As many of you know, I have more than a passing interest in surveys and polls, especially when it comes to health care reform. I’m especially interested in the science behind polling, and how different groups can poll the same topics and get different answers. I noticed this post some time ago from Nate Silver, but didn’t comment on it. I probably should have:
Both critics and defenders of Rasmussen Reports’ polling have frequently cited Rasmussen‘s use of a likely voter model to explain why their polls have tended to show substantially more favorable results for Republican candidates than the average of other surveys. I have often mentioned this myself, for that matter.
The argument goes like this: those people who vote most reliably in midterm elections tend to be older, whiter, and to have higher social status — which are also characteristics of voters that generally lean toward the Republican candidate. When coupled with what also appears to be a Republican enthusiasm advantage this cycle, it is quite reasonable to believe that a poll of likely voters (like Rasmussen‘s) should show more favorable results for the Republicans than one of registered voters or adults (like most others).
But here’s the thing. At some point, you get to check and see how valid the different models are. Those points are called elections. As elections approach, polling firms usually do a fair amount of predicting how the race will develop. Then, when the election occurs, we can all see which firm was correct. You can’t say Rasmussen has a Republican bent if they get the election right. They are just correct; and others have a Democratic bent.
Which is why this is not very comforting:
Rasmussen has been this cycle’s most prolific pollster, by far. As of February:
Yes, Rasmussen Reports has fielded far more polls so far this cycle, both in absolute terms (45 vs. 13) and as a percentage of the total (28% vs 18%).
Rasmussen’s volume hasn’t decreased since then. If you want to see how spammy they are, check out this link.
Yesterday the nation had several hot races, including the House special election in PA-12, primaries in both parties in Kentucky and Arkansas, and the Democratic Senate primary in Pennsylvania.
And somehow, Rasmussen was nowhere to be found. Yet this past week, Rasmussen found time to poll Colorado, California, and those burning Idaho senate and governor races. He even polled the general election in Arkansas, ignoring the imminent primaries — the better to show Arkansas Republican primary voters who their strongest candidate was.
Turns out Rasmussen also did a lot of polling leading up tot he Massachusetts special election back in January, but stopped two weeks before the election. Why? Some claim it’s because Rasmussen is more interested in setting a narrative than reporting the facts, but I don’t know if that’s true.
What I do know is that Rasmussen seems to be avoiding opportunities to prove that their results are more accurate than others. And, in doing so, they seem to be avoiding current important political races in order to comment on far-off and unverifiable races. Why?