So if that is the current understanding of how the vaccine works, isn’t it important to follow up on data that seems to not align with those statements? If 69% of people are vaccinated, I would expect <69% of recent COVID cases to be from the vaccinated.

When we get data that contradicts what we expect, it needs to be taken seriously, it needs to be broadcast, it needs to be followed up on, and the correct conclusions need to be drawn from experts in a transparent way.

> So if that is the current understanding of how the vaccine works, isn’t it important to follow up on data that seems to not align with those statements?

Sure.

> If 69% of people are vaccinated, I would expect <69% of recent COVID cases to be from the vaccinated.

I would expect both vaccination and detection of cases to overrepresent people with better access to and inclination toward medical services

In addition vaccinations are biased towards the elderly, and (because vaccines haven't been approved, even on an emergency basis, for this population) completely exclude young children. This also biases the figures.

Basically, the media reporting of “contradictory” findings is pretty much completely abuse of statistics rather than actual contradictory findings.

A decent discussion is here: https://english.elpais.com/usa/2021-07-27/covid-19-infection...

I am familiar with Simpson’s Paradox but I carefully read your linked article anyway and it truely is enraging to me when statistics is twisted to lie to people. I totally agree there is a very large amount of really bad headlines and non-expert pundit conclusions from right wing media. And it is harmful. I wish more people took the time to understand what statistics are actually saying. Unfortunately I think some portion of the population might just be incapable of properly understanding stats, it really is a hard and tricky subject.

That said, I don’t think the data from the OP’s CDC article is a case of Simpson’s paradox. The quoted statistic in the CDC article is that 74% of infected people were vaccinated. I am comparing that to the vaccination rate in MA of 69% (also from the article). The denominator in one stat is a strict subset of the denominator of the second stat. Simpson’s paradox requires disjoint denominators.

So that being the case, I would still like for actual experts to follow up on this data and come to some conclusion.