Basic Concepts and Novel Designs
Graeme Blair (UCLA)
July 1 - July 3
A central question in EITM is whether a given research design is a “good” design for evaluating empirical implications of a theory. The Basic Concepts and Novel Designs module provides a set of theoretical concepts and computational tools for answering this question. We introduce the model-inquiry-data strategy-answer strategy (MIDA) framework for declaring the features of a research design and then diagnosing their properties. We then learn the DeclareDesign family of R packages for simulating the research design and conducting Monte Carlo studies of its properties. With these conceptual and computational tools in hand, we examine the properties of standard survey experimental designs as well as novel designs that have been recently proposed. For each design, we declare its features in code, diagnose its properties via simulation, and examine common tradeoffs faced by researchers such as the choice between sample size and the number of treatment arms.
Graeme Blair is an Assistant Professor of Political Science at the University of California, Los Angeles. Previously he was a Pre/Post-Doctoral Fellow at Columbia University. Graeme's expertise is on experiments and field research in developing countries with a special emphasis on the causes and consequences of violence.
The syllabus for the course can be found here.