November 25, 2007 - December 1, 2007


"Outliers" for 11-30

Andrew Sullivan sees a flaw in the composition of the Hunter College survey of gay, lesbian and bisexual voters (more here).

Jay Carney crunches the Rasmussen second choice numbers in Iowa.

Jennifer Agiesta looks at how likely Iowa caucus goers rate the importance of various issues and finds only an issue on which one candidate "break[s] out from the pack as the one voters' trust most: Sen. Hillary Clinton on health care."

Jay Cost looks at how the ratings for this week's CNN/YouTube debate stacks up against the other debates this year.

Joe Klein has a first person account of a Frank Luntz debate "dial group" in Florida.

Jonathan Martin and Charles Babbington report that the curtain is rising on "Act Three" of the Republican IA/NH primary campaigns.

Doug Schoen asks who benefits most from a Huckabee surge in Iowa.

AAPOR's Nancy Mathiowetz considers the need for greater disclosure in the context of the controversial Zogby internet poll.

And you've probably seen this already, but The Caucus reports that the Clinton campaign "polled" the impact of a Barbara Streisand endorsement (via Sullivan and many others).

Re: Frankovic on Iowa's "Challenge For Pollsters"

Topics: 2008 , The 2008 Race

Yesterday, I blogged about the possibility that the age estimates in the 2004 Iowa Caucus "entrance poll" may have understated the percentages of younger (under 45) voters. I also wondered if the poll also overstated the percentage of self-reported first time caucus goers (55% on the poll), as I assume past caucus participation to be related to age.

Well, we can confirm that past caucus participation is strongly related to age, thanks to Sarah Dutton at CBS News who ran the data from the 2004 exit poll for us. Here are the numbers:


Various surveys have shown strong relationships between Democratic vote preference and both age (Obama does better among voters under 45) and past-caucus history (Edwards does better with past caucus goers). The data above show these two characteristics are strongly related to each other. And here's the important part: Not only are we uncertain about what the "right" mix of age and past caucus experience will be on January 3, 2008, but the strong likelihood of non-response bias toward younger participants in the 2004 entrance poll should make us treat the estimates of both characteristics from that sample "with caution" (as Kathy Frankovic put it).

Incidentally, the CNN table of the 2004 entrance poll result appears to mislabel the question about past caucus participation. The question was whether voters had attended any caucus before [Update - complete text is: "Before tonight, had you ever attended an Iowa Democratic presidential caucus?"]. So a "no" indicated a first-time caucus-goer.

Typo corrected

POLL: Fox New Hampshire Survey

A new Fox News/Opinion Dynamics statewide survey (Dem story, GOP story, results) of 1,000 likely primary voters in New Hampshire (conducted 11/27 through 11/29) finds:

  • Among 500 Republicans and those who lean Republican, former Gov. Mitt Romney leads Sen. John McCain (29% to 21%) in a statewide primary; former Mayor Rudy Giuliani trails at 19%, former Gov. Mike Huckabee at 7%.
  • Among 500 Democrats and thsoe who lean Democratic, Sen. Hillary Clinton leads Sen. Barack Obama (30% to 23%) in a statewide primary; former Sen. John Edwards trails at 17%, Gov. Bill Richardson at 12%.
  • All other candidates receive less than five percent each.

POLL: SurveyUSA Massachusetts '08 Match-ups

A new SurveyUSA automated survey of 540 registered voters in Massachusetts (conducted 11/9 through 11/11) finds:

  • Presidential Match-ups:

    Clinton 50%, McCain 45%
    Clinton 54%, Giuliani 40%
    Clinton 59%, Romney 35%
    Clinton 61%, Huckabee 31%

    McCain 47%, Obama 44%
    Obama 46%, Giuliani 44%
    Obama 54%, Romney 37%
    Obama 57%, Huckabee 27%

POLL: Rasmussen New Hampshire GOP Primary

A new Rasmussen Reports automated survey of 881 likely Republican primary voters in New Hampshire (conducted 11/29) finds former Gov. Mitt Romney (at 34%) leading former Mayor Rudy Giuliani and Sen. John McCain (both at 15%) and former Gov. Mike Huckabee (14%) in a statewide primary; Rep. Ron Paul trails at 8%, former Sen. Fred Thompson at 3%.

Iowa: Obama Trending Up

Topics: Iowa

Today's new ARG poll of likely caucus goers in Iowa generally confirms the trend reported by other polls that we see in our Iowa chart. Barack Obama has been rising in recent weeks, while Hillary Clinton's support -- which had been trending upward since the summer -- has mostly leveled off.


Keep in mind that the regression based trend lines on our charts are more conservative by design. As Professor Franklin frequently reminds us, that characteristic helps us isolate real changes while filtering out the short term "noise" from differences in methodology or simple sampling error.

Another lower tech way to try to see trends amidst the noise of differing methods is to do what I call "apples- to-apples" averages of comparable polls. In this case, ARG, Rasmussen and Strategic Vision have all fielded surveys in the last week or so, as well as earlier in the month. Whatever qualms we have about their methodologies, their surveys are done the same way each time and are at least helpful in spotting trends. And while the change is relatively small, they are essentially consistent: All three show small increases for Obama. On average, the three surveys indicate a three point gain for Obama and a one-point drop for Clinton.


