The Source for Crime and Justice Data

Bureau of Justice Statistics Workshop on Quantitative Analysis of Crime and Criminal Justice Data

Description

This four-week seminar introduces participants to major data series sponsored by the Bureau of Justice Statistics (BJS), such as the National Crime Victimization Survey (NCVS) and the National Incident-Based Reporting System (NIBRS). The seminar focuses on innovative applications of survey methods and incident-based data in criminal justice. Issues of data collection and analysis are covered in depth. Each participant will design, conduct, and present the results from a quantitative research project that uses BJS data. The course is designed for faculty and research professionals as well as for advanced graduate students in the social sciences who are comfortable with data analysis software and quantitative research. The workshop is limited to 12 participants.

View the workshop listing on the ICPSR Summer Program site

Structure

A crucial part of this course is independent projects by each participant. Therefore, participants are expected to arrive with a project in mind, and a dataset available to use. Presentations on these projects will be given at the beginning and end of the workshop. At the start, participants will describe the rationale behind their proposed analysis, and at the end, update the class on their progress and findings.

Along with a daily seminar, several lunches and guest speakers will be scheduled. Small groups with guest speakers will be arranged, as appropriate. Discussions within the seminars will center on issues raised in the readings, by speakers and by the demands of participants' projects. To provide some continuity as throughout the workshop, specific methodological (e.g. combining contextual and individual levels of analysis, accommodating measurement error) and substantive themes (e.g. establishing and interpreting disparity in criminal justice system decision-making) that are common across data sets will be highlighted.

The seminar will provide an introduction to NACJD and a number of data sets available in the archive. Emphasis will be given to BJS data that describe the principal activities of the criminal justice system:

These data sets will be described briefly since most of the basic descriptive data will be in the readings. Participants will have the opportunity to manipulate these data in lab sessions. More complex issues pertaining to sampling error, measurement error and file structure issues will be described and methods for dealing with these issues will be presented. The goal at the end of the workshop is to have participants ready to use these data correctly to inform theoretical and methodological issues.

Background

For 24 years, the BJS has sponsored a four week intensive seminar on Quantitative Analysis of Crime and Criminal Justice Data. In past years, the seminar has featured an overview of many of the data collections compiled by BJS and the use of these data in an individualized research project. The growth of many data collections programs at BJS has been impressive, but has caused practical difficulties for the organization and content of the seminar. Trying to learn about all data sets compiled by BJS, or even only about the most important ones, results in incomplete consideration of important analytic topics. Therefore, the BJS summer program adopted a more focused, issue-oriented approach beginning in the year 2000 and continuing in future years.

Recent reductions in crime and the growth of imprisonment are substantive issues addressed by a number of BJS data sets. Also evident in these data sets are methodological issues concerning new measures of criminal offending, reliable and valid instrumentation design, accuracy of event reporting, administrative record data sampling, missing data imputation, and standard error calculation for multi-stage sampling designs. Seminar sessions focus on these substantive and methodological issues through the use of didactic lectures, discussions with guest speakers, computerized problem solving exercises, and the development and execution of a research report using the available data.