This is a flaw in their statistics. Just eyeball the map, and you can see the red counties are in areas of remote/economic depravation. Over the course of 100 y

This is a flaw in their statistics. Just eyeball the map, and you can see the red counties are in areas of remote/economic depravation. Over the course of 100 years, you are getting "the thin end of the gene pool" as others move out. So, your baseline samples are not the same. Also the gene-pool will be more startified by education as time goes on (ie, its more "common" to have edu degrees), so as the weaker become educated again only the weakest do not. So their is instability in the relationship between the proxy variables for "class" and success and the genetic ones. This article is a good example why researchers (and reporters) need to be boots on the ground and understand what they are dealing with. None of this would be surprising to anyone who hase been to many of the remote rural areas flagged in red here.