Using Propensity Score Matching and Difference within Differences approaches to evaluate policies and interventions in educational and social sciences Seminar

Event details

Ideally, when we want to evaluate the impact of an intervention, we would use a Randomised Controlled Trial to ensure that we evaluate impact in the most rigorous way possible. However, in many cases policies and interventions are implemented without a rigorous evaluative design of this type. In these cases, the 'next best' approach is to use methods that approximate RCT's as closely as possible. In this seminar I will present two such methods, Propensity Score Matching and Difference-within-Difference designs, using two recent educational policy interventions as examples.