Texas, Virginia, and Vermont are being criticized by public health experts for mixing up their test data, producing far more impressive numbers

• Texas, Virginia, and Vermont have been accused of mishandling their coronavirus testing data, producing large totals but mangling the figures to make them functionally meaningless.
• The states all made the error of combining their totals for two types of tests: Those for active infections, and those for past infections (measured via antibodies).
• The two are not the same. Adding the figures together makes the numbers seem bigger, but experts said they are of little use for understanding the pandemic.
• Some states have since fixed their methods, and say it has not affected the underlying trends in data.
• Other states have been slammed for their handling of data: A Florida official said she was fired to refusing to manipulate the figures, while Georgia was under fire for misleading graphs.

Texas, Virginia, and Vermont have been accused of presenting their coronavirus data in misleading ways that makes inflates the scale of their testing programs, and makes the underlying data meaningless.

The Associated Press reported that officials in Virginia, Texas, and Vermont said they had been combining the figures for two separate types of tests.

The first were tests to find if people were currently infected with the coronavirus, the second were tests looking for antibodies, which are found in people who had the virus but are no longer sick.

The two are not comparable.

Experts told the AP that adding the figures together makes it look like states are doing more testing than they really are. This makes it harder for lawmakers to make informed decisions about easing restrictions.

Jennifer Nuzzo, from the Johns Hopkins Center for Health Security, said that presenting the data in this isn't necessarily malicious, but has important consequences.

She told the AP that with that system "you're not going to be able to make good decisions about reopening and about what level of disease you have in the community."

The AP reported that officials in Vermont and Virginia have since stopped combining the figures in this way.

Gov. Ralph Northam of Virginia, a Democrat, said that the correction had caused "no difference in overall trends."

Texas Gov. Greg Abbott, a Republican, also said on Monday the state was not at that point mixing up its figures. But state health officials said last week that they had been treating the data this way.

The Atlantic had previously reported on the presentation of data in Virginia, which Ashish Jha, the director of Harvard's Global Health Institute, said made it impossible to use the data to make decisions: "It is terrible. It messes up everything."

The AP's report comes after separate accusations of data manipulation in Florida and Georgia.

Rebekah Jones, a top data researcher in Florida who developed the state's coronavirus dashboard, said she was fired as she refused to "manually change data to drum up support for the plan to reopen."

Georgia also came under fire when it published a graph of coronavirus cases in descending order rather than in chronological order, giving the false appearance that the outbreak was easing.

The graph was deleted in less than a day, the AP reported. Republican Gov. Brian Kemp's office denied trying to mislead people.

Candice Broce, his communications director, tweeted: "Our mission failed. We apologize. It is fixed."