rasmussen-screen-grab1.jpg
Remember the bevy of polls that came out in the weeks before Tuesday's election? Murray up by three points in one; Rossi up by a

"/>

NYT: Rasmussen Polls Are Officially Biased Toward Republicans

rasmussen-screen-grab1.jpg
Remember the bevy of polls that came out in the weeks before Tuesday's election? Murray up by three points in one; Rossi up by a single point in another; Murray up by 15 points in yet another. By election day, the general consensus was that all polls suck and people should just wait for the voting results to come in. Well, today the New York Times' Nate Silver--probably the most respected number cruncher in the political world--says he did the math and it appears all your liberal friends were right, Rasmussen really is biased toward Republicans.

In coming to his conclusion, Silver essentially took the surveys that the pollsters published just before the election, then looked at the actual results and quantified how much they were wrong by.

The top performer was Quinnipiac University, which had an error ratio of only 3.3 points on either side of the actual final tally. Rasmussen polls, however, were the worst of the bunch. The firm's polls were off by an average of 5.8 points, and, in almost all cases, missed high on the side of conservative candidates, sometimes by huge margins.

pollacc1.jpg

The news likely comes as no surprise to liberals and Democrats who have long vilified the firm for what many consider to be a known bias against left-leaning candidates.

Silver simply put some math behind it.

The 105 polls released in Senate and gubernatorial races by Rasmussen Reports and its subsidiary, Pulse Opinion Research, missed the final margin between the candidates by 5.8 points, a considerably higher figure than that achieved by most other pollsters. Some 13 of its polls missed by 10 or more points, including one in the Hawaii Senate race that missed the final margin between the candidates by 40 points, the largest error ever recorded in a general election in FiveThirtyEight's database, which includes all polls conducted since 1998.

Silver chalks up the discrepancies to many factors in the way Rasmussen conducts its polls, including that it uses a prerecorded calling system, doesn't call cell phones and uses preordained assumptions about party affiliation in each state.

Then there's the fact that Scott Rasmussen, the head of the firm, is a self avowed conservative and the only pollster who's written a book about the Tea Party and routinely appears on camera to defend his methods.

On that note, here's the man now. You be the judge.

 
comments powered by Disqus