Rodney Stark's strange interview w/ Kathryn Jean Lopez

I can't remember the last time I've seen an interview as strange as the one National Review published recently, where Kathryn Jean Lopez asks sociologist Rodney Stark about his new book on the evidence of the benefits of religious adherence. 

This really reads like an interview where something has gone horribly wrong:

LOPEZ: Religion can keep me from mental illness? My inbox suggests it is evidence of my mental illness.
STARK: Several hundred studies are unanimous that frequent church-attenders are far less likely than non-attenders to suffer from mental illness — I devote many pages to the matter.  

LOPEZ: The 'higher the church membership of a city, the lower its crime rates.' What evidence do you have for this contention?
STARK: I cite many published studies. 

LOPEZ: How can you prove that 'religious parents are better parents, who raise better-behaved and better-educated children'?
STARK: I cite a very large research literature.

It continues from there, with Lopez asking strange questions that seem -- best I can guess -- to be bad imitations of the kinds of questions she thinks atheists and/or liberals would ask. The gag, maybe, is that all of these things are obvious, and so obvious as to obviously not need evidentiary support. But now there is statistical evidence ... so, hahaha. Or something.

Stark, who's done some important work in the sociology of religion but has also been sharply criticized for some of his work, and, generally, seems way over-defensive when questioned about the conclusions he draws from his facts, gives answers that equal the questions in oddity. With some answers, he sounds like your standard resentful conservative ("the media are dominated by the irreligious. So are universities". With others, Stark sounds as cranky someone with a toothache listening to a baby cry.

One of his answers, literally, is "See Scandinavia."

I suspect there are serious problems with this book. The whole thing appears to be based on a basic misunderstanding of evidence, and the kinds of correlation-causation confusions that one would hope an accomplished sociologist would understand and avoid. I'm disinclined to read the book and find out if that's the case, though, after this very weird case of book "promotion."

Update:
Via twitter, Per D. Smith suggests that it's been hard to take Stark seriously as a sociologist for more than a decade, citing the article "Atheism, faith, and the social scientific study of religion" in the Journal of Contemporary Religion from 1999.