ABSTRACT
US presidential candidates have been using campaign websites since 1996 and social-networking sites since 2004, but little research has compared the content of these candidates' online messages to the content of their messages on other channels. political communication scholars have analyzed campaign messages in speeches, debates, ads, and online content by categorizing their tone as either positive, negative, or mixed/contrast, and by categorizing their content as focused on either issues (policies) or image (character) traits. To access the tweets, this chapter utilizes the Twitter Data Collection and Analysis tool on the Data-driven Computational Social Science (DD-CSS) website to download the candidates' Twitter posts over the course of the primary campaign. Once the Twitter, debate, and speech transcripts were gathered and prepared, they were analyzed by DICTION® software, a computer-assisted text analysis (CATA) program that searches for more than 10,000 words, sorts them into 35 categories, and calculates five master variables.
