Rasmussen vs. real polling, the NJ edition

Blog ››› ››› ERIC BOEHLERT

Let's conduct an experiment.

Two new polls are out measuring the job approval rating of New Jersey's new Republican Governor, Chris Christie. The first, from Quinnipiac University, peg's Christie's approval rating at 44 percent. There's also a new poll out from Fox News' favorite firm, Rasmussen.

Based only on the name, and Rasmussen's well-known history of producing utterly predictable GOP-friendly results with his robo calls of "likely voters" only, you ought be able to guess what Christie's approval rating is in the Rasmussen poll.

Go ahead and try. Pick a number.

If you took the Quinnipiac results and added seven points (i.e. the Rasmussen GOP curve), then you nailed it: Rasmussen found Christie at 51 percent. What's revealing though, is looking at the differences in methodology between Rasmussen's Fox News filler, and an authentic poll conducted by Quinnipiac. The differences might explain how the polls came to such different conclusions.

-Number of NJ voters interviewed by Rasmussen: 500

-Number of NJ voters interviewed by Quinnipiac: 1,461

-Margin of error for Rasmussen: 4.5 percent

-Margin of error for Qunnipiac: 2.6 percent.

BTW, in case you missed them from today's WashPost profile of Rasmussen, check out these devastating quotes [emphasis added]:

"The firm manages to violate nearly everything I was taught what a good survey should do," said Mark Blumenthal, a pollster at the National Journal and a founder of Pollster.com. He put Rasmussen in the category of pollsters whose aim, first and foremost, is "to get their results talked about on cable news."

Nate Silver, who runs the polling analysis site FiveThirtyEight, soon to be hosted on the Web site of the New York Times, faults Rasmussen for polling only likely voters, which reduces the pool to "political junkies."

"It paints a picture of an electorate that is potentially madder than it really is," agreed Scott Keeter, director of survey research at Pew Research Center and vice president of the American Association for Public Opinion Research (AAPOR). "And potentially more conservative than it really is."

Bottom line: Rasmussen's work is considered a joke within the polling world.

We've changed our commenting system to Disqus.
Instructions for signing up and claiming your comment history are located here.
Updated rules for commenting are here.