Has Barack Obama "moved ahead" of Hillary Clinton? Neither our chart, this "apples-to-apples" average nor any of the polls individually (if we consider sampling error), shows that yet, but the trend is clearly moving in his direction.

Dueling SC Results

Topics: 2008 , Divergent Polls , Measurement , The 2008 Race

After we posted the results from the Clemson University Palmetto poll of South Carolina on Wednesday, a reader noticed the exceptionally high undecided percentage reported for Democrats (49%) and concluded that "something's wrong with these numbers" (the Republican sample also showed a higher than average 28% in the undecided column). Charles Franklin responded with a comment worth promoting:

Nothing's "wrong"-- the poll just let huge numbers of people say they hadn't made up their mind!

The release quotes the question wording as "If the 2008 presidential primary in South Carolina were held today, for whom would you vote?"

There is no indication whether the question LISTED the candidate choices or forced respondents to think of names on their own. That would make a difference. Further there was apparently no follow up to get "leaners" accounted for.

This is a good example of the kind of academic poll (commercial pollsters seem to NEVER do this) that deliberately chooses to allow respondents to remain"undecided". The academics who do this seem to believe that this is more revealing of how unsettled the preferences are and is therefore a more realistic view of how malleable opinion is. I see their point, but it makes the poll virtually unusable because it is so far out of line with more conventional polls that push for a direction of preference.

Given how volatile preferences can be once the voting actually begins, the academics may have a good point, but the 50% undecided isn't very helpful for knowing how opinion is forming.

Today, of course, we have new set of South Carolina surveys from the American Research Group which shows only 11% in the undecided category among Democrats (10%) among Republicans. This survey shows Hillary Clinton with a 24-point lead (45% to 21%), while the Palmetto poll showed things much closer between the top candidates (Clinton 19%, Obama 17%).

Now, sampling "likely voters" in the low turnout South Carolina primary is nearly as challenging as in Iowa, and there are many potential explanations for this difference beyond the likelihood that the Palmetto poll prompted for undecided while the ARG surveys do not. However, even if we take both sets of numbers at face value, they may be telling us that the Clinton lead in that state is soft; that voters there who currently prefer Senator Clinton (when pushed) are softer in their support than those leaning to Obama or the other candidates.

All of this gets at a point I've made several times recently: A preference is not a final decision.

POLL: American Research Group IA/NH/SC

Three new American Research Group statewide surveys of likely voters in Iowa, New Hampshire, and South Carolina (conducted 11/26 through 11/29) finds:

  • Among 600 Democrats in Iowa, Sen. Barack Obama edges out Sen. Hillary Clinton (27% to 25%) in a statewide caucus; Sen. John Edwards trails at 23%, Sen. Joe Biden at 8%. Among 600 Republicans, former Gov. Mitt Romney narrowly leads former Gov. Mike Huckabee (28% to 27%); former Sen. Fred Thompson trails at 14%, Sen. John McCain and former Mayor Rudy Giuliani both at 9%.
  • Among 600 Democrats in New Hampshire, Clinton leads Obama (34% to 23%) in a statewide primary; Edwards trails at 17%. Among 600 Republicnas, Romney leads Giuliani (36% to 22%); Huckabee trails at 13%, McCain at 11%.
  • Among 600 Democrats in South Carolina, Clinton leads Obama 45% to 21% while Edwards trails at 12%, Biden at 6%. Among 600 Republicans, Giuliani narrowly leads Romney (23% to 21%) while Huckabee trails at 18%, Thompson at 13%, McCain at 10%.
  • All other candidates receive less than five percent each.

POLL: Hunter College Lesbian, Gay, Bisexual National Primary

Via Ben Smith, a new Hunter College national survey of 768 lesbians, gays, and bisexuals (conducted 11/15 through 11/26) finds:

  • Among 501 who plan to vote in a Democratic primary for president, Sen. Hillary Clinton leads Sen. Barack Obama (63% to 22%) in a national primary; former Sen. John Edwards trails at 7%, Rep. Dennis Kucinich at 5%. All other candidates receive less than five percent each.
  • Among 78 who plan to vote in a Republican primary for president, former Mayor Rudy Giuliani leads Sen. John McCain (50% to 23%) in a national primary; former Gov. Mitt Romney trails at 11%, former Sen. Fred Thompson at 10%. All other candidates receive less than five percent each.
  • Favorable/Unfavorable Ratings (n=768)

    82/12 Clinton
    73/16 Obama
    63/16 Edwards
    28/15 Kucinich
    30/45 McCain
    28/55 Giuliani

POLL: Zogby/Kennedy Louisiana Senate

A new Zogby statewide survey paid for by John Kennedy for U.S. Senate (R) of 1,001 adults in Louisiana (conducted 10/10 through 10/14; released 11/29) finds Kennedy leading Sen. Mary Landrieu (45% to 38%) in a general election match-up for U.S. Senate.

