A friend writes, re the recent Slate poll on creation

T: Very sobering points. And, when a pollster wants to get "expertise" on Evo vs Creation and Design etc issues, guess who s/he is likely to call up? No prizes for guessing who, compounding the problems. As for look-it-up on Wiki . . . let's not go there. KF kairosfocus

Learned Hand (2): Yes, people give inconsistent answers to polls. There are two main reasons for this: 1. The average person is not highly gifted intellectually, and is not rigorous and disciplined in thinking, but the average person loves stating his or her opinions. 2. The poll questions are often (i) imprecise or vague; (ii) slanted; (iii) artificially polarizing. Thus, it is quite conceivable that the answer to question 1 might end up inconsistent with the answer to question 2, because (i) the person answering the question does not understand the words of the question in the way that the designer of the poll meant them, or (ii) the person answering might be sucked in by the slanted choice of words in question 3, but might resist the slanted choice of words in question 5; (iii) the person might answer question 8 on the "left" extreme but question 10 on the "right" extreme, in both cases wishing a third option was available instead of two extremes which he disagrees with; having to choose between two equally wrong alternatives at several points in the survey, he might well not be able to answer very consistently. OK, so in sum, the first cause is incompetent poll respondents, and the second cause is incompetent poll designers. Nothing can be done about the first cause. But there is no reason to compound the difficulty, which is what the second cause does. If people will answer imperfectly given the clearest and most useful poll questions, they will answer even more imperfectly given badly conceived and written poll questions. The reason that poll questions are so often bad is that polls are designed by "polling experts" rather than by people who actually have deep knowledge of the subject that the polls are about. The average pollster has only the vaguest knowledge of what "intelligent design" and "theistic evolution" mean; the average pollster has only the most un-nuanced understanding of what "creationism" means. The average pollster probably isn't aware that there is a split between old-earth and young-earth creationists, for example, or that there are ID proponents who accept evolution. So on these polls, there is an abnormal focus on a young age of the earth, and the questions tend to be slanted toward "creation versus evolution" rather than "design versus chance"; thus, if the person answering the questions is a creationist who is fine with a 4.5 billion year old earth, he cannot answer some of the questions either yes or no; and if he believes in "designed evolution" or "guided evolution" he usually has no good choices in answering the questions -- he is continually forced into "evolution, no design," or "creationism, design." These problems could be eliminated if the pollsters would consult with leaders of ALL the camps before designing their polls. People still might answer the polls inadequately, but if they did, it wouldn't be the pollsters' fault. As it stands now, three-quarters of the blame for worthless answers to these polls lies with the pollsters, not the people being polled. Timaeus

A response that the Slate writer mocks was to a 5-option badly worded "inerrancy" type of question: "Which of the following comes closest to your personal beliefs about the Bible?" The first two options were: 1. The Bible is the actual word of God and is to be taken literally, word-for-word. 2. The Bible is the inspired word of God, without errors, but some parts are meant to be symbolic. Despite the Slate writer's mirth at finding people who supposedly think the Bible is literally true but also think some of it is meant to be symbolic, the real problem is the question, not the response. The words "some parts are meant to be symbolic" can be taken as meaning not literally true but are just symbolic, or yes there are symbols in the Bible but they represent literally true things, such as we find in the books of Daniel and Revelation. So it is not surprising that literalists who know that some things in the Bible are represented by symbols may interpret the question differently while pretty much believing the same things. The sloppy question gives sloppy writers like the Slate guy a chance to show their bias rather than their analytical skills. TimT

The polls reflect the lack of reflection of much of the population. I think only 5-15% are real bible believing creationists. like me. The rest just pick things as they like them. God, or man being created, and so on. the point is that eNOUGH people disagree with evolution to demand equal time where the public institutions teach these things. A whole chunk of the pop gets shafted by the present establishment. its truly a immoral, illegal. and tyrannical regime that set in since WW11. Time to take care of business once more. Robert Byers

"ID is the elephant that they did not let into the room." ID is the mouse that has lost its squeak. REC

Vetting the poll wouldn't have fixed the inconsistency he's complaining about. People just give inconsistent answers to these polls--it's a known phenomenon. Learned Hand

Statistics is a hairy business at best. Not only can wrong conclusions happen with convoluted outcomes but the very ways that polls and surveys are taken often become suspect. I personally don't know how one can be absolutely sure a survey is random and in some cases surveys have been shown to be clearly biased. Besides, how truthful are those who are being surveyed and what about the data collector? As my dad used to say, "Believe only half of what you see and a quarter of what you hear." fossil