|
Instructors: Prof. Dr. Steffen Eckhard
Event type:
Seminar / exercise
Org-unit: Politics, Administration & International Relations
Displayed in timetable as:
CSS: Text analysis
Hours per week:
3
Credits:
6,0
Location:
Campus der Zeppelin Universität
Language of instruction:
Englisch
Min. | Max. participants:
4 | 35
Priority scheme: Standard-Priorisierung
Course content:
This seminar offers a non-technical introduction to computational text analysis, requiring very limited previous experience in statistical software. The primary goal is to provide students with an overview of text analytical methods in a broad range of research applications. This includes methods such as dictionary analysis, topic modelling or machine learning. Advanced students get an opportunity to test such methods on a research project of their own.
Methods of computational text analysis and natural language processing facilitate the analysis of such large amounts of text, such as newspaper articles, statements, protocols, blog posts, or posts in social media. Being able to analyze and understand text data is crucial for many industries. At the Chair of Public Administration and Public Policy, research projects use computational text analysis to answer questions about the Politics of Evaluation in International Organizations or the Influence of Speakers in the Debates of the UN Security Council.
Students attending this seminar in the past said:
“I am glad the seminar exposed me to quantitative text analysis methods which are increasingly relevant in everyday politics”.
“Without having taken this class, I would have never realized that coding is not so bad after all
“I am now considering applying text analysis in my Bachelors thesis”
Educational objective:
The seminar will provide students with the knowledge and skills necessary to critically evaluate the methods and impact of text analysis. The seminar is structured around three topics:
- The first part will familiarize students with research applications and two text corpora published by the Chair of Public Administration and Public Policy (One text corpus on speeches in the UN Security Council, and one text corpus on evaluation reports by international organizations). Students learn to access these corpora for potential use for their own projects.
- The second part provides an overview of various text as data research applications. These applications focus on a broad range of political science topics and are structured along various methods of text analysis.
- The third part is devoted to students’ own research project. Students are encouraged to apply own text-analysis projects in this seminar, but it is also possible to write seminar papers that, for example, engage critically with the impact of text analysis on politics and society.
It is possible to pass the class without knowledge of statistical software and without the conduct of an own text analysis. We will explore the use of modern analytical tools on the basis of large language models (such as Chat GPT). Students who also attend Advanced Methods class (12344) “Introduction to computer-based text analysis: Basic text analysis with R” (offered by Daniel Baumann) may conduct their own research project in this seminar.
Further information about the exams:
Oral contribution and presentation (40%); seminar poster or research paper (60%)
Mandatory literature:
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. doi:10.1093/pan/mps028
Grimmer, Justin, Margaret E. Roberts, and Brandon M. Stewart. 2022. Text as data: A new framework for machine learning and the social sciences. Princeton University Press.
Wenn Sie E-Learning Funktionalitäten nutzen möchten, tragen Sie bitte "Ja" ein.:
Ja
|
