In a recent study published in the Journal of Quantitative Description: Digital Media, Jula Lühring, Hannah Metzler, Ruggero Lazzaroni, Apeksha Shetty, and Jana Lasser, explore the utility, limitations, and potential pitfalls of the NewsGuard database as a tool for studying misinformation and news trustworthiness, focusing on its temporal stability, completeness, and the added value of contextual assessments of political orientation and topical coverage.
Given recent discussions about NewsGuard’s alleged bias, particularly accusations from incoming Trump administration regulators and far-right Republicans in Congress about censoring conservative news sites, this paper provides a valuable and nuanced analysis that helps to better understand the situation.
The abstract:
Researchers need reliable and valid tools to identify cases of untrustworthy information when studying the spread of misinformation on digital platforms. A common approach is to assess the trustworthiness of sources rather than individual pieces of content. One of the most widely used and comprehensive databases for source trustworthiness ratings is provided by NewsGuard. Since creating the database in 2019, NewsGuard has continually added new sources and reassessed existing ones. While NewsGuard initially focused only on the US, the database has expanded to include sources from other countries. In addition to trustworthiness ratings, the NewsGuard database contains various contextual assessments of the sources, which are less often used in contemporary research on misinformation. In this work, we provide an analysis of the content of the NewsGuard database, focusing on the temporal stability and completeness of its ratings across countries, as well as the usefulness of information on political orientation and topics for misinformation studies. We find that trustworthiness ratings and source coverage have remained relatively stable since 2022, particularly for the US, France, Italy, Germany, and Canada, with US-based sources consistently scoring lower than those from other countries. Additional information on the political orientation and topics covered by sources is comprehensive and provides valuable assets for characterizing sources beyond trustworthiness. By evaluating the database over time and across countries, we identify potential pitfalls that compromise the validity of using NewsGuard as a tool for quantifying untrustworthy information, particularly if dichotomous "trustworthy"/"untrustworthy" labels are used. Lastly, we provide recommendations for digital media research on how to avoid these pitfalls and discuss appropriate use cases for the NewsGuard database and source-level approaches in general.
Find the full open-access paper here: https://doi.org/10.51685/jqd.2025.003
Cite article:
Lühring, J., Metzler, H., Lazzaroni, R., Shetty, A., & Lasser, J. (2025). Best practices for source-based research on misinformation and news trustworthiness using NewsGuard. Journal of Quantitative Description: Digital Media, 5. https://doi.org/10.51685/jqd.2025.003