The Onion's "Bar Fight" Poll

Topics: 2008 , The 2008 Race

I'm having to spend some time doing some back room work here at Pollster, so I'm relieved I can pass along this video from The Onion summarizing results from their latest "poll" (includes a word or two NSFW):

And yes, for those unfamiliar, The Onion is a humor site; the video is satire. As is this clever news flash of interest to blog readers everywhere.

This video, on the other hand...

POLL: Rasmussen Iowa Democratic Caucus

Additional results from the recent Rasmussen Reports statewide survey of 1,156 likely Democratic caucus goers in Iowa (conducted 11/26 through 11/27) finds Sen. Hillary Clinton (at 27%) narrowly leading Sen. Barack Obama (25%) and former Sen. John Edwards (24%) in a statewide caucus; Gov. Bill Richardson trails at 10%. All other candidates receive less than five percent each.

POLL: Datamar GOP CA Primary

A new Datamar automated survey of 514 likely primary voters in California (conducted 11/23 through 11/27) finds former Mayor Rudy Giuliani leading former Gov. Mitt ROmney (28% to 16%) in a statewide primary; former Sen. Fred Thompson trails at 14%, Sen. John McCain at 10%, former Gov. Mike Huckabee at 8%. All other candidates receive less than five percent each.

POLL: InsiderAdvantage Post Debate Surveys

InsiderAdvantage/Majority Opinion released two new statewide surveys (conducted 11/28) of 1,035 registered Republicans in Iowa and 341 undecided Republicans in Florida who said they watched tonight's debate and were willing to call in after the debate to answer questions.

  • Among Republicans in Iowa, 32% believe former Gov. Mike Huckabee won tonight's Republican debate; former Gov. Mitt Romney gets 16%, former Mayor Rudy Giuliani 12%, Sen. John McCain 10%, former Sen. Fred Thompson 7%, Rep. Ron Paul 6%.
  • Among Republicans in Florida, 44% believe Huckabee won the debate; Giuliani gets 18%, Romney 13%, McCain 10%, Thompson 5%.
  • All other candidates receive less than five percent each.

More on InsiderAdvantage's methodology for their Florida survey can be found here.

Thanksgiving Leftover "Outliers"

Walter Shapiro remains a "a major-league skeptic about Iowa polling."

Janet Elder ponders whether the change in the data on last week's ABC/Washington Post Iowa poll matched the coverage or changed perceptions of the race it generated.

Todd Beeton reminds us of the Iowa and New Hampshire poll results this time four years ago.

Greg Sargent and Kos slam this week's Zogby-internet poll (see also the commentary from our own Charles Franklin).

The Democratic Strategist notices yet another Florida poll inconsistency.

Jennifer Agiesta reports another dip in the Washington Post-ABC News Consumer Comfort Index (CCI), hitting "its lowest point since October 2005."

Jason Zengerlee has a lengthy piece on "electability." Jay Cost reacts. See also my thoughts posted several weeks ago.

Dante Scalia continues to check the log books at New Hampshire's WMUR.

John Sides, David Park and Lee Sigelman, three political scientists at George Washington University, launched The Monkey Cage, a blog devoted to academically-oriented commentary on politics and the latest in political science research.

The Deseret Morning News looks at the "push poll"/message testing controversy surrounding home-town firm Western Wats.

POLL: Suffolk University New Hampshire Primary

A new Suffolk University/7NEWS statewide survey (release, results) of likely primary voters in New Hampshire (conducted 11/25 through 11/27) finds:

  • Among 300 Democrats and those who lean Democratic, Sen. Hillary Clinton leads Sen. Barack Obama (34% to 22%) in a statewide primary; former Sen. John Edwards trails at 15%, Gov. Bill Richardson at 9%.
  • Among 330 Republicans and those who lean Republican, former Gov. Mitt Romney leads former Mayor Rudy Giuliani (34% to 20%) in a statewide primary; Sen. John McCain trails at 13%, Rep. Ron Paul at 8%, former Gov. Mike Huckabee at 7%.
  • All other candidates receive less than five percent each.

POLL: SurveyUSA New York State General Election Match-ups

A new SurveyUSA automated survey testing general election match-ups among 516 registered voters in New York State (conducted 11/9 through 11/11) finds:

    Clinton 56%, Giuliani 37%
    Clinton 56%, McCain 37%
    Clinton 62%, Romney 31%
    Clinton 65%, Huckabee 27%

    Obama 46%, Giuliani 46%
    Obama 49%, McCain 43%
    Obama 55%, Romney 34%
    Obama 58%, Huckabee 29%

POLL: SurveyUSA New York State General Election Match-ups

A new SurveyUSA automated survey testing general election match-ups among 516 registered voters in New York State (conducted 11/9 through 11/11) finds:

    Clinton 56%, Giuliani 37%
    Clinton 56%, McCain 37%
    Clinton 62%, Romney 31%
    Clinton 65%, Huckabee 27%

