This initiative uses automated and human coding to identify negative and negative and hateful speech by prominent people in media and politics against groups. It uses a Python-based tool to search media transcripts (for media personalities) and Tweets (for politicians) and identifies potential units of analysis when negative words and groups (from two dictionaries) are in close proximity. Potential cases are then validated or rejected by trained human coders based on a codebook.

The study is based on an analysis of show transcripts from the top 10 media personalities/shows (on cable news, radio, and YouTube) on the right and left (20 shows total) and Tweets from approx. 1,000 leading US politicians. The study period is from one week before the Nov. 3, 2020, presidential election to one week after the Jan. 20, 2021, inauguration.