2025 Theses Doctoral

The Modern Media Playbook: Elite Strategies for Digital Influence

Burson, Manu Singh

The first paper in my dissertation examines how social media engagement metrics influence perceptions of electoral success for U.S. congressional candidates. Through two online survey experiments with 800 and 600 participants respectively, the research demonstrates that candidates with higher social media engagement receive significantly higher predictions of electoral success, with ratings increasing 9-13.5% compared to control conditions. The effect is most pronounced among inattentive respondents and shows notable partisan differences, with Democratic voters displaying stronger susceptibility to social media signals than Republicans. Interestingly, neither digital literacy levels nor awareness of automated activity on social media platforms moderated these effects. The study also reveals that politically knowledgeable voters, rather than being immune to social media metrics, systematically incorporate such information into their evaluation process. These findings have significant implications for understanding how social media influences modern political decision-making, particularly in low-information environments such as primary elections, and suggest the emergence of a new heuristic in the digital age of political communication.

In the following paper, I examine how fake social media accounts boost politicians' online popularity and this phenomenon's subsequent spillover on traditional news coverage. Using the 'Botometer' algorithm, I assessed the proportion of bot accounts engaging with tweets from 382 U.S. Congress members on Twitter. A policy change to Twitter's API infrastructure in November 2022 was an exogenous shock to the platform that significantly hampered bot functionality. My first-stage analysis demonstrated that this policy change only affected high-bot-engagement politicians, who saw a substantial decline in followers after November 2022. Placebo comparisons show that this decline was not observed in comparable data from Facebook 'likes' or Instagram followers. My second-stage analysis revealed that, following the policy change, high-bot-engagement politicians also experienced a decline in coverage in digital news articles and TV news from December 2022 to February 2024.

In the third paper, I examine how media coverage, both in terms of volume and sentiment, influences pricing and trading behavior in political prediction markets. Drawing on daily candidate-level data from PredictIt and sentiment-scored news coverage, I analyze 39 betting markets, looking at 78 U.S. political candidates during the 2022 election cycle. Using transformer-based sentiment classification, I find that positive and negative media mentions significantly increase prediction market prices and trade volumes, though negative sentiment often exerts a stronger effect. The relationship is nonlinear, with evidence of diminishing returns from ``attention saturation'' where excessive media coverage yields weaker or even negative marginal effects. A pooled event study design also reveals that extreme sentiment intensity shock days drive sustained increases in trading volume, but only negligible price shifts. The effects of media vary by party, candidate profile, and race competitiveness: Republican candidates, challengers, and those in tightly contested races show heightened sensitivity to negative coverage. Notably, Republican candidates' stronger responses to negative coverage suggest coordinated negative media campaigns could artificially inflate their market prices, potentially influencing resource allocation by donors and political organizations who monitor these market signals for strategic decision-making.