I really want to urge Democrats not to believe any Rasmussen poll that can’t be tested by an actual election. Every opinion poll that they put out seems unhelpful and demoralizing to Democrats, and they rarely agree with other polling (when other polling is available). It has gotten to the point that I am suspicious about whether Rasmussen is even conducting the polls that they’re releasing. You can get the result you want from a poll if you know how to tweak the questions or the methodology (e.g., the sample). But, why bother? If your poll results can’t be verified by comparison to actual outcomes, then you’re free to just release made up data. If Rasmussen’s election polls were to come out as badly as, say, Zogby’s, then people would dismiss their polling on health care or wiretapping or the president’s popularity or teabagging birthers. But Rasmussen, in my opinion, is using the credibility that they gained as a decent election polling outfit to dupe people into believing phony polling data on the events of the day. Maybe Nate Silver can figure out a way to test this theory, but I believe I am right about this.
Job One for the Right is to convince Congress that the people are opposed to health care reform despite having voted for it rather explicitly in the last two election cycles. Bogus polling helps that cause. Crazy yelling constituents helps that cause. But those things are distortions. The only polls that should be trusted are internal Democratic polls.