Saturday, August 1, 2009

Science Dies Ugly Death! Only 4 in 10 Americans Believe in Continental Drift! (wait a minute...)

(This post is stolen almost verbatim from Geotripper, but since he is me, it is ok)

Check out a poll and excellent discussion by Devilstower in which Research 2000 asked about plate tectonics (for DailyKos). Worded like similar polls by Gallup about evolution, they asked whether the respondents "believed" Africa and America were once connected ("Do you believe that America and Africa were once part of the same continent?"). If this sounds like a poorly worded question, it was, and they did it on purpose. First, look at the results:

YES NO NOT SURE

ALL 42% 26% 32%
DEM 51% 16% 33%
REP 24% 47% 29%
IND 44% 23% 33%
OTH/REF 42% 25% 33%
NON VOTERS 46% 22% 32%

WHITE 35% 30% 35%
BLACK 63% 13% 24%
LATINO 55% 19% 26%
OTHER/REF 56% 19% 25%

18-29 48% 20% 32%
30-44 40% 28% 32%
45-59 43% 24% 33%
60+ 39% 30% 31%

NORTHEAST 50% 18% 32%
SOUTH 32% 37% 31%
MIDWEST 46% 22% 32%
WEST 43% 24% 33%

Wording in a poll is everything, and for a long time major polling organizations have been asking badly designed questions about science, especially those on evolution, by wording their questions poorly, and then reporting the results with a misleading emphasis. Following the point made by Devilstower, a headline may very well read "Only 4 in 10 people believe..." but this ignores that fact that a full third of the respondents understood that they didn't have enough knowledge in the subject to give an informed answer. The real news in this poll is that only a quarter of the respondents were wrong in their perception of the science and that their ignorance was influenced by region, political affiliation and race (the interesting point in this poll result is how poorly whites did in comparison to blacks and hispanics, if you want to interpret the results literally).

Devilstower does a great job of explaining the inflammatory nature of the use of Africa in the question. Other questions in the poll were highly political ones, including an approval poll for congress and the president. It helps to explain the disparity of the findings in regards to the Southern states. Would the disparity still apply if Europe were substituted for Africa in the question?

The big problem with the poll is the use of the word "believe". People believe in deities. People hold opinions that animal testing is wrong. People believe it's wrong to torture prisoners. But does one believe in gravity? Can a person believe they don't need oxygen to live? They can choose not to believe these things, but it doesn't change the fundamental fact that they will fall if they jump off a cliff, or suffocate if they try breathing water. In the most proper sense scientists don't deal with beliefs. They deal with experimentation and confirmation of physical facts. Hypotheses can't be believed in, they have to undergo testing. They will usually be confirmed or disproven, and it doesn't fall to a vote about belief, whether by the scientists themselves, or by the public at large.

This misunderstanding about being able to pick and choose what science to "believe" is at the heart of issues like human-induced global warming or evolution. I have a lot of respect for the people who responded in this poll by saying they didn't know. I just hope they take the next step and try to learn something about it. Education is everything in facing the complex problems of our society.

No comments:

Post a Comment