February 18, 2004

Need some scary correlations to go with the scary data.

Posted by Arcane Gazebo at February 18, 2004 5:51 PM

After a day's worth of measurements, I have concluded that our chip does not, in fact, work. This at least frees me up to go write the follow-up to Monday's scary poll numbers post.

First off, Polling Report has a nice summary of the data with the actual questions ("word-for-word", as it were). In my initial reaction to the results I wondered to what extent factors like quality of science education and anti-science attitudes contribute to the high rate of biblical literalism. It occurs to me that if the poll results included correlation data one could actually get some insight into this. Clearly some of the Bible stories in the poll have more plausibility in a literal interpretation than do others, and it could be revealing to look at how many people only believe some of the stories are literally true.

The results as presented are basically consistent with almost any set of correlation data; consider for example two populations:
(A) 60% are true biblical literalists, 30% are atheists, and 10% have no opinion, or
(B) 10% have no opinion, and the remaining 90% believe two of the three stories are true, but they are equally divided in which story they believe is false.
The results of the poll for both (A) and (B) would match each other and the actual results to within 5% in every number, but (B) doesn't have a single member who thinks the entire Bible is literally true while (A) has a majority.

It seems to me that a population like (B) is likely only if all three stories are equally plausible (or, I suppose, if respondents are answering essentially randomly). This doesn't appear to be the case to me, although I acknowledge that the very results of this poll are evidence that my standards of plausibility are very different from the average person's.

Anyway, the Noah's Ark story seems clearly to be the least plausible. All it takes is some common-sense thinking about any of the numbers involved to cast a whole lot of doubt on the idea. So this story rightfully gets the lowest number of responses, but only by one percent. This is pretty stunning. I would think that if one is willing to grant that at least some Bible stories may not be literally true, this would be the first one to go. And yet, far fewer people claim to be absolute Biblical literalists than think that the Noah's Ark story is true (see the poll just below this one on the Polling Report page).

By contrast, the creation story doesn't violate any rules of logic (except for the part where there are two creation stories and they contradict each other). If one were unaware of geology and paleontology and so forth, one could at least imagine that it might be true. So we would expect more people to believe this one than the flood story, and the difference perhaps due to lousy science education. We're right about the first part at least, but it's a small effect (that one percent).

Finally, Moses parting the Red Sea is the most plausible of the three (unless you look at a map, and realize that the Red Sea isn't between Egypt and Sinai, but we know how Americans are about geography). If one believes that divine intervention sometimes occurs, or used to occur, there's nothing about this particular miracle that makes it especially dubious. I would imagine that people who disbelieve this story are the sort who would disbelieve any supernatural occurrence chronicled in the Bible. On the other hand, it's hard to believe that 28% of Americans are this sort of person.

The actual results are in fact consistent with this ordering of plausibility, but the differences are small. One possible conclusion one could draw is that the spectrum of viewpoints from wholly literal to wholly metaphorical is not well populated in the middle -- most people are either one or the other. But it would be nice to have the actual correlation data.

Post a comment