Geen categorie

New Project: Personalised Communication

personalisedMy institute, the Amsterdam School of Communication Research, is cooperating with the Institute for Information Law  in a project on Personalised Communication (project website). Broadly speaking, we are investigating the transformation from a mass media landscape to much more individualized communication. I am very excited to be involved in this project – and a nice side effect is that I have an office again in our old historical building in the center of Amsterdam (be it for one or two days per week only; most of the time, you will still find me on Roeterseiland).

Conference presentations

WAPOR_17-06-2015_177_bewerktSome weeks ago (yes, it’s been some weeks, I have been travelling South America afterwards…) I have been giving two talks at the WAPOR conference in Buenos Aires about my recent work on news sharing and on automated content analysis. The slides for both talks are available on slideshare: Packing and unpacking the Bag of Words: Introducing a toolkit for inductive automated frame analysis (co-authored by Jeroen Jonkman) and Viral news: How to predict news sharing based on article characteristics (co-authored by Petro Tolochko and Björn Burscher).

In this conference season, other work in which I was involved was presented at the ICA conference in San Juan, Puerto Rico, and the ECPR conference in Montreal, Candada. For an overview of all conference presentations in which I have been involved as author or co-author, please have a look at my publication list.

A man charges his phone using a wire dangling from a powerpole near a govenment building in the center of Asunción, Paraguay.

A man charges his phone tapping electricity from a loose wire dangling from a powerpole near a govenment building in the center of Asunción, Paraguay.

Student’s research, new style

I have reported here occassionally on what type of research students I supervised were conducting. Admittedly, I have not really kept up and there should be much more to report. Therefore, I am especially happy to announce that we have a new approach of sharing our student’s insights. Every student who writes a Bachelor thesis at the College of Communication at the University of Amsterdam now has to share his results on a project blog called ProjectCW. Feel free to have a look!


Talks, media, discussions

Within academia, it’s important to publish one’s research results in scientific journals to share one’s insight. But I also think that as academics, we have the responsibility to share our knowledge in other ways as well. The last weeks were particularly busy in this respect.

  • Quest Psychologie, a popular scientific journal, interviewed me for an article on news avoidance.
  • I gave a lecture and some hands-on demonstrations about news sharing on social media during the Weekend van de Wetenschap, the National Science Weekend in the Netherlands. Its aim is to invite a lay audience to visit our labs and talk with us about our work.
  • I met a lot of interesting people (journalists, policy makers, researchers) during a brainstorm session on the future of journalism. I was invited by the Stimuleringsfonds voor de Journalistiek, the Dutch Press Funds to join the discussion.
  • The Utrecht Data School asked me to give a workshop on using Python for data analysis in the social sciences, which I happily did.

Always good to see that other’s are interested in your work. And talking about work – now I’m going to work on another academic paper for a journal.


New publication on tweeting about TV debates

It’s always great news if a new publication is out – and especially if it goes so quickly as in this case. I submitted this paper for the first time on March, 3. (To those who are not familiar with publication cycles in the scientific world: It would not be too pessimistic to expect it to be published maybe beginning next year). But, I am very happy to announce that it is already out (and thanks to the Social Science Computer Review, the editor and the anonymous reviewers for a really good publication process) :

Trilling, D. (2014) Two different debates? Investigating the relationship between a political debate on TV and simultaneous comments on Twitter. Social Science Computer Review, Advance Online Publication. doi:10.1177/0894439314537886.

What did I do? I had a look at the TV debate (the so-called #tvduell) between the two main candidates during the German Elections 2013, Angela Merkel and Peer Steinbrück, and compared it to how people tweeted about the debate. I found some interesting relationships. As a candidate, mentioning a topic more frequently makes it more likely that you are mentioned together with this topic on Twitter. The only problem: You are mainly associated with negative aspects, or people make fun of you.   Have a look at the article for detailed findings.

How did I do this? I wrote a series of Python scripts (find them here) to process the textual data, count words, find word co-occurences, and produce files for further analysis in other programs.

What I learned from this semesters’ Bachelor students

This morning, I had the last meeting of this semester’s afstudeerproject, a course that all students whose BA thesis I supervise have to follow and in which we discuss everything related to their theses. I asked them to write in one sentence what the most important insight from their thesis research is. The list shows the impressive variety of interesting empirical studies the students conducted (in Dutch):

Het reageren van bedrijven op Facebook-posts is van invloed op de merkreputatie, ongeacht de aard van de Facebook-post.

Nederlandse populaire- en kwaliteitskranten schakelen naast journalisten ook andere actoren in om nieuws uit het buitenland te verschaffen op verschillende manieren.

Persoonlijkheidskenmerken blijken niet samen te hangen met het Facebookgebruik van mensen. Echter blijkt de demografische variabele leeftijd wel van invloed te zijn op het soort Facebookgebruik.

Lokale politici zijn terughoudend in het gebruik van negative campaigning en de voorspellers van dit gedrag verschillen van de voorspellers voor landelijke politici, zowel nationaal als internationaal.

Tijdens de gemeenteraadsverkiezingen in 2014 kregen vrouwen, politici hoog op de kandidatenlijst en politici met veel volgers op Twitter meer voorkeurstemmen.

De sociale impact theorie is in beperkte mate van toepassing op de interactie tussen gebruikers en organisaties op Facebook.

Promoted posts zijn effectiever dan banners: zij scoren hoger op advertentie herkenning, advertentie herinnering en advertentie attitude dan banners.

De mate waarin een blogger zijn persoonlijke merkbeleving in een artikel naar voren laat komen, heeft een indirecte invloed op merkattitude en koopintentie, met attitude tegenover de blog als mediator.

Iets minder dan een kwart van de twitterberichten van lokale Nederlandse politici een week voor de gemeenteraadsverkiezingen van 2014 is gepersonaliseerd, waarbij linkse partijen en stedelijkheid de meest belangrijke voorspellers bleken te zijn.

Seksestereotypen spelen sinds 1972 tot en met 2003 in variërende mate een rol in televisiecommercials, maar zijn altijd aanwezig geweest.

De social comparison van Festinger is niet van toepassing op Facebook, mensen vergelijken hun eigen leven niet met dat van anderen aan de hand van Facebookposts.

Het tonen van een opiniepeiling voor de verkiezingen heeft geen invloed op de stemintentie van kiezers in het Nederlandse partijenstelsel.

Facebookgebruikers gebruiken het liken van pagina’s meer als expressie van hun actual self, ten opzichte van de expressie van hun ideal self, onafhankelijk van persoonlijkheidsfactoren.