Comment by gryson | Hacker News Reader

Comment by gryson | Hacker News Reader

In regards to the issue of selection bias, this is what the ad looked like:

https://twitter.com/foxjust/status/1251270848075440133

Quote from ad: "We are looking for participants to get tested for antibodies to COVID-19."

A quote from someone who participated in the study:

"I participated in the study because I had been sick the week before and was very curious. In the intake questionnaire they asked if I had recent symptoms. I'm unpleasantly surprised that they seem not to have made an effort to use that data to unbias the study."

https://twitter.com/mattmcnaughton/status/125132223548416819...

Another:

"I was part of this study and that is totally why I signed up! People I talked to who tried to sign up had similar reasons. Lots of subjects at the testing site wearing masks, more than you see at the grocery, more evidence that a lot of us were more conscious about transmission"

https://twitter.com/McSalter/status/1251511091294691328

It's very hard to deny that selection bias may have played a part here.

The symptoms of COVID range from a cold to a flu (for those who are ineligible for viral RNA test kits). If the selection is biased, it is biased toward people who felt cold or flu symptoms in the past few weeks, which is going to be a huge amount of people.

Nontheless, that's still a highly biased sample of the actual population.

"hypochondriacs who think they may have gotten covid" accounts for 100% of Californians

That's just.. a basic misunderstanding of statistics.

So, I'm not getting why selection bias would matter here. If they are counting the total number of participants who were positive for the antibodies, and this number was out of the total number of residents in the county, big deal.

Why would they ask all the healthy people to come get tested? Just assume that all the untested people are negative for the antibodies.

Otherwise you would get a higher infection rate count, which would of course result in a lower mortality rate for the disease.

You're misunderstanding the methodology. They tested a sample and multiplied to get the overall infection rate, under the assumption that their sample was representative.

If the sample was biased, then the results cannot be generalized to the whole population. The purpose of this study was not to find positive cases to treat, it was to estimate the prevalence in the general population.

Sure, but that's not at all how the sampling was done. Just read the article!

Reply on news.ycombinator.com