You do not currently have permission to view the full text for this document. Please sign in to an account with an active subscription or contact support.
In addition to the political controversies that they ignite, the contents of presidential tweets are subject to diverse statistical analyses as evidenced by postings on the internet. As a form of do- it-yourself proof, empirically minded tweet investigators offer downloadable data sets and computer code. A public archive (trumptwitterarchive.com) containing tens of thousands of time-stamped Trump Tweets (TTs) is available. There, an obsessive bot monitoring the presidential Twitter account updates the archive minute by minute.
As a result of linguistic and sensitivity analyses, we know that in terms of emotional content, TTs are most frequently classified as “negative” or as “attacks.” Statistically, Trump's attacks on people and institutions most often focus on: 1) weakness (“lightweight,” “loser,” “poor,” “pathetic,” “weak”), 2) stupidity (“incompetent,” “moron,” “clueless”), 3) failures (“failing,” “failed,” “disaster”), 4) illegitimacy (“fake,” “false,” “hoax,” “witch hunt,” “unfair,” “discredited”), and 5) corruption (“crooked,” “liar”, “dishonest,” “illegal”). His all-time favorite word is “great” (some studies find that he uses “great” more often than his next several favorite words combined). His favorite insult is “fake.” His favorite pronoun, “I.”
Some investigators suggest that there are two distinct sources of TTs. The first group, a small minority of all TTs, are believed to be authored and posted by staff on behalf of the president. These tweets are noteworthy for their use of complete sentences, appropriate capitalization, minimum exclamation points, and logical coherence. They are frequently sent from an iPad or iPhone.
The second and major source, believed to be directly authored and posted by President Trump, typically come from an Android device. These tweets are characterized by incomplete sentences, recurrent single words or short phrases (e.g., “sad,” “bad,” “witch hunt,” “fake news,” “no collusion”