(iSchool) Associate Professor  presented her “Illuminating 2016” project recently at the  at the Columbia School of Journalism, where she is currently a fellow.
Jennifer Stromer-Galley
Stromer-Galley’s project is designed to help political journalists cover social media during the 2016 presidential campaign.
“Although Donald Trump seems to be getting a lot of love from journalists who cover what he is tweeting about, all of the presidential candidates are tweeting, posting, Snapchatting and Instagramming,” said Stromer-Galley. “Their social media not only drives what reporters say about the campaigns, but also mobilizes their supporters, generates needed cash and draws out opponents in extended tit-for-tats.”
While there are other projects that have begun to visualize the frequency of social media postings by presidential candidates, reporting the numbers of followers and changes in follower rates, for example, none are carefully examining what these candidates are actually saying.
Part of the reason for that, says Stromer-Galley, is that while it’s easy to count frequencies of already structured data, such as the number of followers a candidate has, it’s much harder to categorize, analyze and count what they are saying in the unstructured data of tweets and Facebook posts.
And that’s where Stromer-Galley’s “Illuminating 2016” project comes in. Her goal is to advance public understanding of what the presidential candidates are saying through their social media accounts. And the project is doing that by using state-of-the-art computational approaches for studying unstructured text.
“We’ve been collecting all of the announced major party candidates’ Twitter and Facebook posts since they declared their presidential bids,” explained Stromer-Galley. “In all, we have filled six servers with 24 presidential candidates’ social media messages, and of course we’re still collecting. We’ve developed categories for classifying the candidates’ messages, looking at their strategic messages that promote their policies or attack their opponents, their calls-to-action, their conversations with the public and their attempts to inform.”
Stomer-Galley and her team of iSchool researchers has trained computer models to categorize candidate messages along nine categories, and has so far achieved 70 percent accuracy with the classifications (by comparing how the algorithm performs when categorizing data where the “truth” has been established). Categories they are tracking include: attack, advocacy, image, issue, endorsement, calls-to-action, conversation, information and ceremonial.
The next step in the project’s progress is to learn more about what political journalists need to help them understand what the candidates are saying on social media.
Stromer-Galley has been conducting interviews with political journalists to learn how they can best make use of the data that have been collected to support their job of covering what has become a unique election year.
Next month, Stromer-Galley and her team will conduct user tests with journalists to see what they find useful and what they don’t based on their data and visualizations so far.
Stromer-Galley was  in February.
At the iSchool, her research focuses on human interaction with and through digital technologies. Her work explores why people talk politics online, what practical addition deliberation can bring to e-government and the development of a coding scheme to assess the qualities of political discussion.
Her award-winning book, “ (Oxford University Press), provides a history of presidential campaigns as they have adopted and adapted to emerging digital communication technologies.
In addition, Stromer-Galley serves as director for the iSchool’s , and is an affiliated faculty member with the Department of Political Science in the , and vice president of the .
