Who are the creationists?


This year brings us yet another survey trying to diagnose what's wrong with creationists.  Oh goody.  I've commented on such studies before, and I'm always interested in what else I can find from the survey data that the authors of the research didn't really highlight in their publications.  Here's a sample of past "fun with surveys":


So here we go again, this time with Weisberg and colleagues from the March issue of BioScience.  In a survey of 1,100 people, they tried to examine factors that influenced respondents to be creationists and found several of the usual suspects.  For example, they found that the more religious you are, the less likely you are to accept evolution.  Likewise, those who self-identify as liberal or very liberal are more likely to accept evolution, and conservatives and "very conservatives" are more likely to reject it.  No surprise there.

Their study added a few new things, which I thought were very helpful.  The biggest innovation was their Evolution Knowledge Test (EKT), a series of 26 questions that examined participants' knowledge of evolution.  Not surprisingly, creationists tended to have substantially less knowledge about evolution than non-creationists.  Or so it seems.

The other innovation they added to the survey was a revision of the Gallup evolution question, which focuses on human evolution.  Gallup gives three options when asking about the participant's view on how humans got here: humans evolved with guidance from God, humans evolved but God had nothing to do with it, and humans were created by God in their present form.  That's an interesting question, but it's got some limitations (human origins are emotionally charged and might bias answers, for example).

So Weisberg and colleagues re-wrote the question and answers.  Here's their version:

Which of the following best describes how YOU think animals and plants came to exist on earth?
  • Animals and plants were created by God in more or less their current form
  • Animals and plants developed through natural processes, which were guided by God the entire time.
  • Animals and plants developed through natural processes, which were set up by God but continued on their own.
  • Animals and plants developed entirely through natural processes.

Their version of the question takes away the bias of asking about human beings, and it also gives another option, which they call the Deist option ("God set up natural processes and let them go").  With that single question, they divide people into Creationists (first answer), Theistic Evolutionists (second answer), Deists (third answer), and Naturalists (last answer).

I find this strategy better than the Gallup question but still unsatisfactory.  First of all, if you haven't thought a lot about evolution and creation, you probably won't understand the question very well because it's got some specialized jargon.  On the other hand, if you have thought a lot about creation and evolution, you might not be happy choosing any of those categories, or you might even think that more than one category applies to you.  So right off the bat, I have trouble with the rest of the study, and I think some of the results are influenced by this drawback.

For example, much to my surprise (and delight, if I'm being honest) the Creationists and Theistic Evolutionists did equally poorly on the EKT.  You can easily see that in the distribution of scores:

The mean EKT score for all 1,100 participants was 53.84%, which corresponds to about 14 questions.  The mean EKT scores for Creationists and Theistic Evolutionists were 46.33% and 48.46% respectively, but the distribution of their EKT scores was indistinguishable.  So on average, Creationists do as poorly on the EKT as Theistic Evolutionists.

Anecdotally, I think that finding may well be true.  It has been my experience talking with young evangelicals in particular that they accept evolution even though they don't really know much about it.  It's kind of the trendy thing right now.  In this case, however, I'm just dubious of the question that divides the camps into Creationists and Theistic Evolutionists.  I'm not persuaded that the respondents understood the choice they were making.

Not convinced?  There's more.

A surprising 44.7% of Creationists in the study did not identify as "born again Christian," despite creationists being predominantly evangelical Protestants.  Of course, that language isn't used in all Christian traditions, so we should look at other factors.  When asked about church attendance, 30.8% of Creationists said they attended "seldom" or "never," and 9.6% of Creationists said that religion was "not too important" or "not at all important."  The religious affiliation of Creationists included one "Atheist," one "Agnostic," and 29 "Nothing in particular."  So more than 10% of Creationists have no religious affiliation.  In the arena of politics, you might expect Creationists to be stuffy, ultra-right wing Republicans.  If so, you might be puzzled to see 23% of Creationists identifying as "Strong Democrats" and 21% as "Very Liberal" or "Liberal."

Does any of this seem fishy to anyone else?  It's certainly possible to have politically liberal creationists, but that seems like a big rarity in my experience.  I'm also quite dubious of the non-religious creationists and especially the "Atheist Creationist."  I just don't know how that works.

Leaving aside the problem of identification, I also wondered what makes an "above average" Creationist, which in the context of this study means the people who chose the "Creationist" option on the evolution question but still got better than the population average on the EKT.  That's 82 people, by the way, a relatively small group (just 7.5% of the entire survey sample and 29% of the Creationists).

I compared those 82 above average Creationists to 200 below average Creationists in the study, using every factor I thought might be important.  There isn't much that distinguishes them, at least at first glance.


As you can see, the fraction of each group that are women, born again, attend church at least once a week, and claim that religion is important to them are nearly the same.

Nevertheless, there were some demographics that were strikingly different between the above and below average Creationists.


