Welcome
This is a Quarto website for the “Forschungsseminar CSS” course at Leipzig University. It covers different techniques for the aspiring computational social scientist, hence we have dubbed it “Toolbox CSS.” The site and original course materials were developed by Felix Lennert, whose work you can explore on his website. We, Sascha Grehl and Wiebke Krügener, have now taken over the course and maintain and adapt these materials. Many thanks to Felix for generously sharing his work and supporting the transition. For questions about the course or materials, please contact Sascha!
As Felix’s example shows, putting time and effort into CSS can open doors to exciting opportunities:
| Then | Now |
|---|---|
 |
 |
| In the field, harvesting some fresh raw material before transforming it into insights (© Alexandra Gibbons) | From harvesting raw data to cultivating playlists: Thanks for helping us grow, Felix—and all the best for your next chapter at Spotify! |
The Course
Here’s the official description:
The Forschungsseminar in Computational Social Science (CSS) equips you with the tools to analyze human behavior, predict social trends, and tackle complex societal issues using cutting-edge data science techniques. From web scraping to AI-powered text analysis, you’ll learn to use your computer in new ways to gain insights into social phenomena.
Hörsaal 21
The curriculum covers a range of topics including data management, web scraping, speech-to-text, and computational text analysis. Students will hone their R and develop skills in Python, applying these languages to real-world social science problems. The course progresses from fundamental concepts to advanced techniques, including the use of state-of-the-art AI models for text analysis.
The course structure consists of one lecture and one lab session per week, providing a balance of theoretical knowledge and practical application. Throughout the semester, students will benefit from hands-on coding exercises, one-on-one mentoring, and collaborative projects. The course culminates in a research paper, allowing students to apply their new skills to a topic of their choice.
Course Structure
The course consists of lectures introducing each week’s content and a course script that provides hands-on coding examples for the content. It is mostly containing R with some Python mixed in for good measure when no great R alternatives exist (e.g., for web scraping with Selenium, text classification with transformer models).
At the beginning of the course, students are encouraged to form groups based on research interests and general vibes. I require each student group to check in with me at the beginning of each week to report their progress (even if there’s nothing to report – no progress, no problem). This does not count towards any grade but rather serves the purpose of me receiving feedback on the learning experience (this is a new course!) – and will hopefully help me with providing more appropriate guidance.
An overview of the topics covered can either be found in the syllabus or in the lectures overview.
Syllabus
Please click here to download the latest version of the syllabus.
Alternatively, read it here (probably not the best if you’re visiting this page on a mobile device):