JONATHAN CHAIT MAY 25, 2010
Regarding my previous item, a couple friends have suggested that Rasmussen deserves the benefit of the doubt because it accurately predicted the 2008 election. Well, yes and no. Yes in the sense that a record of success carries real weight. But no in the sense that Rasumussen was not producing strong outlier results in 2008. Here, again, is Nate Silver:
The bottom line is this: the sample included in Rasmussen's polling is increasingly out of balance with that observed by almost all other pollsters. This appears to create a substantial house effect, irrespective of whether Rasmussen subsequently applies a likely voter screen.
It also appears to be a relatively new facet of their polling. If one looks at the partisan identification among all adults in polls conducted in September-November 2008, Rasmussen gave the Democrats at 6.5-point edge, versus an average of 8.7 points for the other pollsters; their house effect was marginal if there was one at all.
In other words, the question at hand is the odd tilt in Rasmussen that has appeared since the start of 2009. This has yet to be tested against electoral reality. There's historical precedent to trust Rasmussen's results, but very little to trust the current incarnation of Rasmussen's results. The problem is compounded by Rasmussen's odd habit of not conducting polls of races close to election time:
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.
You see, the thing about Rasmussen is that he cares only about setting the narrative that Democrats are doomed. And it's hard to build those narratives if you screw up polling actual elections.
I excerpt the above from Kos because of its data points, not its argument. To me, the thesis that Rasmussen is deliberately trying to create pro-GOP narratives, and narrowing its profile of polls that can be tested against reality, is one I'm open to but consider far from settled. The point is that Rasmussen's findings have changed dramatically since 2009 and they haven't been tested. For instance, they stopped polling the iconic Brown-Coakley race a week before the election.