Pollster.com

May 10, 2009 - May 16, 2009

 

New Gallup has PID Tied. Yep, It's an Outlier.


pidOutliers.png
The latest Gallup (5/7-10/09) poll has party identification tied at 32-32 and caused an immediate howl of "outlier!" in the comments at Pollster.com.  In this case, the howl is justified. Compared to all recent Gallup polls (so we compare apples to apples) this latest stands out quite a bit from the rest.

The chart above shows the trends since January 2001 for Gallup polls for Dems, Reps and the Rep minus Dem margin. This latest poll is circled for easy reference.  The Dem percentage is a bit below trend, while the Rep percentage is considerably higher than trend. The difference of the two is therefore quite high- a difference of zero compared to a trend estimate of -9.3.

The Dem value is just inside the 95% confidence interval while the Rep value is outside the CI, as is the difference.  (The Dem and Rep estimates are, of course, correlated, though not perfectly so since Independent is an available option as well.)

Is there anything that might explain this outlier? One intriguing possibility is that this Gallup poll is their annual "Values and Beliefs" survey, which spends most of the interview asking questions about many of the classically partisan moral issues including abortion, gay rights, stem cell research, the death penalty and other topics. (Gallup has not yet released the full questionnaire so my list here is based on the content of past surveys in this series and may be wrong about some details.)

We know how powerfully question order can sometimes affect subsequent responses. Since party id is usually asked late in the survey it is plausible that a lengthy set of questions about these social issues would affect independents who might have recently called themselves Republican. Priming these issues might be expected to provoke more pro-Republican considerations, which in turn could shift the partisan balance. That would be interesting as a finding because of the supposed durability of party id and would invert the outlier finding because the outlier would be the result of the content of the survey, rather than a mere random fluctuation in the makeup of the sample.

Happily we can test this using the five previous Values and Beliefs surveys, which are done annually in early May. (No party ID data have been released for the 2001-2003 surveys, so this analysis is based on 2004-2009 only.)  If the question order effect is real, we should see the pid distribution shift in a pro-Republican direction each year, since the content is similar and should provoke similar imbalance in considerations among independents who might lean Rep if they thought about these issues.  The chart below shows the residuals for the pid series, this time circling all the Values and Beliefs polls.

pidOutliersResiduals.png
The data are quite clear: There is no evidence at all that the Beliefs and Values surveys are systematically skewed in party id. Only the new 2009 edition of this series is an outlier. It looks like idiosyncratic sampling fluctuation remains a more plausible explanation.

This "null finding" is worth appreciating for a moment. I had high hopes for the hypothesis, which fits well with research on both priming and question order effects. And yet we find no difference. This too is consistent with the literature: question order effects have proved irritatingly difficult to provoke on purpose. We find large order effects sometimes, yet when we try a different treatment which is seemingly just as likely to have an impact, we get nothing. This seeming randomness in when we can and cannot provoke order effects is a cautionary tale about being too quick with post-hoc explanations for outliers based on question order. These effects may be real, but require substantial evidence before we accept them. Theory alone is a poor guide to empirical reality in this case.

It is easier to be confident about the outlier status of this poll than to account for why it is so clearly out of line with previous Gallup results. At least we can address the outlier status empirically and with some statistical confidence. They "why" of that status must remain the always true maxim: "Outliers Happen."


US: Party ID (Gallup-5/7-10)


Gallup Poll
5/7-10/09; 1,015 adults, 3% margin of error
Mode: Live Telephone Interviews

National

Party ID
Democrats 32, Republicans 32, independents 34 (chart)

(source)


AAPOR09: Elizabeth Dean - Second Life

Topics: AAPOR , AAPOR2009 , Elizabeth Dean , RTI International , Second Life

I interviewed Elizabeth Dean, a survey research methodologist at RTI International, about their efforts to recruit and interview survey respondents in Second Life  at the annual conference of the American Association for Public Opinion Research (AAPOR).

RTI has more information about their research facility in Second Life's online virtual world.


200905160749.jpg

The following links take you to all the videos, the conference program (pdf and online searchable) and occasional Twitter updates from me and others at the conference.


