Recent estimates are based on those 5000-man polls, and seem to be self-reported. Thus the diagram in your diary is consistent (coming from the same polls). The pdf mentions that the poll method, too, has been criticised: differential willingness to answer pollsters' question could be a factor. They argue that these criticisms are countered by some studies on second-generation immigrants, and that weighting can compensate the effect; however, I am not sure I understood their argument. *Lunatic*, n. One whose delusions are out of fashion.
The pdf mentions that the poll method, too, has been criticised: differential willingness to answer pollsters' question could be a factor.
To be more explicit: the article seems to imply thst unwillingness or inability (due to language) to answer polls is a problem chiefly among first-generation immigrants. I don't get how studies on the secularisation of second-generation immigrants enables to control that in any way by weighting. But, I only read via Google trnalate -- Nomad, could you read that section (chapter 4, section 3, page 36-37) more thoroughly and 'report' if you 'got' more out of it? *Lunatic*, n. One whose delusions are out of fashion.
