Rasmussen Sucks

I think we all understood that, and it’s borne out by the evidence: the U.S. pollster has a significant bias towards the Right, and it’s not terribly accurate either.

I can see the argument for having a right-(or left-)wing bias in polling, as far as creating momentum, driving the narrative, dominating the news-cycle — in short, spin. And frankly, a pollster’s honesty can be selective in releasing polls: if your candidate or party is doing poorly, you bury the poll. It’s an interesting if open question how polling and pollsters drive political discourse.

But it seems to me if you’re in the polling game, accurate information is the coin of your business, regardless of political affiliation. If you’re a candidate who’s invested a lot of money in a poll, how can you tell your messaging is working (for example) if you can’t trust the data?

There are two possible conclusions. One is, simply, Rasmussen Reports is a shitty pollster, and should be ignored. The second is that Rasmussen has two polling models in which to input raw data, one for the rubes, and a second, resulting in “internal polls” for actual paying customers. If the second is true, it unfortunately doesn’t speak well for Rasmussen’s integrity as a pollster.

Which we all sort of knew, too.