Extra Fizz: Mallahan, Constantine Ahead in Big-Ticket Races. R-71 Ahead. 1033 Down.

By Josh Feit, Tuesday, October 27, 2009 at 9:24 AM
View Comments

Update/Caveat: The sample size in the poll is not standard. The pollsters report they polled 724 people statewide. That’s solid, but if you break that down for Seattle, you’re talking about 65 people (about nine percent of the state’s population.) That’s not a reliable sample on the Mayor’s race,  for example.

The pollsters do report that they oversampled in Seattle, but then that may have skewed the numbers on I-1033 and R-71.

All in all, lots of questions about the poll. And the PI.com reminds us that in August the Washington Poll had McGinn in fourth place at 9 percent—losing the primary behind Nickels, Mallahan, and Donaldson.

In fact, McGinn won the August 18 primary, 27.7 to Mallahan’s 26.8.

Update: UW political science professor Dr. Matt Barreto, who heads up The Washington Poll, says they polled 400 Seattle voters.

A new poll released this morning by The Washington Poll at the UW has Joe Mallahan leading Mike McGinn in the mayor’s race 44 to 36 and Dow Constantine beating Susan Hutchison 47 to 34 in the King County Executive’s race.

The poll also has a snapshot comparing voters before and after McGinn’s big “I wont be a baby about the tunnel” announcement.

Before: McGinn 35 vs. Mallahan 34.

After: McGinn 29 vs. Mallahan 45.

And the crosstabs have the stuff you already knew: McGinn is ahead among Democrats and young people; Mallahan is ahead among older voters and Republicans; and Constantine is beating Hutchison among Democrats 72 to 7 while losing to Hutchison among independents 57 to 25.

The Washington Poll also looked at I-1033 (Tim Eyman’s state government spending cap) and R-71, the measure to affirm or nix expanded domestic partner rights.

Eyman’s 1033 is losing 46 to 41 among all voters. It’s also losing among likely voters, 49 to 40, and it’s losing among those who have already voted, 56 to 44.

R-71 (remember, a vote to “approve” means gay couples keep their rights) is winning 56 to 39 among all voters, 57 to 38 among likely voters, and winning 55 to 45 among those who have already voted.

  • Black_America
    Who told you African American will break to Mallahan.

    We have nothing in common with the Tunnel Dude.
    McGinn understands our issues, and many of the African American i talk to like McGinn better.
  • South-emder
    Well, Mr X, African Immigrants do vote in large number, but many American, including King County Election Office can not tell the difference between African American and African immigrant-- we all look the same, don't we-- Mr X.

    There are about 40 to 50 thousand African Immigrants in Seattle, most of them are citizen. About 90% of those who are eligable to vote are registered-- because of Obama FeVer. so i am sure-- we will have some surprises on this Nov 3rd.
  • Michael M.
    @72

    If we're going to get into Demographics of how the whole shebang will turn out -

    I agree, McGinn will likely have a slight edge amongst the Asian population. African Americans, I think, will break for Mallahan. European immigrants I also think will break for Mallahan. Gays will give Mallahan an edge as well, as will women. And, of course, the ever important white vote - white women will break Mallahan by a higher percentage than white men, but I'm guessing they all break Mallahan.

    18-25 year olds I see breaking McGinn hard, but 25-40 year olds I think are going to be more evenly split. 40-65 will be Mallahan, by a small amount, while the 65+ overall will break strong Mallahan (except the old school liberals).

    That's just my guess, based on polling, and chatting with people out and about.
  • 70. I agree with much of that, X, though McGinn seems to have the Asian vote down, if recent polls and this video are any indication.
  • Snooklepuff
    @37 - You've got my vote!
  • Mr. X
    The discussion of which candidate the new immigrant SE/SW Seattle population will go for is fascinating, but since white women over 50 in the 43rd and 46th Districts usually decicide Seattle elections, it's likely to be irrelevant.

    My two cents on the Asian vote is that it will go slightly for Mallahan, but will probably be reasonably equally split. It ain't gonna be a slam dunk for McGinn, though. Uncle Bob still carries weight with a lot of folks, and his endorsement will make Mallahan seem safe to a lot of people.

    I have no idea how members of the African immigrant community are voting in this race, but with due respect to the poster @59 I don't think they historically vote in local elections in large enough numbers to have much effect. I truly wish that young people, poor folks, and recent immigrants were more active in municipal races, but by and large they're just not.
  • Michael M.
    @67

    I didn't even know that Peter was from Africa. I just knew he was an immigrant, which the previous poster (as you may have noticed) stated, with great authority, is behind McGinn.

    Truth be told, all of the Vietnamese (I know a few), Thai, Chinese and Korean immigrants and children of immigrants that I know are supporters of Mallahan. Of course, I'm sure there are some people I know that are immigrants that are supporters of McGinn, but I haven't heard them proclaim as such.
  • Kenyan
    Who is Peter Masundire? Is he another lost Boy..
  • Sh-Ali
    Peter Masundire is Simbabwian. west Africa.. Majority of African in Seattle 89 to 90% are East African.. Do you know any East-African who endorses Mallahan.. I gues the answer is NO.
  • Michael M.
    @65:

    Try looking at older posts (I find it's especially prevalent in September), and you'll find that Publicola has had no problem going after McGinn (just as the Times had no problem going after Mallahan).

    They have since both made their endorsements, and is their coverage of the other candidate a bit harsher? Sure. But I've still found that, when called for, they each have no problem being tough on their preferred candidates.

    Is Publicola's reporting perfect? By all means, no. But I do find their "dose" of objectivity to be far more objective than the Stranger, who would NEVER say anything bad about McGinn.
  • chicagoexpat
    @ 56

    if you break that down for Seattle, you’re talking about 65 people

    with "analysis" like that, you show not only that you don't have a clue about modern polling (which you purport to write about).

    you're clearly simply a propoganda vehicle for a real idiot -- not only "printing" all the McGinn PR's virtually intact (as your own web site's analysis), but even inserting editorial comments WITHIN MCGINN QUOTES when you don't think he attacks Mallahan strongly enough.
  • Michael M.
    @63 -

    I can't not respond to everything today. Publicola has become my new addiction. Damn quitting smoking!!!!
  • 61. Some anon dude's trolling, Michael.

    FWIW Asians will make up about 11-13% of voters and those not undecided are leaning 40-32 for McGinn per SurveyUSA's 10/20 release. Blacks and Hispanics make up only 7-9% of likely voters and often there's too little of a sample in polling to note any real trends, but SUSA got a good sample for their 10/20 release that showed Blacks who weren't undecided leaning heavily towards Mallahan (43-25).
  • Michael M.
    @60 - That is a good question. But Mallahan is up, so no matter how it's spun by McGinn, as long as the facts (400 people, MOE of 4.?%) are reported, I'm okay with it.

    And after the results are in, there will be no more spinning. Just an unemployed ex-lawyer, a new mayor, and a couple new city council members.
  • Michael M.
    OMG @ 59 - You sure were able to check with each and every Vietnamese, Japanese, Chinese, Filipino/a, Cambodian, East African and other immigrants quickly. I'm impressed.

    And for those who have citizenship, and are registered to vote, I highly doubt we'll see go 100% for either candidate.

    Hell, I know that Peter Masundire has endorsed Mallahan...so that's at least one that you're leaving out.
  • Pete
    @58, I've asked, with no response (sound familiar?). But I think we know what party has more at stake here and therefore who would be quickest to question the poll. I don't mind that they "go after" candidates, that's actually what I want them to do. I just think we should know if their getting their "scoops" fed to them and by whom. Turn the tables, if the poll had shown McGinn up and they received an angry email from the Mallahan campaign questioning the polls methods, what do you think they would do with it?
  • Asian
    Hey, Don't worry uncle Bob has sold his soul to the busness establishment, so No Asian will take him serious. He is 80s anyways.

    It is true No one called us 30.000 Viatnamis, 1800 Japanes, 50,000 Chines, 20.000 Philipins, 10,000 Cambodian, 45000 East-Africans and other 20,000 immigrants. We can make a big differencs, and sure--you will see a different result on Nov 3rd. futher info call 1800-McGinn for immigrants.
  • Michael M.
    @57 - Or maybe that campaign is ripping off the Publicola? I don't know. I'm a Mallahan supporter, and while it pisses me off every time they trash my man, I do love when they go after McGinn (even if it's been awhile). And that guy that I was commenting to is a bit crazy-ass.
  • Pete
    @56, well unless someone wants to comment otherwise, I have to assume that P-cola is a tool for one of the campaigns here. At least the Stranger quoted the McGinn campaign as the source of the claimed small sample size.
  • Michael M.
    @Chicagoexpat:

    Calm down. While I agree that the gun was jumped a bit here (see my original post re: wanting to see the 1033 and R-71, two campaigns that most, if not all, of the regulars here are in agreement on how they should go), this is also a relative newcomer to the local political scene, and as such, needs scoops (granted, I totally scooped them, mentioning the poll in the thread for the morning fuzz a good ten minutes before this was posted...HA!)

    As more facts came out, they updated their original post. Where they got each update? I don't know. What I do know is that, once they had the 400 person sample, Josh made that clear. A true propaganda machine would call that into question, and stick with the 65 number flouted by the floundering McGinn campaign.

    Just remember - when the Mallahan administration is pissing you off, Publicola will be there to report. And not what the Stranger will be doing, but actual reporting. It's a good thing.
  • chicagoexpat
  • chicagoexpat
    i mean, c'mon, what happened to "this poll sampled only 65 Seattle voters"
  • chicagoexpat
    gawd, youse guys who drank the kool aid for mcginn don't even hae the slightest clue what opinon pollin and "oversampling" for the purpose of getting a good sample in a sub-are ia all about.

    From a far more reputable sourde:
    "The poll sampled 400 likely registered voters in the city of Seattle, with a margin of error of 4.9 percent."


    But y'all keep wishing on that star
  • Michael M.
    @51 - one issue I did have with the poll was the length of it. It was taken over a week long period. That's a LONG time for a poll to be taken (although it was able to take into account the before/after effect of the tunnel flop by McGinn)

    Of course, the most important poll is going on now. King County has just over 10% of the ballots returned. Considering how few undecideds there are, that number should be a crap load higher. I'm assuming everyone here has turned in their ballots.
  • There's one big advantage to automated: There's no asker to frame or load a question in a way designed to coax a particular answer. It helps too that their questions are neutral in scope: Who are you voting for? How old are you? What's your political ideology?

    It's just black and white 'here is the question, give us an answer'.

    While the poll itself was unbiased, their polling sample was likely limited and skewed in scope despite their best efforts to diversify it.
  • Michael M.
    @48 -

    The problem with SurveyUSA is that it's all automated polling.

    What makes me sad is that Elway hasn't done a poll this year. He's had a pretty good track record in this state and region.

    And this poll is a non-partisan poll, done by thinkers at UW, and as such should be given more credence. It's done by people that have a much better understanding of the way local politics work, as opposed to the robo-calling SurveyUSA.
  • What the Puck?
    @45 There's no way to know.....no.

    Actually self-identified party ID is very predictive of vote on a whole variety of things, even with the (very few) super-cool renegade Democrats like you who choose Republican or Independent on robopolls to "screw with their bogus methodology."

    Polling done right is extremely accurate and guess what, it also helps the issue campaigns and candidates you care about spend their limited resources effectively.
  • I will say this about this poll: The only polling service whose results I tend to give more than a grain of salt about is SurveyUSA, since they do non-paritsan polling for all 50 states and their methodology is the closest to airtight. These polls, while their practice may be basically sound, isn't going to tell us too much due to the sample noise and the limited scope of the administrators. Take it for what it is, folks: A small sample of current public opinion. Make too much of it or its faults, either way, at your own risk.
  • What The Pho'?
    @ 42
    Let's see Uncle Bob Santos and other community leaders, most of the dems reps, and most of the ngo in ID are for Mallahan. Which Asian and East African block are you referring to? The "Viatnamist" block of where McGinn has an office down in SE Seattle and East African block of Holly Park?

    @43. Fuck you. Your pseudo-Vietnamese, broken English
    is beyond offensive. I don't care which side you are on.
  • 39. Concrete details, Elaine. Neither of you have given them.
  • Kathryn
    There really is no way to know since we don't register by party. There are the people who pay party dues. There are the people that are PCOs. There are the people who signed in at the caucuses last year affirming their party choice.

    And, there are the people like me. Democrat on all the above measures, but chooses Republican or Independent on robocalls just to screw with their bogus methodology.
  • What the Puck?
    So we know nothing about the Party ID breakdown for this poll - how many D's, I's, and R's - which is crucial in all these races as it is a strong driver of how you vote, even in "non-partisan" races.

    Trying to reverse engineer the data from the King County Executive results suggests that they've got too many Dems in their sample but it's hard to know for certain. And without that info, this poll is largely useless.

    And polling over two weeks this close to the election is ridiculous. I might've been pro-McGinn last week (I wasn't) and pro-Mallahan this week (I'm not). Look at their own results by week for the Mayor's race to see how absurd it is to combine the results. It's one thing if people vote and send in their ballots so they can't change their minds, but I doubt many did.

    And 33% of Republicans are undecided in the County Exec race? Not a chance in hell -- completely inconsistent with other polling.

    They really need to release the full questionnaire with results to be credible.

    Absent that, caveat emptor.

    Bring on the Poll Nerd.
  • Viatnamist
    Those pollesters don't call us. I have been an-active member of the Viatnamist community, and no single person from our community have been called to ask who he/she will vote. So i tell you, McGinn has majority of the immigrant community in his side. and that will change the out-come on Nov 3rd. There are block voting going on in the south-end
  • Asian
    Asians and East Africans are block voting for McGinn. that may change the result
  • Pete
    @39, but the poll wasn't "suspect" as you claimed earlier. In fact, Josh's later post was what was "suspect", as it mimicked the McGinn campaign's email attacking the poll earlier in the morning.

    Lesson: get background information (such as sample size) from the pollster, not one of the campaigns.
  • UW political science professor Matt Barreto said hte sample size in Seattle was 400 and the margin of error was 4.9

    http://blog.seattlepi.com/seattlepolitics/archi...
  • elaineinballard
    @29 What Fat-tailed said in #34
    btw, informed polling analysis benefits all of us, regardless of the candidate or issue we support.
  • Michael M.
    Speaking of Port Commission - don't forget to mark the circle next to MAX VEKICH.
  • @36 I appreciate the support, but I'd prefer a write-in vote in John Creighton's race against himself.
  • Brian K
    @34 I think I'm writing in Fat-tailed for school board. We need kids to have better math and spelling skills once they get to write polls about "Hutchinson" and "Nichols" over at UW.
  • Pete
    Josh,

    Did you get the info questioning the Washington Poll from the email that the McGinn campaign began circulating this morning? Did you see McGinn's email before your post?

    Also, did the McGinn campaign tip you off to the Jared Smith "story" of yesterday? I noticed that McGinn was alleging a connection between the viaduct video and the Mallahan campaign via Smith in a story printed in the Times this morning.
  • Publicola could really stand to hire someone with a passing knowledge of how to look at a poll. Or at least have one of you there learn the vocabulary. It's just so *frustrating* that after all your combined years of political reporting, you still haven't grasped what it means for a poll to "oversample." Hint: sampling and weighing are separate.

    For better or worse, polling analysis is an important part of political reporting these days. Please oh please study up. Or get a math nerd. Or something.
  • Michael M.
    @31 -

    And if that's the case, that's a decent sample size.
  • Dorkestra
    #30 for Comment of the Day
  • gloomy gus
    Slog heard back from the pollster that the Seattle sample size was 400.
  • johnmocha
    Its hard to figure out what McGinn is for other than being self righteous.
  • 23. I know whose side you're on, but I'm not sure what point you're exactly trying to make. What's wrong with the poll exactly?
  • Michael M.
    @21 -

    Normally I would agree with you, but this year's election is so...different...from most. And while 10% may be high, I definitely think it will be up around 8%, if not 8-10.

    Or there will be a lot of under-votes for Seattle Mayor.

    In the end, the SurveyUSA poll, this poll, the Publicola poll - they're all spelling out a bad bit of news for McGinn. If he can't break 40%, that's a problem this late in the game. I'm not saying Mallahan being stuck in the low to mid 40's is good for him, but McGinn's numbers are especially bad for him.
  • 13. That is based on a consistent set of samples from SurveyUSA, which asks each person for a party affiliation every time they conduct a survey.
  • Transpo guy
    Josh: I'm sure Barreto weighted the results to ensure that Seattle responses don't distort the overall results. But, your point about the Seattle sample size still holds.
  • Stacy
    Thanks for digging into the #'s Publicola.
  • Mikos
    Their method of oversampling for Seattle has not proved to be an accurate way to poll this race so far. Better to do three separate polls, Seattle, King County, Washington state. Also, the margin of error for the county and city race is given as 3.6%. That's hard to believe given the method of polling.
    That said, the numbers for the mayoral race aren't that far off from the Survey/USA numbers.
  • elaineinballard
    Josh, I hope that this is a lesson on how to report poll results in the future. Copying and pasting the numbers doesn't tell the story. There are so many reasons that this poll is suspect, which your update now accurately reflects. Please do a thorough review of this poll and post another story. For help on analyzing polls, this link is very useful:
    http://www.aapor.org/Questions_to_Ask_When_Writ...
  • Pete
    I was wondering the same as @19, is the 65 number you added as a note your computed sample size or did you get that from the Washington Poll? I also expect (!) that they took the KC oversample in to account in reporting statewide results.
  • Transpo guy
    Michael M @12:

    Ther is no wag that write-ins will come in at 10%. That just doesn't happen, even in a race like this. More likely around 3%.