AAPOR09: Jon Krosnick

Topics: AAPOR , AAPOR2009 , Internet Polls , Jon Krosnick

I interviewed Jon Krosnick, Stanford professor of communications, political Science and psychology, on challenges in assessing the quality of newer survey methodologies, including online panel surveys, at the annual conference of the American Association for Public Opinion Research (AAPOR).

Follow the links to all the videos, the conference program (pdf and online searchable) and occasional Twitter updates from me and others at the conference.


AAPOR09: Scott Keeter

Topics: AAPOR , AAPOR2009 , Pew Research Center , Scott Keeter

I interviewed Scott Keeter, director of survey research at the Pew Research Center, on a paper he presented on the perils of pre-election polling in 2008.at the annual conference of the American Association for Public Opinion Research (AAPOR).

Follow the links to all the videos, the conference program (pdf and online searchable) and occasional Twitter updates from me and others at the conference.


AAPOR09: Chase Harrison

Topics: AAPOR , AAPOR2009 , Accuracy , Chase Harrison

I interviewed Chase Harrison, preceptor at the Program on Survey Research at Harvard University, on his research into the accuracy of pre-election polling forecasts in 2008 at the annual conference of the American Association for Public Opinion Research (AAPOR).

I also interviewed Harrison last year on the first stage of this research.

Follow the links to all the videos, the conference program (pdf and online searchable) and occasional Twitter updates from me and others at the conference.


AAPOR09: Tom Guterbock

Topics: AAPOR , AAPOR2009 , Message Testing , Push "Polls" , Tom Guterbock

I interviewed Tom Guterbock, director of the Survey Research Center at the University of Virginia, on a paper he presented on tests he conducted over the last year to improve message testing poll at the annual conference of the American Association for Public Opinion Research (AAPOR).

Follow the links to all the videos, the conference program (pdf and online searchable) and occasional Twitter updates from me and others at the conference.


AAPOR: Masahiko Aida

Topics: AAPOR , AAPOR2009 , Democracy Corps , Greenberg Quinlan Rosner , Likely Voters , Masahiko Aida

I interviewed Masahiko Aida, director of analytics for Democratic polling firm Greenberg, Quinlan Rosner (which conducts surveys for Democracy Corps) on a paper he presented on their likely voter validation study at the annual conference of the American Association for Public Opinion Research (AAPOR).

Follow the links to all the videos, the conference program (pdf and online searchable) and occasional Twitter updates from me and others at the conference.


AAPOR09: Reg Baker

Topics: AAPOR , AAPOR2009 , Online Panel Surveys , Reg Baker

Up next in our series of interviews at the AAPOR 2009 conference is Reg Baker, Chief Operating Officer of Market Strategies International discussing his work as chair of AAPOR's Online Panel Task Force

You can see all the videos from this year's AAPOR conference here (and last year's too).  I may also post occasional updates via Twitter.


Robert Groves Census Confirmation Hearing on C-SPAN

Topics: Robert Groves

Bob Groves' confirmation hearings were held today (5/15). C-SPAN has the hearings at the link below. Grove's opening statement is well worth listening to and begins at about 19 minutes into the video. He strongly speaks to the need for a non-partisan approach to the census and maintaining a focus on the scientific issues that confront the census and it's many data gathering activities.

Click here to link to the C-SPAN video.

The C-SPAN Summary:

CENSUS NOMINEE SAYS NO TO STATISTICAL SAMPLING

At his confirmation hearing in front of the Senate Homeland Security & Governmental Affairs Cmte., Census Director nominee Robert Groves said that statistical sampling in the 2010 census will not take place. Mr. Groves was nominated by Pres. Obama in April.


AAPOR09: Chris Borick

Topics: AAPOR , AAPOR2009 , Chris Borick , Muhlenberg College

The next interview in our series of interviews at the AAPOR 2009 conference is with Chris Borick, director of the Institute of Public Policy at Muhlenberg Colllege about the paper he presented on the "incumbent rule" in 2008.

You can see all the videos from this year's AAPOR conference here (and last year's too).  I may also post occasional updates via Twitter.


US: Pro-Life/Choice (Gallup-5/7-10)


Gallup Poll
5/7-10/09; 1,015 adults, 3% margin of error
Mode: Live Telephone Interviews

National

With respect to the abortion issue, would you consider yourself to be pro-choice or pro-life?

