Introduction to Statistics, Programming in R, and Graphical Design

This course serves as the foundation course and will introduce basic statistics, common statistical distributions, hypothesis testing, and programming in R. Course examples will include data from randomized controlled trials studying poverty interventions among low-income populations. While providing a foundational overview of statistics, the class will familiarize students with new research in the field of development studies. The class will introduce students to R—used throughout the certificate program—and learn to use R for statistical analysis and professional-looking graphics.

Students take courses consecutively with subsequent classes building on material in previous classes.

Experimental Methods

Understanding the experimental method is key to estimating impacts of an array of development programs and policies. This course will give students exposure to experimental methodology, laboratory experiments, and field experiments. Students will learn how to collect and measure both objective and subjective phenomena and to run experiments in a variety of settings and for a broad range of interventions. It will also provide an introduction to sampling, creation of questionnaires, tablet survey software, and the design of indices used for various social, economic, and psychological applications.

Introduction to Econometrics

This course will introduce students to data management and regression analysis, focusing on Ordinary Least Squares (OLS) regression, the assumptions of the OLS model, multivariate regression, how to address violations of the assumptions of the basic OLS model, and how to build and test regression models with continuous, discrete, and categorical variables. Examples will focus on social impacts from health, education, microenterprise, and infrastructure programs.

Causal Econometrics and Machine Learning

Valid program evaluation and testing of treatment effects require a strong understanding of causal statistics. In this course, students will learn how to identify causal effects of development programs using methods such as interrupted time series, difference-in-differences, pipeline methods, covariate matching, instrumental variables, and regression discontinuity design. Students will learn about machine-learning algorithms such as LASSO and ridge regression and how they can explain how social impacts vary among program beneficiaries.