Using Multilevel Growth Models to Evaluate a Community-Level Intervention to Prevent Youth Violence
The Chicago Center for Youth Violence Prevention (CCYVP) is a CDC-funded National Center for Excellence on Youth Violence Prevention. Because the problem of youth violence is determined by multiple factors and requires coordinated effort at multiple levels and across systems, the activities of the CCYVP include a range of interventions to address individual, family, and community levels and include both universal and targeted interventions within a single community in Chicago. The purpose of this presentation is to describe the evaluation approach and initial impact on violence outcomes.
Publicly-available crime incident data for 2008-2013 were obtained from the Chicago Data Portal. Monthly counts of crimes were calculated by police beat. Data were analyzed using an interrupted time series design implemented under a generalized linear mixed model using a Poisson distribution. In these quasi-experimental analyses, trajectories of crimes for the five police beats within the target CCYVP communities were compared to trajectories for 75 other similar beats within the City of Chicago.
The CCYVP beats, compared to other Chicago beats, had more rapid decrease in total violent crime (p<.01) and shootings (p<.05). In addition, the CCYVP beats showed a marginally more rapid decrease in homicides (p<.10).
The results of these quasi-experimental analyses demonstrate that the CCYVP was successful in altering the trajectory of violent crime in the five police beats included in the target neighborhood. These analytic methods proved to be a powerful approach to evaluating complex, multi-year community-level interventions.
Learning Areas:Conduct evaluation related to programs, research, and other areas of practice
Public health or related research
Formulate hypotheses describing changes growth trajectories that can be attributed to community-level interventions. Discuss the use of linear mixed models and interrupted time series designs to the evaluation of community-level interventions. Discuss how this approach allows for the absence of a control group.
Keyword(s): Youth Violence, Evaluation
Qualified on the content I am responsible for because: I have been Co-I on multiple federally funded grants focusing on interventions to prevent youth violence, including the grants supporting the research being presented. Among my scientific interests is the evaluation of community-level interventions.
Any relevant financial relationships? No
I agree to comply with the American Public Health Association Conflict of Interest and Commercial Support Guidelines, and to disclose to the participants any off-label or experimental uses of a commercial product or service discussed in my presentation.