A Positive Step for the National Science Foundation

Since 1979, the National Science Foundation (NSF) has been producing a study entitled Science and Engineering Indicators. It is a quantitative review of science and engineering progress in the United States and the rest of the world. One chapter from that report is called “Science and Technology: Public Attitudes and Understanding,” and it attempts to assess how the people of the United States view and understand science compared to the people in the rest of the world. The way they try to gauge the public’s understanding of science is to produce a survey that asks questions such as, “How long does it take for the Earth to go around the Sun?” and “True or False: The center of the earth is very hot.”

For 20 years now, two of the True/False questions on that survey have been:

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

The universe began with a huge explosion.

According to the journal Science, two expert panels formed by the NSF’s governing body, the National Science Board, have suggested changing these two true/false questions to:1

According to evolutionary theory, human beings, as we know them today, developed from earlier species of animals.

According to astronomers, the universe began with a huge explosion

The National Science Board has decided to ask the NSF to make that change on half of the surveys given out next time to see what effect it has on the results. This suggestion has infuriated some, but I see it as a very positive step for the NSF.

Why did the expert panels suggest the change? Because there are plenty of people (and plenty of scientists) who understand evolutionary theory very well but do not think it is correct. We look at the data and see that it argues against evolutionary theory. Thus, while we understand what evolutionary theory says about human origins, we know the data well enough to understand that what it says is not correct. As a result, while we would answer “false” to the original version of the question, we would answer “true” to the modified version. Since the NSF is looking for “true” as the correct answer, the modified versions of the questions are better gauges of a person’s scientific knowledge of those subjects than the original versions of the questions.

As I said, the proposed change has infuriated some people. For example, the Science article I referenced quotes Dr. John Miller, a science literacy expert at the University of Michigan as saying:

If you are altering the questions in that way, you are doing it for religious reasons…We don’t make statements like ‘According to some economists, we had a recession’ or ‘According to the weatherman, we had a tsunami.’

In fact, Dr. Miller’s statement shows that he doesn’t understand evolutionary theory very well. We don’t need to say, ‘According to the weatherman, we had a tsunami,’ because a tsunami is easily verified. We can look at the data and quickly decide whether or not a tsunami occurred. Things aren’t so easy for evolutionary theory. The data must be heavily interpreted, and they often provide evidence against the theory. This, of course, is why the panels didn’t suggest qualifying other questions, such as the one about how long it takes for the earth to go around the sun. The fact that it takes the earth one year to orbit the sun is easily verified by many observations. Thus, no qualifier is needed. Since the same statement cannot be made about evolution or the beginning of the universe, scientific statements regarding such topics need to be qualified.

The reason I think this is a step forward for the NSF is that the survey is designed to gauge scientific knowledge, and you don’t measure that by finding out how many people mindlessly believe what they are told to believe. Instead, you measure it by finding out how much the people actually know about science topics. I know more about evolution than many who believe in it, and that knowledge makes me realize that the data speak heavily against it. If the NSF really wants to see what people know about science, it needs to word its questions so that critical thinkers who are willing to evaluate what they are told are not penalized for their critical thinking.

I expect the NSF will find that when they review the surveys with the modified questions, they will see a measurable increase in the number of correct answers to those two questions.


1. Yudhijit Bhattacharjee, “New NSF Survey Tries to Separate Knowledge and Belief,” Science 333:394, 2011.
Return to Text

23 thoughts on “A Positive Step for the National Science Foundation”

  1. Wouldn’t the answer to The universe began with a huge explosion be False? Changing the question to according to astronomers still leaves the correct answer as false. Most astronomers I’ve met say the universe began with a very rapid expansion of space and matter, not an explosion where something is blown apart into pre-existing space. I’m surprised the NSF worded the question the way they did both times.

    1. WSH, while the word “explosion” is oversimplified, I do think it is an adequate description of the Big Bang. I don’t think the word implies something that is blown apart into pre-existing space. Instead, it just means a rapid, hot expansion, and that is how the Big Bang models the beginning of the universe. It is a common usage. A NASA website that discusses a gamma-ray burst says, “The energy released in a cosmic gamma-ray burst detected in December 1997 is the most energy ever detected from an explosion in the Universe, perhaps making it the most powerful explosion since the creation of the Universe in the Big Bang.” The Free Online Dictionary even defines the Big Bang as, “The cosmic explosion that marked the origin of the universe according to the big bang theory.”

      So while a specific usage of the term “explosion” doesn’t fit the Big Bang, if you think of the broadest possible definition of the word, I think it fits the theory adequately.

