Semistructured interviews involve an interviewer asking some prespecified, open-ended questions, with follow-up questions based on what the interviewee has to say. This Reflections on Methodology post describes a semistructured interview protocol recently used to explore how children who experience poverty perceive their situations, their economic status, and public benefit programs.
Data from management information systems, direct observations, and the reactions of staff members can help programs understand themselves, identify areas for improvement, and set goals. This infographic presents examples of how programs in the Building Bridges and Bonds study used data from different sources to gain insights.
An essential step in the child support process is delivering legal documents to the person named as a parent. This infographic summarizes results from a Georgia intervention that aimed to get parents to come in and accept documents voluntarily instead of using a sheriff or process server to deliver them.
MDRC launches the first of a five-part web series from the Chicago Community Networks study — a mixed-methods initiative that combines formal social network analysis with in-depth field surveys of community practitioners. It measures how community organizations collaborate on local improvement projects and how they come together to shape public policy.
Easing the Transition to Adulthood for Vulnerable Young People
This infographic describes MDRC’s results from the largest random assignment evaluation of a program serving youth people aging out of the foster care and juvenile justice systems. After one year, YVLifeSet, a program run by Youth Villages, boosts earnings, increases housing stability and economic well-being, and improves outcomes related to health and safety.
Using an alternative to classical statistics, this paper reanalyzes results from three published studies of interventions to increase employment and reduce welfare dependency. The analysis formally incorporates prior beliefs about the interventions, characterizing the results in terms of the distribution of possible effects, and generally confirms the earlier published findings.