Sunday, May 13, 2012

One biased conservative poll does not an election make

By Richard K. Barry 

A couple of conservative commentators are very excited about a recent poll (published May 11) by Rasmussen Reports that finds Mitt Romney ahead of President Obama by a seven-point margin, 50% to 43%, which is actually based on what they call a "rolling three day average." They randomly survey 500 voters each night.

To put this in perspective, RealClearPolitics lists seven polls on the race, including Rasmussen, and calculates an average of 47% to 45% in Obama's favour. All of these polls were conducted within the past 11 days.

Aside from Rasmussen, only one other poll listed by RealClearPolitics has Romney ahead, and that is the Politico/GWU/Battleground poll that has Romney up by 1. Associated Press/GfK has Obama up by 8, although with a relatively small sample size.

So, is there a problem with Rasmussen?

According to Digital Journal:

Rasmussen polls were cited 94 times by Fox News prior to the 2010 midterm elections and over-stated potential Republican votes, according to Nate Silver of the New York Times as reported by Media Matters. Media Matters goes on to say that this should raise suspicion about motivation with respect to Rasmussen polls.

In discussing Rasmussen, along with the Cato Institute, the article had this to say:

The Rasmussen poll, like the Cato Institute, is seen by some analysts also as a conservative, largely Republican group whose poll questions are set up to solicit, and obtain, conservative opinions.

As Silver put it, perhaps treading somewhat carefully on the question of bias:

What it does mean, though, is that when you're evaluating the polls in a particular race, you need to take an especially long look at who's conducting them. If you tell me that the latest poll has the Republican up by two points in Colorado, that's going to mean one thing if it's Research 2000 telling me that and quite another if it's Rasmussen. 

Digging a bit deeper, regarding the 2010 midterm elections, and perhaps being a bit less careful this time, Silver had this to say:

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.

Moreover, Rasmussen's polls were quite biased, overestimating the standing of the Republican candidate by almost 4 points on average. In just 12 cases, Rasmussen's polls overestimated the margin for the Democrat by 3 or more points. But it did so for the Republican candidate in 55 cases — that is, in more than half of the polls that it issued.

And this:

The discrepancies between Rasmussen Reports polls and those issued by other companies were apparent from virtually the first day that Barack Obama took office. Rasmussen showed Barack Obama's disapproval rating at 36 percent, for instance, just a week after his inauguration, at a point when no other pollster had that figure higher than 20 percent.

Not that this should necessarily mean anything, but Rasmussen has in the past issued polls under a subsidiary on behalf of Fox News, that "fair and balanced" news outlet.

There is a lot more in the Silver piece if you really want to read a good critique of faulty methodology, but you get the point.

So, when cites the Rasmussen poll and then asks if it might be time for President Obama to panic, the answer is, ah, no. Like everything else that goes on at that sewer of a website, it's bullshit.

(Cross-posted at Lippmann's Ghost.)

Labels: , , , ,

Bookmark and Share


Post a Comment

<< Home