Why partisans feel hated: Distinct static and dynamic relationships with animosity meta-perceptions

Jeffrey Lees*, Mina Cikara, James N. Druckman

*Corresponding author for this work

    Research output: Contribution to journalArticleAcademicpeer-review

    1 Citation (Scopus)
    92 Downloads (Pure)

    Abstract

    Partisans hold inaccurate perceptions of the other side. What drives these inaccuracies? We address this question with a focus on partisan animosity meta-perceptions (i.e. how much a partisan believes opposing partisans hate them). We argue that predictors can relate to meta-perceptions statically (e.g. at a specific point in time, do partisans who post more about politics on social media differ in their meta-perceptions relative to partisans who post less?) or dynamically (e.g. does a partisan who increases their social media political posting between two defined time points change their meta-perceptions accordingly?). Using panel data from the 2020 US presidential election, we find variables display distinct static and dynamic relationships with meta-perceptions. Notably, between individuals, posting online exhibits no (static) relationship with meta-perceptions, while within individuals, those who increased their postings over time (dynamically) became more accurate. The results make clear that overly general statements about metaperceptions and their predictors, including social media activity, are bound to be wrong. How meta-perceptions relate to other factors often depends on contextual circumstances at a given time.

    Original languageEnglish
    Article numberpgae324
    Number of pages9
    JournalPNAS Nexus
    Volume3
    Issue number10
    DOIs
    Publication statusPublished - Oct-2024

    Keywords

    • elections
    • intergroup conflict
    • meta-perceptions
    • polarization
    • social media

    Fingerprint

    Dive into the research topics of 'Why partisans feel hated: Distinct static and dynamic relationships with animosity meta-perceptions'. Together they form a unique fingerprint.

    Cite this