    Obama 46%, Giuliani 46%
    Obama 49%, McCain 43%
    Obama 55%, Romney 34%
    Obama 58%, Huckabee 29%

Frankovic on Iowa's "Challenge For Pollsters"

Topics: 2008 , Exit Polls , Iowa , The 2008 Race

Kathy Frankovic, the director of surveys for CBS News, devotes her latest column to a subject regular readers know well: The Iowa caucuses and the special challenges they create for pollsters. Frankovic has long been a central player in the network consortium that conducts exit polling, so we should pay particular attention to one particular comment she makes about the limitations of past Iowa Caucus entrance polls:

In typical exit polls, tallies of the gender, race and apparent age of those not responding to the interview request are kept, and it is possible to adjust (and correct) the responses for those demographics. Historically, the largest non-response in exit polls has come from older voters. This often doesn't matter; it shouldn't matter if there is no difference in whom older and younger voters support. But historically that correction hasn't been made in the Iowa entrance polls. And this year, on the Democratic side, there is an age difference in candidate support. In the CBS News/New York Times Iowa poll, younger caucus goers -- those under the age of 45 -- favor Barack Obama: he gets 39 percent of their support, and Hillary Clinton 24 percent. The oldest caucus-goers -- those over 65 -- favor Clinton over Obama, 30 percent to 11 percent. Pre-election polls show less of an age difference among Republican caucus-goers.

The point here, in case it is not obvious: Non-response bias may have exaggerated the percentages of younger (under 45) caucus goers the 2004 Iowa entrance poll (something I wondered about a month or so ago). And since I'm assuming that age is strongly related to having attended a caucus in the past, the entrance poll estimate of the number of caucus newcomers in 2004 may be exaggerated as well.

PS: CBS has also recently launched a new feature, in partnership with the Roper Center, that provides a free, searchable archive of all questions asked on all CBS News surveys conducted since January 2007. That is a resource worth bookmarking.

POLL: Clemson University South Carolina Primary

A new Clemson University Palmetto Poll of 900 likely primary voters (story, results) in South Carolina (conducted 11/14 through 11/27) finds:

  • Among 450 Republicans, former Gov. Mitt Romney edges out former Sen. Fred Thompson (17% to 15%) in a statewide primary; former Gov. Mike Huckabee trails at 13%, Sen. John McCain at 11%, former Mayor Rudy Giuliani at 9%, Rep. Ron Paul at 6%.
  • Among 450 Democrats, Sen. Hillary Clinton narrowly leads Sen. Barack Obama (19% to 17%) in a statewide primaryl former Sen. John Edwards trails at 12%.
  • All other candidates receive less than five percent each.

Later Tonight: An Innovative Post Debate Survey

Topics: 2008 , Debates , The 2008 Race

The pollster/public relations firm known as InsiderAdvantage announced today that they will be conducting an "unprecedented" poll, sponsored by the Florida Chamber of Commerce, to determine "the winner" of tonight's CNN/YouTube debate:

Because public-survey phone calls are illegal after 9 pm, InsiderAdvantage placed over 100,000 calls to registered Republicans in Florida over the past week-and-a-half, with the criteria of finding undecided registered Republican voters who said they planned to watch the debate, and who agreed to call a special toll-free number immediately after the debate to answer questions about which candidate won the contest.
Data should be collected by 10:20 pm,” said Jeff Shusterman of Majority Opinion, InsiderAdvantage’s research partners. “We will weight the data for age, gender, and geographic location, and should have results between 10:30 and 10:45.”
The results will be announced by live webcast on www.southernpoliticalreport.com, and www.insideradvantage.com from the debate site in St. Petersburg, and will be released to the media immediately after.

I was curious about the mechanics of this project, so I called Gary Reese of InsiderAdvantage. According the Reese, they made those 100,000 calls using an interactive voice response (IVR) method in which respondents hear recorded questions and answer by pressing keys on their touch-tone phones. Although they dialed 100,000 of Florida's 3.9 million registered Republicans, they were able to identify only about a thousand households willing to dial back in.

So for those who see (or are tempted to report) the results, keep in mind that this survey can only truly represent a tiny sliver of registered Florida Republicans: The roughly 1% with no preference who were planning to watch the debate and willing to call back. Of course, the audience itself is likely to be not much bigger as a percentage of U.S. households. The CNN/YouTube Democratic debate in July drew an estimated 2.5 million viewers (the US has roughly 113 million television households).

This is an interesting, innovative project, and we will certainly link to the results when available. I sent an email to Republican pollster David Hill (also a weekly columnist for The Hill), partly because of his past skepticism of instant reaction surveys of debate watchers based on "panel-back surveys." Along with what he described as "the usual caveats about robo-calling and weighting procedures," he too considers this a "useful and interesting exercise." He added:

I have done this sort of advance placement calling to test advertising embedded in programming. It gives a qualitative flavor of viewer reaction that's invaluable. Let's just be careful about making blanket statements suggesting that this definitively answers the question, "Who won the debate?" But pollsters need to try new strategies like this.

