So, is the Republican wave there or not? Some pollsters say yes, others say no. Why? As I wrote in September, the difference is in predicting who will actually vote in this election.
A Wall Street Journal article written by the assistant director of Quinnipiac polling explains the situation very well:
When Gallup questioned potential voters recently, it found the country could either be headed for a Republican landslide, or for just a normal off-year election loss for the party controlling the White House — the Democrats.
The difference depended on who the nation's oldest and most venerable polling firm decided was likely to actually go to the polls. In one scenario, Gallup saw the GOP with a slim three-point lead nationally; in another, the margin was 17 percentage points.
How Gallup, and every other pollster, makes that determination dictates their forecasts for November and explains much, if not most, of the fog surrounding the size of expected Republican victories.
I am convinced this is why the Elway and CNN polls show Patty Murray with a comfortable lead, while Rasmussen shows Dino Rossi ahead slightly, and SurveyUSA calls it a dead heat.
The same thing is happening in the 3rd Congressional District. SurveyUSA shows Republican Jaime Herrera well ahead, while other polls show the race very close.
So who should we believe? In defending his results, Stuart Elway argues that automated polls, such as those conducted by SurveyUSA and Rasmussen, are less reliable than polls using live interviewers. Recent history, however, doesn’t show that to be the case.
The only way to measure a pollster’s accuracy is to compare their final poll to the actual election results, if they release a poll within a week or so of election day. SurveyUSA routinely releases a poll just before the election, and in 2008 they were within 1 percent of the actual Washington state results for governor and president. Rasmussen’s final polls came earlier than SurveyUSA’s, but their results were also largely accurate.
The Elway poll, on the other hand, often has been somewhat off. In September 2004, the Elway poll showed Christine Gregoire 11 percent ahead of Rossi, and Murray 20 percent ahead of George Nethercutt. This was in sharp contrast to what other polls were showing at the time, and dramatically overstated the vote the two Democrats would receive two months later. In 2008, the last two Elway polls showed Gregoire ahead of Rossi by 16 percent and 12 percent while other polls showed a close race. Gregoire won by 6 percent.
There will be more polls between now and when we receive the truth from the voters. It will be fascinating to ultimately learn who was right and who was wrong.