Hm... I get your points.
My intervention started following a discussion with a friend (and then with a friend of his) on facebook about the matter.
Somewhere you will see our discussion. Also there are a couple of intriguing references by Elizabeth Eisenstein, my mentor at RAND and one other provided by a participant (by Walter Russell Mead )
It is amazing the extrapolations one can do about mass behaviour with only a fraction of the total population.
From experience I can tell you, that nationalities et al do not play the role that... nationalists :) would expect. I mean I used the work I did for the Scottish Referendum, to the English and the Welsh when I predicted their elections, then to the Greeks for the Referendum and Elections, then back to the Brexit and then again for your 2016 elections.