Evaluation of Safe Start: Promising Approaches for Children Exposed to Violence (RAND Proposal 2005-553)

RAND Corporation has been selected as the national evaluator for the Safe Start Promising Approaches for Children Exposed to Violence program. RAND project staff will work with OJJDP staff to finalize and implement a design that addresses the following evaluation hypotheses:
(1) Interventions or programs that are evidence-based or theory-driven and implemented with fidelity will show positive outcomes (less depression/anxiety, developmental delay, school failure, antisocial or delinquent behavior, and post-traumatic stress disorder).
(2) Interventions or programs that lack a theoretical or evidence base or that are delivered without fidelity will show no changes in outcomes or negative outcomes.
(3) Certain elements of the intervention will be instrumental in their effectives and necessary for replication. For instance, programs focused on child witnesses of IPV that include the parents in the program will be more effective than those that focus solely on the child
(4) Certain elements of the intervention will be seen as necessary for replication by program staff members and consumers, while other elements will be seen as unnecessary. For instance, consumer involvement (via steering committees or management groups) in the planning and ongoing monitoring of the program will be viewed as a necessary and effective means to insure the intervention is responsive to the target population.
(5) More effective programs will be those offering a wider range of services, and more intensive services, and will cost more per participant than less effective programs.
(6) Certain interventions will be produce change at the individual level. Other interventions will produce change at the organizational or system level. For instance, programs focused on providing counseling services for children exposed to violence will produce change at that level while programs focused on changing how law enforcement organizes their response will first produce change at an organizational or system level
(7) Certain interventions will be more effective in certain populations (depending on the interventions and populations under study). For instance, programs focused on children exposed to community violence will be more effective for older children.
(8) More intervention or exposure to the program will be associated with more positive outcomes.
(9) Programs or interventions that improve symptoms and behaviors will also improve resiliency.
(10) Improved protective factors will at least partially mediate the intervention effect within effective programs.
(11) Baseline resilience and dynamic improvements in resilience over time in the control/comparison group children will buffer them from the impact of re-exposure to violence.

An important component of this project is building and sustaining relationships with the sites to be evaluated and collaborating with technical assistance providers, where necessary, to assist the sites in collecting data and meeting other evaluation related goals.
CA/NCF