Only 32 percent of Republicans believe in evolution. And most Democrats (51 percent) don’t know both that the Earth goes around the Sun and that it takes a year to do so.

These are just some of the findings that emerge from a closer analysis of the underlying data in the new National Science Foundation Study. A lot has been written about the drop in the proportion of Americans (55 percent) who think that astrology is “not at all scientific” and about the 26 percent of the population who don’t know that the Earth goes around the Sun. Yet what has entirely escaped comment are some more interesting findings that can be gleaned only from crunching the numbers.

1. The Earth goes around the Sun and it takes a year to do so

In the NSF study, the National Opinion Research Center (NORC) at the University of Chicago asked:

Now, does the Earth go around the Sun, or does the Sun go around the Earth? [EARTHSUN]

That would seem to be an easy question, and if you didn’t know the answer, there was a 50-50 chance of guessing it right anyway.  Yet only 74 percent got that one correct.

For those who did answer correctly, a follow-up question was asked:

How long does it take for the Earth to go around the Sun: one day, one month, or one year? [SOLARREV]

The question is a multiple choice one, and there is a 33 percent probability of guessing the right answer. In the 2012 survey, only a bare majority of adults were able to answer both questions correctly: 55 percent of adult Americans responded both that the Earth goes around the Sun and that it takes a year for that to occur.

More disturbingly, in 2012 a majority of Democrats (51 percent) could not correctly answer both that the Earth goes around the Sun and that this takes a year. Republicans fare a bit better, with only 38 percent failing to get both correct.

As with astrology questions, conservative Republicans fare the best (67 percent correct on both questions), followed on this issue by Republicans overall (62 percent correct) and liberal Democrats (62 percent correct).

At the bottom are non-liberal Democrats—conservative Democrats (27 percent correct) and moderate Democrats (44 percent correct). For the full political breakdown, see Table 6 in “Who Believes That Astrology is Scientific?” The margin of error for the party and political orientation groups is 3.3-4.0 percent, while the margin of error for the four combination subgroups just mentioned is 4.5-7.8 percent.

2. Humans evolved from other animals

In the 2012 NSF survey, only 48 percent of Americans endorsed evolution as true:

Human beings, as we know them today, developed from earlier species of animals. [EVOLVED]

About 39 percent considered the statement false and 13 percent said that they don’t know.

The Republican and conservative refusal to recognize evolution is well known, but the extent of it may not be. As if the numbers for all adults (48 percent) aren’t depressing enough, only 28 percent of conservative Republicans believe that humans evolved from earlier species. In the next three spots are 32 percent of Republicans believing in evolution, 34 percent of conservative Democrats, and 37 percent of conservatives. (For comparison, 28 percent of fundamentalist Protestants believe in evolution, as do 27 percent of those who believe that the Bible is the literal word of God.)

At the other end of the spectrum, the political group most likely to embrace evolution is moderate Independents (68 percent), followed by liberal Democrats (66 percent) and liberals overall (66 percent).

In 2012 the NSF did a fascinating little experiment. On both evolution and the origins of the universe in a Big Bang, the NSF study used a split ballot. On evolution, a random half was asked the standard evolution question quoted above and the other half was asked a different question probing not belief but knowledge.

(The NSF report is somewhat confusing on how the experiment was done (p. 7-21), but on Thursday I confirmed with Tom W. Smith, who directs the General Social Survey for NORC, that an ordinary split ballot approach was used, one half getting one version and the other half getting the other version.)

The alternative evolution question asked whether it was true or false that:

According to the theory of evolution, human beings, as we know them today, developed from earlier species of animals. [EVOLVED1; emphasis added]

Because a rejection of evolution is mostly a belief, but is typically analyzed as a scientific knowledge question, the second version was designed instead to test the respondent’s knowledge of the theory of evolution. The results were quite different.

Overall, 71 percent of people agree that the theory of evolution involves humans evolving from earlier species, compared to only 48 percent believing in it. For Republicans the reversal is dramatic: on evolution only 32 percent of Republicans are believers, well below Independents (53 percent) and Democrats (53 percent). But in understanding the gist of evolution, Republicans (76 percent) are insignificantly ahead of Independents (71 percent) and slightly, but significantly ahead of Democrats (68 percent).

What this NSF experiment suggests to me (though other interpretations are possible)  is that some standard scientific knowledge questions do not actually measure what one knows, but rather what one chooses to endorse. For this reason, in its report the NSF did not include the evolution and Big Bang questions in its index of scientific knowledge.

UPDATE (WED, 6:33pm ET):

I am working on a substantial manuscript on the political correlates of scientific knowledge and plan to explore some of my larger findings in future posts when I’m a bit farther along. But I’ll reveal a bit about the NSF’s main 9-question index of scientific knowledge.  Like so many things related to education, the highest scoring political groups on this index in 2012 are on both the right and on the left: conservative Republicans, liberal Democrats, liberals overall, and Republicans overall.