Tracking Michigan protesters raises privacy, COVID-19 spread questions
Lansing — The tracking of hundreds of Michigan Capitol protesters’ cellphones from two rallies this spring has raised questions about protecting privacy rights, while testing limits of what location data can predict about the spread of the novel coronavirus from packed public events.
A liberal advocacy group most recently pursued the anonymized cellphone location data from more than 400 devices from individuals who attended the American Patriots Rally on April 30 in Lansing against Gov. Gretchen Whitmer’s stay-home orders and used it to gauge their locations the next day in a bid to determine the potential spread of the virus from the demonstration.
The location of 422 cellphones within a roughly four-block radius of the Capitol showed individuals traveled back to areas throughout the state, including West Michigan, Metro Detroit, Northern Michigan and the Indiana border after the event, according to data obtained by VoteMap on behalf of the progressive group Committee to Protect Medicare and Affordable Care.
"They ended up pretty much scattered across the state," said Dr. Rob Davidson, executive director of Committee to Protect Medicare, a west Michigan emergency room physician and a Democratic former congressional candidate.
“I think you can absolutely argue that a likely high-risk group of individuals all gathering in close quarters have a high risk of propagating the virus,” said Davidson, whose group did not track protesters at the police brutality protests across Michigan over the weekend.
But the group pulled similar data at an April 15 rally called Operation Gridlock.
The tracking, known as "geoharvesting," is when data is gathered from a smartphone app on a device connected to the internet. That data contains geolocation information that can be queried to show movement on a map.
The Michigan Department of Health and Human Services’ contact tracing initiative has come across no reports where a COVID-19-positive individual reported attendance at an event, including protests, in their travel history, said Lynn Sutfin, a department spokeswoman.
The tracking has raised the hackles of Capitol demonstrators, such as Phil Robinson, a Barry County resident who said he considers the data gathering during the April 15 and April 30 protests a violation of his rights. The committee at least could have alerted attendees they were being tracked, he said.
"I’d have still showed up, but I would have left my phone at home. I wouldn’t have gone live on Facebook," said Robinson, founder of the Michigan Liberty Militia who attended both protests.
He also wondered why the Committee to Protect Medicare, which is supposed to be concerned about health care, would fail to track demonstrators at the police brutality protests in Grand Rapids, Lansing and Detroit.
"That doesn’t make sense," Robinson said. "They’re going to track us at the Capitol but nowhere else?"
Ryan Kelley, a west Michigan resident who helped organize the American Patriot Rally, said he was uncomfortable with the cellphone location data being gathered there.
"That sounds like an invasion of privacy," Kelley said. "That sounds like it should be illegal.”
Were protesters' privacy violated?
The protesters agreed to be tracked because the information was collected from "opt-in" cellphones, in which the user at some point agreed to terms allowing for such data to be collected by firms such as VoteMap, said VoteMap CEO Jennifer McEwan.
The cellphone users remained anonymous and their locations were instead culled from the publishers of the opt-in apps they were using, said McEwan, CEO of the Austin-based startup Datum that is the parent company under which VoteMap falls. People gauging cellphone locations the day of the April 30 protest and the day after would not be able to see to whom the phone belonged, she said.
While VoteMap is used primarily for Democratic-leaning campaigns, Datum also offers Moblyze for Republican campaigns and YouTarget for commercial endeavors, McEwan said.
"Data is always being transmitted," McEwan said. "You just can’t see it. We sort that data that’s already there and then query it to answer questions.”
Davidson, a Democrat who lost to U.S. Rep. Bill Huizenga, R-Zeeland, in the 2018 election, said his group hired VoteMap to monitor location data for both the April 15 Operation Gridlock and the April 30 American Patriot Rally.
Between 800 and 1,000 individuals attended the American Patriot Rally, according to Michigan State Police, with dozens entering the state Capitol during House and Senate session. Some protesters were armed and chanted at State Police troopers outside the House chamber to “let us in.”
