This Thursday I’m giving a talk at Oakland University in Michigan on the evidence for evolution and the reason why Americans reject it. Drop by if you live in the area, and I’ll be glad to sign copies of WEIT (books will be on sale there). There will also be a secret word, announced later, that will get you a cat drawn in your book.

In my talk I’ll show the following slide, which reveals a negative correlation between belief in God and acceptance of evolution among 30 European countries, Japan, and the U.S. The correlation is highly significant and, as you see, the U.S. is second lowest in accepting evolution, with only Turkey below us. I’ve shown this figure here before, and won’t dilate on it again. You can find the relevant data, statistics, and discussion in my Evolution paper from last year about the relationship between evolution, science and society (free download).

Here’s a plot of the data on acceptance of human evolution by country; the figure is from Miller et al. (2006; reference below). The arrow shows the U.S. at the bottom, deeply shamed by our position relative to Japan and the countries of Northern Europe.

This got me wondering whether there was a similar correlation between religiosity and acceptance of evolution among the states in the U.S. I felt that there must be, simply because religiosity (as is well known) is higher in the southern U.S. than in the north, and the south also contains more people who reject evolution. Doing a bit of Googling, I found data on acceptance of evolution by state at Subnormal Numbers. Whoever writes that site took raw data from a 2010 Pew Forum survey and converted them into a bar graph similar to the one above for countries. (I haven’t dug up the original data yet.) Here it is:

Just to check my intuition, I found a “State of the States” Gallup Poll from 2009 that broke down American religiosity by state, and also listed the ten most and ten least religious states. The survey assessed “religiosity” by asking the question, “Is religion an important part of your daily life?” Of course states vary widely: at the top for non-religiosity are states like Vermont (only 42% “yes”) and New Hampshire (45% “yes”), while southern states like Mississippi and Alabama are highly religious (85% and 82% “yes,” respectively). I then put colored arrows representing the ten most religious states (red) and ten least religious states (blue) on the “acceptance of evolution” bar graph.

You can see the results above: there’s no overlap between the arrows. The least religious states show the highest acceptance of evolution, while most religious states show the most denial of evolution. This is, of course, statistically significant just by a ranking test: the chances that the arrows would separate into two such non-overlapping groups by chance is minuscule. I’ll eventually do a full correlation when I get my hands on the raw evolution data.

The figures among states clearly aren’t independent, for there is geographic correlation of religiosity (and evolution denial, which comes from religion), so each state does not represent an independent test of the question “Is there a correlation between religiosity and acceptance of evolution among American states?” Nevertheless, these data do support a “yes” answer to that question, and that answer in the expected direction: the more religious a state, the higher the proportion of its inhabitants that reject evolution.

The correlation is no surprise, of course because all opposition to evolution stems from religion. But it was still surprising to me how strong these data are. (The correlation for the 32 countries at the top is correlation is −0.608, and the probability that this could arise by chance less than 0.0001).

Along with everything else that religion poisons, we can include acceptance of evolution. That holds not just across the globe, but also within the U.S.

____________

Miller, J. D., E. C. Scott, and S. Okamoto. 2006. Public acceptance of evolution. Science 313:765-766.