2026/01/08 by Anders Olof Larsson · 1 voice
Social Sciences · #Computational and Text Analysis Methods #Misinformation and Its Impacts #Social Media and Politics
paper · pdf · doi:10.5210/fm.v31i1.14448
openalex publication_date 2026/01/08 · openalex created_date 2026/01/09 · openalex updated_date 2026/05/21
This study investigates the dynamics of online political communication on Facebook, focusing on the Facebook posts made by the Norwegian Progress Party and its recent leaders, Siv Jensen and Sylvi Listhaug, over a decade-long period. Answering the call for longitudinal insights into political communication, we utilize a novel hybrid content analysis approach combining large language models (LLMs) and human classification, to assess the sentiment of Facebook posts — categorized as negative, positive, or neutral. Moreover, we assess the levels of engagement reached by posts featuring different sentiment. Our analysis reveals increased negative sentiment and engagement over time, particularly under Listhaug’s leadership, confirming the hypothesis that negative content drives higher engagement. Posts with negative sentiment consistently garnered more shares and comments, reflecting the strategic value of negativity in amplifying political messages. However, likes were more frequently associated with positive content, suggesting more nuanced engagement patterns. These findings contribute to our understanding the evolving landscape of digital politicking, highlighting the interplay between political actors’ communication strategies and audience engagement.