Gallup survey: Republicans underestimate COVID risks while Democrats overestimate them

A six-month survey of 35,000 Americans carried out by Gallup and Franklin Templeton concluded that there was COVID misinformation circulating on both sides of the political aisle. Here’s how the NY Times described the findings:

To many liberals, Covid has become another example of the modern Republican Party’s hostility to facts and evidence. And that charge certainly has some truth to it. Yet the particular story with Covid is also more complicated — because conservatives aren’t the only ones misinterpreting scientific evidence in systematic ways. Americans on the left half of the political spectrum are doing it, too.

The errors on both sides tend to reinforce the perception that the virus is either not much of a threat or far more dangerous than it actually is. For instance, here’s a chart from a Brookings write-up of the results showing how people in each party believe COVID deaths are distributed by age. As you can see, the truth (yellow bar) is that 80% of deaths have been among those 65 and up. Both parties tend to underestimate old age as a factor in COVID deaths but Democrats’ estimates are consistently worse (further from reality).

On the other hand Democrats were more likely to know that COVID causes many more deaths than car accidents or the flu in a given year. They also were more accurate on whether or not the virus could by spread by those with no symptoms.

But Democrats got it wrong again when it came to rates of hospitalization. The actual rate is 1-5% but 41% of Democrats put the hospitalization rate at 50+ percent, which is way off. Meanwhile, Republicans were 2 1/2 times as likely to give the correct answer.

The survey added an extra wrinkle to this question. What if people were getting this wrong simply because they don’t really understand how percentages work to estimate risk. So they asked respondents to take a test called the Berlin Numeracy Test. If you’re interested in what the test measures you can read this BBC explanation or if you want to test yourself the BBC points to this online test created by The University of Oklahoma. I’m guessing most of our readers can ace it but you do have to stop and think a bit to get it. In any case, the point of including it was to ask how whether or not basic familiarity with statistical risk was the main driver behind so many people answering the question about hospitalization rates wrong.

And the answer was no. Even among those with high numeracy skills, Republicans were twice as likely to get the right answer and Democrats were nearly twice as likely to vastly overestimate:

Getting the wrong answers to these questions directly connects to how much risk you are willing to tolerate. So, for instance, it’s not a coincidence that Democratic teacher’s unions are holding up the reopening of schools primarily in blue states. That’s happening, in part, because the estimate of risk on the left is out of whack. And of course you could argue the same thing about people in red states re-opening businesses despite the risk from the spread of more contagious COVID variants.

The point of all of this is that a lot of the arguments we’ve been having now and for the past year involve people’s general misestimation of risk. There has been a lot of focus in he media on Republicans placing that risk too low (with Gov. DeSantis becoming the target of frequent criticism) and relatively little of Democrats placing it way too high.