    51% Pro-life
    42% Pro-choice

    Women
    49% Pro-life
    44% Pro-choice

gallup1510.gif

gallup2510.gif

"With the first pro-choice president in eight years already making changes to the nation's policies on funding abortion overseas, expressing his support for the Freedom of Choice Act, and moving toward rescinding federal job protections for medical workers who refuse to participate in abortion procedures, Americans -- and, in particular, Republicans -- seem to be taking a step back from the pro-choice position. However, the retreat is evident among political moderates as well as conservatives.

It is possible that, through his abortion policies, Obama has pushed the public's understanding of what it means to be "pro-choice" slightly to the left, politically. While Democrats may support that, as they generally support everything Obama is doing as president, it may be driving others in the opposite direction."

(source)


AAPOR09: Paul Donato

Topics: AAPOR , AAPOR2009 , Nielsen Company , Paul Donato

The latest interview from our series of interviews at the AAPOR 2009 conference is with Paul Donato, Executive Vice President and Chief Research Officer for The Nielsen Company. Donato was one of two speakers at last night's Plenary Session, "The Role of Traditional Survey Research in a World of Electronic Measurement and Changing Information Needs."

Unfortunately, a combination of logistical and technical challenges prevented me from completing an interview with Columbia University Professor Kenneth Prewitt, who shared the stage with Donato.

You can see all the videos from this year's AAPOR conference here (and last year's too).  I may also post occasional updates via Twitter.


AAPOR2009: Christopher Wlezien

Topics: AAPOR , AAPOR2009 , Christopher Wlezien , Prediction Markets , Robert Erikson

Yesterday, I also interviewed to Temple University Professor Christopher Wlezien, who long-time Pollster readers may remember from the guest contribution he co-authored with Joseph Bafumi and Bob Erikson. We talked today about the work Wlezien and Erikson have done on the comparative accuracy of polls and political prediction markets:

Wlezien sent a copy of the paper he co-authored with Erikson that he presented yesterday. Their earlier article, "Are Political Markets Really Superior to Public Opinion Polls," appeared in Public Opinion Quarterly in 2008.

You can see all the videos from this year's AAPOR conference here (and last year's too).  I may also post occasional updates via Twitter.


AAPOR09: Lou Harris

Topics: AAPOR , AAPOR2009 , Harris Poll , Lou Harris

The second interview in the series I am conducting at this week's AAPOR conference has a bit of a back story. I had the real honor of interviewing 88-year-old Lou Harris, the founder of Harris and Associates in the mid-1950s and creator of the Harris Poll. Harris was also one of the first pollsters to serve as a senior advisor to a presidential candidate -- John F. Kennedy and 1960.

The unfortunate back story is that about 30 minutes after we finished our interview and just before he was scheduled to speak at the AAPOR conference, Harris fell and hurt himself badly enough that he had to be taken to a nearby hospital. I am told, however, that the injury is not serious although he is spending the night for observation. We hope he is feeling better soon.

We talked for a little over twelve minutes. In this first segment, Harris recounts the work he did for the Kennedy campaign prior to the West Virginia Democratic primary in 1960:

In this second segment, I asked Harris about the apparently apocryphal story that had candidate Kennedy snapping, ""Just give me the numbers, Lou, I know what they mean." Apparently, that story is something of a myth, although he goes on to describe Kennedy as a perceptive but demanding client:

.

You can see all the videos, including those from last year's conference, here.

Update:  Twenty five years ago, the Washington Post's Louis Romano reported the following about the mythical "I know what they mean" story (via Nexis, 1/15/1984):

THE STORY IS SAID to be apocryphal, but political pollsters still love to tell it:

John F. Kennedy was soaking in a tub during the final days of the 1960 presidential campaign.

Lou Harris, the first in a new breed of specialized presidential pollsters, was on the edge of the tub discussing his latest survey results, complete with a complicated package of political advice and in-depth analysis for Kennedy.

"Just give me the numbers, Lou," Kennedy is said to have snapped. "I know what they mean."

"Oh, we used to always meet in the bathroom," says Lou Harris today, one of the deans of political polling. "But the story is part of the mythology . . . I was certainly the first polltaker who . . . served on a super-strategy committee."


