Advanced Research Methods in Global Health
1. Develop competence in quantitative research skills needed for dissertation and beyond.
Major topics covered
This course is intended for doctoral students preparing for their comprehensive exams and working towards their dissertations. The focus is on providing quantitative skills for conducting program, impact or other forms of evaluation using econometric methods, particularly for use with observational data and non-experimental or quasi-experimental using the Stata 12.0 (or 13.0) statistical software package. Key topics that will be covered are: Linear regression models with their assumptions and limitations; Limited dependent variable models (logit, probit tobit, multinomial logit/probit); Structural Models; Instrumental variables and two-stage least squares to correct for endogeneity; Sample selection models; Multilevel models and models with complex sample designs; Matching methods: Propensity score matching, exact matching, nearest neighbor matching; Applications of program evaluations; Time series analysis with pooled and longitudinal data; Regression discontinuity designs.
Classroom lectures, problem sets, independent reading, research projects