Treating Political Texts as Data: Introduction to Text Mining with R - Einzelansicht

Treating words as data has become a popular approach to analysing text documents in the social sciences. The main goal of this seminar is to introduce students to the possibilities and pitfalls of automated content analysis with R. Beginning with a brief overview of text-as-data methods, this seminar delves into specific text-mining techniques, including algorithms for supervised and unsupervised ideological scaling. The course will include theoretical sessions introducing and discussing the conceptual frameworks described in the reading material, as well as hands-on exercises applying the main preprocessing and text-mining methods.

We will cover:

• Introduction to text mining, R and RStudio.
• Feature extraction and the basics of visualisation.
• Text preprocessing and its pitfalls.
• Sentiment analysis.
• Dimensionality reduction with correspondence analysis.
• Wordfish.
• And more!

The last session(s) will be dedicated to presenting your mini-projects. Together with the homework and active participation in the discussion throughout the semester, they will form the exam in this course. Feedback on the submitted assignments will be communicated individually via Moodle; a sample solution will be discussed in class.

• Grimmer, J., Roberts, M. E., and Stewart, B. M. (2022). Text as Data: A New Framework for Machine Learning and the Social Sciences.
• Grolemund, G., and Wickham, H. (2016). R for Data Science.O'Reilly Media, Inc.
• Silge, J., and Robinson, D. (2017). TextMining with R: A Tidy Approach. O'ReillyMedia, Inc.

Completing the research design seminar is strongly recommended.

Proficiency in R is not a prerequisite for participation in this seminar, although basic knowledge may be helpful.

Assessment Methods:

1. Seminar:

• Home assignments
• Project presentation
• In-class discussions

2. Module exam: seminar paper

Registration and withdrawal deadline for the module final examination: 17.10.2024 - 30.03.2025

Further details will be discussed in class. 

MA students in Political Science, Sociology, and National and International Administration and Policy (MANIA).