Self-Report and Police Arrest Data in Gang Program Evaluation: Methodological and Policy Implications

Kwai Ming Wa, University of Chicago
Rolondo Villarreal Sosa, University of Chicago
Irving A. Spergel, University of Chicago

Almost all gang program research has used either self-report offenses/arrests or police arrest histories to evaluate program outcomes. In preliminary evaluations of an OJJDP comprehensive, community- or area-wide gang crime reduction programs in three cities, we separately analyze self-report and police arrest histories of gang members and highly at-risk delinquent youth and young adults in Bloomington-Normal (101 program) and Champaign-Urbana (79 comparison), Illinois; Mesa, Arizona (107 program and 98 comparison); and San Antonio, Texas (101 program and 120 comparison). Early findings indicate that self-report and arrest data may or may not produce congruent results. Issues of sampling, matching of program and comparison groups, as well as community interests and concerns, and distinctive police policies and practices have to be considered. Self-reports may or may not be superior to police data in determining what the gang problem is in a particular community or area. Complementary use of both sources of data better explains the nature and scope of the gang problem than use of one type of data alone, particularly in measuring possible program effects.

(Return to Program Resources)

Updated 05/20/2006