One of the "theories" I've had for a long time is that the smaller a proportion of a society's population atheists are, the stranger and more deviant they are going to be. A reason I came to this position is that read an account by an atheist American scientist who had some interactions with Soviet religious dissidents during the Cold War. His position was that in many ways American atheists and Soviet religious dissidents exhibited similarities in terms of personality, likely because they were generally not conformists. One of the peculiarities of the massive re-confessionalization of Russian society after the fall of the Soviet Union is the reality that these Communist era dissidents are now being marginalized in many congregations by recent converts who had a background as apparatchiks in the old regime, and were sometimes even actively involved in persecuting their current coreligionists! In any case, what about my hypothesis? Do I have any evidence for it? Not in any substantive manner. So I thought it might be interesting to look in the World Values Survey, naturally. How do attitudes of atheists and religious people vary within a society as a function of the proportion of each group?

I limited the sample to males, because men are more secular on average and exhibit more variance between nations. Additionally, because so many nations have very few atheists I put a lower bound of N = 20 for "convinced atheists." I mollified my own concerns about such a low N with the hope that if an N in a society is that low, the atheists may be strange enough indeed that their deviation from the social median may still swamp the noise. As before, the means for a class were calculated. So, the mean political self position of atheists and the religious is on a 1-10 scale. Below are are the charts for the results of a set of questions which exhibit a 1-10 level of agreement along a spectrum. The position is less important than the difference. First is a simple scatterplot which shows the attitudes of both the religious and atheists by nation. The expectation is a strong correlation between the religious and atheists, because most of the variation is naturally between nations. The second chart shows the difference between the two groups, "Religious persons" and "Convinced Atheists." I excluded those who were "Not religious" from the sample (so those who don't consider themselves religious, but neither are they professed atheists). Lastly, I plotted the difference between atheists and the religious as function of the ratio of religious to atheists. So, for example, the ratio of religious to atheists for Iraq is very high, atheists are a small minority (though to my surprise the N was large enough to stay above the threshold I put). In China the number of convinced atheists and religious are at parity, though those who are without religion and are not atheists are a plural majority.


Looking at these results I'm going to withdraw my model.

* For the "justifiable" questions 1 = never, 10 = always.
* Competition is good = 1, competition is harmful = 10.
* 1 = everything determined by fate, 10 = people shape their fates.
* 1 = gov. more responsibility, 10 = individual more responsibility.
* 1 = incomes more equal, 10 = we need larger differences for incentives.
* 1 = private ownership should be increased, gov. ownership should be increased.
* 1 = science makes world worse off, 10 = better off.
* 1 = Left, 10 = Right.

Labels: data, World Values Survey