Hacker News Reader

Does this study make the same mistakes as the Standford study?

One particular criticism of the Stanford study was that they recruited via Facebook adverts, and that method may be influenced in recruitment by the "desire to get a test if you had reason to believe that you, or someone near you, had the virus"

In more detail: https://statmodeling.stat.columbia.edu/2020/04/19/fatal-flaw...

This USC study seems to have recruited differently: "Participants were recruited via a proprietary database that is representative of the county population. The database is maintained by LRW Group, a market research firm" Via: http://www.publichealth.lacounty.gov/phcommon/public/media/m...

The flaw was that participation was voluntary, with the effects you describe. I think this is also the case with this study. I'm not sure how a non-voluntary study could be conducted, you can't really force people to participate in this.

edit: typo

The biggest flaw is that they didn't use the confidence intervals associated with the specificity in their calculations properly.

Given the CI's of the specificity all 50 positives could have been false positives so we can't conclude anything from the study.

Some of the authors are the same, so most likely yes.

Reply on news.ycombinator.com