To echo that thought, remember that with a single-digit audience size, the post-debate coverage will likely have far more impact on voters' preferences than the initial reaction of the debate audience. Also, the audience for a primary debates at this stage in the campaign tends to be better educated and better informed. It includes many more voters who have an initial candidate preference but may still be open to change. By focusing only on those who were initially "undecided" (18% of Republicans as reported on the just released InsiderAdvantage poll), we may miss reactions among the soft supporters of candidates who are not absolutely certain about their choice.

POLL: Rasmussen Iowa GOP Caucus

A new Rasmussen Reports automated survey of 839 likely Republican primary voters in Iowa (conducted 11/26 through 11/27) finds former Gov. Mike Huckabee narrowly leading former Gov. Mitt Romney (28% to 25%) in a statewide caucus; former Mayor Rudy Giuliani trails at 12%, former Sen. Fred Thompson at 11%, Rep. Ron Paul at 5%. All other candidates receive less than five percent each.

POLL: SurveyUSA California General Election Match-ups

A new SurveyUSA automated survey testing general election match-ups for president among 502 registered voters in California (conducted 11/9 through 11/11) finds:

    Clinton 52%, Giuliani 41%
    Clinton 53%, McCain 41%
    Clinton 57%, Romney 34%
    Clinton 58%, Huckabee 31%

    Obama 49%, Giuliani 42%
    Obama 52%, McCain 39%
    Obama 57%, Romney 32%
    Obama 58%, Huckabee 29%

POLL: InsiderAdvantage Florida GOP Primary

A new InsiderAdvantage statewide survey of 675 likely Republican primary voters in Florida (conducted 11/25 through 11/26) finds former Mayor Rudy Giuliani leads former Gov. Mike Huckabee (26% to 17%) in a statewide primary; Sen. John McCain trails at 13%, former Gov. Mitt Romney at 12%, and former Sen. Fred Thompson at 9%. All other candidates receive less than five percent each.

POLL: CNN Florida GOP Primary

A new CNN statewide survey (story, results) of 300 likely Republican primary voters in Florida (conducted 11/25 through 11/26) finds former Mayor Rudy Giuliani (at 38%) leads former Gov. Mitt Romney (17%), Sen. John McCain and former Sen. Fred Thompson (both at 11%), former Gov. Mike Huckabee (9%), and Rep. Ron Paul (5%). All other candidates receive less than five percent each.

POLL: Strategic Vision (R) Iowa Caucus

A new Strategic Vision (R) statewide survey of likely caucus goers in Iowa (conducted 11/23 through 11/25) finds:

  • Among 600 Republicans, former Gov. Mitt Romney edges out former Gov. Mike Huckabee (26% to 24%) in a statewide caucus; former Mayor Rudy Giuliani trails at 14%, former Sen. Fred Thompson at 10%, Sen. John McCain at 7%, and Rep. Ron Paul at 5%.
  • Among 600 Democrats, Sen. Hillary Clinton and Sen. Barack Obama both run at 29%, former Sen. John Edwards trails at 23%, former Gov. Bill Richardson at 6%.
  • All other candidates receive less than five percent each.

Anxious about "Single Anxious Women"

Topics: 2008 , The 2008 Race

[Today's Guest Pollster's column comes from Margie Omero, President of Momentum Analysis, a Democratic polling firm based in Washington, DC.]

It's almost official. Single women are poised to be the "Security Mom" or "Soccer Mom" of the 2008 election. They even have their own easy to remember moniker: the "Single Anxious Female." At first blush, it seems like a good thing for women. A woman top-tier candidate, a focus on women's issues and women's voters - it must be a good thing, right?

Actually, much of what you read about single women and voting is not borne out by the data. There is indeed a "Marriage Gap" among women. Married people vote at a higher rate than non-married people. But the marriage gap is actually larger among men. According to Census reports from the 2004 election, married men are as likely to vote (63% turnout) as married women (65%). But unmarried men (which includes single, divorced, separated, and widowed) are substantially less likely to vote (46%) than unmarried women (55%). The marriage gap is 10 points among women, and is nearly twice that (18 points) among men.

If you look specifically at single, never-married adults, this pattern holds. A majority of single women voted in 2004 (52%), compared to fewer single men (43%). This is even true with 18 to 24 year-olds (47% of single women in that group vote, compared to 40% of single men). The table below shows the turnout rate by gender and marital status. [Note: at the time of this post, the Census table contained an error, in that Row 88 (widowed men 18-24) should be blank, and all data currently in Rows 88-91 should be moved down one row. The error was corrected by email from the Census, but has not yet been updated on the site.]


You would never know about women's higher turnout by examining the press coverage. A CNN piece this month called single women, particularly younger single women, "notoriously difficult to get to the polling booth." An entire organization is devoted to closing the marriage gap among women . And women's advocates hypothesize about why single women don't turn out, making their own gender-based assumptions about women not recognizing their power.

More disturbingly, however, is what this focus on single women has wrought. Dubbed "Single Anxious Female," that cringe-inducing name has stuck, and has generated a sizable amount of press devoted to the caricature of the single woman. This group has become defined not by political views, but by their lack of gravitas. Several have called them the "Sex in the City voter". Feminist icon Naomi Wolf says they are more like Melanie Griffith in "Working Girl," as compared to Hillary's Sigourney Weaver. The CNN piece said this group is "more interested in showing off than in true political activism" and cited others who called single women "slutty" or "stupid." Advocacy has begotten dismissiveness.

Now, encouraging non-voters to vote is obviously important, and the point here is not to object to women-specific voting programs. And certainly campaigns should continue to reach out to women. But we need to change tactics. First, let's use the data correctly. Women vote at a higher rate than men. Unmarried women, however defined, vote at a higher rate than unmarried men. And this pattern holds across age groups. Second, it does not further the cause to allow women to be called anxious, show-offs, bubbly, stupid, or confused. These characterizations only perpetuate stereotypes about women, rather than work to improve our status.

New Strategic Vision (R) Iowa Coming Tonight

NBC's First Read is reporting a new Strategic Vision (R) Iowa survey showing:

    Romney 26%
    Huckabee 24%
    Giuliani 14%
    Thompson 10%
    McCain 7%

    Clinton 29%
    Obama 29%
    Edwards 23%

We will post the full results later this evening when they are available.

Zogby Internet Poll Trial Heats are Odd

Topics: Divergent Polls


A new Zogby Interactive poll, conducted using volunteers over the internet, has produced some odd results for trial heats involving Senator Clinton against all four top Republican opponents. What makes this especially odd is that the results are not equally unusual for Obama.

This poll was reported by Reuters' John Whitesides, who also reports on the Reuters sponsored polling Zogby does by conventional telephone methods. The similarities in the reports make it hard to tell, but apparently these results are not part of the Reuters-Zogby polling partnership, but are independent work by Zogby Interactive. Likewise Zogby's website posts the results without mention of who sponsored the work, so presumably Reuters did not.

The Zogby poll was conducted 11/21-26/07 with 9150 respondents who had agreed to take part in Zogby's online polling. This is not a normal random sample of the population. More on the technical issues below.

The hugely surprising result is that the Zogby poll finds Sen. Hillary Clinton losing to all four top Republicans in head-to-head trial heats. What makes that surprising is that Clinton LEADS all four of those Republicans in the trend estimates based on all other polling by between 3.8 and 11.6 points. Zogby also has Clinton losing to Arkansas Governor Mike Huckabee by 5 points. There are too few Clinton-Huckabee trial heat polls from other organizations for me to compute a trend estimate for that comparison.

The chart above shows all the trial heat data from national polling and the estimated trend lines for each pairing. The data points for the new Zogby data are indicated in the charts as "Zogby Inet" in blue for Clinton and red for each Republican.

What is immediately clear is that the Zogby Clinton numbers are well below the estimated trend for Clinton in each of the four comparisons. Clinton is consistently 8-10 points below her trend estimate based on other polling.

In contrast, the Republican results are quite close to the trend estimate in most cases: Giuliani is at 43 in Zogby, with a trend of 44. Romney is 43 in Zogby, 38.3 in trend; Thompson is 44 in Zogby, 41.3 trend, and McCain is 42 Zogby, 42.7 trend. Those Republican numbers are about the kind of normal noise we see around the trend estimate, so don't seem out of line.

Why then is Clinton so far down in comparison to other polls? The Reuters story doesn't note that these results are far from other polling, and instead uses the theme that Clinton is declining to frame these Zogby results:

The results come as other national polls show the race for the Democratic nomination tightening five weeks before the first contest in Iowa, which kicks off the state-by-state nomination battles in each party.

Some Democrats have expressed concerns about the former first lady's electability in a race against Republicans. The survey showed Clinton not performing as well as Obama and Edwards among independents and younger voters, pollster John Zogby said.

While this is certainly a theme of recent reporting, boosted by a pre-Thanksgiving ABC/WP poll showing Obama leading Clinton in Iowa, it is striking that no other poll has found recent results as far from the trend estimates as are Zogby's results and that the Reuters story fails to note that fact.

One answer to why Clinton does so badly MIGHT be that the poll has too few Democrats and thus biases its results. But if that were so, we'd expect Obama to also underperform his trend estimates. That doesn't happen, as the chart below makes clear.


The Zogby results for Obama are all quite close to his trend estimate from all polls:

Zogby has Obama at 46% vs Giuliani, while the trend puts him at 44.3. Against Romney Zogby has Obama at 46%, while trend says 46.6. Against Thompson Zogby has Obama at 47, while trend is 47.0, and against McCain Zogby has Obama at 45 while trend puts him at 43.4.

This is clearly not consistent with a general anti-Democratic bias in the Zogby Internet poll. It is also clear from the graph that the Obama pairings find Republicans doing quite close to the trend estimates as they did against Clinton.

(Trial heats against Edwards are not very common recently, so the Zogby results for him lack much polling for comparison.)

