D06: Introduction to Quantitative Text Analysis

This course covers basic methods of handling large volumes of text data. A mix of interactive lectures and guided coding sessions will enable you to implement your first text as a data research design.

In the era of digital information, the vast volumes of text data available present both challenges and opportunities for researchers. Automated text analysis methods, which leverage computational power, have become essential tools for social scientists aiming to extract reproducible and valid insights from textual data. These methods complement traditional manual content analysis approaches and are increasingly prevalent in computationally oriented disciplines.

This course provides a comprehensive introduction to text-as-data methods, focusing on both theoretical foundations and practical applications. Participants will explore the entire workflow of text analysis, including data collection, processing, quality control, and critical interpretation of results. The course is structured into five days, with each day featuring two sessions that combine lectures and hands-on coding exercises. The topics covered include:

Day 1

Session 1: Motivations and applications for using text-as-data methods, and an introduction to text representation.

Session 2: String manipulation and text representation using R.

Day 2

Session 1: Feature selection techniques for text analysis.

Session 2: Concepts of data collection and manual coding.

Day 3

Session 1: Introduction to dictionaries and their role in text analysis.

Session 2: Practical dictionary creation and validation using R.

Day 4    

Session 1: Unsupervised text analysis methods & the importance of validation

Session 2: Topic modelling implementation and validation.

Day 5    

Session 1: Text classification and using large language models (LLMs).

Session 2: Practical applications of LLMs for text analysis.

The course combines lectures, interactive discussions, and guided coding sessions. Lectures will introduce the basics of each topic, highlight recent advancements in methods, and provide resources for further exploration. Practical sessions will involve working through prepared code in R, with additional coding challenges assigned on Tuesday and Thursday to reinforce learning. Participants are expected to have basic knowledge of R and RStudio prior to attending.

By the end of the course, participants will be able to:

1. Make informed decisions about suitable text analysis methods for specific research scenarios.
2. Apply basic text-as-data methods in practice.
3. Critically evaluate the results of text analysis.

This course is open to researchers and practitioners seeking to enhance their research capabilities. It is particularly suited for individuals with little or no prior experience in automated text analysis who wish to incorporate these methods into their projects or deepen their understanding of the field.

Readings

Grimmer, J., & Stewart, B. M. (2013). Text as data: The promise and pitfalls of automatic content analysis methods for political texts. Political analysis, 21(3), 267-297. 

Van Atteveldt, W., Trilling, D., & Calderon, C. A. (2022). Computational analysis of communication. John Wiley & Sons. http://cssbook.net

2 Credits

For completion of all work before and during the course, as outlined by the Instructor, and 90% participation and attendance of the course.

2 Additional Credits

Course specific extra assignments. These can include submitting assignments before the course, daily assignments, and/or a final assignment to be completed after the course as decided by the Instructor.  This description is subject to change at the discretion of the Instructor.