Level: Graduate
Taught: S26, S25, S24, S23
Role: Instructor of Record
Description: In the social sciences, quantitative analyses depend critically on the availability of high-quality data, but these data are rarely readily available at the click of a mouse. Instead, they are often messy, in difficult formats, or in forms not immediately useable in statistical software programs. In this class, we will learn how to find, curate, and create high-quality social science data. We will balance the teaching of practical tools—such as web scraping, crowd-sourcing techniques, online surveys, massive database management, and dataset merging—with the examination of theoretical perspectives on the role of measurement error in the analysis of causality. Code will also be treated as a form of data, and some teaching emphasis placed on how to make software packages that can improve reproducibility and enhance research impact. Projects completed during the course will give students an opportunity to create new, high-quality data contributing not only to their research projects but also to the broader cumulation of knowledge in the social sciences. Geared towards those with some coding background in R or Python.
–
Past Student Projects (Selected)
All 2024 US Political Candidates & Staff, by Will Schrepferman
Dataset of Trade-Related Congressional Press Releases, by Elle Diether
Dataset on Internal Displacement in Africa, 2008-2023, by Joseph Amoah and Chukwuemeka Ohajionu
Identity Appeals in 2020 U.S. House Campaign Tweets and Replies, by Young Seok Kim, Jiyoung Park, and Jeong Hyun Kim
UNDP Internal Audit Reports, by Jongwon Oh and Hyunjin Yim
–