Wisconsin exit poll: What happened?
By Jon Cohen,
Governor Barrett, meet President Kerry.
Exit poll numbers released to subscribers just before polls closed in the Wisconsin recall election Tuesday dangled the possibility that Milwaukee Mayor Tommy Barrett (D) could win.
The numbers seemed to pop off the screen — 50 percent apiece for Barrett and Republican Gov. Scott Walker, the subject of the recall effort. Walker had a clear lead in independent pre-election polls, so the tie score sent analysts scrambling and buoyed Democratic hopes when the numbers were widely reported elsewhere minutes later at the official poll close time.
Just a half hour later, the exit poll shifted to 52 to 48 percent, tilting in Walker’s favor. (The final margin appears to be seven percentage points.) A potential Gov. Barrett era had ended before it started, and a fresh round of bash-the-exit-poll commenced.
For the exit poll, it was reminiscent of 2004, when leaked midday results showing Democratic contender John F. Kerry with leads in key states led his own pollster ask the candidate “Can I be the first to call you Mr. President?” These aren’t lone examples: Recall then-senator Barack Obama winning the New Hampshire primary?
On Tuesday, as in the other instances, the fault is less about the exit polls themselves, than it is about a widespread, albeit understandable misrepresentation of the numbers.
The exit poll is, after all, a poll, complete with a margin of sampling error and other foibles.
One issue with the exit polling for the recall election was that there was no telephone survey of absentee voters. NBC News estimates at least 15 percent of all voters voted that way, and that they favored Walker over Barrett. The first exit poll numbers to include estimates of the vote breakdowns for absentee voters was the release a half-hour after poll-close, perhaps accounting for the shift from 50-50 to 52-48.
Another, easily forgotten aspect of early numbers is that they are preliminary. The exit poll includes several rounds of interviews with randomly selected voters as they leave polling places (sometimes augmented with telephone polls of early and absentee voters). Different types of people vote at different times of day, with results from morning interviews varying from those at other times.
As it happens, the first round of interviews had Walker way up, the second round had Barrett at 50 percent and Walker at 49 and the third had Walker up again.
When actual precinct level results start to come in, exit polls are adjusted accordingly.
One way to avoid Election Day confusion is to focus on what exit polls are good for — the tally of how different groups voted in an election, and their relative size in the overall electorate — not what they’re not: predicting results.
Even though exit poll subscribers have an obligation not to misuse the topline numbers (The Washington Post does not report them), they are what most election watchers want to see. Burned by the early release of numbers in 2004, the exit pollsters instituted a quarantine room, with no data at all getting out before 5 p.m. EST. How, if at all, might they react to continued problems?