After twenty-four hours of solidly negative coverage of the exit polls that led many to believe that John Kerry would cruise to an easy victory, the pollsters themselves spoke out in their own defense. Today's Washington Post reports that the data fed to the networks -- and released without authorization to Internet sources such as Drudge -- carried a specific warning of its incomplete status:

Results based on the first few rounds of interviewing are usually only approximations of the final vote. Printouts warn that estimates of each candidate's support are unreliable and not for on-air use. Those estimates are untrustworthy because people who vote earlier in the day tend to be different from those who vote in the middle of the day or the evening. For instance, the early national sample Tuesday that was 59 percent female probably reflected that more women vote in the day than the evening.

That is why the early leaks anger Lenski. "The basic issue here is the leaking of this information without any sophisticated understanding or analysis, in a way that makes it look inaccurate," he said.

After the survey is completed and the votes are counted, the exit poll results are adjusted to reflect the actual vote, which in theory improves the accuracy of all the exit poll results, including the breakdown of the vote by age, gender and other characteristics.

I wrote about this problem in my midday post on Tuesday. The problem with taking results from the middle of the survey is the incomplete nature of the sample. When the demographic of the polls showed that 59% of the responders were women, people cried "Fraud!" A more likely answer is that by mid-afternoon, when the numbers leaked, more women had voted than men. This simple analysis eluded the people who at first latched onto the premature raw data as a validation of John Kerry as well as those who used them to beat up exit polling in general.

Not that I carry any water for the idea of exit polling -- lately, we've had two elections where the exit polling impacted election coverage negatively, and some might argue actual voting as well. Exit polling has never been terribly predictive, as pollster Pat Caddell explained on Fox News this afternoon. It can give interesting demographic information about the vote after it's complete, but as a predictive model, it's a disaster.

Blaming the pollsters in this case looks more like shifting the blame for bad coverage. ABC's Charlie Gibson bluntly told viewers yesterday morning that the polling information "got it wrong," which isn't accurate in itself. Once the polls were complete, the information they yielded came close to matching the overall vote -- but the talking heads had already seized on the earlier, incomplete raw data for their punditry and the producers were using it to make calls on states. A computer crash that delayed the final results didn't help, either.

Putting an end to exit polling may be desirable, and if states upgrade their voting process to modern machines able to quickly and accurately tally ballots, they may be unnecessary for predictions. The best poll, after all, is the ballot itself. However, the pollsters do not deserve all of the blame for Tuesday's blown calls.

Trackback Pings

TrackBack URL for this entry is

Listed below are links to weblogs that reference Exit Polls Misused, Not Inaccurate, Pollsters Say:

» A Million Gays for Bush? from A Stitch in Haste
So, on the one hand, everyone is acknowledging the absolutely disgraceful inaccuracy of the exit polls in the presidential election. Unless, of course, you're a gay blogger... [Read More]

Tracked on November 4, 2004 7:41 AM

» LAST HURRAH FOR THE EXIT POLLS? from Right Wingnuthouse
Somehow, it was so very satisfying to watch as the returns started coming in, Kos & Co. begin to melt like the Wicked Witch of the West when Dorothy threw water on her. [Read More]

Tracked on November 4, 2004 9:39 AM

» Early Exit Polls Fakes from Josh Ferguson
This was obviously a way to not only give encouragement to Kerry supporters and suppress Bush supporters, but to also try and discredit the new media. Unfortunately, too many bloggers as well as the Drudge Report were so eager for data to post, that ... [Read More]

Tracked on November 4, 2004 1:08 PM