For best experience please turn on javascript and use a modern browser!

Growing numbers of people form and express their opinions on social media platforms. How should we study these debates? How can we detect meaningful patterns in massive amounts of data? This afternoon features two speakers who address these questions head-on, in the light of the 2016 Brexit referendum.

Detail Summary
Date 16 May 2019
Time 14:00 - 16:00
Location Spui 25
Twitter, social media aangeleverd door Odycceus
© Odycceus

How was public opinion formed on Twitter in the period leading up to the Brexit referendum? Marco T. Bastos (Media and Communication at City, University of London) presents his research on identifying ideological shifts during the Twitter debate of those momentous months. As politicans, voters, and news media shared their perspectives, automated Twitter bots - often unidentified as such - fuelled partisan sentiments .   

In  his study, Bastos relied on geographically-enriched Twitter data of more than half a million tweets in the UK’s 650 parliamentary constituencies and a deep-learning algorithm to detect nationalist, globalist, populist, and economist sentiments in tweets. This afternoon's event unpacks the Brexit classifier, the tests performed to train the algorithms, and its performance in identifying the four key ideological coordinates explored in the study. It concludes with a discussion of his findings that nationalist and economic concerns dominated the debate on Twitter and were only offset in the last days before the referendum – when a globalist upsurge brought the Twittersphere to a divide between nationalism and globalism. 

Rens Vliegenthart (Communication Science, University of Amsterdam) presents several computational methods to detect meaningful features of media articles, such as the (sub)topics and frames. Vliegenthart will discuss these approaches in relation to social scientific questions on the British immigration debate and relations between the Dutch parliament and media. 

A short panel discussion will follow on the broader theme of using social media data to study political debates and opinion dynamics. This discussion is moderated by Justus Uitermark (Sociology, University of Amsterdam). 

This evening is organized in collaboration with the H2020 research project ODYCCEUS (Opinion Dynamics and Conflict in European Space).

About the speakers

Marco T. Bastos is Senior Lecturer in Media and Communication at City, University of London. His research sits at the intersection of communication and computational social science and addresses the cross-effects between online and offline social networks. His work has appeared in leading journals in communication and social sciences, including Journal of Communication and New Media & Society, and has been covered by major media outlets such as the BBC, New York Times, Guardian, and Washington Post. Topics of Marco Bastos' publications include the diffusion of opinions and political user behavior on Twitter.

Rens Vliegenthart is Professor of Communication Science at the University of Amsterdam. His research into the relations between politics and media is at the intersection of communication science, political science, and sociology. Topics of Rens Vliegenthart's publications include framing and the role of news media in anti-immigration attitudes.

Justus Uitermark (moderator) is Associate Professor of Sociology at the University of Amsterdam. A relational sociologist with a background in human geography, Uitermark has a long-standing interest in the roles of cities as terrains of political struggle and more recently has been studying how social media platforms facilitate challenges against the status quo or breed conformity.

Logo Odycceus


You can sign up for this program for free. If you subscribe for the program we count on your presence. If you are unable to attend, please let us know via | T: +31 (0)20 525 8142.


Spui 25

Spui 25-27
1012 WX Amsterdam