    Stacy @14: Good point. This poll has a much better track record at the statewide level, which is great news re: 1033 and R-71.
  • Polling somebody pre and post McGinn straddling the tunnel is not a good sign for him.
    That might have something to do with him walking that back to more opposition talk, as compared to the press conference.
  • Michael M.
    @ Publicola - are ya'll going to email/call the folks at WaPoll to get a better idea of the sampling size in Seattle, and how they weighted the oversample to make the statewide and KC polls more...accurate?
  • rob
    The "undecideds" are still the 100 pound gorilla in the mayor's race.
  • Michael M.
    Of course, the big question - if Mallahan wins, will the McGinn supporters call shenanigans on the voters, saying that it was not a good poll? ;-)
  • Pete
    @14, the primary was a much different story. McGinn and Mallahan both surged late in the primary picking up undecideds in large numbers and getting votes from the other candidates as voters realized they weren't viable. It's interesting to note that Nickels numbers moved very little throughout the primary campaign. This tells me that voters had already decided not to vote Nickels and waited until the very end to decide among the rest.

    But there could be good news in these numbers for McGinn, since there was still a large number of undecideds and the poll was conducted prior to most of the debates. Looks like a nail-biter.
  • Michael M.
    @14 -

    And it also had Nickels at 23%, with a host of undecideds. I'm not sure if you remember, but the primary was all about late choices. 37% were undecided in that poll.
  • Stacy
    This poll is a complete joke. It doesn't even provide the sample size for Seattle! These same "pollsters" had McGinn at 8% two weeks before the primary, and we all saw how that turned out.
  • Michael M.
    @9 -

    We don't register. Where do you get your numbers?
  • Michael M.
    @7

    It's all about personal identification. I know plenty of people who ALWAYS vote for Democrats, but consider themselves Independents. I also know a few people who oftentimes vote for Democrats, but consider themselves Republicans. As Ivan would complain about, many people, even in "liberal" Seattle, like to consider themselves above partisanship, even if their voting history says otherwise.

    That being said, the Hutch even scored nearly 15% of the vote in the 43rd, the highest performing (for Democrats) district in the State, and that was her lowest performance. Considering how many Democrats and Independents voted for her, I would venture a guess that even the 43rd is 8-9% Republican. The 37th, 34th and 46th would all be higher (possibly even by significant amounts). Overall, it wouldn't surprise me if at least 15-20% of Seattleites considered themselves Republicans.

    At the same time, I would venture a guess that around 40% would consider themselves Democrats, the rest Independents (or Greens, Libertarians, Socialists, etc). 40% may even be a bit high.

    Additionally, while McGinn is up by a large margin amongst declared Democrats, the undecideds are still significant (28%).

    More importantly, however, older voters, 41+, are breaking Mallahan, and have fewer undecideds than McGinn's base of the 18-40 y/o voters (of whom, according to this poll, are 44% undecided).

    I still think this will be 50% Mallahan, 40% McGinn, and 10% write-in, in the end. I just have a feeling that this poll is grossly underestimating the write in vote.
  • As for the results, if this poll accurately reflects the voter sway from the McGinn announcement, then he's toast. Gaining no extra votes is bad. Losing voters who were on your side is worse.
  • @6, Sorry. Flipped the numbers. Fixed that. Thanks.
    @5, That's 39 not 49. Sorry. Thanks. Fixed.
  • 7. Around 60-65% of likely voters are Democrats, ap. Many Seattle voters aren't registered Democrats. A few are Republicans, and many others don't affiliate with either the party or a strictly liberal ideology.

    Hard to believe given your peers and chosen news sources, I know.
  • Character is the biggest issue of the Seattle Mayoral race. Can long-term progressive results emerge from an unethical campaign? A promising if inexperienced candidate, Mike McGinn, is testing this proposition. wwww.lightandair.wordpress.com
  • ap
    if mcginn is up among democrats, by 9 points, theoretically he should be able to win.. there aren't that many independents or republicans in seattle, are there? i can't believe there's still such a huge undecided vote at play.
  • Michael M.
    Oh, and Josh - it's losing among likely voters, 49-40. Minor detail ;-)
  • Jarvis
    Josh, there must be an error in the R-71 results. 56% + 49% = 105% (???)
  • Michael M.
    @2,

    I'm completely relaxed. I just like exclamation points (!!!!!)

    Okay, not completely relaxed...all of these polls got me excited, especially the cross tabs. :-)
  • dubious
    Gimme a break. Why should we give any credence to folks who don't even know how to spell Hutchison or Nickels?
  • @1,

    You totally need to relax. I wanted to get the Mayor and Exec up fast.
  • Michael M.
    You totally left out 1033 and R-71!!!!
blog comments powered by Disqus