New publication about Japanese candidates’ adoption of campaign media in the 2017 elections in Japan

Telematics and Informatics published our research about Japanese candidates’ adoption of online media services (website, Facebook, Twitter, weblog and YouTube) for campaigning purposes. Titled Individual and contextual determinants of adoption of online media services in the 2017 lower house election campaign in Japan, the study analyses patterns of adoption, using Multiple Correspondence Analysis and Mokken Analysis, and aims to identify what candidate characteristics and contextual characteristics explain not just single online media services but multiple online media services simultaneously.

I thank Leslie Tkach-Kawasaki and Junku Lee for their collaboration on this study. The study can be found here:

This study was made possible by a grant provided to Maurice Vergeer by the Japan Society for the Promotion of Science (JSPS) International Research Fellows Program (Grant ID: L17568).


  • Candidates adopted online media services in a consistent order;
  • Multiple Correspondence Analysis and Mokken analysis are useful scaling techniques to measure adoption of technology;
  • Individual and contextual characteristics both explain adoption of online media services;
  • Candidates on the PR list adopted more online media services than those in districts;
  • Incumbent candidates adopted more online media than new and former candidates;
  • Younger candidates adopt more online media for campaigning than older candidates;
  • In densely populated areas, candidates adopted more online media services for campaigning.

New publication about artificial intelligence in the Dutch press

My latest publication is published in Communication Studies, titled “Artificial Intelligence in the Dutch Press: An Analysis of Topics and Trends”.

The present study focuses on newspaper articles published in all
newspapers in the Netherlands from 2000 up to 2018 on how they
report on artificial intelligence. The study showed how reporting
changes over time and how different types of newspapers report
differently about various topics in the field of artificial intelligence.
The findings show that newspapers increasingly report about AI from
2014 onward. Newspapers do so on a wide range of topics related to
AI. Several types of newspapers showed distinct coverage of many
artificial intelligence-related topics. Although robots and their football
capabilities appear to be ever-present in newspapers, recent
articles about tech giants and fake news are more prominent in
newspapers. One of the most notable findings was that religious
newspapers published less about AI topics, even though some topics
on artificial intelligence have religious connotations (cf. singularity,
consciousness). Sentiments about AI in newspaper articles remained
balanced between positive and negative sentiments over the years.

Thank you to Seungahn Nah, Jang Hyun Kim, Jasmine McNealy, and Jungseock Joo who initiated the special issue “Communicating Artificial Intelligence: Theory, Research, and Practice” In communication Studies. The publication can be downloaded as Open Access using this link.

And now for something completely different: Football and Twitter

So, my studies mostly focus on political communication, journalism and ethnic minorities and migrants in the media. But some time ago, my former student, Leon Mulder, collected cool data on football players’ positions, their scoring capabilities and their basic track record. We then collected their Twitter data. The study attempts to answer the question of what explains online popularity: performance on the football pitch or performance on Twitter? To be honest, we were surprised nobody ever took on this question before. But here we are, a new publication:

Vergeer, M. & Mulder, L. (2019). Football players’ popularity on Twitter explained: Performance on the pitch or performance on Twitter? International Journal of Sport Communication.

The link to the article is Below is the abstract.


This study tested football players’ performance on the pitch against their performance on Twitter as explanations for Twitter popularity. Guided by network theory, social-identity theory, and basking in reflective glory and using data of all players of all teams in the Dutch premier league (“Eredivisie”), the multilevel models show that players with a Twitter account were more popular when they scored more goals, were non-Dutch, were on loan at another club, and were networkers actively following others on Twitter. The findings also show that context matters. Players under contract with a successful club receive an automatic bonus: Irrespective of their performance on the pitch or on Twitter, they automatically acquire more followers on Twitter. Players in general do not need to put a lot of effort into communicating on Twitter because sending tweets is unrelated to having more followers. Advertisers’ best options to reach larger and homogeneous audiences through football players are to choose attackers, scoring players, those out on loan, and foreign players, as well as players from successful teams in general. The study also identified which player characteristics do not add to a larger audience reach, such as tweeting behavior and experience on Twitter.