A Framework for Analyzing Social Media Influence on Public Opinion Formation

Thesis Abstract

The rapid proliferation of social media platforms has significantly transformed the landscape of public opinion formation, prompting concerns about the mechanisms through which social media influences collective perceptions, attitudes, and behaviors. This study addresses the pressing need to develop a comprehensive analytical framework that explicates the dynamics of social media's role in shaping public opinion, particularly in the context of political discourse, social movements, and public health campaigns. The primary aim is to construct and empirically validate a theoretical model that delineates the key factors and interactions underpinning social media influence on public opinion processes. Specific objectives include identifying the core components of social media influence, examining the interrelationships among message characteristics, user engagement, and opinion change, and evaluating the applicability of existing communication theories, notably the Social Cognitive Theory and the Framing Theory, within this context. Employing a mixed-methods research design, the study integrates quantitative and qualitative approaches to provide a nuanced understanding of social media influence mechanisms. The quantitative component involves administering a structured survey to a representative sample of 750 active social media users across urban and rural regions within the country, recruited through stratified random sampling to ensure demographic diversity. The survey instrument is designed to measure variables such as message exposure, perceived credibility, user engagement levels, and attitude shifts regarding selected social issues. The qualitative component comprises in-depth interviews with 20 key informants including social media influencers, political communicators, and media scholars, aimed at exploring contextual insights and heuristics that underpin observed phenomena. Data collection instruments are subjected to rigorous validity and reliability testing, including Cronbach’s alpha assessments and pilot studies, to ensure robustness. Data analysis employs multiple regression analysis and Structural Equation Modeling (SEM) to test hypothesized relationships among constructs and to validate the proposed theoretical framework. Thematic analysis is utilized to interpret qualitative data, complementing the quantitative findings with contextual richness. The expected outcomes include identifying salient factors such as message framing, source credibility, and peer influence that significantly impact the degree of opinion change attributable to social media exposure. The model aims to demonstrate that interactive features, narrative framing, and perceived authenticity serve as mediators in the influence process, while demographic variables moderate effect sizes. The findings are anticipated to reveal variances in influence effectiveness across different social groups and issue types, underscoring the importance of context-sensitive communication strategies. This research makes a substantive contribution to knowledge by offering an integrated framework that encapsulates psychological, communicative, and technological factors influencing public opinion in the social media era. It advances theoretical understanding by operationalizing and empirically testing the applicability of established theories in a novel digital context, providing a basis for future model refinement and scholarly inquiry. The main conclusion emphasizes the multifaceted nature of social media influence, highlighting that user engagement and message framing are critical levers for opinion formation. Based on these insights, the study recommends strategic communication practices aimed at fostering transparency, authenticity, and user participation to enhance constructive public engagement. Moreover, it advocates for policy interventions that support media literacy and critical engagement to mitigate misinformation and manipulation. Future research directions include longitudinal studies to track influence trajectories over time and explorations of emerging social platforms to extend the model’s generalizability across digital contexts.