|
Abstract:
|
Annually since 1973 the National Crime Victimization Survey (NCVS), and its predecessor the National Crime Survey (NCS), have collected self-report victimization data. Typically trend analyses have been restricted to data collected after 1992, when the survey underwent a large methodological redesign. However, in 2012 the Bureau of Justice Statistics (BJS) and Research Triangle Institute (RTI) initiated the Historical Trends project to integrate the historical NCS and contemporary NCVS into a single cross-sectional concatenated data file spanning 41 years. The project was a methodologically complex multiyear undertaking involving challenging decisions on-data linkage, variance estimation, continuity of variables over time, and importantly, analytical approaches for discussing 40-year trends in a meaningful way. As data collections from federal statistical agencies like BJS continue to mature, there will be increased demands to apply formal methods suitable for analysis of long term trend data. This presentation examines the use of Joinpoint Analysis to analyze long term trend data on violent and property victimization. Considerations and implications of using JPA are discussed.
|
