Jones-ing for COVID Data

by Jessica Weinkle, Department of Public and International Affairs, University of North Carolina Wilmington

Last week, Rebekah Jones, the scientist ousted from her position formatting COVID data for Florida Department of Health (DOH), said she will be returning to Florida to face criminal charges related to cybersecurity. 

Over the course of 2020, Jones developed a name for herself as a whistleblower and activist, earning the ire of Florida's governor, Ron DeSantis.  Of her return, Jones tweets,  

The Governor will not win his war on science and free speech. He will not silence those who speak out.

All told, the controversy between Florida and Jones has earned her extensive media buzz and $300k in a GoFundMe

Jones is credited with building Florida's public GIS interface for COVID data.  In an interview with CNN's Cuomo Prime Time (reported on here in the Tampa Bay Times), Jones explains that the problems between her and DOH began in April 2020 when the COVID data provided politically uncomfortable knowledge to policymakers looking to "reopen" the economy:

Jones: Well, the first time I was asked to do something improper was in April. When I brought basically what I thought the results [sic] on whether or not each county could open to superiors, they essentially told me they did not like the results.

Cuomo: How so? What does that mean?

Jones: The results didn’t match the report for reopening that had already been written. Basically a lot of rural counties for a wide range of reasons, didn’t meet the criteria that the state had outlined in order to qualify for reopening. Whereas some more populated counties did meet that criteria, and I was told that specifically, and this is a quote, ‘We can’t tell Jackson and Franklin counties that they can’t reopen, but Broward and Miami-Dade can.”

To place this exchange in context, the Florida Governor's reopening phases were described in his executive orders as a "data-driven" strategy (e.g. here) whereby reopening was, at least implied, to be contingent on "several critical benchmarks."  

Considering that the quarantine measures put in place in the early part of the year had significant economic and political repercussions the world over, there were incredibly high stakes attached to reopening decisions.  In turn, there was a high likelihood that hinging re-opening decisions on COVID data would result in political pressure being placed on processes of data collection and analysis. 

Jones continues to explain the disagreement that came up in May between her and the DOH ultimately leading to her being fired,

Jones: They calculated the number of positive, or the percent positive people, and changed it to new cases over total tests per day, which is also deceptive.

Cuomo: Why is that deceptive?

Jones: So let’s say I give you 100 apples, right? And 50 are rotten. And I ask you what percent are rotten. You’d say, 50 percent. If I then tell you that 30 were rotten yesterday, and 10 were rotten two days before that and that five are almost completely rotten but not quite, and I’ve cut the other 50 non-rotten apples into hundreds of tiny little pieces, what percent of the apples are rotten? It’s still 50 percent.

Cuomo: It’s still 50 percent.

Jones: Yeah. You just made it extremely complicated and convoluted. So what we used to do for percent positivity, which was one of the benchmarks that each county had to meet in order to qualify to reopen, it had to be below 10 percent and decrease over two weeks. So normally, when we people think of percent, they take the number of positive people divided by the people tested. That seemed honest and fair. They changed it to the number of new cases per day over the number of negative tests per day, so if you decide you want to get tested five times today, you count five times toward the negative if those are all your results. If I’m positive, I count once.

On her website, Jones notes that political pressure on the data reporting process was well known in the DOH: 

It was no secret in our office that Shamarial Roberson asked me to compromise my integrity and use my dashboard to mislead the public about the safety of reopening each county. 

When Jones went public about her firing things unraveled in what has come to be a fairly typical fashion: 

  • scientists and professional groups rallied
  • political opponents (DeSantis) launched credibility attacks (here and here)   
  • the matter was framed as a choice between pro-science or anti-science

Thus, when Jones again made headlines in December after the Florida Department of Law Enforcement raided her home for cyber security violations (a video recording of which went viral), Jones announced leaving the state because,

The governor of Florida seems have an irrational and passionate hatred for who I am and what I represent: defiance in the name of science of human decency. 

And now the polarization of the issue is complete: you can side with Jones and be pro-science or side with the state of Florida and be anti-science.  

Released email from Jones to her supervisor

A more practical look at the situation though shows that at the heart of the conflict between Jones and the state of Florida is a disagreement over the appropriate methodology to use to calculate the percent of positive COVID cases. 

I can't speak to which statistical method is more appropriate. 

However, it is fair to say that there are often many methodological options available for analyzing data and choice of method is a reflection of priorities and perspectives.  

Policymakers could have better handled the situation by publicly explaining why their chosen method is most appropriate for the problem they are seeking to manage rather than lodging insults at Jones and firing up the scientific community.  An explanation would have contributed to public discussion about what sorts of things matter to the public for managing the pandemic in Florida. 

For her troubles, Forbes named Rebekah Jones Tech Person of the Year because "she is the latest technologist who stepped up to fill the vacuum left by governments during COVID-19."  And this is the overarching problem of the whole affair. 

State agencies and the Federal government were ill prepared to handle the data demands of a pandemic response. Without Federal leadership and a coherent system for defining data needs and enabling its collection, decisions about data were inconsistent and lacked transparency.  As Nature reports, 

Without up-to-date, reliable information on who is infected, why and where, US scientists, policymakers and the public must instead rely on media reports and independent efforts to consolidate data.

Jones' appears to be a victim of this national failure by stepping in, largely on her own, to 'fill the vacuum' in Florida.  

As is stated nearly everywhere Rebekah Jones is discussed, Florida's COVID-19 dashboard is her own creation and she had extensive leeway in how data was presented.  In her own words

While no one operates without the support of their team, I was the sole architect and manager of the dashboard and its published underlying data. 

The dashboard itself received praise from the Dr. Deborah Birx, the former White House Coronavirus Response Coordinator.  And Jones reports receiving praise from professional organizations for the dashboard.  

However, that the gravitas of the politics around a failed Federal response to COVID data management and reopening the Florida economy would come to bear down on one lowly geographer, was a disaster in the making.  

This is not a critique of Jones' expertise, but on the dynamics of science advice in place.  

Whether or not the Florida DOH is a strong enough institution to resist political meddling in data reporting is a secondary issue- though a worthy one to ponder.  The immediately alarming matter is the collegial dynamics within the DOH itself.  

Given the nature of the uncertainty around the virus and data relevancy, it was entirely reasonable for analytical methods to evolve over time.  Decisions about how data would be analyzed and presented should have been a collaborative discussion within DOH from the beginning so that no one individual could hold-up the show or shoulder the blame if things went awry.  

This should have been the case in the best of political scenarios.  But it was ever more important given the severe politicization of US pandemic response.

I am not the only one that has found the controversy around Jones more telling about problems at the DOH than about Jones herself.  Writing in Slate, cybersecurity scholar, Josephine Wolff, points out that the cybersecurity charges brought against Jones is telling of deeper concerns about the way the state secures its emergency systems.

In the end, making this story about Rebekah Jones takes attention away from bigger picture issues about accountability at Florida's DOH.