Cognitive Dissonance

Truly odd Gallup poll result. The question: "Next, we'd like to ask about your views on two different explanations for the origin and development of life on earth. Do you think [see below] is definitely true, probably true, probably false, or definitely false?" They rotated two different answers into the blank space. One was "Evolution -- that is, the idea that human beings developed over millions of years from less advanced forms of life." The other was "Creationism -- that is, the idea that God created human beings pretty much in their present form at one time within the last 10,000 years."

In the Evolution sample, 18 percent said evolution is definitely true, and an additional 35 percent said it's probably true. In the Creationism sample, however, 39 percent said creationism is definitely true and 27 percent said creationism is probably true.

We've all heard of "framing effects" in polls, and that's what you're seeing here -- people seem inclined to agree with the questioner -- but the scale of the effect seems enormous here, especially since the question isn't particularly obscure.

Comments

Indeed - the result is odd and should cause us all to take poll results with a grain of salt.

My hypothesis for why we see this result: A large majority of people believe in God and believe that he had an active role in the creation and direction of life on Earth. There is less consistency in peoples' beliefs as to the mechanism of that action.

When presented with this question most people - who, rightly or wrongly, care a helluva lot more about self-affirmational warm n' fuzzies than they do about clean data sets - will be motivated to "score well" the second statement, as it offers them an opportunity to affirm their belief in God.

My guess is that many people will do this even if they believe that evolution is the mechanism through which God created human being and that the universe is billions of years old. Again, they simply care more about hitting the "Yes I believe in God" button than they do about giving Gallup - or you and I - clean public opinion data.

If my hypothesis is true then I would also suspect that the "evolution" question understates the % who give credence to the Darwinian explanation of human origins.

Posted by sd | June 10, 2007 11:34 AM

If the effect is caused by people's inclination to "agree" with the interviewer, it's not a framing effect, it's a survey effect. An example of a framing effect would be if they asked everyone about evolution but rotated "less advanced forms of life" and "monkeys," and that affected their responses.

Posted by Aaron S. Veenstra | June 10, 2007 11:37 AM

The other thing I see is that both questions revolve around how one sees humans in general or oneself in particular. The creationism question implies God creating humans in his image while the evolution question implies development from pond scum. Because of this, evolution will never have a high acceptance as long as these question are only framed in "human" terms.

Also, the creationism question relates to what is called "young earth" creationism which also rejects other sciences: geology, physics, astronomy. They should try axing the time references and determine the number of people who support creationism or evolution just from a biology point of view.

Posted by Carl | June 10, 2007 11:45 AM

I'm not descended from a monkey, no Sir.

Posted by abb1 | June 10, 2007 12:02 PM

Carl: Technically, the creationism question does not ask when the earth was created... it just asks about when humans were created. So the respondents think, yes, I believe in God, and 10,000 years is a pretty long time, so sure it makes sense that God would create us somewhere in there. I bet many of the respondents never even thought of evolution when answering... we know that code words like "pretty much in their present form at some time" are meant to mean "not evolution," but much of the public might not really make that connection...

Also, I bet more people would reject Creationism if the question said "Creationism- God created the world in 7 days and humans on the 6th day" or something like that. I dunno how many more would, though.

Posted by AP | June 10, 2007 12:17 PM

Actually, scratch my comment-- MY's "rotated" phrasing made me think the respondents were asked about either creationism or evolution but not both, but they were actually asked about both, so never mind.

Posted by AP | June 10, 2007 12:21 PM

I'm not entirely happy with questions that are phrased this way, because in a sense they're comparing apples and oranges. That is, the "definite truth" of Creationism isn't to be interpreted in the same way as the "definite truth" of scientific theories. In the former case, people believe it's true in the same way that they consider moral beliefs to be true (at least, that's my impression). In the latter case, I think that the survey is getting at whether there's any room for reasonable doubt about the applicability of the theory of evolution to human development. I'm perfectly willing to say there's no room for reasonable doubt, but that's different from saying that something is definitely true. (I'm just nitpicking, of course; I'd answer that evolution is "definitely true" for this question. Still, it seems to put science in the wrong light.)

