I've pointed to the World Values Survey before. It comes in 5 waves spaced out over 2 decades, and has substantial, if not total, coverage. Additionally, for many non-developed countries the educational data to me suggest some high SES skew in terms of representativeness (though spot checking the American data that looks very representative, as there have been other national surveys you can cross-reference it with). On some of my blogs a few commenters have started to follow up posts and use the WVS to answer questions, instead of offering of speculations. It's not as complicated of an interface as the GSS, but it isn't as flexible either. Nevertheless, there are some obvious questions one might ask.

For example in general within societies the religious have more offspring than the non-religious. Even controlling for variables there is often a significant effect. That implies that over time if religiosity is heritable (whether biologically or culturally) societies should become more religious. So a priori assertions such as Mark Steyn's that Turkish secularism is doomed because the rural religious have outbred the citified secularists seem plausible. The WVS can help us answer this sort of question.

For example, if the religious are outbreeding the non-religious and religion is substantially heritable so as to counteract any rate of defection than younger age cohorts should be noticeably more religious, right? Are they in Turkey? I use Turkey as an example to illustrate how useful the WVS can be.

So first go to http://www.worldvaluessurvey.org/

I've circled some areas red to click through.

Click the area where I've circled read. You need to jump through some hoops (it uses POST to go from page to page).

I've broken down the importance of religion as a function of age. There is no trend toward greater religiosity among the young.

I've now broken down by both and age & sex. As in most societies secularism is more pronounced with youth among males.

I went back and looked at another question in regards to the influence of religious leaders on voting. There is no trend of younger people being more supportive of this. There are plenty of other religion & government related questions you can ask. When Steyn made that assertion I made sure to remember to poke around Turkey's WVS results, and they don't seem to support it. The theory is coherent, but the facts do that match. I hope this is a lesson for readers. Theory provides free information. But since there are tools to check inferences one makes from assumptions one should do so before taking theory as a given (all the above took me 3 minutes, excluding screen capture & Photoshop).

Labels: Religion, World Values Survey