"More on Lead and Crime, but Really on How to Make Decisions"
This could explain a lot, part 2

"IRBs Gone Wild"

I had one--thankfully, just one--experience similar to Peter Klein's.

So, even social-science researchers must receive IRB training and have individual research projects — yes, every research project that involves “human subjects,” which includes research using secondary data — approved by the campus IRB.

The data I used had already been gathered and was completely anonymous. There was a probability lower than the sun will go supernova tomorrow that anything I did with the data would harm, or even offend, the subjects. But I had to jump through IRB hoops. It's as if stupid people are running our government and universities.