Sarah Hoyt wrote a new version of her critique of epidemiological models (see my post A Study in Denial) for her column in the far-right outlet PJMedia: https://pjmedia.com/blog/modeling-covid-19-and-the-lies-of-multiculturalism/ It’s basically the same points she made in her original essay, nobody really knows and therefore Hoyt knows and therefore she knows that it is all down to culture or population density etc. It is at best wild guesses and half-formed opinions where her credentials are established like so:
“However, as the mother and wife of STEM people for whom physics is a game and who create such models for fun, I know that the accuracy of the model depends on how much you put into it and how much of the real factors on that day, in that place, you can put in.”
Her theory is, of course, another kind of model and it shares with any model all the flaws plus the additional ones of being half-arsed opinion based on a weak grasp of the news.
“For instance, my friend in Albany, Georgia, tells me he assumes part of the reason it got so bad in his neighborhood (the worst per capita in the U.S. last I looked) is that “we are the touchiest, most social people I know,” i.e., there is a lot of touching and hugging. At a guess, this is the reason it got so bad in Italy, too, but not nearly as bad in Germany, where, frankly, people aren’t that touchy/feely/huggy.”
Quite how the UK fits into the Hoyt-Covid-Hugs model of infection I don’t know but I can’t say us Brits have ever been accused of being a very hug-prone nation. It does help resolve what Hoyt things ‘culture’ might be: national stereotypes. Maybe France has a high infection rate because they wear berets whereas Brazil has a different pattern because they wear bikinis? That’s about the level of Hoyt’s analysis.
So if there is nothing new in Hoyt’s PJMedia piece (one of four, two behind paywalls) why am I mentioning it. Well Rush Limbaugh (who apparently still exists even though I’d forgotten about him sometime around 2010) has been praising the column by Hoyt.
“Now, this is a tough case to make. And Sarah Hoyt does a great job in the piece, a very long piece. We will link to it at RushLimbaugh.com. It was published yesterday. But I want to try here because her point is that we get these models projecting how many people are gonna get sick, how many people are gonna die, assuming everybody’s identical, everybody’s the same, gonna behave the same.”https://www.rushlimbaugh.com/daily/2020/04/13/the-covid-19-models-cant-account-for-culture/
Hmm, so what would Mr Limbaugh or Ms Hoyt suggest instead for a pandemic response? Now consider, neither of them deny that there is a pandemic even if they question the severity of it. Yet consider, there is no reason to believe that Covid-19 is the worst possible novel viral disease that could occur. Notably, the measures that have been seen to work require them to be implemented BEFORE the full severity of the pandemic is known – the earlier the better. So there is no viable scenario in which the response to a novel pandemic can be made with full and accurate models. Nor is there any possible way of creating models that account for every person’s unique individuality (and what a rabbit hole that would be — implying a level of surveillance state of dystopian proportions).
[ETA: For a better discussion of the limits and value of models in this pandemic see this extended cartoon/discussion at FiveThirtyEight by Zach “SMBC” Weinersmith https://fivethirtyeight.com/features/a-comic-strip-tour-of-the-wild-world-of-pandemic-modeling/ ]