R. Predicting Crime through Incarceration: The Impact of Prison Cycling on Crime in Communities in Boston, Massachusetts, Newark, New Jersey, Trenton, New Jersey, and Rural New Jersey, 2000-2010. Inter-university Consortium for Political and Social Research [distributor], 2017-03-22. https://doi.org/10.3886/ICPSR35014.v1

Export Citation:

  • RIS (generic format for RefWorks, EndNote, etc.)
  • EndNote

Funding

United States Department of Justice. Office of Justice Programs. National Institute of Justice (2009-IJ-CX-4037)

Subject Terms

Geographic Coverage

Smallest Geographic Unit

census tract

Distributor(s)

Inter-university Consortium for Political and Social Research
Hide

Time Period(s)

2000 -- 2010 (Boston), 2007 -- 2010 (Newark), 2009 -- 2010 (Trenton), 2003 -- 2010 (New Jersey Rural)

Date of Collection

2009-12-01 -- 2013-09-13 (Boston), 2009-12-01 -- 2013-09-13 (Newark), 2009-12-01 -- 2013-09-13 (Trenton), 2009-12-01 -- 2013-09-13 (New Jersey Rural)

Data Collection Notes

  1. These data are part of NACJD's Fast Track Release and are distributed as they were received from the data depositor. The files have been zipped by NACJD for release, but not checked or processed except for the removal of direct identifiers. Users should refer to the accompanying readme file for a brief description of the files available with this collection and consult the investigator(s) if further information is needed.

Hide

Study Purpose

The purpose of this project was to estimate the impact of "prison cycling" (the flow into and out of prison) on crime rates in communities, with special concern about areas that have high rates of prison cycling.

Study Design

The original electronic data files for this project were sourced from nine different state and local criminal justice agencies, four in Massachusetts (Boston Police Department, Suffolk County House of Correction, Massachusetts Department of Correction, and Massachusetts Parole Board) and five in New Jersey (Newark Police Department, Trenton Police Department, New Jersey Department of Corrections, New Jersey State Police, and New Jersey State Parole Board).

Each agency provided individual level data on offenders (incarceration data) or offenses (crime data), in most cases, from their data information systems. State agencies provided state-wide data and local agencies provided local data. These original individual level data files were then cleaned and relevant street address elements were geocoded. Once all individual level data had been geocoded the various admissions, release, and crime incident files were then merged with census data (from the 2000 decennial census - Summary File 3) and aggregated to the block, block group, and census tract level.

Sample

Upon receipt of data files from one of our partner agencies, any personal identifying information in each file was removed and replaced with unique study identification numbers before subsequent cleaning, geocoding, and analysis. After cleaning, all files were submitted to Justice Mapping, a partner organization, for geocoding.

Prior to geocoding, the research team made sure that each data file contained fields for a street address, city, state, and zip code (when available) for geocoding. Justice Mapping geocoded and mapped the crime and incarceration datasets that did not already have geocodes. Records without a mappable address were removed. Addresses were then geocoded using the Geocode.com service. The service assigned a code to each address and returned the latitude and longitude, State FIPS, County FIPS, Census Tract, and Census Block Group IDs for each address. Only matches with a code of 1 to 5 were used. A match code of 1 indicated a match to the primary name of the street segment, a match code of 2 indicated an intersection match "from end of the segment from the first street in the address," a match code of 3 indicated an intersection match "to end of the segment from the first street in the address," and a match code of 5 indicated a match to alternate or secondary name on a segment.

Time Method

Longitudinal: Panel

Universe

Prison admissions and releases in Boston, Massachusetts, Newark, New Jersey, Trenton, New Jersey, and rural New Jersey from 2000 to 2010.

Unit(s) of Observation

New Jersey Rural: census tract, Trenton: census tract, Boston: census tract, Newark: census tract

Data Source

Massachusetts Department of Correction

Massachusetts Parole Board

Trenton Police Department (New Jersey)

Boston Police Department (Massachusetts)

New Jersey State Police

Newark Police Department (New Jersey)

New Jersey Department of Corrections

Suffolk County House of Correction (Massachusetts)

New Jersey State Parole Board

Data Type(s)

Mode of Data Collection

Description of Variables

The Boston (44 variables, n=6,862) and Newark (45 variables, n=1,440) files have nearly all the same variables. The Newark file has one additional variable (variable name: county ; variable label: County (FIPS)). These data files include such information as admission rates, quarterly crime rates, median income, tract population, and release rates. They also contain measures of concentrated disadvantage, poverty rate, and unemployment rate. Additionally, the data files include census tract.

The Trenton (32 variables, n=66) and New Jersey Rural (32 variables, n= 1,170) files have the same variables. These data files include such information as admission rates, median income, tract population, and release rates. They also contain measures of concentrated disadvantage, poverty rate, and unemployment rate. Additionally, the data files include census tract.

Response Rates

Not applicable.

Presence of Common Scales

none

Hide

Original Release Date

2017-03-22

Version History

2018-02-15 The citation of this study may have changed due to the new version control system that has been implemented. The previous citation was:
  • Clear, Todd R., Natasha A. Frost, Michael Carr, Geert Dhondt, Anthony Braga, and Garrett A.R. Warfield. Predicting Crime through Incarceration: The Impact of Prison Cycling on Crime in Communities in Boston, Massachusetts, Newark, New Jersey, Trenton, New Jersey, and Rural New Jersey, 2000-2010. ICPSR35014-v1. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 2017-03-22. http://doi.org/10.3886/ICPSR35014.v1
Hide

Notes

  • These data are part of NACJD's Fast Track Release and are distributed as they were received from the data depositor. The files have been zipped by NACJD for release, but not checked or processed except for the removal of direct identifiers. Users should refer to the accompanying readme file for a brief description of the files available with this collection and consult the investigator(s) if further information is needed.

  • The public-use data files in this collection are available for access by the general public. Access does not require affiliation with an ICPSR member institution.