Rain, Church Attendance, and Crime

Yesterday, the University of Texas at San Antonio web site published an article about one of their economics professor's research on the relationship between rain, church attendance, and crime.  See the article here.  This is the economic approach to religion in action!

The research question asks what is the effect of church attendance in a community on crime in that community.  To answer this question, you cannot just merely look at the correlation between church attendance and crime across communities because there are many other factors that may muddy those correlations.  Ideally, we would run an experiment in which we (the experimenters) randomly assigned some community members to attend church and others to not attend church, but this experiment is not really possible (or ethical) to run.

When social scientists are unable to run the ideal experiment, they often look for situations in which there is something in the real world that has some features of an experiment.  In this case, Professor Moreno-Medina uses rainfall as a type of experimental treatment because rainfall during the times of a church service causes a decrease in church attendance.  The rainfall provides a kind of exogenous, random experimental treatment -- what economists call a natural experiment.

Read the article to see what he finds.  It is a short article, so read it entirely.  Pay attention to how the article describes the potentially complex relationship between church attendance and crime.