Anyone can explore patterns in Twitter content using two readily available tools: Mozdeh, a free PC application for big data text analysis, and Microsoft Excel.

This video shows how to set up Mozdeh to capture @RealDonaldTrump tweets, import the captured tweets into Excel, and use Excel to search, filter and categorize the tweets.

Meanwhile, this video demonstrates more advanced analytical techniques using tweets about Tide Pods captured during Superbowl LII in February of 2018.

An Excel data file and .pdf-formatted PowerPoint are available to accompany the second video.

A cautionary note: There are many reasons to avoid assuming that patterns you see in Twitter content reflect actual public opinion. One such reason is that Twitter can be chock full of content produced by “bots,” or automated accounts that pose as actual human Twitter users. The Pew Center recently estimated that bots produce two-thirds of tweeted links to popular websites. One way to find content likely generated by a bot is to aggregate your data by source, then look for sources that are posting with unrealistic frequency, like a tweet every minute over the course of an hour or more. A more sophisticated approach would be to take a random sample of your tweets and run each tweet’s source through the Botometer. The Botometer also offers an API in both free and paid-access versions.