  2. Thanks Dr. Wile! This article on the whole really surprised me. Let’s see how long NSF makes it until they retract this due to the outcry.

  3. I don’t get this. The way the question is framed, it seems that NSF is implying that there are alternative scientific theories to Evolution on the origin of human species or at the very least conceding that there are valid scientific objections to common descent. Is that really the intention of NSF? If not, changing the statement doesn’t make any sense. It is similar to changing “Earth revolves around the Sun” to “According to Heliocentric theory, Earth revolves around the Sun”.

    1. NoOne, I don’t think the NSF is ready to concede that there are alternative theories to evolution. However, they are willing to concede that there are very knowledgeable people who do not accept evolution. This, of course, is the first step along the road of self-correction in science. When a few knowledgeable people express skepticism about a well-loved theory, they are written off as crackpots. As more and more knowledgeable people express skepticism about a theory, they can no longer be written off as crackpots. As a result, the theory starts receiving more serious scrutiny. If those knowledgeable people are correct, the serious scrutiny causes the theory to be overthrown.

  4. NoOneKnows, is it a scientifically accepted fact that the Earth goes around the Sun? Or are they mutually attracted to one another with equal forces, though the movement of the Sun is much smaller due to its greater mass? While for a planet and star orbit the distinction is almost impossible to see, for other orbits (such as pairs of stars) it is readily apparent that they are both going around some point between their individual centers of gravity. Perhaps it would indeed be wise to be a bit more cautious about the declaration.

  5. Two things based on this news article: http://www.bbc.co.uk/news/science-environment-15391388

    Disproving Clovis-First is a paradigmatic paradigm shift.
    How does YEC disproving evolution compare to it?

    The article mentions new radio atomic accelerators that have very small error bars. In this case 13,800 years ago, plus or minus 20 years. Is there a YEC critique of this new dating method? How does YEC estimates of the age of the earth, 10,000 plus or minus 4,000 years as far as I can tell, compare?

    1. Thanks for the questions, Eric. I have never contended that YEC disproves evolution. In fact, in my textbooks (and in a blog post) I make it clear that science can’t prove anything. Since I cannot prove YEC, I cannot disprove evolution. Any theory about what happened in the past is very difficult to prove or disprove, because the evidence is heavily dependent on interpretation. For example, with the piece of evidence in the story you linked, I could say that the thing in the mastodon’s rib is not really a weapon at all. It is an unusually-sharp rock that the mastodon fell on. It is a desperate explanation to be sure, but you cannot refute it with 100% certainty. Thus, you cannot disprove the Clovis-First model. You can say it is hard to believe based on this evidence, but you can’t disprove it.

      Using accelerator mass spectrometry for radiometric dating is not new. The lab where I got my PhD (back in 1989) was one of the leading accelerator mass spectrometry labs, and it did a lot of radiometric dating. Thus, the technique has been around for a long, long time. The error bar on the date has nothing to do with the uncertainty in the date. It only tells you the uncertainty in the measured amount of carbon-14 in the sample. They simply convert that to date. However, the conversion requires that all the assumptions necessary to go from amount of carbon-14 to time are correct, and we know that some of them are not.

      Probably one of the most devastating pieces of evidence against carbon-14 dating is that samples (including diamonds) which are so old that they should have NO carbon-14 in them have significant amounts of carbon-14 in them.

      I personally think that the YEC estimates are more reliable, as they are based on more valid assumptions. Nevertheless, either way you go, there are assumptions involved.

  6. I’m surprised, Dr. Wile. I thought you were a fan of Karl Popper’s philosophy.

    I agree that there are some situations where something can neither be proven nor disproven simply because there is not sufficient evidence, especially after time and weathering set in against what evidence there is. However that doesn’t imply that science is unable to disprove anything on principle.

    1. Josiah, the website you give simplifies Popper’s philosophy significantly. Remember, the reason science cannot prove anything is that all the evidence is based on experiments or observations, and those experiments or observations can be flawed. Well…how do you falsify a theory? You do so with experiments or observations. Since they can be flawed, a falsified theory is not disproven. It is simply falsified given the level of evidence currently available. If the experiment or observation that falsified the theory is shown to be flawed, then the theory is no longer falsified.

      Now once again, just as you can have a theory that has so much evidence that it is hard to believe it to be wrong (such as the heliocentric theory), you can have a theory that has been falsified by so much evidence that it is hard to believe it could be right (such as the phlogiston theory). Nevertheless, if you cannot prove something with science, the logical conclusion is that you cannot disprove something with it. You can only pile up evidence one way or another.

