Show simple item record

dc.rights.licenseCC-BY-NC-ND
dc.contributor.advisorSchaefer, Mirko
dc.contributor.authorHielkema, Sascha
dc.date.accessioned2021-11-09T00:00:28Z
dc.date.available2021-11-09T00:00:28Z
dc.date.issued2021
dc.identifier.urihttps://studenttheses.uu.nl/handle/20.500.12932/198
dc.description.abstractThe rise of the digital world led to a shift from Public Service Broadcasters towards Public Service Media, where content is offered both offline and online. The Dutch Public Broadcaster NPO aims to produce content that has an impact on a diverse audience by making them feel connected to the society around them, increase their knowledge and touch their feelings. Therefore an impact score has been created by the NPO. On the other hand, social media websites became an important new part of the viewing experience of citizens. Now that viewers can express their opinions live during episodes, more information about the experiences of viewers can be measured and this could potentially be used as a metric to evaluate the content of television programmes. To gain more knowledge about the relation between online audience engagement and the impact of PSM television programmes, this research has been conducted. Impact scores of talk show episodes have been compared to the associated user engagement on Twitter, live during three NPO talk shows. Natural Language Processing tasks, such as sentiment analysis and text classification extracted sentiments and meanings from live posted tweets. These characteristics were added to a linear mixed model in which the impact score was predicted and talk show titles were added as mixed-effects. Results showed that the number of tweets and the percentage of negative tweets was related to a lower impact score, while the percentage of tweets about content indicates a higher impact score. The mixed-effects did explain some of the variance in the impact scores. Future research is needed to gain more knowledge about the relation between audience engagement and the impact of PSM content on individuals.
dc.description.sponsorshipUtrecht University
dc.language.isoEN
dc.subjectThis research is focused on the impact of Public Broadcaster talk shows from the NPO and the associated viewer engagement online. The goal was to discover whether audience engagement online could be used to evaluate the impact of talk show episodes on humans. Using natural language processing tasks, meaning from tweets posted by viewers during three talk shows were extracted. The tweet characteristics were added to a linear mixed model to find their relation to the impact scores.
dc.titleThe Relation between the Impact of Public Service Broadcaster’s Talk Shows and Social Media Engagement
dc.type.contentMaster Thesis
dc.rights.accessrightsOpen Access
dc.subject.keywordsNatural Language Processing; text classification; BERTje; RobBERT; logistic regression; Linear Mixed Model; Public Service Broadcaster; Public Service Media; Social media engagement; Twitter
dc.subject.courseuuApplied Data Science
dc.thesis.id806


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record