It's subtle, but a claim in this story is misleading: "At the same time, 31% said they were able to get tested when they sought to do so."

I'd say that 31% sought to get tested *and* were able to.

56% were able to get tested *when* they sought to do so. https://twitter.com/statnews/status/1361284351565193219

The key point is that only 55% sought to get tested. When looking at the ability or inability of people to get tests, we should be looking at just that 55%.

So if 55% wanted to get tested and 31% got tested, that means 56% of those who wanted to were able to get tested.

Why does it matter?

It matters because the claim that 24% were unable to get tested is really an underestimate. That's 24% of the fraction who *wanted* to be tested.

The more important figure is that 44% of those who wanted to be tested were unable to.