Multiple Imputation for the Supplementary Homicide Reports: Evaluation in Unique Test Data, 1990-1995, Chicago, Philadelphia, Phoenix and St. Louis (ICPSR 36379)

Published: Mar 31, 2016 View help for published

Principal Investigator(s): View help for Principal Investigator(s)
John Roberts, University of Wisconsin-Milwaukee; Aki Roberts, University of Wisconsin-Milwaukee

https://doi.org/10.3886/ICPSR36379.v1

Version V1

This study was an evaluation of multiple imputation strategies to address missing data using the New Approach to Evaluating Supplementary Homicide Report (SHR) Data Imputation, 1990-1995 (ICPSR 20060) dataset.

Roberts, John, and Roberts, Aki. Multiple Imputation for the Supplementary Homicide Reports: Evaluation in Unique Test Data, 1990-1995, Chicago, Philadelphia, Phoenix and St. Louis. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 2016-03-31. https://doi.org/10.3886/ICPSR36379.v1

Export Citation:

  • RIS (generic format for RefWorks, EndNote, etc.)
  • EndNote
United States Department of Justice. Office of Justice Programs. National Institute of Justice (2013-R2-CX-0038)

City

Inter-university Consortium for Political and Social Research
1990 -- 1995
1990 -- 1995

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.

These data were constructed from the Evaluating Supplementary Homicide Report (SHR) Data Imputation, 1990-1995 (ICPSR 20060) dataset. For instructions on how to replicate these data please see the appendix of the codebook.

The purpose of the study was to evaluate the effectiveness of multiple imputation strategies in addressing missing data in the Supplementary Homicide Report (SHR) data. The specific goals of the study were as follows:

  1. In the test population, evaluated multiple imputation methods by comparing imputed and actual distributions of victim-offender relationship.
  2. In the test population, evaluated multiple imputation methods by comparing results obtained for imputed and actual data in analysis of substantive models of criminological interest.
  3. Extended the work pursuant to the first and second goals to include the impact on these comparisons of choices such as particular methods for multiple imputation and the number of imputations used, and evaluate the Uniform Crime Reports [United States]: Supplementary Homicide Reports With Multiple Imputation, Cumulative Files 1976-2007 (ICPSR 24801) data in this context.

New Approach to Evaluating Supplementary Homicide Report (SHR) Data Imputation, 1990-1995 (ICPSR 20060) dataset was selected as it included incidents in which offender information was recorded as missing in the SHR data, but was recovered from police records. This test population enabled the evaluation of multiple imputation strategies and comparisons of imputed and actual data in various ways.

Not applicable.

Cross-sectional

Homicides known to the police in Chicago, Illinois, Philadelphia, Pennsylvania, Phoenix, Arizona, St. Louis, Missouri, from 1990 to 1995.

Homocide

New Approach to Evaluating Supplementary Homicide Report (SHR) Data Imputation, 1990-1995 (ICPSR 20060).

administrative records data

Main Data Version 1 (n=6,573) includes 18 variables on the details of the incident, demographics of the victim, demographics of the offender, victim and offender relationship, and source of victim and offender relationship information.

Main Data Version 2 (n=6,573) includes 15 variables on the details of the incident, demographics of the victim, and demographics of the offender. This dataset includes the same incidents as Main Data Version 1 with some of the missing data filled in with information from police reports.

Imputation Version 1 to 4 datasets each contain 1,000 imputations generated using IVEWare software. They differ in the number of variables used to create the imputations. Only variables used for the imputations are included in each file.

  • Imputation Version 1 data (n=6,573,000) includes 5 variables on year, victim/offender relationship, imputation number, jurisdiction, and ID.
  • Imputation Version 2 data (n=6,573,000) includes 9 variables on year, calendar season, victim/offender relationship, victim demographics, imputation number, jurisdiction, and ID.
  • Imputation Version 3 data (n=6,573,000) includes 11 variables on year, calendar season, circumstance of incident, victim/offender relationship, victim demographics, imputation number, jurisdiction, and ID.
  • Imputation Version 4 data (n=6,573,000) includes 14 variables on year, calendar season, circumstance of incident, victim/offender relationship, victim demographics, offender demographics, imputation number, jurisdiction, and ID.

SAS MCMC and SAS MCMC No Bounds datasets each contain 1,000 imputations generated using the normal-MCMC routine in SAS. Each dataset (n=6,573,000) includes 31 variables on year, calendar season, circumstance of incident, victim/offender relationship, victim demographics, offender demographics, imputation number, jurisdiction, and ID. The no bounds version of the SAS MCMC dataset does not impose bounds on the numerical value of the victim and offender age variables.

2016-03-31

2016-03-31

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:
  • Roberts, John, and Aki Roberts. Multiple Imputation for the Supplementary Homicide Reports: Evaluation in Unique Test Data, 1990-1995, Chicago, Philadelphia, Phoenix and St. Louis. ICPSR36379-v1. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 2016-03-31. http://doi.org/10.3886/ICPSR36379.v1

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.

NACJD logo

This dataset is maintained and distributed by the National Archive of Criminal Justice Data (NACJD), the criminal justice archive within ICPSR. NACJD is primarily sponsored by three agencies within the U.S. Department of Justice: the Bureau of Justice Statistics, the National Institute of Justice, and the Office of Juvenile Justice and Delinquency Prevention.