Oliver Rittmann, Mannheim Center for European Social Sciences (MZES): Attention during parliamentary debates (January 2025)

Oliver Rittmann is a Lorenz-von-Stein Research Fellow at the Mannheim Center for European Social Sciences (MZES). In his research, he uses computational methods for the analysis of audio, image, and video data to answer political science questions in the areas of political representation, inequality, and communication.

What is your current research topic?

With an interdisciplinary team of computer and political scientists, I ask whether women in parliaments receive less attention for their parliamentary speeches than men. To answer this question, we analyze video recordings from TV cameras installed in the Landtag Baden-Württemberg, a large state-level parliament in Germany. The recordings from these cameras capture members of parliament while following parliamentary debates. We use computer vision techniques to automatically recognize whether and for how long MPs are attentive during their colleagues' speeches. Our results show that, on average, female MPs receive less attention for their speeches than their male colleagues. This difference in attention can be entirely attributed to the men in the audience, who pay less attention to women's speeches. In contrast, women are equally attentive to the speeches by men and women.

For those who have not yet delved deeply into the topic of Data Science: How would you explain to a child what you are working on?

Imagine that at school, the boys are always listened to more than the girls. That would be unfair. The girls have already complained about it, but they can't prove it. It's similar in parliaments, a place where politicians discuss and decide on important issues. Unlike at school, however, there are cameras that show exactly whether someone is paying attention to others. In our research project, we use computers to analyze the recordings from these cameras to find out whether women in parliament are listened to less than men.

Everyone talks about Data Science – how would you describe the importance of the topic for yourself in three words?

Exciting, challenging, valuable

What points of contact with Data Science does your work have? Which methods do you already use, and which would be interesting for you in the future?

As an empirical and quantitative political scientist, data science is an integral part of my work. The points of contact are diverse and range from classic data analysis to the application of complex AI models to analyze videos. For me, new data science methods become interesting when they can help me answer social science questions.

How high is the value of Data Science for your work? Would your research even be possible without Data Science?

My work might be possible without data science, but it would be entirely different from what I do now. When I deal with parliamentary debates, I analyze thousands of speeches and try to identify patterns. This would be impossible without data science methods. Instead, I would have to focus on a few selected speeches. Such work can also lead to interesting and important results, but the knowledge gained would be fundamentally different.

What development opportunities do you see for the topic of Data Science in relation to your field?

Although audiovisual impressions significantly shape social experience, political science has paid little attention to audiovisual data in the past. This is because such data was considered too complex and unstructured, and the field lacked analytical methods to make sense of it. Data science has the potential to provide methods for the quantitative analysis of audiovisual data and, thus, make audiovisual data accessible to political science research. As a result, data science expands the field of questions we can effectively answer in political science. There has been a very similar development concerning text data in the past. Like video and audio data, text has long been considered too unstructured to be a valuable data source. But thanks to data science research, quantitative text analysis is now an established field in political science methodology.

Back