Human communication is increasingly recorded as digital text, which constitutes big data that can be
used to study numerous scientific and real-world problems. The goals of this course are to (i) provide
an introduction to quantitative methods designed to analyze text, (ii) give an overview over common
applications of these methods in economics and the social sciences, and (iii) illustrate the potential of
text-as-data methods to ask new research questions and find new answers to existing problems.

On completion of the course, students will be able to:

1. Demonstrate knowledge of prominent methods to manage and analyze (large) text, including
natural language processing, text classification, sentiment analysis, and topic modeling.
2. Recognize typical research applications using these methods in students’ field of study.

3. Carry out preliminary steps necessary to analyze text, such as creating a corpus; removing
white space, numbers, and stop words; word stemming; and lemmatization.
4. Use existing packages in a programming environment (e.g., Quanteda in R) to describe
patterns in text.
5. Write and execute scripts to implement data-as-text methods.
6. Visualize results using word clouds and appropriate diagrams.
7. Exploit online resources (e.g., Stack Overflow) to find solutions for common methodological
problems.

8. Select the optimal (mix of) method(s) for a given problem.
9. Draw adequate conclusions from text analyses.

The course provides an overview over the most common text-as-data methods as well as their typical
areas of application:
- Prerequisites (text import, creation of corpora, pre-processing, creation of document-term
matrices, lemmatization)
- Text statistics (e.g., frequency analysis, measures of readability, similarity indices)
- Generic and customized dictionaries
- Sentiment analysis
- Text classification using reference texts and supervised learning
- Topic modeling

The course uses lectures, seminars, tutorials.

The teaching is conducted in English.

Admitted to a doctoral program in Economics, Business Administration, or a related subject of a recognized business school or university. Knowledge in basic statistics and previous experience with R are recommended, but not required.

The course is graded Fail (U) or Pass (G).

? Active seminar participation in the form of discussing selected research papers (20%) fulfils
ILOs 1 and 2.
? A written assignment in the form of an R script fulfils ILOs 3-9.
The grades are “pass” or “fail”. All examination elements need to be completed for a “pass”.

A course evaluation will be conducted at the end of the course.

Connection to Research and Practice
The course helps its students to understand research designs in economics and related disciplines
(e.g., communications, finance, management, marketing, and political science) that rely on text-asdata
methods. In addition, students can directly apply these methods in their own research. The
methods can be used to answer research questions in JIBS’ focus areas, such as “What text
characteristics determine the success of a crowdfunding campaign?” (entrepreneurship), “Can social
media influencers affect the adoption of green technologies?” (renewal), and “What media slant do
newspaper owners provide?” (ownership). As these examples illustrate, the course draws on
contemporary problems in business and society, while providing a sound scientific basis of the
relevant methods.

Beattie, G. (2020). Advertising and media capture: The case of climate change. Journal of Public
Economics, 188.
Garz, M. (2020). Quantitative methods. In M. B. von Rimscha (ed.): Management and Economics of
Communication, pp. 109–128. De Gruyter, Boston, MA.
Garz, M., & Martin, G. (2020). Media Influence on Vote Choices: Unemployment News and
Incumbents’ Electoral Prospects. American Journal of Political Science, forthcoming.
Gentzkow, M., Kelly, B., & Taddy, M. (2019). Text as Data. Journal of Economic Literature, 57, 535–
574.
Gurun, U. G., & Butler, A. W. (2012). Don’t Believe the Hype: Local Media Slant, Local Advertising,
and Firm Value. Journal of Finance, 67, 561–598.
Jetter, M. (2017). The effect of media attention on terrorism. Journal of Public Economics, 153, 32–
48.
Jurafsky, D., & Martin, J.H. (2008). Speech and Language Processing, 2nd edition. Prentice Hall, Upper
Saddle River, NJ.
Lee, D., Hosanagar, K., & Nair, H. S. (2018). Advertising Content and Consumer Engagement on Social
Media: Evidence from Facebook. Management Science, 64, 5105–5131.
Taeuscher, K., Bouncken, R., & Pesch, R. (2020). Gaining Legitimacy by Being Different: Optimal
Distinctiveness in Crowdfunding Platforms. Academy of Management Journal, forthcoming.
All course literature will be provided at the beginning of the course.