Articles and Analysis


Eisinger: Do We Really Know How Much Citizens Like or Don't Like Senator Clinton?

Topics: Barack Obama , CBS/New York Times , General , Hillary Clinton , Newsweek , Robert Eisinger , Time , USAToday Gallup

Today's Guest Pollster's column comes from Robert M. Eisinger, a political science professor at Lewis & Clark College and the author of The Evolution of Presidential Polling (Cambridge U. Press).

The Obama "phenomenon" is a product of many things, most notably a superbly smart candidate and a sharp, disciplined campaign team, both of whom clearly articulated a resonating message that mobilized voters. What we don't know if how many of those supporters galvanized around Senator Obama simply because he is not Senator Clinton. This is not to say that Senator Clinton is without fans. To the contrary - she has many. They are devoted and dedicated, imparting the kind of loyalty that any political candidate would desire.

But as evidenced by the reaction to her recent mentioning of Robert Kennedy's assassination, Senator Clinton appears to be a lightning rod - people are either repelled or attracted to her.

Arguments for an Obama-Clinton dream ticket suggest that Senator Clinton's keen intelligence, legislative acumen and support among Democrats outweigh her negatives. However reasoned this claim is, it is potentially flawed in a critical way currently understudied by the public opinion and political cognoscenti. The problem is that we do not know enough about her positives and negatives, especially among voting Democrats and swing voters.

By itself, the "favorability question" is crude and insufficient indicator of likeability. It does not claim to measure the intensity underlying that favorability or lack of favorability. For example, a May 23, 2008 Newsweek poll asks, "Who would you MOST like to be nominated as the Democratic Party's presidential candidate this year...Hillary Clinton (or) Barack Obama (choices rotated)?" "Do you support (INSERT CHOICE) strongly or only moderately?" The poll then asks, "We'd like your overall opinion of the presidential candidates. As I read each name, please tell me if you have a favorable or unfavorable opinion of this person - or if you have never heard of them before this interview. What about (INSERT - READ AND RANDOMIZE). Do you have a favorable or unfavorable opinion of him (or her)?"

Favorability questions specifically asked about Hillary Clinton in the past have been worded in numerous, thoughtful ways, including the following: 1

CBS News/New York Times: Is your opinion of Hillary Clinton favorable, not favorable, undecided, or haven't you heard enough about Hillary Clinton yet to have an opinion?

Gallup/USA Today/CNN: I'd like some overall opinion of some people in the news. In general, do you have a favorable or unfavorable opinion of Hillary Clinton?

Yankelovich/Time/CNN: Please tell me whether you have generally favorable or generally unfavorable impressions of [Hillary Clinton], or whether or not you are familiar enough with [Hillary Clinton] to say one way or another.

Each of these questions is carefully written, but they do not capture the intensity that may lie beneath the answer. In fact the Yankelovich/Time/CNN question employs the phrases "generally favorable" and "generally unfavorable", allowing the respondent to articulate her overall impression, but in doing so, diffuses the potential passion or force embedded within that answer.

If respondents were asked to place their favorability/un-favorability on a seven point scale, then one might get a better sense of the potential polarizing nature of the answer, or put another way, the strength of that favorability and its opposite. Such an option is costly in that it requires an additional question to be asked, and more nuanced data analysis.

Anecdotal conversations in the blogosphere, in the taxi cab and the around the water cooler - reveal that many citizens - men and women, Democrats, Republicans and Independents - have a palpable and deep disdain for Senator Clinton. Different blog, cab and water cooler discourses tell us that Senator Clinton is revered. Scholars of public opinion and savvy journalists are appropriately suspicious of these unrepresentative remarks. Sure, the plural of anecdote is data, but we wonder if selection bias (i.e., we surround myself with like-minded folk; we listen more carefully only to extreme answers) taints our perspective and analysis.

The Obama campaign should be privately measuring the favorability intensity for all prospective Vice Presidential nominees. If not, then they are avoiding a datum that may be critical to their electoral success. Media polls should explore this question as well; given the dearth of the intensity question, the answers will undoubtedly surprise us. //END

1 From Barry C. Burden and Anthony Mughan, "Public Opinion and Hillary Rodham Clinton," Public Opinion Quarterly, Vol. 63, No. 2 (Summer 1999), 237-250.



This confirms my belief that Pollster.com is in the Obama camp.



what a ridiculous comment to make. If you don't like this website, then don't post here.



I don't think pollster.com is in the Obama camp, but I do think there may be a tiny bit of selection bias going on just in terms of what information is deemed interesting... I still keep hoping to see some analysis of the women's vote - Gallup has some stuff on it - but it just doesn't seem to be happening. Likewise, while we've gotten a lot on the potential impact on racial bias, we've gotten zero on issues related to gender (for example, the paper on social desirability response and race was featured; a paper on the same topic but measuring reactions to gender was not. I have no idea if the second paper is any good though). I can only assume that gender somehow seems like a "side issue." Also, I'm certain that there are fewer polls and papers to draw from. But who knows, maybe I'm totally wrong. Overall I think this site is a sterling example of the web at its finest - allowing thousands of people access to expertise and information that is presented in a transparent, efficient, intellectually stimulating manner. In short, it's f-ing great.

