Opinion-based Homogeneity on YouTube

Combining Sentiment and Social Network Analysis

  • Daniel Röchert University of Duisburg-Essen
  • German Neubaum University of Duisburg-Essen
  • Björn Ross University of Duisburg-Essen
  • Florian Brachten University of Duisburg-Essen
  • Stefan Stieglitz University of Duisburg-Essen
Keywords: machine-learning, echo chamber, social network analysis, computational science


The growing complexity of political communication online goes along with increasing methodological challenges to process communication data properly in order to investigate public concerns such as the existence of echo chambers. To cover the full range of political diversity in online communication, we argue that it is necessary to focus on specific political issues. This study proposes an innovative combination of computational methods, including natural language processing and social network analysis, that serves as a model for future research on the evolution of opinion climates in online networks. Data were gathered on YouTube, enabling the assessment of users’ expressed opinions on two political issues. Results provided very limited evidence for the existence of opinion-based homogeneity on YouTube. This was true even when the whole network was divided into sub-networks. Findings are discussed in light of current computational communication research and the vigorous debate on echo chambers in online networks.

How to Cite
Röchert, D., Neubaum, G., Ross, B., Brachten, F., & Stieglitz, S. (2020). Opinion-based Homogeneity on YouTube. Computational Communication Research, 2(1), 81-108. Retrieved from https://computationalcommunication.org/ccr/article/view/15