Bugsy...
As I said before, the urban legend version contains no particular indication of where the population of inmates is from.

I would also advice taking your data from a table that represents all the inmates, not just the serious crime inmates...keep reading and you will see it is around 112,000

More investigation shows this difference to be well outside the standard error for the statistics....

Grey,

If you think that sample size has a "standard error" on surveys, then I think you need to finish your remedial stats classes.

The 'urban legend' version also does state where the data comes from, but there's no independent or verifiable publications that back it up, that's what makes it unreliable.

The reasons that you fabricated for dismissing the data are insufficient and erroneous.

So, I accept your apology on your error.

Grey: "such study MUST do a multivariate analysis"

OK, OK. I am not disagreeing. I was just pointing out it seems to be generally accepted by sociologists the correlation between crime rates and religion is lower than for other variables.

Its not a particularly relevant point. Its hard to say anything about the religious as they are a pretty broad bunch. In particular in most societies saying you belong to a church the default position of anybody who hasn't thought about it too much. That is why I choose people who specifically declared themselves to be atheists. Atheists seem to be a tightly self selected mob in comparision. They are people who have almost certainly thought about their moral position, and it isn't derived from any absolute source. Try as you might, I think you will have trouble finding any figures showing them to be over represented in prisons. This is in direct contradiction to what the article says.

Grey: "Regarding your hypothetical dialog."

The dialogue wasn't written by me, if that is what you mean. It was posted by the Graham - the author of the article. It is his favourite line of argument. He says a mind that doesn't believe in absolute morals is not compelled by logic to follow any morals, and ergo they don't. The problem with the argument is the mind is not a logic machine. People do things that aren't logical all the time. That in turn means it's irrelevant whether objective derivation of morals is possible.

This is a different like of reasoning to my last post, but arrives at the same conclusion. And it is of course it is again the reverse of what Graham argues in his article.

I am surprised nobody has bothered to quote another set of figures. Unlike the ones I gave, they are pretty clear. There is a strong negative correlation between how secular a society is and its crime rates.