That said, I think the column here is atypically poor.

First, I thought the typical way of judging "intensity" is posing a follow-up that asks if the voter feels the "very strongly, somewhat strongly.." etc. Is Professor Eisinger suggesting that this is not being done in the polls commissioned by the big dogs? Or is he suggesting that this type of measure is inadequate, and should be replaced by the 'seven point scale' or somesuch?

Per the issue of cost, I would think that rather than spending money on a longer questionnaire drawing out the nuances of the intensity ("do you just kind of hate her, really hate her, really hate her a lot, or hate her to infinity + 1..") of your whole sample, it would be better to pay for a larger sample that would allow you to split out subgroups. Just saying.

Professor Eisinger also overlooks a crucial piece of the puzzle per the 'dream ticket.' A test of HRC negatives is meaningless in this context unless you test them against the numbers of other possible nominees. For example, HRC may provoke intensely negative feelings among elite white males inclined to vote Democratic - a crowd BHO has locked up. Sen. Webb, perhaps, would not provoke those negative feelings. But Sen. Webb, perhaps, could drive away considerable numbers of female voters. Now which subgroup does BHO need to work on more? That's the kind of question I assume the campaign is asking itself. My point is that Professor Eisinger's analysis is a gross oversimplification.

I also take issue with the statement, "The Obama "phenomenon" is a product of many things, most notably a superbly smart candidate and a sharp, disciplined campaign team, both of whom clearly articulated a resonating message that mobilized voters." Actually, His Supersmartness only persuaded roughly half the voters. The other half prefer(red) someone else. And that "sharp, disciplined" campaign team has had enormous help from the media. How hard is it, really, to cast a powerful woman as a villain in our culture?

Last, if the Professor is saying that his watercooler / taxicab conversations are roughly evenly divided between pro and anti HRC sentiments, what's the point of bringing them up at all? I'm pretty sure we're all aware that some people like HRC and some people don't.




I don't think Professor Eisinger is giving specific examples of water-cooler talk he has personally heard. It's just a way to bring out the point that such anecdotal evidence is not to be trusted. Do you have a beef with that?

The Newsweek poll example shows that a coarser 5-point scale was asked as a follow-up to the "who would you like MOST to be nominated" question. The 5-point scale is: Strongly favor, moderately favor, neutral, moderately disfavor, strongly disfavor. It's quite easy to picture a similar follow-up to the "favorability" question; but the Newsweek poll did not ask it.

I look at the Clinton-negatives in a different way. A strength of Senator Obama is that he's drawing support from affluent White folks, who might have otherwise voted for Republican. You may think he has that vote locked up; but if these voters have voted for Obama over Clinton, and are faced with a choice of Obama+Clinton vs McCain, is there a chance they might stay away from any potential Clinton-drama?

I am not saying that will happen; but it's a possibility to consider. It's the same way, for sure, with Women voters. One potential answer would be to nominate someone who (a) does not drive voters away (polarize them); and (b) satisfies Clinton-supporting Women voters.

I can picture a few candidates who might fill these criteria easily. If this candidate were to be a woman, people just have to realize that Senator Clinton is NOT the only accomplished female politician around.



I actually think that the idea that Hillary "polarizes" voters more than Obama is a myth. Sure a segment of the mostly right wing still get excited over Hillary, but they also have been getting to know Obama and have now directed their focus on him... like having fresh meat dangled in the lion cage. And I think that these people were not going to vote for a Democrat anyway.

The third leg of the "polarization stool" is that among Democrats there is a large division between loyal Obama and Hillary supporters. And it seems that that more Hillary supporters are "polarized" against Obama than vice-versa.

At this point in time no other "accomplished female politician" can be substituted for Hillary. There would be a huge outcry and backlash. Hillary has earned a well-deserved reputation of being smart, tenacious and a champion for the populist message. I cannot think this moment of any potential female candidate equal to the task.



"Hillary has earned a well-deserved reputation of being smart, tenacious and a champion for the populist message."

...who as VP would cause the P to hire a bunch of food-tasters ;-)

Janet Napolitano, Kathleen Sibelius (two-term Governors) easily fit the bill. Sibelius even has Ohio roots, and is a Catholic, FWIW. Keep thinking - given five months, folks can get used to the idea of someone else.



Sebelius ruined her potential "national" political career with her dull dull Democratic response to this year's State of the Union. She may be physically appealing to the guys, but her speaking abilities and presentation techniques are boring, boring, boring.

