Skip to main content
Markkula Center for Applied Ethics

Muddying the Data, Refracting the Message

Illustration by prof. Michael LaBossiere

Illustration by prof. Michael LaBossiere

Welcome to the 2020 campaign, with its “data hauls" and accidental transparency.

Irina Raicu

Irina Raicu is the director of the Internet Ethics program at the Markkula Center for Applied Ethics. Views are her own.

On June 14, Brad Parscale tweeted, “Just passed 800,000 tickets. Biggest data haul and rally signup of all time by 10x. Saturday is going to be amazing!” Parscale is president Trump’s campaign manager, and he was referring to ticket requests for the president’s first mid-pandemic rally in Tulsa. The “data haul” he boasted of was the information collected from registrants. Such data is much valued by all political campaigns, but especially by the Trump campaign, which has been using it, much more assiduously than its competitors, it to hone is targeted messages to supporters on social media.

On June 18, Newsweek published an article referencing Parscale’s tweet, and quoting a data scientist from NGO TacticalTech who argued that the data mentioned was “not just data on a million voters, it's data on a million high-value voters, very reliable, strong supporters.” The article, however, also cited political science professor Eitan Hirsch, who commented that some registrants “could have signed up as ‘an expressive act.’”

Hirsch turned out to be right. By Saturday, following the rally that drew around 6,200 attendees, various media outlets were reporting that TikTok users and K-pop fans had “registered” for the rally with no intention to attend, precisely in order to inflate—and then deflate—the attendance expectations, highlighting the gap between triumphant campaign rhetoric and reality. The New York Times, among others, detailed the way in which the “effort to deprive Mr. Trump of a large crowd spread from Twitter and TikTok across multiple social media platforms, including Instagram and Snapchat.”

By Sunday, Brad Parscale had commented that “online trolls doing a victory lap, thinking they somehow impacted rally attendance, don’t know what they’re talking about or how our rallies work,” and adding that “[r]egistering for a rally means you’ve RSVPed with a cellphone number and we constantly weed out bogus numbers, as we did with tens of thousands at the Tulsa rally, in calculating our possible attendee pool.”

Many of the phone numbers, however, were not “bogus”—or, at least, many of those who registered had been advised to sign up for a Google Voice or similar phone number to use for the registration. Others, however, had signed up with their own numbers, but still with no intention to attend. Neither of those groups would have been “weeded out.” The data collected, at least in those cases, does not come from “reliable, strong supporters.”

Ironically, some of the people involved in the activist effort had had no intention to muddy the campaign’s data by volunteering their own. In fact, one of the organizers commented that “several people who took part in her campaign complained that once they signed up for the rally with their real phone numbers, they couldn’t get the Trump campaign to stop texting them and sending them messages.”

Those misdirected Trump campaign messages might themselves have a powerful, unintended effect.

As an extensive article in The Atlantic details, the Trump campaign has spent millions of dollars to make its targeted messaging to supporters as effective as possible. “People have marveled that Trump never stopped running Facebook-ad campaigns,” write Ian Bogost and Alexis Madrigal. “And the reason is, he couldn’t. The whole point is that the campaign has to keep fresh data flowing through the system. … [T]he ads will drive the users to Trump websites, where Facebook pixels will slurp up revised data that will help refine the Trump campaigns’ advertising even more.”

What happens, then, when the messages “optimized” for a particular audience reach very different people—such as the social media users who signed up for the rally tickets and who otherwise, most likely, wouldn’t have come across those messages? They shock, and their effect is the opposite of what was intended. They reveal the chasm between what “very reliable, strong supporters” hear and what others do. They unintentionally pop the filter bubbles that we often live in.

This is not to suggest that you should sign up for the invasive apps that various campaigns are now deploying, or register on websites aimed at supporters of politicians that you actually oppose. Some people do that, but fair warning: the effect of the repeated “refracted” messaging can be both enraging and exhausting. Still, one accidental outcome of muddied data and misdirected messages may be a better-informed citizenry, with a broader understanding of our polarized community and of the tools that amplify that polarization. So perhaps the actual number of registrations announced by Brad Parscale in his tweet will indeed make a difference.

Illustration by Professor Michael LaBossiere

Jun 26, 2020

Subscribe to Our Blogs

* indicates required
Subscribe me to the following blogs: