That Washington
Polling Guru: WA is Polling's "Bermuda Triangle"
Polling guru Nate Silver of FiveThirtyEight.com, writing in the New York Times, says polling in Washington State's US Senate race has been "all over the map," with some polls leaning heavily toward incumbent Sen. Patty Murray even as others favor her challenger, Republican Dino Rossi.
This, Silver notes, isn't a new phenomenon. For example, Rasmussen has strongly favored Republicans in the past---overestimating the performance of the Republican candidate by an average of four or five points during the past decade---while SurveyUSA and Elway Research have overestimated the performance of Democrats by an average of three to five points.
Why is Washington State so unpredictable? One possibility Silver raises is our higher-than-average percentage of cell-phone users (pollsters only call land lines); another is the state's large population of Asian immigrants, who may have language barriers that prevent them from responding to English-language polls.
Silver's most interesting suggestion, however, is that Washington "is one of two states, along with Oregon, where voting takes place almost entirely by mail ." This, Silver writes,
It's an interesting analysis, but I'd add one caveat: SurveyUSA polls, at least, ask respondents whether they've already voted, which seems to address Silver's final point.
This, Silver notes, isn't a new phenomenon. For example, Rasmussen has strongly favored Republicans in the past---overestimating the performance of the Republican candidate by an average of four or five points during the past decade---while SurveyUSA and Elway Research have overestimated the performance of Democrats by an average of three to five points.
Why is Washington State so unpredictable? One possibility Silver raises is our higher-than-average percentage of cell-phone users (pollsters only call land lines); another is the state's large population of Asian immigrants, who may have language barriers that prevent them from responding to English-language polls.
Silver's most interesting suggestion, however, is that Washington "is one of two states, along with Oregon, where voting takes place almost entirely by mail ." This, Silver writes,
can wreak havoc with traditional likely voter models, which often ask questions like, “Have you voted in the election precinct before?” and “Do you know where people in neighborhood go to vote?” — questions that are nonsensical in the context of an election that takes place by post. Also — probably because of mail balloting — turnout in Washington and Oregon has generally been very high, so targets that might work well in other states could fail there. Finally, since many voters in Washington return their ballots well in advance of Election Day, a pollster surveying the race close to Election Day will encounter another type of voter — those who claim to have voted already — which traditional likely voter models are not well designed to handle.
It's an interesting analysis, but I'd add one caveat: SurveyUSA polls, at least, ask respondents whether they've already voted, which seems to address Silver's final point.
