Chair: Hans-Christoph Hobohm
Omission of information: Identifying political slant via an analysis of co-occurring entities
Jonas Ehrhardt1, Timo Spinde2, Ali Vardasbi1, Felix Hamborg2
Abstract
Due to the strong impact the news has on society, the detection and analysis of bias within the media are important topics. Most approaches to bias detection focus on linguistic forms of bias or the evaluation and tracing of sources. In this paper, we present an approach that analyzes co-occurrences of entities across articles of different news outlets to indicate a strong but difficult to detect form of bias: bias by omission of information. Specifically, we present and evaluate different methods of identifying entity co-occurrences and then use the best performing method, reference entity detection, to analyze the coverage of nine major US news outlets over one year. We set a low performing but transparent baseline, which is able to identify a news outlet’s affiliation towards a political orientation. Our approach employing reference entity selection, i.e., analyzing how often one entity co-occurs with others across a set of documents, yields an F1-score of F1=0.51 compared to F1=0.20 of the TF-IDF baseline.
Keywords: media bias; bias by omission; news articles; co-occurrences
Short Paper
Does the General Public Share Research on Twitter?
A Case Study on the Online Conversation about the Search for a Nuclear Repository in Germany
Steffen Lemke1, Paula Bräuer2, Isabella Peters2
1ZBW – Leibniz Information Centre for Economics, Germany; 2ZBW – Leibniz Information Centre for Economics & Kiel University, Germany
Abstract
The search for a final nuclear repository in Germany poses a societal and political issue of high national medial presence and controversy. The German Repository Site Selection Act demands the search to be a “participatory, science-based […] process”. Also, the repository search combines numerous scientific aspects (e.g., geological analyses, technical requirements) with broad societal implications. For these reasons it constitutes a promising background to analyze the general public’s habits regarding referencing research on Twitter. We collected tweets associated with the conversation around the German nuclear repository search based on keywords. Subsamples of the resulting tweet set are coded regarding sending users’ professional roles and types of hyperlinked content. We found the most vocal group participating in the conversation to be activists and initiatives, while journalists constituted the follower-wise most influential accounts in the sample. Regarding references to scientific content, we found only very few cases of direct links to scholarly publications; however, several kinds of indirect references to academic findings could be identified, e.g., links to paraphrases of studies in news articles or blog posts. Our results indicate participation from a fairly diverse set of users in the observed communication around the German repository search; exchanges of research findings however appear to have happened rarely and been limited to very few particular studies. The findings also illustrate a central problem regarding the expressive power of social media-based altmetrics, namely that a large share of signals indicating a scholarly work’s influence will not be found by searching for explicit identifiers.
Keywords: Twitter; research references; scholarly use of social media; scholarly communication; nuclear repository search; altmetrics
Long Paper
Open practices of early career researchers
A qualitative study on research and teaching behavior
Tamara Heck1, Ina Blümel2
1Information Center for Education DIPF | Leibniz Institute for Research and Information in Education, Germany; 2German National Library of Science and Technology | Technische Informationsbibliothek (TIB) & Hochschule Hannover | University of Applied Sciences and Arts
Abstract
Many researchers have a positive attitude towards open science and are motivated to apply them. However, applying them requires a change in one’s daily practices. Different factors might challenge a behavioral change. The introduced study wants to get deeper insights into the reasons and influences that lead early career researchers to apply open practices in their daily research and teaching work. The participatory design let ten participants choose open practices they wanted to learn and adapt in either research or teaching scenarios. The study accompanied them and collected their positive and challenging experiences via diverse methods like interviews, diary entries and workshops. This paper introduces the study design and preliminary results.
Keywords: open science practices, early career researchers, research behavior, user study
Short Paper
Information Behavior towards False Information and “Fake News” on Facebook
The Influence of Gender, User Type and Trust in Social Media
Thomas Schmidt1, Elisabeth Salomon2, David Elsweiler2, Christian Wolff1
1Media Informatics Group, University of Regensburg, Germany; 2Information Science, University of Regensburg, Germany
Abstract
In this paper, we present a survey study with 119 participants conducted in German, which investigates respondents’ Facebook behavior. In particular, the survey provides insight into how the individual factors gender, user type and trust in social media influence information behavior with respect to false information on Facebook. Our participants’ Facebook use is predominantly passive, the trust in social media is mediocre and most users claim to encounter false information on a weekly basis. If the truthfulness of information is verified it is mostly done by checking alternative sources and for the most part, users do not react actively to false information on Facebook. Of the different categories of Facebook users studied, more active and intensive users of Facebook (posters and heavy users) encounter false information the most. These users are the only user group to report posts with false information to Facebook or interact with the post. Participants with higher trust in social media tend to check the comments of a post to verify information.
Keywords: false information, fake news, social media, gender, user type, trust in social media
Long Paper
Event Timeslots (1)
Track 1
-