Evaluation of Appropriate Weighting Methodology for the School Crime Supplement in the Seven Largest States

• NCJ Number: 251862
• Author(s): Andrew Moore; Bonnie E. Shook-Sa, M.A.S., RTI International; Philip Lee, RTI International; Barbara A. Oudekerk, Ph.D., BJS Statistician
• Date Published: April 2020
• Length: 207 pages
• Grant Number(s): 2011-NV-CX-K068
• Agencies: BJS-Sponsored
• Publication Type: Third-Party Report; Report (Grant-Sponsored)

Description

This report evaluates a methodology to produce estimates of bullying in the seven most populous states from data collected in the National Crime Victimization Survey's School Crime Supplement (SCS). Because the SCS sample was not designed to produce state-level estimates, the SCS data were reweighted based on counts of students from administrative records in the Department of Education’s Common Core Data (CCD) collection. These recalibrated weights were used to produce estimates of bullying in California, Florida, Illinois, New York, Ohio, Pennsylvania, and Texas from 2007 to 2013. Those estimates were then compared to estimates produced using the original SCS weights. This report describes how recalibrating the original SCS weights to the CCD student counts led to representative bullying estimates for public-school students in the seven states.