Debate Structures Around COVID-19 on Twitter (A)

Analysing the Tweets of German politicians concerning COVID-19 on Twitter in regards to Twitter-inherent measures as well as NLP- and SNA-related methods.

Members: Jonas Puchinger, Theresa Strohmeier, Sabine Roth

Keywords: COVID-19, social media, Twitter, politics

Description

Many politicians use Twitter as a tool for active and direct communication. The combination of Twitter's simplicity and popularity make it an effective tool for politicians to communicate their ideas to the general public, with the added possibility of direct feedback. Especially during a public crisis, politicians are the center of attention. Insights into the communication of German politicians on Twitter regarding COVID-19 can provide meaningful knowledge. Also differences between the political parties become clear. Therefore, we analyse the behaviour of German politicians on Twitter through quantitative and qualitative content-based methods such as sentiment analysis and topic detection, as well as social network analysis methods. All these tools are used with the distinct features and characteristics of Twitter in mind. This combination of approaches helps us to create a more holistic picture of the way politicians use Twitter. We show how to identify tweets regarding COVID-19, how to analyse their content, compare them to non-COVID-tweets, and how to contextualise these findings through real world data.

Goals

Finding differences and similarities in sentiments, emotions, contents and network structures between German politicians in COVID-19 related discussions on Twitter. Also differences between the political parties become clear. These findings may help inform politicians on how to effectively use Twitter, and the general public on how to better understand government communication on social media, especially in combination with real world data.

Updates

Final Reflection (2021-03-26)

Summary of the work on our project (more...)


Study results (2021-03-10)

Presentation of the results of the sentiment analyses, content analyses and network analyses (more...)


Analysis methods (2021-02-08)

Description of the methods for the analyses (more...)


Data update and extension (2021-02-01)

Updating the data set by collecting the twitter data 3 weeks after the last relevant date and adding the followers connections (more...)


Initial Data Collection (2021-01-13)

A short overview of the initial scraping process, in which the tweets of 555 german politicians over the year 2020 were collected. (more...)


Current state (2020-12-18)

Update of the project status (more...)


Roadmap (2020-11-28)

Defining the research questions, the methods we want to use and the time schedule. (more...)


Definitions (2020-11-24)

Working definitions of keywords and concepts relevant to the project (more...)


Initial literature research (2020-11-17)

A short overview of the research process and first findings of the initial literature research, as well as the current state of the research we found. (more...)


About this project (2020-11-16)

Outlining the scope of the project, discussing necessary steps before kick-off, and listing team members and their motivations. (more...)


Further Resources