AAPOR09: Conference Chair Michael Link

Topics: AAPOR

This post is the first of a series of video interviews I'll be doing over the next few days at the annual conference of the American Association for Public Opinion Research (AAPOR). We begin with Michael Link, Chief Methodologist/VP for Methodological Research at The Nielsen Company, who is serving as this year's AAPOR conference chair. He explains what this conference is about and what we can expect over the next few days.

You can see all the videos, including those from last year's conference, here.

Credit for the animated graphic title that introduces each video goes to Lisa Mathias at the Winston Group (and her colleague Kristen Soltis for recommending her). Thank you Lisa and Kristen!


US: National Survey (FOX-5/12-13)


FOX News / Opinion Dynamics
5/12-13/09; 900 registered voters, 3% margin of error
Mode: Live Telephone Interviews

National

Obama Job Approval
60% Approve, 30% Disapprove (chart)
Dems: 93 / 5 (chart)
inds: 57 / 30 (chart)
Reps: 23 / 63 (chart)

Congressional Job Approval
41% Approve, 49% Disapprove (chart)

Party ID
Democrat 42, Republican 30, independent 24 (chart)

(story, results)


NY: Gay Marriage (Quinnipiac-5/5-11)


Quinnipiac University
5/5-11/09; 2,828 registered voters, 1.8% margin of error
Mode: Live Telephone Interviews

New York State

Would you support or oppose a law that would allow same-sex couples to get married?

    46% Support
    46% Oppose

Would you support or oppose a law that would allow civil unions for same-sex couples?

    65% Support
    27% Oppose

Some people say that not allowing same-sex couples to get married is discrimination. Do you agree or disagree?

    49% Agree
    46% Disagree

(source)


NJ: 2009 Governor (Rasmussen-5/12)


Rasmussen Reports
5/12/09; 412 likely Republican primary voters, 5% margin of error
Mode: IVR

New Jersey

2009 Governor - Republican Primary
Christie 39, Lonegan 29 (chart)

(source)


Pre-AAPOR "Outliers"

Topics: Outliers Feature

Nate Silver takes a closer look at Gallup's massive tabulation of party ID by birthyear.

Patrick Ruffini grapples with Barack Obama's personal popularity.

Chris Cillizza ponders the polls on torture.

Joshua Tucker ponders the ethical responsibilities of social scientists in the torture debate.

Andrew Gelman revises his maps of vote by income.

DemFromCT rounds up H1N1 Flu polling.

RealClearPolitics interviews Democratic pollster Sergio Bendixen.

Mark Mellman condemns a study showing a lack of concern about global warming.

Stephen Kinney says California's surge in minority turnout will persist.   

James Vega takes another look at the questions asked by Resurgent Republic.

Neil Malhotra and Yotam Margalit  poll on anti-semitism and the economy (via Jelveh/TNR)

Lake Research Partners shares a poll on gay marriage in Massachusetts (sponsored by Marriage Equality Works).

Flowing Data links to maps of the seven deadly sins.


Twitter Updates

Topics: AAPOR

I am headed off to the airport in few minutes for Hollywood, Florida and this year's conference of the American Association for Public Opinion Research (AAPOR).  I will be video-blogging for the rest of the week (like last year), but will otherwise try to keep updated via Twitter

The box below displays my last five Twitter updates. You can also follow posts by other AAPORites on Twitter using the hashtag #AAPOR09.

;


US: National Survey (CBS-5/6-12)


CBS News
5/6-12/09; 1,874 adults, 2% margin of error
Mode: Live Telephone Interviews

National

Obama Job Approval
63% Approve, 26% Disapprove (chart)
inds: 59 / 28 (chart)
Reps: 34 / 54 (chart)

Economy: 56% Approve, 33% Disapprove (chart)

State of the Country
45% Right Direction, 48% Wrong Track (chart)
Economy: 32% Getting Better, 23% Worse (chart)

Do you have confidence that Barack Obama will nominate good justices to the Supreme Court, or are you uneasy about who he will choose?

