The Truth About Polling
Hugh Hewitt interviewed pollster Scott Rasmussen on his show today (transcript to come). The discussion yielded an important point that simply can't be emphasized enough, as Republicans are barraged with a series of polls that predict electoral disaster three weeks from today.
Here's the point: The polls are only as good as their models. The turnout models, to be precise. In other words, if the pollsters aren't accurately predicting who's going to turn out to vote, they're not accurately predicting the outcome, either.
In some cases, there's reason to believe that turnout models are flawed. Perhaps the best example is Virginia. Scott Rassmussen has estimated Republican turnout at 39%, and Democratic turnout at 39%. With that turnout model, George Allen leads by only 3 points. As Hugh pointed out, however, Virginia Republicans historically turn out in much greater numbers than Virginia Democrats. If that's the case again, it's likely that George Allen's margin of victory will well exceed three points.
Republicans can't control a lot going in to this election. We can't control an MSM that's ideologically inclined toward the Democrats. We can't control events in Iraq. We can't control Democratic dirty tricks. But we can, at least, have some control over whether we and our friends and compatriots turn out. And we all need to do our part to make it happen.
Here's the point: The polls are only as good as their models. The turnout models, to be precise. In other words, if the pollsters aren't accurately predicting who's going to turn out to vote, they're not accurately predicting the outcome, either.
In some cases, there's reason to believe that turnout models are flawed. Perhaps the best example is Virginia. Scott Rassmussen has estimated Republican turnout at 39%, and Democratic turnout at 39%. With that turnout model, George Allen leads by only 3 points. As Hugh pointed out, however, Virginia Republicans historically turn out in much greater numbers than Virginia Democrats. If that's the case again, it's likely that George Allen's margin of victory will well exceed three points.
Republicans can't control a lot going in to this election. We can't control an MSM that's ideologically inclined toward the Democrats. We can't control events in Iraq. We can't control Democratic dirty tricks. But we can, at least, have some control over whether we and our friends and compatriots turn out. And we all need to do our part to make it happen.
3 Comments:
If GOP voters do not vote then who can be blamed if the Dems win? No even if the GOP is not perfect what makes Dems so right? The dictators will cheer for the Dems and hope for retreat of the US so that the forces of oppression will win.
One time when Regan was President I got a call from Gallop (Galup?) and they asked me the following poll question: Was Ronal Regan doing a, (a) poor job, (b) complete failure or (c) unsatisfactory job. Since I thought he was doing a good job I asked if there were any other choices and I was told I could volunteer an answer, but it was unlikely they'd use it. Since then I've never trusted polling data.
A lib columnist, whose name escapes me at the moment, (a fine looking black chick, who always hacks me off) wrote about the fact that some people wrongly say that voting is a privelige when the Constitution says its our right. A really stupid semantic game. It is both those things as well as our duty as Americans to exercise that right. It is equally our duty to encourage everyone we know to register and then to vote, even if you know they'll vote against your candidate. All that's required is the encouragement, you can't force them, and those who truly care about the United States of America will do what's necessary to get it done. Oh yeah. You CAN encourage them to vote Republican.
Post a Comment
<< Home