CRPL Research Methods Seminar Series - Ben Weidmann (Harvard): Missing Presumed Different

Missing Presumed Different

This presentation will focus on the risk posed by attrition in quantitative education evaluations. It is based on a forthcoming paper which estimates the magnitude of attrition bias for 10 Randomized Controlled Trials (RCTs) in education. Are children and schools who dropout of studies different in observed and unobservable ways from those who stay in evaluations? If so, how big a problem is this? And should we do more than the standard missing value imputation to test whether our results are robust?

About the Speaker

Ben Weidmann is a final year PhD student at Harvard and a Senior Researcher at the Education Endowment Foundation. His research focusses on the quantitative methods we use to accumulate knowledge in education. In particular, he is interested in how well different designs work in schools and in quantifying various biases.