As best I can tell, this article looks at resale data on e-readers and attempts to extrapolate that to use of e-readers. I doubt this extrapolation is appropri

As best I can tell, this article looks at resale data on e-readers and attempts to extrapolate that to use of e-readers. I doubt this extrapolation is appropriate. If, for instance, e-reader use was a constant proportion in all the cities, and the proportion of users who resell their e-reader was a constant proportion, but the proportion of users who resell their TVs was higher in the non "cultured" cities, using the authors' logic you would incorrectly reach the conclusion that the non "cultured" cities have greater e-reader adoption.