Posted by RSA | June 10, 2007 12:29 PM

I think sd is probably right. I don't know how you'd present it in a survey form, but I think you need to give people three options to really break out the total creationists from the "I believe in God and evolution" types. The third option would be something like: evolution is the method that God used to create life on earth.

Posted by mw | June 10, 2007 1:01 PM

Umm...OK, I's afraid I'm being dense here. But I don't understand what Matt means by the "Evolution sample" and the "Creationism sample." Isn't it the case that the respondents were presented with both options, and that the "options rotated" insertion only indicates that the order in which the options were presented was rotated?

The other interpretation, that each respondent was only presented with one option, but that the options were rotated from respondent to respondent, hardly makes sense given the preamble "we'd like to ask about your views on two different explanations for the origin and development of life on earth."

The strange thing about the result, then, is that 66% of respondents said that Creationism was either definitely or probably true, while 53% of respondents said that Evolution was either efinitely or probably true. So clearly a fairly significant number of people fall into both categories, and think that both hypotheses are at least probably true - this despite the fact that they are formulated in such a way as to appear mutually exclusive.

So this doesn't seem to be a framing effect. My guess is that it has something to do with tendencies toward suggestibility and submissiveness in certain groups of people. Some people don't like to disagree with any statement that is formulated in an articulate and coherent manner, and will give their assent to mutually incompatible statements rather than adopt even a mildly confrontational stance that requires disagreeing with some people.

Posted by Dan Kervick | June 10, 2007 1:55 PM

Does anyone know if pollsters do a positive or a negative control of their subjects? Something really, really anodyne like "True or false: I am standing on the ground" or "True or false: there is a gigantic chicken on my forehead" would be an interesting and useful experiment, I think.

Posted by Klug | June 10, 2007 3:27 PM

I'm less concerned about the 19% of the population who appear to believe in both evolution and creationism, than I am about the 47% of the population who apparently don't believe in evolution at all.

Posted by John | June 10, 2007 3:36 PM

Funny how lefties think everyone who refuses to believe one part of Darwin's theory of evolution (the part where humans and apes share a common ancestor) is a knucklehead, but lefties refuse to acknowledge the other part of Darwin's theory, that different races of humans evolved to have different capabilities. Both parts of Darwin's theory are glaringly obvious, and both challenge cherished beliefs (in divine creation and egalitarianism, respectively).

Posted by Harry | June 10, 2007 4:51 PM

If you want to see bigger effects of how questions are framed on responses, study the polls cited by the media about how the public really favored the Bush-Kennedy Axis of Amnesty.

Funny how that worked out!

Posted by Steve Sailer | June 10, 2007 4:57 PM

lefties refuse to acknowledge the other part of Darwin's theory, that different races of humans evolved to have different capabilities.

Perhaps because there is no such "part" to the theory. Later crap from social darwinists and eugenicists was invented independently of Darwin himself.

Posted by jimBOB | June 10, 2007 5:57 PM

Ah, look, Steve Sailer has found another nail.

Posted by Hammer | June 10, 2007 6:33 PM

the other part of Darwin's theory, that different races of humans evolved to have different capabilities

b-but, but, the Bell Curve!!!

Posted by drjimcooper | June 10, 2007 6:35 PM

Darwin on race, from The Descent of Man, pp.175-176::

"...the various races, when carefully compared and measured, differ much from each other -- as in the texture of hair, the relative proportions of all parts of the body, the capacity of the lungs, the form and capacity of the skull, and even the convolutions of the brain. But it would be an endless task to specify the numerous points of difference. The races differ also in constitution, in acclimatization and in liability to certain diseases. Their mental characteristics are likewise very distinct; chiefly as it would appear in their emotions, but partly in their intellectual faculties. Everyone who has had the opportunity of comparison must have been struck by the contrast between the taciturn, even morose aborigines of South America and the light-hearted, talkative negroes."

Posted by Harry | June 10, 2007 10:58 PM

Well, I have to admit that Harry's got the goods on poor old Darwin. But the quote is irrelevant. Darwin is talking about the data he thought he had; it's not integral to the theory.

