Abstract: http://works.bepress.com/psheaton/1/
Paper: https://pdf.yt/d/THWiTlgHIKcbfZwg

I'm an economist, so I don't trust the empirical methods of sociology. While I was reading the post, I was instantly skeptical of the findings because the degree of religiosity in the community could be itself related to the level of crime.

Simple regressions don't handle that kind of simultaneous relationship no matter how many control variables you throw in. This paper tries to take care of that problem and finds no relationship between religion and crime.