If you're reading for the humor, this is not a post for you.
I was browsing through some of the exit poll questions (from the Casey/Santorum race) and ran across this one:
ABORTION SHOULD BE...
| TOTAL | Santorum | Casey, Jr. |
|---|---|---|
| Always Legal (24%) | 16% | 84% |
| Mostly Legal (35%) | 28% | 72% |
| Mostly Illegal (24%) | 59% | 41% |
| Always Illegal (14%) | 72% | 28% |
I'll ignore the obvious part of the numbers (more legal means more Casey, more illegal means more Santorum, duh). I'd like to know if that was the way the the question/answers were worded. Do 24% and 14% of the people surveyed really think that "Abortion should be "always legal" or "always illegal"? Do people understand the meaning of the word ALWAYS.
Who thinks abortions should be legal late in the ninth month of pregnancy?
And who thinks ectopic pregnancies should be illegal to abort?
So what's the deal? Do people not think about stuff like this? Do people not understand the difference between "Mostly" and "Always"? Or are they answering a different question?
4 comments:
Hmmm... it is problematic since we are not sure how they worded the question. I think that if it was not specified, many people might assume that they were talking about first trimester abortion. If my day was interrupted by an annoying poll taker, and I was eager to get on with my day and priorities, not theirs, I can see myself jumping to this conclusion and not actually taking the time to think about it -- and as you know, this issue is near and dear to my heart and mostly will rank near the top (maybe even *always* at the top?) of the critical issues that determine my candidate choices.
Good insight. After do much more thinking (much more than I would have time for while taking a poll) I wonder if people aren't equating these 4 levels into publicly debated "levels" like life-of-the-mother, rape/incest, early term abortion-on-demand, and so called partial birth.
Having had to design survey questions, you do your best to make sure that most of the people undersatnd the question. The problem is that if you make the responses too long, or give too many choices, then the respondant either tunes out, or gets confused. Generally, going with the 'always, most of the time, seldon, never' is the best you can do. You have to accept that some people will mis-understand, but it will be less than if you try to be too detailed.
Generally, survey questions are a lesser of the evils type of thing.
mess
mess - Another good viewpoint. I'll try that out in a new post.
Post a Comment