POL42050

Automated text analysis has become very popular in political science over the past years. With the massive availability of text data on the web, political scientists increasingly recognize automated text analysis (or “text as data”) as a promising approach for analyzing various kinds of social and political behavior. This module introduces students of political science to the quantitative analysis of textual data. We discuss the underlying theoretical assumptions, substantive applications of these methods, and the respective implementations in the R statistical programming language.

Each session combines lectures with practical, hands-on exercises to apply the methods to political text, dealing with practical issues in each step of the research process. Most of these methods can be reduced to a three-step process: first, identifying texts and units of texts for analysis; second, extract quantitatively measured features from these texts and converting them to a quantitative feature matrix; third, analyse this matrix with statistical methods, such as dictionary construction and application, scaling models, and topic models, to draw inferences about the texts. Students will learn how to apply these steps to various types of texts. The course will also introduce advanced methods, including word embeddings, speech transcription, machine translation, and computer vision.