i am still confused about the article you posted, even more after looking into it a little...the basis for my suspicion really is the use of aggregated data to characterize what appears to be a socially diverse area. i found a crime data map here:
http://pgcrime.info/
and looked at the homicide data since january of this year (before i got distracted and went outside to look at 15 ducks wandering around and a boat that was drifting up the river) and they're concentrated in a tight ring around the edge of washington. assaults are concentrated in the same area--which makes me wonder what's up in the immediate area around washington dc.
somehow i think the article is just way to simple--milloy even blames hip hop at the end of it for all this. which is nonsense.
i'm not saying that the simple reverse of your argument is always necessarily the case--i just think this information is curious--and getting past aggregation effects is kinda tough in the internet from essex massachusetts (presumably from elsewhere as well).
on the prison population of new hampshire--cynically, because i grew up there, nothing really surprises me. but that's just cynicism. now i have another datapoint to think about.
but on the pg county thing, i just am not sure of what's actually going on. the reporter seems to do a court beat, so works outward from police information--which is always a dicey affair. he also doesn't seem to like mentioning economic class very much--i read a bunch of his articles and it just doesn't figure in his reporting. i can't say why it isn't a variable exactly. but it's curious nonetheless.
perhaps class is out of fashion amongst bourgeois journalists.