Partisan bias describes systematic differences in how individuals from opposing political parties consume, evaluate, and share news. This includes both supply-side effects (asymmetries in fake news production favoring one party) and demand-side effects (users' selective exposure to ideologically congruent content and rejection of opposing-party sources).
Key observations
Asymmetric fake news production: In the 2016 U.S. election, pro-Trump fake news articles vastly outnumbered pro-Clinton articles (115 vs. 41), and pro-Trump content received substantially more engagement on social media. This supply-side asymmetry reflects both platform affordances and partisan motivation.
Asymmetric sharing behavior: Osmundsen et al. (2021) document that Republicans are significantly more likely than Democrats to share from sources with a history of publishing false claims, with strong Republicans 15+ percentage points more likely to share pro-Republican fake news compared to strong Democrats' sharing of pro-Democratic fake news (when such sources exist).
Ideological selectivity: Users selectively share news that portrays the opposing party negatively, a pattern observed across both fake and real news sources. Pro-Republican fake news sources tend to focus on attacks against Democrats, while Democratic users preferentially share real news sources that attack Republicans—suggesting partisan goals, not accuracy, drive sharing decisions.
Key papers
- 2017-allcott-social-media-fake-news — Documents partisan asymmetry in fake news production (pro-Trump articles 2.8× pro-Clinton articles) and engagement in 2016 election
- 2021-osmundsen-partisan-polarization-fake-news — Shows strong party identifiers selectively share ideologically congruent false content; Republicans more likely than Democrats to share from fake news sources; partisan affect (especially out-party hostility) predicts sharing better than ignorance or cognitive factors