Why do private schools do a better job of teaching science, when most of them are religious? This isn’t just a question that challenges our preconceptions about science and religion. It shows us where we need to rethink education policy as well.

To listen to the guardians of the education status quo, you’d never think it possible that religious schools could teach science—much less teach it better. Rob Miller, the superintendent of Bixby Public Schools, has scoffed that private school choice programs are for “a parent who wants to use the Bible as your child’s biology text.” Last year, groups representing public school special interests freaked out about theocracy in response to a completely anodyne bill to protect teachers from interference in “helping students understand, analyze, critique and review in an objective manner the scientific strengths and weaknesses of existing scientific theories.” The bill specifically said it would only protect “teaching of scientific information,” but special interests portrayed it as the second coming of Torquemada. Such views are common among public-school spokespeople and interest groups.

But federal data are clear: private schools, which are mostly religious, outperform public schools in science scores. On the National Assessment of Educational Progress, the so-called “Nation’s Report Card,” 8^th grade science scores in private schools average 151, in public schools 142. Private schools also outperformed public schools in each subcategory: physical science, life science, and earth science.

Why might that be? At the risk of stating the obvious, one factor is that private schools are better at teaching in general. A large body of top-quality experimental studies confirms that private schools outperform public schools even when serving the same students. These studies take advantage of school choice programs that admit applicants through random lottery, allowing us to compare how public and private schools perform when serving students who are differentiated only by random chance. Since private schools teach math and reading better, why not science as well?

The science results intersect with this area of education research in an interesting way. There is an old “creaming” canard, which holds that private schools only look better because their students are less disadvantaged (i.e. private schools only take the “cream,” the most advantaged students). This has long been undermined not only by the demographics of school choice programs—which tend to be more disadvantaged than average, yet perform better in private schools—but by this experimental research. Now, it also has to face the fact that private schools produce higher science scores. If the prevailing views among the special interests are right, then if anything, religious schools ought to do whatever is the opposite of “creaming” (“dredging”?) when it comes to students who are good at science. But their scores are higher.

On a deeper level, the plain fact is that most religious people, and most religious schools, are effectively indistinguishable from their less-religious neighbors when it comes to beliefs about science. Meanwhile, the minority of religious people and schools that really do dissent from some aspects of scientific orthodoxy do not dissent from all of it—and even where they do dissent, they still want their children to understand the claims of modern science. So, the majority of religious parents and schools approach science education the same way everyone else does, and the minority approach it in a way that is at least similar enough to equip students to understand the material.

Let’s start with the majority. Nobody can say precisely what it means to “believe in science.” That phrase contains paradoxes upon paradoxes, for the religious and the secular mind alike. Yet if the average American who doesn’t attend worship weekly “believes in science,” so does the average American who does attend worship weekly.

This is partly because the extent to which less-religious people “believe in science” is overrated. Consult your daily horoscope for guidance on whether secular reason and revealed religion are the only belief systems in modern America. Don’t worry, if you can still find a newspaper, you’ll have no trouble finding a horoscope—nearly every U.S. paper has printed them for generations, in spite of unanimous opposition to astrology from the world religions. If you’re a Libra, you can weigh the evidence and find that secular Americans are imperfectly rational. If you’re an Aquarius, you can pour cold water on the illusions of secular rationality. If you’re a Gemini, you can pour it twice.

The more important factor, however, is that ignorant people have vastly understated the extent to which religious people and institutions in the modern world “believe in science.” None of the foundational commitments of science—that nature works regularly, that the human mind is capable of discovering and describing that regularity—are in conflict with religion. That is why all the world religions have embraced modern science; indeed, the Christian assumption that nature and the human mind were made by a rational God was, historically, an essential precondition for the emergence of modern science.

Belief that miracles have sometimes occurred is no hindrance to science. On the contrary, you can’t believe in miracles as exceptions to the ordinary course of nature until you believe that the ordinary course of nature is rational and regular. And you can’t believe miracles serve to demonstrate visibly to their observers that nature is being disrupted—which is what miracles are for in the first place—unless you believe that the human mind is capable of knowing the regularity of nature (and hence knowing when it has been disrupted). Belief in miracles, far from contradicting the view that nature is regular and that we can know its regularity, presuppose this view.

And what of that great scientific-debunker’s boogeyman, evolution? The Roman Catholic Church, the nation’s largest private school operator, teaches authoritatively that God used biological evolution to create the human race. Huge numbers of Protestants, Jews, and Muslims believe the same. Hindu and Buddhist teachings accommodate evolution even more comfortably. I don’t share that view myself, but let’s be clear: I’m in the doctrinal minority.

These observations force us to recognize that we have to do better at distinguishing two questions. One is whether people ought to “believe in science” given their worldview, and the other is whether those people do in fact “believe in science.” Whatever you think about what people ought to believe, as a point of empirical fact the relationship between people’s beliefs about religion and their beliefs about science simply does not justify the confident assertions made about these beliefs.

And what about that troublesome minority that objects to scientific orthodoxy? To start with, the objections tend to be sharply limited. Yes, some of your neighbors believe the earth is only a few thousand years old (a belief that even most evolution skeptics, like me, find foolish). If you’ve never talked to such people and learned the full scope of their beliefs—it is possible to do that, they usually don’t bite, or at least not very hard—you might be shocked at how little effect that dissent has on how those people think about all other scientific questions. For 99 percent of science class, they’re just like everyone else.

And even on the points where they dissent, they virtually always want their children to understand the claims of orthodox science. They want their kids to function in the modern world, and they want their kids to be able to explain their beliefs. If you were moving to Mars and discovered that the Martians believed their planet was the husk of a giant magical coconut eaten millennia ago by an ancient god, you’d want your children to understand what the Martians were claiming, both so they could understand real geology and so they could live in Martian society.

Science is not a religion. That’s what makes it science. And that’s why it’s possible for people to accept the basic foundations of science—that nature is regular, that we can know its regularity—and almost all the particular claims of science, even if they dissent on a few specific points out of deference to what they believe God has said.

There are several lessons here for education reform. One is that too many people don’t bother to find out what their neighbors actually believe, because they’re too busy deciding what people of their worldview ought to believe. Another is that we don’t actually know in advance what systems and policies will produce the best results, and we should follow the evidence. And a third is that we should be skeptical when a self-protective educational monopoly tries to scare us out of allowing alternatives.