Thanks for the lnk Prof
I don't doubt the validity of the study at all, or its findings, but I do wonder... Drawing statistical relationships is fine, and very interesting, but what is the controll group / data? I struggle with this whenever any claims are made for the roll of religious belief in having beneficial or limiting affect on a persons moral behaviour. If it's reduced to an individual level then how can we possibly know if a person would be 'better' with or without a religious faith? Based on anecdotal evidence we might expect a conversion too or from religion to have a similar effect, depending on the person.
The danger, I think, is in taking a statistical likelyhood and using it to form a generalised views on the value of religion. Reasons for participation in religion are far more complex than simply for moral betterment. I am excited by what this study reveals but a little uncertain about how relevant or meaningful it can be on its own. It would be good to see a compilation of all the various studies done on the impacts of religion, as I'm sure there must be loads - social, economic, psychological, educational... Sorry this was a bit rambling.
And listen, I am not a researching scientist and am happy to concede to your better informed views. These are just my thoughts.