Development of Data Products and Adjustment Methods to Allow for the Evaluation of Historical Trends in the National Crime Victimization Survey

• Activity Number: 459 - Time-Trend Analysis with Complex Survey Data
• Type: Topic Contributed
• Date/Time: Wednesday, August 2, 2017 : 8:30 AM to 10:20 AM
• Sponsor: Survey Research Methods Section
• Author(s): Andrew Moore* and Marcus Berzofsky
• Companies: RTI International and RTI International
• Keywords: Trends ; NCVS

Abstract

One challenge with conducting trend analysis for long running studies is organizing and adjusting the data to account for changes over time to key measures of analysis. The Bureau of Justice Statistics (BJS) has been collecting data on personal and household victimization since 1973 through the National Crime Victimization Survey (NCVS) and its predecessor, the National Crime Survey (NCS). During this time, the NCVS and NCS have undergone a number of survey protocol and questionnaire changes. In addition, although the victimization survey data are publicly available, working with more than 40 years of data can be challenging, making it difficult for researchers to generate historical estimates. To alleviate this burden on analysts, BJS undertook the NCVS Historical Trends (NHT) project to provide BJS stakeholders with ready access to the entire series of victimization survey data. We discuss the challenges faced during this undertaking, including the development of common constructs to account for questionnaire changes, the handling of missing and/or incorrectly coded data, and adjustment methods to account for a break in the comparability of estimates.