Big Data

Election Tech: The animosity index - why the internet loves and hates Donald Trump

Donald Trump is the Republican presidential nominee. Through the entire primary process, Trump has trounced the competition on social media. Big data analysis tells the story of his success.

candidatestwitterlogzio.png
Mentions of presidential candidates on social media in the days leading up to the Indiana primary. | Image: Logz.io

Donald Trump is the nominee of the Republican party. Trump clobbered the competition at the polls in Indiana, and on social media all primary season. He survived #marcomentum, a #ContestedConvention blip, and the vaporware #stoptrump movement.

In the week leading up the the Indiana primary TechRepublic collaborated with Israeli firm Logz.io to scrape and analyize social media data about the Indiana primary. Logz.io works primarily with corporate IT and uses machine-generated data to find software bugs. Founded by enterprise data veterans Tomer Levy and Asaf Yigal, the Logz.io machine learning algorithm also ingests, analyzes, and visualizes information to help companies make better decisions with internal information assets.

SEE: UX battle: Enterprise software vs. consumer software (Tech Pro Research story)

The algorithm uses social media to learn, predict, and visualize trends and keywords. "Once you analyze this data correctly," said a spokesperson for the company, "you can learn so much meaningful information about how people talk about and interact with your brand, or the things you care about. Data doesn't lie and the more decisions you make based on data, the less biased those decisions will become, as well as more manageable and adaptable."

TechRepublic used Logz.io's analytical machine to get a picture of how the internet reacted to Trump's nomination-clinching victory.

The GOP's #stoptrump moment never materialized. Though the idea of a strong establishment coalition was speculated about ad nauseum by traditional news outlets, the idea of a grassroots, conservative anti-Trump groundswell was mostly that: an idea, not a movement. In the week prior to the Indiana primary, no trend emerged, and the keyword "stop trump" spiked on election night.

stoptrumpmentionsovertimelogzio.png
The #stoptrump movement over time, on social media. | Image: Logz.io

Indiana election issue index:

  • Economy
  • Immigration
  • Terror
  • Health care
  • Christian [religious context]

National election issue index:

  • Economy
  • Immigration
  • Terror
  • Health care
  • Diplomacy
  • Christian [religious context]

indianaissues.png
Issues social media users care about, nationally and in Indiana. | Image: Logz.io

Trump's online popularity should not be confused with positive sentiment by Twitter users. According to analysis performed by Logz.io, though he was the candidate mentioned the most, Trump was also widely disliked by Twitter users. In the days leading up to the Indiana primary Trump scored much higher than his competition in Logz.io Animosity Index, a the number of times the word "fuck you" was tweeted to each candidate.

animosityindex.png
Candidate Animosity and Liar index. Image: Logz.io

Trump was a frontrunner in the GOP contest, but is an underdog in the general election. It is important to remember that the candidate was successful with the GOP electorate, a slice of registered Republican voters, and not the general public. Trump still trails rival Hillary Clinton in most polls. And while his digital presence is undoubtedly powerful, his online success is a product of his celebrity, more than a coordinated strategy.

Clinton's digital team is a formidable force. Though her Democratic rival Bernie Sanders famously used social media to crowdsource campaign funding, Clinton has outperformed Sanders on Twitter—in terms of account mentions, total followers, and relative growth—through the entire primary season. Her social media and big data staffers have access to the world's best technology and are advised by The Groundwork, a stealthy data company funded by Alphabet chairman Eric Schmidt.

In 2008 the social web was a promising but speculative venture for campaigns, and for business. The 2016 campaign cycle has demonstrated, if nothing else, that social media is a mature medium that, with the right tools, can be mined with great success for meaningful insights.

Over the course of the campaign, we will continue to perform simple data analysis. We hope to uncover unique insights and find ways business can benefit from campaign innovations.

If you're a data scientist, social media professional, or inquisitive TechRepublic reader we'd love your ideas on how to inspect campaign social media data. Please leave a comment below or ping us on Twitter @TechRepublic.


Read more

About Dan Patterson

Dan is a Senior Writer for TechRepublic. He covers cybersecurity and the intersection of technology, politics and government.

Editor's Picks

Free Newsletters, In your Inbox