URL 2 IRL: Social Media Followership and Electoral Success

• In Event: Research in Five (Undergraduate Students)

Abstract

Abstract
And what is in a graph? The field of political science has been transfixed by the abundance of readily available data presented by social media platforms like Facebook, Twitter, and Instagram.
From ordinary citizens to trained scholars, individuals from all backgrounds are currently struggling to elucidate, precisely, how social media trends and politics interact. One common misstep is to imply causality — that what Kirsten Gillibrand, Kamala Harris, or Ilhan Omar tweets has direct political consequences — with euphemistic bar charts tracking, simply, retweets or followership against polling numbers. While our intuition may be onto something, the problem of endogeneity looms large — there are simply too many factors which have not been accounted for.

To combat imprecision, we must get creative and devise new indices and modes of measurement (Cowen, 2019). Building off of previous scholarship on voter mobilization, political communication, and digital culture, I propose that social media followership plays a role in determining electoral outcomes through a combination of three specific dimensions: (1) political visibility (2) message transmission cost discounts and (3) informational subsidies (Hemphill & Shapiro, 327; Bode & Dalrymple, 312). In conjunction, these dimensions of influence are what endow social media platforms with an enormous disruptive ability. In recent years, we have seen a great surge in heterodox candidates — from those with unique ideologies to those with non-traditional backgrounds. Most noticeably, women like Alexandria Ocasio-Cortez and Mckayla Wilkes have taken full advantage of social media platforms in a way that history has never seen before.

Methodologically, my paper offers a novel contribution to the field of political communication. Little quantitative research has been done to determine if there is truly a causal relationship between the size of a politician’s social media followership and his or her electoral outcomes. Indeed, at first blush, it may seem like a kind of tautology to assert that candidates who amass legions of social media followers are, consequently, more politically successful; however, the threat of endogeneity looms large, particularly, when we consider that some politicians may enjoy a significant popularity which predates their social media presence.

This study helps clarify the true relationship between followership and electoral success by posing the following question: “Is there a causal relationship between social media followership and electoral outcomes?” To test this, I instrument for the effects of a political candidate’s Twitter followership on his or her vote share using join-date — the number of days since the account’s inception up to the election date. The instrument represents a source of exogenous variation in follower-count and, consequently, legitimizes my finding of a significant relationship between Twitter followership and electoral outcomes.

Notably, I find that the effect size is significant and, notably, that there is a high level of instrument validity. Indeed, followership is a statistically legitimate indicator of electoral outcomes and, for the first time, this assertion can be made with a true causal interpretation backing it.


References

Bode, Leticia, and Kajsa E. Dalrymple. "Politics in 140 characters or less: Campaign communication, network interaction, and political participation on Twitter." Journal of Political Marketing 15.4 (2016): 311-332.

Cowen, Tyler. Nothing Measured Nothing Gained: The first step toward solving some of society’s most intractable problems is to quantify them. Bloomberg. 2019.

Hemphill, Libby, and Matthew A. Shapiro. “Appealing to the Base or to the Moveable Middle? Incumbents’ Partisan Messaging before the 2016 U.S. Congressional Elections.” Journal of Information Technology and Politics 16.4 (2019): 1-18.

Author

• Jasmine Cui, Emory University