We've all been treated to countless stories about how much Americans do not know about science, so the Associated Press decided not to trample over that ground yet again. Instead, it conducted a poll in which it asked respondents how confident they were about the science that they do know.
The results are interesting–and confusing.
"Few Americans question that smoking causes cancer. But they express bigger doubts as concepts that scientists consider to be truths get further from our own experiences and the present time," is how the AP story–by science writer Seth Borenstein and pollster Jennifer Agiesta–began.
As they dug into the results in the next graf, we saw a glass half empty. "Americans have more skepticism than confidence in global warming, the age of the Earth and evolution and have the most trouble believing a Big Bang created the universe 13.8 billion years ago," they wrote.
It's not clear what is being measured here. The skepticism regarding the age of the Earth, evolution, and the Big Bang all conflict with some people's religious views, so we're not getting a clean read on people's confidence about the validity of the science. People's confidence in their beliefs about global warming would seem not to belong here, because those beliefs are much less likely, I think, to conflict with religious teachings.
Borenstein and Agiesta do not include too many poll numbers in their story, but the numbers are spelled out in the following chart. Take a look, and we'll reconvene below for further discussion:
The first thing you will notice is that it's difficult to find percentages that add up to 100. I had to email Agiesta to find out how to do that. She explained that the figures in the first column, in bold, are the sum of the figures in the second and third columns. And the figures in the fourth column are the sum of the figures in the fifth and sixth column. So to get 100 percent of the responses on smoking, for example, you would add 82, 12, 4, and the 2 who refused to answer.
Now that we have that sorted out, take a look at the global warming question. Only 33 percent of the respondents were extremely or very confident that "the average temperature of the world is rising" because of human activity. But if you add in the "somewhat confident" group, you discover that 61 percent of respondents are at least somewhat confident that the statement is true. Almost two-thirds.
So which is it? Only a third of Americans are very confident that human-induced global warming is occurring? Or nearly two-thirds are at least somewhat confident?
The problem with this poll, in my view, is what to do with the "somewhat confident" group. Borenstein and Agiesta wrote that Americans have trouble believing in the Big Bang, but add the highly confident and somewhat confident respondents, and you discover that 46 percent–nearly half–are at least somewhat confident that the Big Bang occurred, despite the contrary teachings of some religions. Plenty of them do believe, even if some have reservations.
This poll was an interesting idea, and I think the AP should get credit for its creative approach here. But I don't think it got this quite right.
I hope Borenstein and Agiesta will take another cut at this. I think it's a fertile area for more polling and reporting.