    55% Confidence
    35% Uneasy

(story, results)


NH: 2010 Senate (Dartmouth-4/27-5/1)


The Nelson A. Rockefeller Center at Dartmouth College
4/27 - 5/1/09; 403 registered voters, 5% margin of error
Mode: Live Telephone Interviews

New Hampshire

Favorable / Unfavorable
Gov. John Lynch (D): 54 / 9
Sen. Judd Gregg (R): 43 / 21
Sen. Jeanne Shaheen (D): 33 / 30 (chart)

2010 Senate - General Election (all trends)
Hodes (D) 31%, Bass (R) 30%
Hodes 38%, Sununu (R) 35%

(source)


US: Woman Justice (Gallup-5/7-10)


Gallup Poll
5/7-10/09; 1,015 adults, 3% margin of error
Mode: Live Telephone Interviews

National

Do you think it is essential that the next Supreme Court justice be a woman, is it a good idea, but not essential, does it not matter to you, or do you think it is a bad idea?

    6% Essential
    26% Good idea, not essential
    64% Doesn't matter
    3% Bad idea

    Among Women
    7% It is essential
    31% A good idea, but not essential
    58% It doesn't matter

(source)


NYC: 2009 Mayor (Marist-5/5-7)


Marist Poll
5/5-7/09; 578 registered voters, 4.5% margin of error
375 registered Democrats, 5% margin of error
Mode: Live Telephone Interviews

New York City

Mayor Bloomberg (R/i) Job Approval
59% Excellent/Good, 39% Fair/Poor (chart)

2009 Mayor - Democratic Primary
Weiner 34, Thompson 29, Avella 8 (chart)

2009 Mayor - General Election
Bloomberg 50, Weiner 36 (chart)
Bloomberg 51, Thompson 33 (chart)
Bloomberg 52, Avella 27

(Bloomberg Ratings, 2009 Mayor)


NY: 2010 Gov (Quinnipiac-5/5-11)


Quinnipiac University
5/5-11/09; 2,828 registered voters, 1.8% margin of error
1,238 registered Democrats, 2.8% margin of error
Mode: Live Telephone Interviews

New York State

Favorable / Unfavorable
Andrew Cuomo (D): 51 / 13
Gov. David Paterson (D): 24 / 55 (chart)
Rudy Giuliani (R): 56 / 32

Gov. Paterson Job Approval
28% Approve, 61% Disapprove (chart)

2010 Governor - Democratic Primary
Cuomo 62, Gov. Paterson 17 (chart)

2010 Governor - General Election
Giuliani 54, Gov. Paterson 32 (chart)
Cuomo 47, Giuliani 41 (chart)

(source)


More on the Edwards 2008 What-If


As a follow-up to Monday's post on the "what-if" speculation about John Edward's 2008 campaign, I did a an online "Q&A" yesterday with NPR.org's Liz Halloran. While much of it reviews the ground covered on Monday, I did expand a bit on one point worth repeating:

Bottom line: What effect, if any, did Edwards have on the outcome of the Democratic primary process?

We should not overlook that John Edwards was an important part of the campaign dialogue throughout 2007 and early January 2008. He was the most aggressive in pressing the case against Clinton, especially — as one Edwards supporter reminded me yesterday — when she flubbed her answer regarding drivers' licenses for illegal immigrants in the Philadelphia debate in October 2007. Without Edwards in the race, the "dynamics" might have been different in ways we simply cannot deduce from poll numbers.

And while we are on the subject of what-if hypotheticals about last year's Iowa Caucuses, one question I thought more intriguing is what might have happened had Clinton and strategist Mark Penn had followed the advice of her deputy campaign manager, Mike Henry, and skipped the Iowa caucuses altogether. I had long assumed that had Clinton not contested Iowa, much of her support would have gone to Edwards, thus depriving Obama of his most important victory. An Edwards win in Iowa would have effectively ended the Obama campaign right there, and in a fight against Edwards, Clinton's firewall of support from African-American, a crucial base of support for President Clinton would likely have held.

Turns out, according to Des Moines Register pollster Ann Selzer, I assumed wrong: In May, the likely caucus goers that supported Clinton as their first choice split almost evenly when asked for their second choice, narrowly favoring Obama over Edwards (46% to 44%). In December just before the caucuses, Clinton's supporters again split, this time preferring Edwards over Obama (39% to 35%). So while an Edwards win in Iowa without Clinton in the race was plausible, it would have been far from automatic.

