Scott Rasmussen’s Conservative-Friendly Question-Wording

Politico says that Barack Obama’s foreclosure-mitigation plan is in political trouble because we’re “a nation of Santellis”. The basis for the piece is a Rasmussen poll which asks “Some people say that having the government subsidize mortgage payments for financially troubled homeowners puts the government in the position of rewarding bad behavior. Is the government rewarding bad behavior when it provides subsidies to those who are most at risk of losing their homes?” The results are 55 yes and 32 no.


Bad news for Barack Obama. But at the same time, when The Washington Post asked if people support Obama’s plan over 60 percent said yes. And The New York Times got a similar result. The difference is striking:


The difference, obviously, is question-wording. Josh Marshall remarks parenthetically:

On the question of the quality of Rasmussen polls in general, I’ve been watching them closely now through at least two cycles. The toplines tend to be a bit toward the Republican side of the spectrum, compared to the average of other polls. But if you factor that in they’re pretty reliable. And the frequency that Rasmussen is able to turn them around — because they’re based on robocalls — gives them added value in terms of teasing out trends. But the qualitative questions, in terms of their phrasing and so forth, are frequently skewed to give answers friendly toward GOP or conservative viewpoints. All of which is to say that his numbers are valuable. But they need to be read with that bias in mind. On the separate question of whether robocalls are as ‘good’ as traditional live question polls, I think they’ve held up quite well over the last two cycles. I see little evidence that SurveyUSA’s poll haven’t stood up as well as those done by live phone callers.

To really understand this dynamic, you need to read my friend Dave Weigel’s story on Rasmussen for The Washington Independent. Dave doesn’t editorialize in his reporting, but if you read between the lines I think a pretty clear picture emerges. Rasmussen is a pretty good pollster whose results are within the range of accuracy one wants from a pollster. But polling is a crowded business. And Rasmussen doesn’t also have a daily newspaper or a television network to tout his results. His business, however, requires attention. So how does he get that attention? Well in part he gets it with issue polling that, while basically methodologically sound, has question-wording that’s designed to lead to conservative-friendly results.

Then the results come out and conservatives tout the results as vindicating their position. It’s free PR for Rasmussen, it’s a morale booster and message-driver for the right. And because the basic horserace polling is accurate enough, these kind of shenanigans don’t get Rasmussen dismissed as a surveyor.

And I don’t really think he should be dismissed. But I think we need to understand his “issue” polling as more like message testing than like normal public opinion research. What we’re learning from his result isn’t that there’s a “nation of Santellis” out there outraged about Obama’s plan. We’re learning that support for the plan isn’t so rock-solid as to be immune to leading questions or negative characterizations.