The second lecture in the lecture series Digital Media in Politics and Society is online. It is available here and wherever you get your podcasts.
Link to script:
Computational social science (CSS) is an interdisciplinary scientific field that studies human behavior and social systems using computational methods and research practices. This includes developing and testing theoretical assumptions but also the systematic description of the behavior of people, organizations, institutions and complex socio-technical systems. A defining characteristic of CSS is the close interdisciplinary cooperation between social sciences, computer science, and natural sciences. Different research fields complement each other in the investigation of social phenomena and processes through their different perspectives and core competencies. The aim of CSS is both to understand new social phenomena triggered by digitization and to develop new perspectives on traditional research interests in social science. Both endeavors are significantly shaped by the use of new data sets and analytical methods made available through digital technology.
In this episode, we will be laying some foundations for using CSS to better understand the impact of digital media on politics and society. We will discuss what the term computational social science means, what makes it different from other approaches in the social sciences, and try our hands at a definition.
00:00:00 – Introduction
00:02:31 – The promise of computational social science
00:12:27 – Computational social science: A definition
00:21:20 – Conclusion
While CSS projects come in a stunning variety of data sets used, methods employed, and questions asked, more often than not, these projects share a pipeline of tasks, problems, and decisions that is typical for CSS. Examining this pipeline allows us to think about engaging in CSS as a practice, while at the same time providing you with a blueprint for potential research projects that might lie in your future.
00:00 – Introduction
01:25 – Research design
05:38 – Data collection
08:41 – Data preparation
10:00 – Linking signals in data to phenomena of interest
12:30 – Data analysis
16:33 – Presentation
21:33 – Conclusion
Text analysis is a prominently used approach from computational social science. In this episode, we examine three recent studies closely, that are using text analysis in interesting and constructive ways. Text is a rich medium reflecting cultural and political concepts, ideas, agendas, tactics, events, and power structures of the time. Large text corpora open windows and allow comparisons across countries, cultures, and time. Different types of text contain representations of different slices of culture, politics, and social life. They therefor are of interest to various subfields in the social sciences and humanities. By collecting and preparing for analysis large text corpora scholars can access and make available vast troves of knowledge on various questions and in different subfields. The tremendous collective efforts in digitizing and making available text corpora are a massive accelerating factor in this effort. Let’s have a closer look.
00:00 – Introduction
00:22 – Text analysis in political science
05:43 – Making sense of party competition during the 2015 refugee crisis with a bag of words
11:51 – Who lives in the past, the present, or the future? A supervised learning approach
15:48 – Political innovation in the French Revolution
22:15 – Conclusion
In computational social science there are great hopes and enthusiasms connected with the availability of new data sources. In this episode, we will be talking about working with one of these new data sources: digital trace data.
Once people interact with digital devices (such as smart phones and smart devices) and services (such as Facebook or Twitter), their digitally mediated interactions leave traces on devices and services. Some of those are discarded, some are stored. Some are available only to the device maker or service provider, some are available to researchers. This last category of digital trace data, those that are stored and available to researchers, has spawned a lot of research activity and enthusiasm over a new measurement revolution in the social sciences. But somewhat more than ten years into this “revolution”, the limits of digital trace data for social science research are becoming just as clear as their promises. Before we look at studies using digital trace data, it is therefore necessary that we look a little more closely at what they are, what characteristics they share, and how this impacts scientific work with them.
00:00 – Introduction
00:18 – Digital trace data
12:11 – Digital trace data in political science
15:42 – Making sense of online censorship decisions
21:24 – It’s attention, not support!
30:03 – Learning about the world with computational social science
33:49 – Conclusion
Link to podcast site:
Link to YouTube Channel: