As the accompanying report tells us, GOPers did better than Dems on just three of the 11 questions asked, at least beyond the margin of error. Dems didn't outscore Republicans on any question beyond the margin of error, though it got a bit close on the obesity question.
So our conclusion is that Republicans are politically smarter than Democrats, at least a little? Maybe.
To test this conclusion, I plucked the three questions out of the data that demonstrated a Republican advantage (Republicans control House, source of electricity, reason for Wisconsin protests) and subjected them to a quick-and-dirty analysis that controlled for one obvious factor -- education.
It's an accepted truism in hundreds of academic studies that education remains the most persistent and significant predictor of what people know. The greater the formal education, the greater the political knowledge, at least in how it's measured in most polls.
So how'd it come out? Through the magic of multivariate analysis we find that if we enter education first in a logistic regression model, being a Republican (versus Democrat) still predicts greater knowledge in all three questions.
- What part of Congress do Republicans control? In this question, even controlling for education, being a Republican makes you twice as likely to answer the question correctly.
- What's the cause of the Wisconsin protests? Same analysis as above, being Republican makes you 1.5 times more likely to answer correctly.
- On the primary source of electricity, same as the Wisconsin results, with GOPers 1.5 times more likely to be right.
Our final takeaway? Republicans do a wee bit better on a few questions, most of them explained by the nature of the questions themselves, but controlling for education doesn't make these differences go away.
If I were scoring this, I'd call it mostly a tie, but with the slightest of nods toward the Republicans.
No comments:
Post a Comment