Posted by AlanC9 | June 11, 2007 3:11 AM

No, race is absolutely integral to Darwin's view of the evolution of humanity -- hundreds of pages of The Descent of Man are about the evolution of racial differences. To deny that is a particularly dopey form of creationism.

http://www.isteve.com/Darwin-EnemiesonLeft.htm

Posted by Steve Sailer | June 11, 2007 4:23 AM

I thought creationism was the belief that God created the whole universe a few thousand years ago, including humans. I agree with AP that the wording of the question is odd as it is limited to humans: so you could say yes if you accepted the universe is 13 billion years old as long as humans only arrived 10,000 years ago. But I guess there aren't many such.
BTW, Darwin's theory of natural selection is an explanation of the fact of evolution - a succession and progression of species over millions of years - found and accepted by many scientists before him.

Posted by James Wimberley | June 11, 2007 10:21 AM

Ah, yes. If Darwin said it, then it must be part of the modern scientific consensus on evolutionary theory, right Steve Sailer?

Except it isn't.

It is now well understood that humans have among the lowest levels of intraspecies genetic variation of any species ever studied, presumably due to our relatively short history as a distinct species, our ability to cross geographic barriers and our long history of trade between neighboring groups. There's more genetic diversity within any race as there is between the races, and there is not a single allele in the entire human genome that is exclusively linked to race.

Modern Homo sapiens have been around for less than 200,000 years and no populations were completely isolated from trade and genetic transfer for more than about 30,000 years. There are significant variations among the races in regard to certain traits, as a result of geographic localization. But the notion that the human races are subspecies which evolved to fill distinct adaptive niches has been completely and thoroughly discredited by genetic research.

I hate to disappoint you, Steve, but Descent of Man is fundamentally irrelevant to modern genetic research. It was a smart, perceptive work based on the data available to Darwin at the time and the biases native to his society. It set the ball rolling for biological research. But the ball didn't end up where Darwin thought it would.

Posted by LaFollette Progressive | June 11, 2007 1:37 PM

...Interpreting the nature and prevalence of religious opinions is tricky, particularly if you depend on polls. Respondents can be lacking in seriousness, unsure what they believe, and evasive. Spiritual values and practices are what pollsters call “motherhood” issues: everybody knows that he is supposed to be in favor of them. Thus sociologists estimate that maybe only half of the Americans who say that they regularly attend church actually do so. The World Values Survey Association, an international network of social scientists, conducts research in eighty countries, and not long ago asked a large sample of the earth’s population to say which of four alternatives came closest to their own beliefs: a personal God (forty-two per cent chose this), a spirit or life force (thirty-four per cent), neither of these (ten per cent), don’t know (fourteen per cent). Depending on what the respondents understood by a “spirit or life force,” belief in God may be far less widespread than simple yes/no polls suggest.

In some religious research, it is not necessarily the respondents who are credulous. Harris has made much of a survey that suggests that forty-four per cent of Americans believe that Jesus will return to judge mankind within the next fifty years. But, in 1998, a fifth of non-Christians in America told a poll for Newsweek that they, too, expected Jesus to return. What does Harris make of that? Any excuse for a party, perhaps. He also worries about a poll that said that nearly three-quarters of Americans believe in angels—by which, to judge from blogs and online forums on the subject, some of them may have meant streaks of luck, or their own delightful infants.

The Bible is a motherhood issue, too. Harris takes at face value a Gallup poll suggesting that eighty-three per cent of Americans regard it as the Word of God, and he, like Dawkins and Hitchens, uses up plenty of ink establishing the wickedness of many tales in the Old Testament. Critics of the Bible should find consolation in the fact that many people do not have a clue what is in it. Surveys by the Barna Research Group, a Christian organization, have found that most Christians don’t know who preached the Sermon on the Mount....

from "Atheists with Attitude"
Book review by Anthony Gottlieb
May 21, 2007 New Yorker

Posted by artappraiser | June 11, 2007 5:54 PM