Conservatives Losing Touch With Reality vs. Liberal Media Bias: A Wee Experiment
As you know, it is now a theme among some conservatives that the polls have a liberal bias (warning! TownHall link! Not an actual source of information!)--largely, apparently, because many of them are done by allegedly liberal media organizations such as the WaPo.
Of course it's rather hard to see what the incentive would be for polling organizations to lie... Not only would it mean intentionally ruining their own reputations, it's not at all clear what they could hope to accomplish. Pumping up Obama's numbers is more likely to hurt Dems than help them. But I digress...
The interesting thing is that this provides us with the opportunity to do a little experiment. With the election coming up, we'll be able to get some idea who was closer to being right, the preponderance of polls, or Rasmussen and Hew Hewitt.
Of course with [respect to just the Presidential popular vote] it's just one case. But if we take the polls state-by-state, add in Senate and House polls, and then ask, after the results are in, who was closer to being right [in more cases], we'll be able to bring [significant] experimental evidence to bear on the question.
If the consensus of non-Rasmussen polls are right, Obama wins by (on current predictions) around 2-3%, and wins by close to the predicted margins state-by-state, then this is evidence that the relevant folks on the right were wrong, the polls were not biased, and charges that they were were unwarranted. If, on the other hand, Romney wins by around 7%, and so on, this is evidence that the polls were biased.
Personally, I'm willing to not only put money on this, but willing to stick with the results of the experiment, and loudly proclaim likely media bias if the experimental results confirm Hewittesque predictions.
I wonder how many on the right are willing to make an analogous commitment?
Of course it's rather hard to see what the incentive would be for polling organizations to lie... Not only would it mean intentionally ruining their own reputations, it's not at all clear what they could hope to accomplish. Pumping up Obama's numbers is more likely to hurt Dems than help them. But I digress...
The interesting thing is that this provides us with the opportunity to do a little experiment. With the election coming up, we'll be able to get some idea who was closer to being right, the preponderance of polls, or Rasmussen and Hew Hewitt.
Of course with [respect to just the Presidential popular vote] it's just one case. But if we take the polls state-by-state, add in Senate and House polls, and then ask, after the results are in, who was closer to being right [in more cases], we'll be able to bring [significant] experimental evidence to bear on the question.
If the consensus of non-Rasmussen polls are right, Obama wins by (on current predictions) around 2-3%, and wins by close to the predicted margins state-by-state, then this is evidence that the relevant folks on the right were wrong, the polls were not biased, and charges that they were were unwarranted. If, on the other hand, Romney wins by around 7%, and so on, this is evidence that the polls were biased.
Personally, I'm willing to not only put money on this, but willing to stick with the results of the experiment, and loudly proclaim likely media bias if the experimental results confirm Hewittesque predictions.
I wonder how many on the right are willing to make an analogous commitment?
posted by Winston Smith at
8:32 AM
1 Comments:
But if the polls end up being accurate, that will just prove to conservatives that the conspiracy WORKED and that the media/pollsters depressed Republican turnout to win it for Obama.
A significant percentage of Republicans are convinced to this day that Obama only won in 2008 because of ACORN-organized voter fraud. Now that ACORN is gone, they need a new scapegoat to delegitamize Obama's presidency.
Post a Comment
Subscribe to Post Comments [Atom]
<< Home