At Operation Gridlock, the committee gathered the locations of roughly 506 cellphones, but the locations collected April 30 from the 422 individuals at the American Patriots Rally was a more useful data set since protesters were in close quarters inside the Capitol instead of segregated in their vehicles, Davison said.
He argued the individuals in close quarters were more likely to spread the virus and more worthy of closer study.
“The activities taking place indoors that we all saw on TV certainly would be higher-risk activities for anyone who had the virus to spread it to others,” Davidson said. “The odds would argue that some people had the virus, didn’t know it and spread it to other people.”
Public health vs. individual rights
VoteMap pulled device locations from Operation Gridlock for the next two days, the company said. But it only pulled data through the next day from the American Patriots Rally.
Similar data was collected by the committee at recent protests in Florida, Wisconsin, Illinois and Colorado.
The data collected April 30 can’t conclusively determine whether protesters contributed to flare-ups in their respective communities, Davidson said, though he argued it could be inferred given what’s known about the transmission of COVID-19.
In some cases, it might take weeks or several iterations of the infection to realize the full impact the gathering had on the virus’ spread, he said.
"I think it is extremely reliable for what it is," Davidson said.
Since the cellphones were anonymous, it’s impossible to know whether those "pinged" at the April 30 rally and later queried for location were protesters, police, lawmakers, journalist or interested bystanders.
Collecting phone data for use in public health studies or operations has become a hot issue because of the coronavirus pandemic, said Dr. Mona Sobhani, director of research and operations at the University of Southern California’s Center for Body Computing. Regular citizens might not realize it, but apps commonly collect location information and barter the data to companies interested in targeted marketing, she said.
"There’s this whole world of data brokers that buy and sell data," Sobhani said. "A lot of the times the default is that it collects location and you have to turn it off.”
There have been efforts to better digitize coronavirus contact tracing and symptom monitoring through apps developed at various universities, Sobhani said.
The apps are not especially popular because of privacy concerns, she said. And, without a large swath of users, the data the apps gather is not very meaningful to reach public health conclusions, she said.
"It does raise this really interesting question about public health concerns versus individual freedoms and liberties," Sobhani said.
Whitmer dives into group's data
Kelley, the April 30 American Patriot Rally organizer, said he worries that Whitmer would use the cellphone location information "to push her agenda."
The Democratic governor appeared to refer to VoteMap data from a Capitol protest in a May 13 interview with "The View" when she said protests in Michigan’s Capitol showed “the risk of perpetuating COVID-19 is real.
"We’ve seen it happening," she said.
Later that day in a state press briefing, Whitmer backpedaled from the claim, noting she didn’t have proof protesters brought coronavirus back to their hometowns.
“I’m not following everybody home and taking their temperatures and watching them for two weeks,” said the governor, who noted “a group” had assembled a report tracking phones to areas where cases increased.
“I don’t know the group,” she said. “I have not vetted the data. I can’t vouch for it. But I think that would not be a surprising outcome if that was the case.”
Whitmer has consistently expressed concern that the Lansing protests against her stay-home order would exacerbate the spread of the virus and warned it could lead to a longer shutdown.
On Monday, Whitmer expressed the same worries about the police brutality protests, noting she had a "high level of concern" about the protesters who refused to wear masks.
"I will say this, a lot of the early-in-the-day events, where it really was a peaceful protest, people were wearing masks," Whitmer said during a Monday press briefing.
"My hope is that we're fully re-engaged by the middle of the month to the end of the month ... we're going to follow the numbers and if we see things start to spike we're going to slow down, we may even have to take a step backward."
Similar technology was used and publicized in Lansing by the Republican Governors Association in February when Whitmer gave the Democratic response to President Donald Trump's State of the Union Address.
The association aired an ad called “Broken Roads, Broken Promises” an hour before the GOP president gave his speech, and used geolocation data to target Facebook and Instagram profiles within a 10-mile radius of the Capitol building in Lansing.
The ad featured media coverage of Whitmer’s struggle to pass a 45-cents-a-gallon gas tax increase through the Republican-led Legislature and includes a clip from Trump’s December rally in Battle Creek, where he noted “she’s not fixing those potholes.”