Also finally, one more anecdote on second choices: PPP's Tom Jensen yesterday reminded us of the second choices of Edwards supporters in North Carolina is an automated survey they conducted on February 8, 2008, about a week after Edwards left the race: 41% to Obama and 41% to Clinton.


Understanding the Political Distinctiveness of the Cell Phone Only Public: Results from the 2006 and 2008 CCES


A few weeks ago, I highlighted some preliminary findings from a paper written by myself and Stephen Ansolabehere for this week's AAPOR conference. The paper is now finished and you can check out a copy here. The data we use for the paper is the 2006 and 2008 Cooperative Congressional Election Study.

One of the major themes in the paper is that understanding the cell-only population is about more than just age. In fact, residential mobility has a strong influence on whether someone has shed their landline. Even after controlling for age and a litany of other demographic variables, we find that respondents who moved within the last year were 24 percentage points more likely to be cell only than those who had lived in the same residence for at least five years. Renters, singles, and those without children were also much more likely to be cell-only.

Our explanation for this pattern:

"There are several reasons that highly mobile Americans may be more likely to go without landlines. First, whenever someone moves from one residence to another, they have an opportunity to reassess their phone needs. Thus, the act of moving provides an opportunity for individuals to shed their landlines. Second, mobile Americans may choose a CPO lifestyle because cell phone numbers tend to be more portable than landlines. When moving from one metropolitan area to another, individuals must change their landline phone number, but do not need to change their cell number. This may provide an incentive for choosing not to maintain a landline in a new residence. Third, those with fewer family and community ties may feel less of a need to have multiple phone lines on which they can be reached by members of their social networks. "

The fact that the cell-only public tends to be more mobile has some important political consequences. Some highlights:

  • The difference in the percentage of landline and cell-only respondents who reported being registered was fairly small--over 95% in both groups. However, there was a much larger gap in actual registration rates (66.8% versus 53.9%). Since cell-onlys are more likely to have moved recently, they may not have successfully registered to vote at their new addresses despite the fact that they may think they are registered.
  • Cell-only respondents were significantly more likely to have problems with their registration when attempting to vote. In 2008, over 7% of cell-only respondents indicated that there was a problem with their registration when they attempted to vote, compared to fewer than 4% of respondents with landlines.
  • Cell-only respondents were more than twice as likely as those with landlines to report that neither campaign contacted them. In short, this group is much less likely to be subjected to mobilization efforts from the campaigns.
  • Cell-onlys are politically distinct on a variety of measures. However, this distinctiveness is somewhat muted when demographic controls are taken into account. Interestingly, the largest differences between cell-only and landline respondents are not on issues or ideological self-placement, but on reported vote choices.
  • Ultimately, we argue that weighting for standard demographic measures such as age, education, income, and race may not be sufficient. Pollsters relying on landline samples may want to consider weighting by other factors such as time in residency, renter/home owner, and marital status. But check out the full paper for a more detailed discussion of all of these points.


    The Edwards 2008 What-If

    Topics: Barack Obama , Hillary Clinton , Iowa , John Edwards , Mark Penn

    The publication of a new book by Elizabeth Edwards, and especially the revelation that she advised her husband against running in 2008, has created a lot of "what if" speculation about how the race might have been different. Two weeks ago, for example, Hillary Clinton's former pollster Mark Penn speculated that a 2008 presidential race without John Edwards would have been "very different," adding "if he had come out and dropped out of the race particularly early, I think a lot of voters would have taken a good fresh look at Hillary Clinton."   

    Then this weekend, George Stephanopoulos reported that "by late December, early January of last year" several members of Edwards' "inner circle" began to believe the rumors about their candidate's affair and "devised a 'doomsday' strategy of sorts...basically, if it looked like Edwards was going to win the Democratic Party nomination, they were going to sabotage his campaign." Mickey Kaus read the report and wondered about the impact of the votes Edwards "drained" while "staying in the race through South Carolina."

    Now with the stipulation that Edwards strategist Joe Trippi is calling the "doomsday" story "Complete BS -- fantasyland - not true" (update: CNN has more),  what can we say about where Edwards' support might have gone had he left the race earlier than he did?

    Let's work backwards.