      The reason Popper thought that falsifiability (not disprovability) is necessary for something to be scientific is that if you can produce evidence for a theory, you logically must be able to produce evidence against the theory. Any theory that is so plastic that it can accommodate any data is not (in Popper’s mind) a scientific theory. While nearly every scientist agrees with Popper that you cannot prove anything with science, not all scientists agree that falsifiability is the proper test to determine whether or not a theory is scientific.

  7. I think the change to sticking “according to evolutionary theory” in would be a good idea. It’s a reminder that evolution is a *theory* not a law,and that there are plenty of other theories out there.

    This is a completely honest question, not one intending to stir up trouble,I promise. Why does it even matter how old the earth is or how God created it? I believe that God created the Earth and the entire universe, and honestly it makes no difference to me how He did. That He did is enough for me, and I honestly fail to see how trying to figure out if the world is 6000 or 6 billion years old is important.



    1. Vivielle, I agree with you that knowing how God created is not nearly as important as recognizing that He is the Creator. However, I do think the age of the earth is an interesting question for at least two reasons. First, there are several ways the creation account has been interpreted by orthodox Christians over the years. If science is able to come up with a well-supported answer for how old the earth is, that might help us understand which of the orthodox interpretations is correct. Second, it is important to know how old the earth is so that we can understand certain natural processes that occur. For example, if the earth is ancient, then the idea that the majority of the geology we see today is the result of uniformitarian processes makes sense. As a result, the general history of the earth as given by the geological column might be accurate. However, if the earth is young, then uniformitarian processes cannot be responsible for most of what we see in geology. As a result, we need to understand earth’s history in terms of catastrophism. If nothing else, then, the age of the earth is an important scientific question. In fact, that’s what fascinates me about the subject. I am theologically comfortable with an ancient earth. I’m just not comfortable with it from a scientific point of view.

      I think determining how God created is an important scientific question as well. It doesn’t really affect the main points of orthodox Christianity, but if we know that God created specific kinds of creatures and they simply vary within that kind, we can start to figure out the limits of biological change. If we know that God created by some macroevolutionary process, then we know that given enough time, there are very few limits to biological change. Once again, then, I think our understanding of science will be helped if we understand how God created.

  8. Thank you for your respectful and well-thought out answer! I really appreciate all the time you take to answer all the questions that blog readers ask you.

  9. “Since I cannot prove YEC, I cannot disprove evolution.”
    That particular statement is true, but the general I cannot prove X, therefor I cannot disprove Y” where X and Y are competing theories is not necessarily true. You are stretching the non-provability of scientific theories too far.

    For instance, “Since I cannot prove the earth goes around the sun, I cannot disprove the sun goes around the earth.”
    Well, yes you can disprove geocentricity. There are observations that disprove it. The correct opinion that scientific results are provisional does not mean we cannot disprove all kinds of things. In fact, the essence of science is disproving things. I would expect you to agree to that.

    In your case, science has disproven YEC. The earth is older than 20,000 years old. Of course, it’s a lot older than that, but for the purposes of disproving YEC, that is enough.

    Which brings me back to my implicit question that you avoided: Why hasn’t YEC science narrowed down the age of the earth more than error bars of 40%? And what is a consensus on the median estimate? How many believe 10,000? How many believe 6,000? Why do YECers favor one or the other?

    1. Eric, I don’t think I am stretching the non-provability of scientific theories too far. I would strongly disagree that you can disprove the sun goes around the earth. You can stack up reams and reams of evidence to show that the idea is wrong, but as you so correctly point out, that evidence is based on observations. The problem is, observations can be flawed – sometimes quite seriously! The history of science is full of experiments that claimed to demonstrate the incorrect nature of a view, but many of those experiments were later shown to be flawed, vindicating the view that they supposedly “proved” incorrect. While I seriously doubt the observations that demonstrate geocentricity to be incorrect are flawed, a good scientist must leave open that possibility.

      And, of course, science hasn’t come close to even offering serious evidence against YEC, much less “disproving” it. In fact, the more we learn about nature, the more the YEC view seems to be supported by the data. Science has not shown that the earth is older than 20,000 years old. However, even if it had, that would not “disprove” YEC. I know many YECs who think the earth is older than that. For example, John C. Sanford, the brilliant geneticist, thinks that the earth is between 5,000 and 100,000 years old.

      I did not avoid any of your questions. Here is what you asked:

      How does YEC estimates of the age of the earth, 10,000 plus or minus 4,000 years as far as I can tell, compare?

      I answered that with the following:

      I personally think that the YEC estimates are more reliable, as they are based on more valid assumptions. Nevertheless, either way you go, there are assumptions involved.