Above average Creationists were far less likely to be Catholic, liberal, or Democrats than their below average counterparts.  Most strikingly, the above average Creationists were twice as likely to have at least some college education than their below average counter parts.

Weisberg and colleagues asked four questions about factors that influence the respondent's beliefs about evolution.  The responses were on a four-point scale of not at all important, not too important, somewhat important, and very important.


The below average Creationists were more likely to say that family beliefs or what they learned in school were somewhat or very important in shaping their beliefs about evolution.  The above average Creationists were more likely to say that the quality of scientific evidence was somewhat or very important in shaping their beliefs about evolution.  We'll talk about what all that might mean a little later.

Weisberg and colleagues also included a series of questions that attempted to gauge the respondents' understanding and knowledge of science in general.  There were several groups of these, and I looked at three: ten questions of science trivia, two questions about scientific theories, and two questions on reasons for belief.

The ten questions of science trivia (I examined only the multiple choice questions) are there ostensibly to evaluate the respondents' science knowledge, which I suppose is a good thing.  I call them trivia because that's basically what they are.  They ask whether the earth orbits the sun and how long that takes, and that sort of thing.  If this had been the only gauge of the respondents' understanding of science, I would have been dissatisfied, but together with the other questions (discussed below), I think these are useful.  I found a substantial difference between the above and below average Creationists in their responses to these questions.  Above average Creationists correctly answered an average of 7.29 questions, while the below average Creationists correctly answered an average of only 5.35 questions.  The average correct answers for all 1100 respondents was 6.87.  So not only are the above average Creationists better than the average of all respondents on the EKT but also on general science knowledge.

Two questions also probed the respondents' understanding of scientific theories.  Respondents were asked to rate their agreement on a scale of 1-5 (strongly agree to strongly disagree) to these two statements: "Once a scientific theory has been established, it is never changed" and "Scientific theories are just scientists’ guesses."  Neither statement is true, so higher scores (stronger disagreement) represent a better understanding of science.  I added them together to give a score range of 2 (strongly agree with both) to 10 (strongly disagree with both).  Above average Creationists scored an average of 6.96, while below average Creationists scored an average of 5.74.  The above average Creationists' score is just about the same as the population average of 6.93.

Two final questions examined the respondents' decision making about beliefs.  They were asked their responses to these two statements, "There is good scientific evidence for it" and "I feel it is true in my gut," on a scale of 1-5, with 1 being "terrible reason for belief" and 5 being "excellent reason for belief."  In this case, since we're evaluating scientific ideas, evidence is good and gut feelings are not good.  So I inverted the numerical values of the "gut feeling" question and added the two responses together to get a 2-10 scale just like on the previous question.  Higher numbers mean more esteem for evidence and less for gut feelings.  Above average Creationists scored an average of 6.22, and below average Creationists scored an average of 5.85.  That's not a dramatic difference, and both groups are below the population average of 6.51.


What does all this mean?  My interpretation of this data is that there are two big groups mixed into this survey that are not easily separated given the present survey questions.  One group is knowledgeable about science and evolution and makes informed decisions.  The other group does not understand science well and maybe doesn't even understand the questions they're being asked.  So you end up with a bunch of nonreligious Creationists, which doesn't make much sense to me.

I think this is supported by the evidence of the demographic differences between the Creationists who scored above average EKT scores and those that did not.   The above average Creationists were much closer to the stereotypical Creationist: religiously committed, politically conservative Protestants.  We also see that the above average Creationists had more education and better scores on the science trivia questions and the questions about scientific theories.  Above average Creationists were also less likely to attribute their views on evolution to their education (which they might disagree with) and more likely to attribute it to the quality of the evidence (which they may be acquainted with and judge as poor or inadequate).  One very intriguing part of this is their scores on the two reasons for believing questions.  Above average Creationists scored about average in their assessment of evidence vs. gut feelings, which makes sense for a group acquainted with evolution (as judged by their above average EKT scores) but who nevertheless choose to reject some or all of evolution as a model for the origin of life's diversity.  They are not necessarily persuaded by scientific evidence alone.

My interpretation also fits well with my own anecdotal experience talking with people about creation and science.  I've found that people respond very well to my personal testimony or my Christian convictions on creation, but when I start talking about the science, things dramatically change.  Some people zone out.  Others try to change the subject.  I've even had a few people get angry with me.  Science is hard, and people have a hard time understanding it.

And that brings me to my final point on this Weisberg survey:  What's in this Evolution Knowledge Test anyway?  Speaking from my own personal experience as an educator, I thought these questions were difficult, like college-biology-class difficult.  That's my personal "gut reaction," though.  Let's look question by question at this test to see if we can find any patterns where the Creationists went wrong.  Here are the fractions of each group that chose the correct answer for each question:


Notice anything interesting?  Everybody does about the same on almost every question?  Yeah, I noticed that too.  There are only a handful of questions where the four groups of respondents answered in substantially different ways.  The biggest of those that I see is question 21:

During the industrial revolution, the English countryside became covered in soot and ash, and the native moth species became darker in color. How do scientists explain this change in color?

