Predicting Washington: polling in a vote-by-mail world

Jeff Alworth

The balance of power in the US Senate may come down to the West Coast. And with Barbara Boxer apparently solidifying her lead over Carly Firorina in California, it may come down to Washington. The good news is that recent polling shows incumbent Patty Murray beginning to edge Dino Rossi. An Elway Research poll (a local firm) put Murray ahead by 15 points among likely voters. That's an outlier, but a CNN/Opinion Research poll also gave her an 8-point lead.

However, as Nate Silver points out, Washington state polls haven't been so accurate over the past decade:

One reason could be that it is one of two states, along with Oregon, where voting takes place almost entirely by mail. This 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.

A lot of this is familiar to Oregonians, but here's an Evergreen State quirk: both Rasmussen and SurveyUSA have given numbers biased toward the GOP by about 4.5 points over the past decade. And Elway, the group that gave Murray the 15-point lead this week has, over the same period, been biased toward the Dems by about 3%.

Fascinatingly, in Oregon, the GOP biases of Rasmussen and SurveyUSA are far smaller (1.5% and .8%), even though the two states are similar in voting method, political orientation, and demographics. So riddle me this pollman: what is it about the two states that differs and makes the Washington results so much harder to predict?

I'm not completely convinced by Silver's explanation:

A few other things about Washington State are worth considering. First, Washington, a technologically savvy state, has a somewhat higher than usual number of cellphone-only adults, which pollsters may not be capturing. It also has a fair number of Asian-American voters, who can sometimes be hard to reach because of language or cultural barriers. These things could make a difference at the margins.

Anyone have a better idea of what's going on here?

  • (Show?)


    This is a surprisingly error filled posting by Silver. We've responded to it here, but in brief:

    1) WA did not go "full" VBM until 2006 (and is not there yet) so Silver's comparisons for WA going back to 2000 are wrong.

    2) A does not connect to B in Silver's posting. If higher turnout and higher use of mail balloting leads to inconsistent poll results, then there are quite a number of other states where we should see the same phenomenon.

    3) Isn't it the case that, under high turnout, a likely voter model should be increasingly irrelevant? Yet Silver makes the opposite claim, that higher turnout should lead to less predictive models. That is backwards by any argument I can muster.

  • (Show?)

    Although Washington isn't "full" vote by mail; it is primarily so. I live in the only county, Pierce, that still operates polling stations on election day and even here 90%+ voters vote by mail. Beginning in 1999 Washington allowed anybody to register as a permanent absentee. The percentage voting by absentee was over 60% in 2000 and has increased every year since.

  • (Show?)

    Mary yes but WA is not unique in having a lot of no excuse Absentee voters. 70% of CO ballots came back through the mail in 2008.

    WA did not legislate full (required) VBM until 2006. Kari, yes pierce is the last holdout.

  • (Show?)

    Paul, thanks for the clarification. Still, it doesn't all add up. The consistency with which WA polls have been biased toward the GOP is impressive (which I think does eliminate mail as the answer--as the state moved more toward that form, it should have affected the polling).

    Do you have any insight as to why WA would be not only biased toward the GOP, but different from Oregon's polling?

  • (Show?)

    UW gives Murray another 8 pt. lead:

  • (Show?)

    Jeff, sorry, been traveling on college visits with my daughter. I'll try to look closer at the polling biases, but right now, I am having to collect early vote numbers. The votes are coming in fast!

connect with blueoregon