Homepage

A friend of mine mentioned to me that hard scientific evidence (e.g. a causal relationship clearly demonstrated by a controlled experiment). would be something they would deem as necessary, but even then not necessarily sufficient, to instantiate a policy.

So, even though burden of proof is not on me to provide resources, I am still more than able to do some. Not only is there a lack of evidence that masks can assist in stopping transmission of a virus, but there is actually the presence of evidence that masks do not help.

Let's start with just a few

  • Refleciton on whether masks are helpful (nejm)
  • A small study that fails to reject H0.
  • A review of multiple studies that still didn't find anything conclusive.

    I may add more studies here as I see fit. But for anyone interested, these are a good place to start.

    Why am I rejecting observational studies?

    Observational studies are vulnerable to cherry-picking and inaccurate conclusions. For example, let's say that I wanted to keep dogs out of my old apartment 809 in Campus View. In an effort to do this, I put a sign on my door that said "No dogs allowed." After 6 months, I record that no dogs entered my apartment and conclude that my sign was effective. Would you agree?

    Of course not - that's absurd! There are no pets allowed in Campus View, and no one that visited me even owned a dog. It would be entirely ureasonable to conclude that my sign has done that much. And this is my beef with most of the studies that claim masks are effective. Here is an example of that. If you look hard enough at any set of data, you can always cherry pick the subset you want.

    However, whenever we take our time and do a controlled experiment, we always fail to find anything. So, when I put up a sign on my apartment CV809 and my neighbor doesn't put up a sign, we both count that no dogs visit our apartments.

    The sign doesn't seem very effective to me.

    But to any reader, don't get caught up in this argument. Yes, it's true that the jury is out on how effective masks are. Don't let this make you lose sight of some of the other injustices in question, namely the contact tracing app and the slowrolling clause.