And so we are left with a puzzle: What is it about these respondents that so strongly affects Clinton support but no one else?

We can probably rule out one easy explanation: That Clinton has suddenly collapsed and Zogby is just the first to find it. The reason is internal to the Zogby result. If Clinton really has suddenly become 10 points less attractive, we'd expect all four Republicans paired against her to do BETTER than their trend estimates when facing her. But what happens is Clinton goes down and they don't do any better. That is hard to reconcile with a real change in Clinton's support. (A tortured version would say Clinton must have collapsed among Dems who now say they are undecided while refusing to move towards any of the Republicans. But that isn't usually what happens in real data when one candidate declines sharply. Usually the other moves up at least a bit, drawing not only from unhappy partisans but especially from independents who now are disenchanted with the former front-runner. So while you could make the math work with this story, it doesn't seem very well supported by the data.)

The Zogby Internet polling has a questionable track record in statewide races for Senate and Governor in 2006, where they often far over-estimated the competitiveness of races compared to conventional phone polls taken at the same time. One way to make sense of those problems turns out not to help much here. It is reasonable that the people who volunteer to take political polls over the internet are considerably more interested in politics (and likely more strongly partisan) than is a random sample of likely voters. That should be expected to lead to fewer people with "don't know" responses as better informed and more partisan respondents are likely to both know more about the candidates and to have made up their minds sooner than a proper random sample. That helps explain why Zogby's 2006 internet polls looked as they did.

But this does no good in Clinton's case. What we see is that MORE internet respondents are undecided about their vote between Clinton and four Republicans than the trend estimates based on less involved and partisan phone samples show. The Zogby undecided rates for the Clinton pairings are 20, 17, 17 and 16% (plus 17% undecided in the Huckabee comparison.) The comparable undecided rates based on the trend estimates are 8.2, 12.8, 9.0 and 10.6. That is an average undecided rate of 17.5 in Zogby vs 10.15 in the trends. Likewise the undecided rate is slightly lower for Obama pairings than it is for Clinton: 17, 13, 14, 13, and 14 for Huckabee. How could it be that a sample that is almost certainly more involved, knowledgeable and partisan can be LESS decided about Cinton, the single best known figure in the race? Again, a tortured story might be constructed, but I think a simpler explanation is that this result is not consistent within the Zogby data itself, or in comparison with outside polling.

Where does this leave us? Puzzled. If these results came from voting machines, I'd suspect that something in the ballot design or the recording mechanism caused a modest but consistent undercount of the Clinton support. The effect seems confined only to that one candidate, and not to any others, Democrats or Republicans. And there was no boost in support for the Republicans paired against Clinton. In this case, I'm similarly inclined to wonder if there is the possibility that the Zogby online survey had a glitch that caused a systematic "undervote" for Clinton. Certainly if my research assistant brought me these results, I'd want to check the software for mistakes before I published it.

Let's assume the Zogby organization has checked for any such possible mistakes or glitches and has ruled that out. (One would assume they were as surprised by the data as anyone and since their reputation is on the line, would have checked very carefully before releasing the data.) Is there any reasonable model of how candidate preferences are evolving that might explain this result, and the stability of Republicans paired against Clinton AND the stability of Obama support and that of his Republican pairings?

Without access to the raw data it is impossible to test any speculation here. But here is one possibility: Internet polls, presumably including Zogby's, use weighting to adjust for non-representativeness in their volunteer respondents. (There is a huge debate about whether this, and more sophisticated approaches, can produce generalizable population estimates with good statistical properties, but we'll leave that for another day.) Clinton has more support among women and somewhat older people. Both those groups are likely to be underrepresented in any pool of internet respondents. As a result the responses of those with these characteristics who ARE present in the sample are likely to be weighted up quite a bit to reach population proportions in the weighted sample. If the relatively few older women who are in the sample are ALSO atypical in other ways that both make them volunteer for internet surveys AND be less disposed to support Clinton than are non-internet volunteering older women, then weighting these respondents up won't properly capture Clinton's support and will lead to a systematic underestimate of her support.

That could do it, but it sounds pretty tortured to me.

I'd check the software one more time.

And based on the large outliers the Clinton results produce, I'd hold off on the Reuters headline until I saw some confirmation from other polls.

Cross-posted at Political Arithmetik.

Polling Over the Holidays?

Topics: 2008 , The 2008 Race

The most important news of the last week was something pollsters tracking the races in Iowa and New Hampshire were probably not so thankful for. New Hampshire Secretary of State Bill Gardner officially confirmed that his state will hold its primary on Tuesday, January 8. The parties in Iowa had already confirmed the Thursday, January 3 as the date for their caucuses. This new calendar, several weeks earlier than in past elections, with fewer days than ever (5) than ever, creates two big challenges for pollsters that give us even more reason to be cautious about the results we will see in the final weeks before the Iowa Caucuses and New Hampshire Primary.