  • Predators were able to see the lighter moths better and ate them, leaving the darker ones to reproduce.
  • All of the moths needed to blend into their environment in order to survive, so they all gradually became darker as they got older.
  • Dark-colored moths learned how to hide from predators by landing only on sooty trees.
  • All of the baby moths born after the trees became sooty were darker than their parents.

This is a widely known example of evolution among creationists, and it is widely believed to be bunk.  It is one of Jonathan Wells's Icons of Evolution, and knowledgeable creationists would probably deliberately choose a different answer than the expected one.  That question should probably be dropped from the EKT.

As I said, there are lots of EKT questions where everyone answered poorly, like questions 6, 8, 13, 18, and 23.  Three of those (6, 8, 18) were choices on questions that asked respondents to choose all that applied, so I'm not surprised some of the correct responses were missed.  Question 23 is of special interest to me because I think it reveals a common misunderstanding on the part of the public:

Scientists think that, over the course of time, all species advance from more primitive to more advanced states.
  • True
  • False
  • I don’t know
The statement is false.  In evolutionary thinking, most species will go extinct over the course of time.   Nevertheless, when the public hears about evolution in the popular culture, it is very common to emphasize how fossil forms were primitive and gave way to the more advanced forms we see in the present.  This question is ripe for misinterpretation due mostly to the way science and the science media describe their work.  Only 18.9% of all respondents answered it correctly.

So are there any other questions where Naturalists do substantially better than Creationists in the survey?  To figure that out, I took the fraction of Naturalists who got the right answer and divided it by the fraction of Creationists who got the right answer for each question.  That tells me how many times better the naturalists were than the creationists in answering questions.  Here's what I found:


Looking at this graph, we can see a number of questions (1, 6, 9, 13, and 21) where the Naturalists did at least twice as better than the Creationists.  What strikes me the most about this result is that the questions where Naturalists are so much better than Creationists are the same questions where everyone did pretty badly.  Questions 1, 6, 9, 13, and 21 were answered correctly by only 35%, 20%, 38%, 27%, and 39% of all 1100 respondents respectively.  So on the questions that confused most people, the Naturalists were confused the least.  Or to put it another way, the general confusion about science that caused some people to select the Creationist option on the Evolution question also caused them to do especially poorly on these EKT questions.

So what have we learned?  Weisberg and colleagues want us to emphasize the role of science education in persuading people that evolution is true, and I'm sure that is a factor.  I'm also sure that eveolution education will not persuade all people that evolution is true, and I'm very skeptical that they have correctly differentiated "real" Creationists from people who are just confused about evolution.

What I'd love to see is a battery of questions that help diagnose a person's position on creation and evolution.  Minimally, we need a question like this:

How important is your belief about evolution and creation?
  • Extremely important
  • Important
  • Not important
  • I don't care at all
I suspect that a Creationist who doesn't care about the issue might also have a poor understanding of science and may well be easily persuaded to adopt a different view.  Alternatively, those Creationists who think their view on evolution is very important will likely be very different from the ones who are just confused (as I believe Weisberg's data hint at).

As to diagnosing a person's position on creation and evolution, I think it would be perfectly reasonable to ask them to choose a word that most closely resembles their own position on creation/evolution: Young-earth creationism, old-earth creationism, intelligent design, skeptic, evolutionist, etc.  That should be supplemented with a series of other questions probing acceptance or rejection (strongly agree to strongly disagree) of a variety of statements about origins:
  • All living things on earth have descended from a common ancestor.
  • The earth is billions of years old.
  • The universe is billions of years old.
  • All life originated by a miraculous intervention of God.
  • Species never change into other species.
  • Humans evolved from primitive, non-human ancestors.
  • There is good evidence for evolution.
  • My religious beliefs tell me how God created everything.
  • My religious beliefs are incompatible with evolution.
  • My religious beliefs are incompatible with billions of years of earth history.
  • I am certain that my beliefs about evolution are correct.
  • God guides evolution directly.
  • God established natural laws, and the universe developed on its own.
  • God established evolution, and life developed on its own.
These questions would help distinguish the various camps in the creation/evolution debate, and that would be incredibly helpful in comparing their EKT scores.  As I have argued here, multiple groups are at play here in the creation/evolution debate, including people who just don't understand and probably don't care.  If we don't correctly distinguish these groups, our survey work will not help us to genuinely understand anyone.

Check out the original paper (if you can get it):


The original survey data are publicly available here (check my work):
Public Understanding and Acceptance of Evolution

Feedback? Email me at toddcharleswood [at] gmail [dot] com. If you enjoyed this article, please consider a contribution to Core Academy of Science. Thank you.