This course explores the intersection of policy analysis and computational social science (CSS), and introduces the Python programming language as a tool for harnessing and making sense of the wealth of policy-related digital data available. Policy research examines how public policies are formed, how they are implemented, and how they change or remain stable. CSS is an interdisciplinary research paradigm that uses computational methods and new data sources to answer traditional social science research questions and to understand emerging phenomena, such as those arising from the political use of social media platforms.
Online and social media, public records and ‘open data’ provide a wealth of information about politics and policy. Computational methods and approaches can help to mine these sources and understand the information they contain. Key issues at the heart of debates in policy analysis are the ‘3Is’ of institutions, interests and ideas, and can be related to the field of CSS: How do different democratic systems affect the polarisation of political networks on social media? In what ways are interest groups and other policy actors connected to each other, and what are the possibilities for constructing policy networks out of (big) public data? How can topic modeling be used to analyse the media framing of policy issues?
We will discuss these use cases of policy research and CSS. Participants will learn the technical skills needed to process, visualise and analyse political data for policy analysis. They will be introduced to the functionalities of Python and explore its possibilities. Our methodological focus is on network analysis and text mining. The course also includes a short introduction to web scraping and APIs as means of data collection. Programming skills are not required, but a willingness to learn on your own, which must go beyond the mere consumption of content in the seminar sessions. Please bring your own laptop.
- Kursleiter/in: Erik Wolfes-Wenker