11-26 ia nh calendar.png

The first issue is that both Christmas and New Year's Day fall in the last two weeks before the Iowa Caucuses. Most pollsters prefer to avoid interviewing in this period because so many Americans are traveling, away from home or otherwise unlikely to participate in a survey. As should be obvious, any random sample will be representative of those who participate. If certain kinds of voters are less likely to be home and reachable during the holiday week, and if those voters have different political preferences than those more likely to be reachable, the results of the survey may be skewed.

Two years ago, I blogged on the topic of December surveys, and reported on a nearly thirty year-old analysis that found much lower response rates in December than other months. However, a more recent study based on the massive data available from government sponsored health surveys (Losch, et. al., Public Opinion Quarterly 66: 594-607) found no difference in contact and cooperation rates in December as compared to other months of the year. However, that study applied only to early December: "All December interviews were completed by the end of the second week of the month" (p. 595).

The Iowa/New Hampshire challenge is made worse by the way the weekends bracket each of these events. Pollsters disagree about whether interviews conducted on weekend evenings (Friday and Saturday nights) introduce similar problems (an issue I discussed in June 2005). Christmas Eve falls on a Monday night, so the last weekday evening before the holidays is Thursday December, December 20. The first weekday evening after New Year's is Wednesday, January 2. So that leaves a two-week window of arguably less than optimal polling conditions right before the Iowa Caucuses.

The next issue, as illustrated best by NBC's First Read, is that the time between Iowa and New Hampshire has been reduced to just five days, including weekend. In the past, polls conducted between Iowa and New Hampshire have tracked big swings over the seven days between the two events, changes that often render the pre-Iowa polls as largely meaningless. In 2004, six different organizations conducted rolling-average nightly tracking that required at least three nights of interviewing to get a clear sense of the changed, post-Iowa landscape. Assuming the same method, those same surveys would not report on pure post-Iowa samplings until Sunday or Monday before the primary, and those results would depend heavily on weekend interviewing.

We will have to wait and see whether pollsters alter their methods and field periods in response to the new calendar. However, the already considerable challenge of sampling likely voters for these two early primaries and measuring their late breaking vote preference will be even tougher in 2008.

POLL: Gallup General Election Match-ups

Additional results from the most recent Gallup national survey of 897 registered voters (conducted 11/11 through 11/14) finds:

Nationwide Match-ups:

    Clinton 49%, Giuliani 44%
    Clinton 50%, McCain 44%
    Clinton 54%, Romney 38%
    Clinton 53%, Thompson 40%

    Obama 45%, Giuliani 45%
    Obama 47%, McCain 44%
    Obama 51%, Thompson 38%
    Obama 52%, Romney 35%

POLL: Zogby National '08 Match-ups

A new Zogby Interactive online survey** (story, results) of 9,150 likely voters (conducted 11/21 through 11/26) finds:

National General Election Match-ups:

    Huckabee 44%, Clinton 39%
    Thompson 44%, Clinton 40%
    McCain 42%, Clinton 38%
    Giuliani 43%, Clinton 40%
    Romney 43%, Clinton 40%

    Obama 47%, Thompson 40%
    Obama 45%, McCain 38%
    Obama 46%, Huckabee 40%
    Obama 46%, Romney 40%
    Obama 46%, Giuliani 41%

    Edwards 45%, Thompson 42%
    Edwards 44%, Romney 42%
    Edwards 44%, Giuliani 43%
    Edwards 43%, Huckabee 42%
    Edwards 42%, McCain 42%

** Note: This survey used a non-probability sample. Respondants had previously volunteered to be interviewed online.

POLL: SurveyUSA Four Statewide General Election Match-ups

Four new SurveyUSA automated surveys of registered voters in Washington State, Kentucky, New Mexico, and Kansas (conducted 11/9 through 11/11) finds:

Washington State (n=517):

    Clinton 50%, Giuliani 43%
    Clinton 52%, McCain 43%
    Clinton 57%, Romney 37%
    Clinton 57%, Huckabee 35%

    McCain 46%, Obama 45%
    Obama 48%, Giuliani 46%
    Obama 49%, Romney 40%
    Obama 54%, Huckabee 34%

Kentucky (n=560):

    Clinton 48%, McCain 47%
    Clinton 48%, Giuliani 44%
    Clinton 54%, Romney 39%
    Clinton 55%, Huckabee 36%

    McCain 56%, Obama 34%
    Giuliani 52%, Obama 38%
    Romney 44%, Obama 43%
    Obama 44%, Huckabee 42%

New Mexico (n=506):

    McCain 48%, Clinton 45%
    Clinton 48%, Giuliani 44%
    Clinton 52%, Romney 40%
    Clinton 53%, Huckabee 39%

    Giuliani 49%, Obama 41%
    Obama 49%, Romney 41%
    Obama 50%, Huckabee 38%
    McCain 50%, Obama 40%

Kansas (n=542):

    McCain 55%, Clinton 38%
    Clinton 48%, Romney 44%
    Giuliani 49%, Clinton 43%
    Clinton 49%, Huckabee 43%

    McCain 53%, Obama 37%
    Obama 46%, Romney 43%
    Giuliani 47%, Obama 42%
    Obama 47%, Huckabee 41%