fbpx Empirical Methods II | Harvard Kennedy School
Michela Carlana Photo

Michela Carlana

Assistant Professor of Public Policy
Dara Kay Cohen Photo

Dara Kay Cohen

Professor of Public Policy

Tim Layton

30th Anniversary Associate Professor of Health Care Policy, HMS

Liz McKenna Photo

Liz McKenna

Assistant Professor of Public Policy

This course is a 6-week module that follows on API-202M, allowing students to further specialize in a particular area of evidence for policy making. First year MPP students can select among the following three options. 

Section A (Michela Carlana) Causal Inference and Prediction: This module will delve deeper into quantitatively evaluating the impact of programs through quasi-experimental study designs, and assessing the strengths and weaknesses of different approaches. Students will also be introduced to the language and techniques of prediction and machine learning. Analyses of datasets will be conducted using R.

Section B (Dara Kay Cohen) and Section C (Liz McKenna) Qualitative and Mixed Methods: This module will cover qualitative and multi-method research design and the basic principles and particular techniques of qualitative data collection and analysis. Research techniques will include research with human subjects, including interviews, focus groups, and ethnography, and research using documents, as well as how these combine with quantitative approaches. Topics will include measurement and conceptualization, research ethics, mixed methods research designs, and evaluating qualitative analyses. 

Section D (Chase Harrison) Survey Methods: Survey research is used in a variety of fields to create original data and answer questions that otherwise couldn’t be answered. This module is designed to teach students the skills necessary to create, use, and interpret survey data. The module will cover questionnaire design, sampling, modes of data collection, and other topics. Students will complete applied assignments in survey methods.

Section Z (Timothy Layton):  Similar to Section A, this module will delve into quasi-experimental techniques most commonly used for causal inference, laying out the mathematical foundations for each approach, and assessing the underlying mathematical assumptions required for each method to produce causal estimates. Students will also be exposed to modern methods for prediction and machine learning, with a detailed description of the statistical theory behind these approaches along with instruction in their practical application. All analysis will be conducted using R.  Please note, students in Section Z of API-202M will continue on to Section Z of API-203M.

Please note, API-203M is offered February 27 - April 14, 2023.