    After Edwards Actually Dropped Out - Probably the best evidence we have is from the tracking surveys that showed what happened when Edward actually departed the race on January 30, 2008. See our chart below. The change in aggregate vote preference surged to Barack Obama.

    2009-05-11_primarytrend.png

    Let's be a little more specific. The regression line smooths things out (as it is designed to do), but note the change in the dots, the individual survey results. The biggest lurch in support over the course of the two year campaign occurs for Obama just after Edwards dropped out (when pollsters stopped including his name on vote preference questions). Just before the Edwards announcement, most polls showed Obama's support in the mid-30s. Just after, his support surged the mid-40s. Over the same period, Hillary Clinton's aggregate support held mostly steady.

    Now my pollster handbook requires the warning that correlation is not causation. The big spike in Obama's support may have also been the result of the other things happening in that period (endorsements from the Kennedy family, a surge in paid media and organizing in Super Tuesday states, positive coverage of Obama's successes in South Carolina and, eventually, on Super Tuesday).

    Although others with access to respondent level data from Gallup or the Annenberg surveys may be able to check these trends more carefully, from what we can see, the evidence we have says that Edwards' actual departure helped Obama more than Clinton.

    Before South Carolina? - Edwards "drained votes from somebody" in the South Carolina primary, says Kaus. True. But Obama won South Carolina with 55% of the vote, so even if every Edwards vote there had been drained from Clinton, Obama still would have prevailed by double digits. Moreover, keep in mind that the final polls in SC massively understated Obama's ultimate share of the vote (a bigger error than New Hampshire), while getting the percentages for Edwards and Clinton about right. Given that understatement, and if we assume for the sake of argument that all of Edwards' SC votes would have immediately shifted to Clinton had he left the race right after New Hampshire, we would have seen a final round of SC polls showing a neck-and-neck Obama Clinton race and a big "surprise" Obama victory on January 26.

    Before New Hampshire? - Probably the least plausible hypothetical involves Edwards dropping just after Iowa and before the New Hampshire primary. But if he had, what might have become of the 17% of the New Hampshire vote he received? Of the four polls that provided raw, respondent level data to the Roper Archives, only the Fox News poll included a second choice question. Those who said they were supporting John Edwards on the final survey (n=94, conducted January 4-6, 2008) reported their second choices as follows: Obama 38%, Clinton 14%, Richardson 13%, Kucinich 4% with the rest choosing another candidate or unable to provide a second choice. Yes, that final round of New Hampshire polling notoriously understated Clinton's support, but this result is not exactly strong evidence that Edwards was "draining" more support from Clinton than Obama.

    Before Iowa? - Just after Penn's speculation about Iowa appeared, Tom Beaumont of the Des Moines Register produced results from the second choice question on the final Register survey conducted by pollster Ann Selzer:

    The Des Moines Register's Iowa Poll, taken in the closing days of the caucus campaign, showed that in fact Obama was the second choice of more Edwards supporters than Clinton was. The numbers? Among Edwards' supporters, 41 percent said Obama was their second choice, compared to 25 percent who said Clinton was their second choice.

    Since Penn's speculation was more about how the long Iowa campaign would have transpired had Edwards dropped out at the very outset (rather than in its final days), I asked Selzer about their earlier poll conducted in May 2007. The survey showed a very close race: Edwards ahead with 29% followed by Obama at 23% and Clinton at 21%. But had Edwards not been a candidate, his supporters would have split decisively for Obama. With the second choices recalculated into the total, Obama would have led Clinton by nine percentage points on that first Iowa poll.

    Penn was  right to concede that it is "unknowable" how the Iowa campaign might have played out without Edwards in it.  Who knows what would have happened had Edwards dropped out before the Iowa campaign got underway.  Other candidates -- Bayh, Warner, Vilsack -- may have gotten or stayed in.  But the Register data strongly suggest that Clinton's problems in Iowa were about much more than just John Edwards.


     

    MAP - US, AL, AK, AZ, AR, CA, CO, CT, DE, FL, GA, HI, ID, IL, IN, IA, KS, KY, LA, ME, MD, MA, MI, MN, MS, MO, MT, NE, NV, NH, NJ, NM, NY, NC, ND, OH, OK, OR, PA, RI, SC, SD, TN, TX, UT, VT, VA, WA, WV, WI, WY, PR