IPSA World Congress

Last week, I attended the IPSA World Congress in Poznan, Poland. The conference was a massive gathering of political scientists bringing together more than 2.000 participants. These large scale events have the tendency to limit the topic specific exchange of ideas as well as networking opportunities due to their sheer size, but in this case the integration of the section "Electronic democracy" was exemplary, providing a number of panels brining together papers at the forefront of social media research in political science. 

On the first day, I was chairing a panel on methodologies of internet research. The panel featured excellent contributions which showed that political debates on social media take place in fluid and highly mediated sociotechnical environments which exacerbates the analysis of online political communication by social scientists. The paper of Simon Hegelich on the identification and impact of social botnets is particularly relevant showing that follower counts of, mentions of and attention to political actors online are heavily skewed by computer-generated activity. 

My take on this important issue is twofold: a) studies inferring from Twitter activity to political phenomena in the offline world are deeply affected by social bots and should try to correct for their effects since non-human activity obviously distorts the relationship between social media and offline metrics like surveys or election results. Since making predictions based on Twitter data is increasingly contested on conceptual and methodological grounds, researchers mostly restrict their analysis to political communication within the political communication spheres of Twitter or Facebook. b) in these studies, social bots are also relevant because they shape the political debates at hand by distorting recommender systems or by diverting attention to certain actors instigating the botnets. For instance, Hegelich shows that in the Ukrainian-Russian conflict, a social botnet pushed the contents of a nationalist Ukrainian faction while mixing non-political messages in between to make the propaganda less obvious. Yet when studying political communication within social media environments, social bots should be explicitly included in datasets because they are part of the mediated reality on social media platforms and shape the perception of political debates by users. This example in itself and other papers during the conference showed that social media research remains a vibrant research field with a lot of potential in political science but also with many remaining challenges.

My own conference paper on the Equalization hypothesis revisited (together with Wolf Schünemann and Stefan Steiger) concentrates on temporal patterns in policy debates on Twitter. In contrast to collective action events (e.g. Bennett/Segerberg 2012; Margetts et al. 2016) or election campaigns (Jungherr 2014), policy debates unfolding over longer time frames have not received much attention in social media research. Our paper examines patterns of relative actor importance during the course of events in two different policy fields: climate change, a globalized agenda with a traditionally high degree of participation by non-governmental actors and the recent regulatory debates on net neutrality which have mobilized transnational advocacy coalitions in several countries. We extracted tweets on both topics and then coded the 500 most central actors in both debates as a) individual activists, NGOs and citizen journalists or b) actors belonging to established groups like political actors (belonging to political parties or institutions) or legacy media. We then apply social network analysis to compare the actor constellations and temporal networks of both debates. The results show temporally fluctuating network structures which converge to elite actors during high attention periods in the policy process, as the traditional gatekeepers in media and politics still maintain central roles in policy networks. While this speaks against an equalization of political communication on social media, an additional analysis demonstrated the positive externalities of these hierarchical network structures in that the increased focus of audiences on gatekeepers serves an integrating function in policy networks on Twitter. The more hierarchical bigger network components during the identified high attention periods are more integrated and feature more ties between different network clusters and actor groups. Feedback on the paper is very welcome.

Next stop: APSA Annual Meeting in Philadelphia

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s