Weighting analysis by prison poplation

[shapironick]

how can we incorporate the prison population into our analysis? we have both capacity and population variables, so we could just use capacity if we are looking for where the big prisons are. it would be interesting also to know if the over capacity prisons are closer to exposure sites.

There are so so many small facilities that my guess is are farther away from toxic sites than the larger ones. so parsing that difference seems key to me. What do y'all think?