An analysis of replies to Trump's tweets

TL;DR

This study analyzes how Americans respond to and consume Donald Trump's tweets, revealing demographic and political patterns in replies and the influence of Twitter's reply curation on perception.

Contribution

It provides a detailed analysis of reply demographics and the role of algorithmic curation in shaping the perception of Trump's tweets, with comparative insights on Biden.

Findings

A small, specific demographic accounts for most replies.

Replies tend to reflect users' political leanings due to Twitter's curation.

Twitter's reply visibility is influenced by follow relationships, affecting perception.

Abstract

Donald Trump has tweeted thousands of times during his presidency. These public statements are an increasingly important way through which Trump communicates his political and personal views. A better understanding of the way the American public consumes and responds to these tweets is therefore critical. In the present work, we address both consumption of and response to Trump's tweets by studying replies to them on Twitter. With respect to response, we find that a small number of older, white, left-leaning, and female Americans are responsible for the vast majority of replies to Trump's tweets. These individuals also attend to a broader range of Trump's tweets than the rest of the individuals we study. With respect to consumption, we note that Trump's tweets are often viewed not in isolation, but rather in the context of a set of algorithmically-curated replies. These replies may therefore color the way Americans consume Trump's tweets. To this end, we find some evidence that Twitter accounts see replies in line with their political leanings. However, we show that this can be largely, although not entirely, attributed to the fact that Twitter is more likely to show replies by accounts a user follows. As a basis for comparison, all results for Trump are compared and contrasted with replies to Joe Biden's tweets.