The Blog

Pollsters Get Points for Accuracy This Cycle

10:27 AM, Nov 7, 2008 • By GARY ANDRES
Widget tooltip
Single Page Print Larger Text Smaller Text Alerts

The 2008 campaign produced a record number of surveys, and it turns out that most of the major public polls were pretty accurate. That's the assessment of Fordham political scientist Costas Panagopoulos, who writes:

On average, pre-election polls from 23 public polling organizations projected a Democratic advantage of 7.52 percentage points on Election Day, which is only about 1.37 percentage points away from the current estimate of a 6.15-point Obama margin in the national popular vote.

Most final predictions turned out pretty close to the final results--in part because this wasn't a nail-biter. But Rasmussen, Pew, Polimetrix/YouGov and Harris Interactive were among the most accurate. Many of the network polls, such as CBS, NBC and ABC, were near the bottom of Panagopoulos's list. You can read his analysis here.

Internet polling also did well, at least at the national level. Some of the most accurate on Panagopolulos' list (Polimetrix/YouGov and Harris Interactive) utilize samples drawn from online panels and deploy Internet technology as opposed to random digit dialing (RDD) telephone calls. But as Carl Bialik points out in the Wall Street Journal, some of these surveys were less accurate at the state level:

The only pollster active in every state in the final week, YouGov/Polimetrix, surveyed people from among a panel who were randomly selected and invited to participate in its political polls. In many states, the sample size was under 500. And YouGov's median error of 4.1 percentage points--including, notably, an eight-point margin for Sen. McCain in Indiana--was higher than for each of the active phone pollsters.

"We had quite a good night," Polimetrix President Douglas Rivers says, pointing out that the firm's prediction of an Obama victory by six percentage points in the popular vote was borne out. Difficulty getting funding for polls in states where the race wasn't expected to be close explains the small sample sizes, he says.

Finally, the "polling aggregators," such as Fivethirtyeight.com, Pollster.com, and Real Clear Politics also performed well. Bialik writes this:

The biggest winners may have been poll aggregators, who were combining disparate polls as far back as 2002, but gained new members and reached a new level of national prominence this time around. Their advantage is twofold: Their composite results may dilute the effect of any error in one poll, and their results are more expansive, including regions that no one pollster can typically afford to cover. A dozen or so Web sites combined polls to forecast the election, and just about all of them put Mr. Obama's electoral-vote total at between 338 and 393; he likely will finish with 364 or 375. (Those that also forecast congressional races generally foresaw Democratic gains.)