Building the Evidence that Research Evidence Matters

Can a professional development intervention increase child welfare agencies’ use of evidence and improve permanency for children in foster care?

Foster care agencies are increasingly called upon to use research evidence to better serve youth, but they lack effective strategies for increasing their staff’s knowledge and skills to improve research use. This study will test a structured intervention designed to increase the capacity, motivation, and opportunity to use research evidence when responding to local problems. The intervention uses classroom and booster sessions to develop participants’ capacity to use research, motivates use through a web-based platform and consultation that demonstrates the utility and value of evidence-based decision making, and identifies opportunities and resources to support research use. The team will engage private foster care agency managers as they work to reduce length of stay in foster care, discharge children to permanent homes, place children in less restrictive environments, and respond to other local challenges. Wulczyn and colleagues will use a cluster randomized trial, with 60 agencies randomly assigned to either an intervention or comparison group, to examine whether the intervention improves agency staff’s capacity, motivation, and use of research evidence, as well as whether changes in the use of research evidence are linked to better outcomes for children placed in out-of-home care. Phone and in-person interviews with study participants and an agency representative will be used to assess changes in relevant skills and knowledge, evidence sources, and motivation to use research. This information will be supplemented with meeting observations and document reviews to identify changes within the agency that advance research use, such as investments in information technology, relevant meeting agenda items, or the creation of supportive positions or policies. Child permanency outcomes will be estimated with administrative data and multi-level models that include an agency-level measure of research use.