Monday, March 14, 2022




The New York Times emails its subscribers every morning with text and links to prominent news stories. This morning, the email poses a question:

These stark differences [in how red and blue areas responded to covid] have created a kind of natural experiment: Did Omicron spread less in the parts of the U.S. where social distancing and masking were more common?

The answer is surprisingly unclear.

Actually, if you look at the charts, the answer could hardly be more clear.

Nationwide, the number of official Covid cases has recently been somewhat higher in heavily Democratic areas than Republican areas, according to The Times’s data.

Imagine that! Here are the charts:

The lack of a clear pattern is itself striking.

On the contrary, I would say there is an obvious pattern, it just isn’t the one the Times wanted to see.

Remember, not only have Democratic voters been avoiding restaurants and wearing masks; they are also much more likely to be vaccinated and boosted (and vaccines substantially reduce the chances of infection). Combined, these factors seem as if they should have caused large differences in case rates.

They have not.

The Times piece is full of factual and logical errors. What is the evidence that Democratic voters are “much more likely” to be vaccinated than Republican voters? Are there any survey data? I haven’t seen any, and the Times cites none. In fact, the largest demographic that lags in vaccination rate is blacks, who vote heavily Democratic. As a result, my guess is that Republicans are more likely to be vaccinated than Democrats.

But, in any event, the statement that vaccines substantially reduce the chance of infection is false, which is why we have just experienced the highest covid case load in history at the same time that a large majority of Americans are vaccinated.

More logical errors to come:

The first lesson is that Covid vaccines are remarkably effective at preventing severe illness. Here are the same four states from the above charts, this time with death rates instead of case rates:

The messiness of the previous charts has given way to an obvious pattern: Covid death has been far more common in red America. Over the past three months, the death rate in counties that Donald Trump won in a landslide has been more than twice as high as the rate in counties that Joe Biden won in a landslide, according to Charles Gaba, a health care analyst.

The Times says that these discrepancies are explained by vaccination rates. So why didn’t it simply compare data on death rates of vaccinated vs. unvaccinated people? Such data are available, although some agencies go to considerable length to obscure the numbers. Such a comparison would show, I think, that vaccination provides some protection against death, although not as much as is commonly claimed.

The red county-blue county comparison is simply ridiculous. The counties that went 60%+ for Donald Trump in 2020 are overwhelmingly rural. The counties that went 60%+ for Joe Biden are overwhelmingly urban. Rural counties have a relatively high percentage of elderly residents, while urban areas have relatively high proportions of young residents. Covid is dangerous almost exclusively to the elderly. If death rates in red and blue counties were compared within a given age band–e.g., 61 to 70, 71 to 80, 81 to 90, etc.–I doubt that you would see any difference. Take Georgia, for example: the death rate is 0.0006 in red counties and 0.0004 in blue counties. I would be shocked if that difference is not accounted for entirely by the age profiles of the counties. In any event, that comparison would have to be made to generate a meaningful result.

Factual errors and logical fallacies aside, we should probably be grateful that the Times, America’s leading source of misinformation, is finally recognizing the futility of shutdowns and mask mandates:

The second lesson is that interventions other than vaccination — like masking and distancing — are less powerful than we might wish.

That’s too delicately stated, but still: better late than never.

No comments:

Post a Comment