      I also explained how the “error bar” on the carbon-14 date you mentioned is not an error bar on the date at all. Instead, it is an error bar solely on the detected level of carbon-14. Thus, it does not take into account any of the inaccuracies in the assumptions required to turn carbon-14 level into an age.

      Now that you are asking a completely different question, I am happy to answer it as best I can. YECs are honest when they approach the data. Thus, they do not report error bars that are essentially meaningless, such as the error bar listed with the carbon-14 date you mentioned. Instead, they try to realistically discuss the errors involved in their data. Given the fact that we are trying to extrapolate over long periods of time compared to the time over which actual measurements have been made, any realistic error bar will be very large.

      I don’t know that there is a consensus on the median estimate for the age of the earth. I talk with many YECs, and there seem to be a wide range of estimates. As listed above, for example, the best that JC Sanford can say is somewhere between 5,000 and 100,000 years old. That seems to be a good representation of many YEC scientists. The problem is that several dating methods give several different answers for the age of the earth. Unlike many scientists, YEC scientists don’t simply choose the ones they like and ignore the rest. They actually try to see what the data really indicate. This leads to a broad range of opinion, since the data indicate a broad range of ages.

      Most YECs favor one dating method over another because they are more comfortable with the assumptions of the dating method or the evidence that stacks up in favor of the dating method. I personally am rather open to various dating methods, because I see strengths and weaknesses in them all.

  10. Before I forget – I use Firefox and the comment box is larger than the space allowed, so that after the words reach the right edge, they disappear until I’ve typed enought that the words pick up at the left again. Very annoying.

    I think your mind is so open that it has fallen out. A scientific answer would be to state your estimate, provisional as it certainly is, and then provide evidence (you seem to overuse “data” which could be irrelevant to to a given hypothesis, when what you mean is evidence) for your estimate.

    “Given the fact that we are trying to extrapolate over long periods of time compared to the time over which actual measurements have been made, any realistic error bar will be very large.” That is hogwash.

    Rather, there is an issue when the error bar is the same order of magnitude as the estimate. Saying the earth is 52,500 years old plus or minus 45,000 years is ridiculous.

    1. Eric, you seem to be missing the point that the carbon-14 error bar is not a real error bar at all. It has no bearing on how well the date is known. In my view, the error bar on that carbon-14 date is essentially 100% of the value of the date, since the carbon-14 dating system is so hopelessly flawed.

      I am not sure you have been paying attention, because I have stated an estimate for the age of the earth. I tend to agree with Dr. Sanford that the earth is somewhere between 5,000 and 100,000 years old. I expect it is much closer to 10,000 years old, but that is probably because I tend to favor using the earth’s magnetic field as a dating method. It enjoys a wide range of support and has a lot of predictive power, which is why I favor it. This method gives an upper limit of 10,000 years for the age of the earth. However, as a scientist, I am bound to follow the data. There is a wide enough range in the data that I must allow for a wide range in the age of the earth.

      You might think that being scientifically careful is “hogwash,” but I do not. Extrapolation is a very tricky thing, and a careful scientist only extrapolates over a small range compared to the range of measurement. When the range of extrapolation is large compared to the range of measurement, the conclusion becomes less and less reliable. Most serious science textbooks will discuss the limits of extrapolation. If you don’t want to bother to look in a textbook, you might read this article.

      It is quite impossible for a scientist to overuse data! The data must be followed, whether you like what they indicate or not. Since the majority of the scientific data indicate a young earth, I am bound to follow that data, whether I like it or not.

      I use Firefox as well and have no problems.

  11. Oops, plus or minus 47,500. 90% of the estimate. As opposed to the carbon-14 error bar, which is .1% of the estimate.

  12. I used to have those problems with firefox on this comment box but it’s sorted itself out, I think in an update to a recent version. There do seem to be technical issues going on though. For example there seem to be phantom comments; it reports “one comment to read” but doesn’t actually show any when you read the post!

    1. Josiah, the annoying “missing comments” are pingbacks. Somehow, when a blog links to one of my articles, it appears on the control panel as a comment. I can’t seem to make it show up on the user’s side, and I can’t keep it from being counted as a comment. Of course, it isn’t really a comment. It is just someone linking to the article.

  13. I am using Firefox 4. It’s too bad you can’t display pingbacks or trackbacks. I would like to see what other people say about your posts.

    1. I agree, Eric. I would love to be able to display them. As for the comment box problem, I am using Firefox 5.0, so that might be why I don’t have the trouble.

Comments are closed.