A plea: please be extremely humble when considering any kind of policy impact evaluation.

Impact evaluation requires a lot of things to come together just right, the right designs to capture it, and the experience to know when it doesn't.

Now is not the time to learn by error.

I've been seeing a ton of garbage COVID impact eval studies from all fields, including people with the credentials to know better.

By the time people like me see it and try to correct the record, it's too little, too late.

Journal editors, I am looking at you here.

The more impact the study might have in the news, policy, research, behaviors, etc., the more sure you should be that the methods hold up.

That means getting the right people to review, and having a heavy hand on the reject button.

At some point I should do a long thread aimed at common problems in the impact evaluation methods (i.e. the interrupted time series, diff-in-diff, synthetic control-type suite of methods).

Don't have time for it today, but maybe.

Who are the right reviewers?

Get impact / policy evaluation specialists. These folks are commonly in health policy/econ departments. Many can be found in epi departments, but this is less common (epi tends to be more individual-scale, policy tends to be population-scale).