Neither Napolitano or Sebelius can hold a candle to Hillary in any way. And most importantly, neither has a clue about the federal government processes and neither does Obama. Obama needs someone who knows how things work (e.g. Al Gore for Bill Clinton, LBJ for JFK).



"sooyapi: This confirms my belief that Pollster.com is in the Obama camp."

What if it is? It still presents a reasoned analysis. Blindfold yourself if it makes you feel better, but you are not persuading anybody of anything.

I am Democratic supporter of Obama, but at tthe beginning of the process I was very well disposed to HRC, but have become increasingly negative toward her. Why? Probably because of her effort to increase Obama's negatives amounted to running as a Republican. By this I mean running a campaign that made her "vetting" argument a self fulfilling prophecy, ie treating politics as a rough and tough stick it to the bad guys campaign. She plays this game well because it conforms to her experience and, probably, her autocratic personality.

My recollection is that the most intense part of this negative campaign against Obama was in the Ohio to Pennsylvania phase, and her negatives rose more than Obama's. Now she's paying the victim again, and advancing patently cynical arguments about the "popular vote." How could her negatives fail to be significant? The only difference is that they have risen among Democrats like me.



You only talk about Hillary negatives. It is Obama who has "increased" negative approval ratings, whereas Hillary's are about the same as always. Several political analysts predicted that Obama's negatives would rise as voters got to know him better.

I do not see how anything Hillary has done or said hurts Obama. His negatives are the result of his own actions (e.g. "bitter comments") and revelations about his pastor Wright and associations with Ayers and Rezko.

And if you do not think that the Obama campaign is not playing the same political game as Hillary (and all candidates) do, then you have not been paying very close attention.

Finally, the popular vote does mean something when half of over 35 million voters did in fact vote for Clinton and not Obama.



TK77: Clinton's negatives are on par with her negatives from other points in her career; Obama's have risen significantly. According to Gallup, 37% of the general public think the media has been too easy on Obama; that compares to 21% who say the media has been too easy on Clinton and 33% who think they've been too easy on McCain. And 30% say the media's been too hard on Clinton.

So, Obama's negatives are now roughly equal (at least in Gallup's estimation) to Clinton's, with a third of all adults saying he's been treated softly. Hillary's negatives are roughly as high as they can go, barring she admits to eating puppies or something. Barack's negatives may very well go higher; 1/3 of the public appears to have an appetite for seeing the media challenge him more.

And I agree with Nickberry. Yes Hillary drives the right wing crazy, yes she does the same to the Dennis Miller/Bill Maher vote (elite swing voting white men). But they went rabid over Gore and Kerry, too, and I doubt that when push comes to shove they'll behave differently with Obama. But Hillary has something people overlook when they talk about the right wing - yes, they hate her, but some of them, on a certain level, are okay with her because they think of her as strong. I think this is why you see occasional positive-to-HRC noises coming from Ann Coulter, Carly Fiorina, Bill O'Reilly, etc. They may think everything she says is wrong, but they know she's no wimp. Obama on the other hand - wimp and crybaby (in their eyes). No way will they support him.

I would say all signs point to us knowing a lot more about Hillary's negatives than Barack's - but that's quite obviously not the answer the good professor was shooting for.

Also, keep in mind that even with the most- favorable-to-Obama way of adding up the popular vote (excluding FL and MI but including projected totals for IA, NV, ME, WA), the spread is only 1.6% Counted a different way (including actual votes cast in FL and MI and not including projected totals for IA, NV, ME and WA), he's -.45%. This narrow a spread is bound to spark disagreement; there's nothing "patently cynical" about it.

Finally, I would say that its Obama who has hurt Obama. No one forced him to behave in an arrogant and dismissive manner. No one forced him into his jihad against mandates. And no one forced him to run before he had any real legislative accomplishments to speak of.

You may want to examine the ways your own bias could be affecting the way you are interpreting information.



@ RS,

I agree that ultimately the professor is saying the water cooler opinions cancel each other out, but his phrasing tells me something different. For example, this: "Anecdotal conversations in the blogosphere, in the taxi cab and the around the water cooler - reveal that many citizens - men and women, Democrats, Republicans and Independents - have a palpable and deep disdain for Senator Clinton. Different blog, cab and water cooler discourses tell us that Senator Clinton is revered."

Now read this completely fictitous statement:

"In my travels as a journalist I've heard from voters - Democrats, Republicans, Independents - in states such as Florida, Minnesota and Iowa - that Barack Obama is both far too arrogant and too naive to be President. Of course, I've also heard some voters say they are highly favorable to Obama."

Does that sound neutral to you?

Its the equivalent of what the professor wrote - 38 words in the first sentence, two negatives, three locations, three types of voter; 14 words in the second sentence, one positive.

Also, the problem with Obama is not that he is not female. The problem is that Obama is Obama. Seriously, this "women are voting for Hillary just because she's female" or "women resent Barack because he's not female" needs to put to rest. Not only is there zero evidence to back up this claim, but past voter performance contradicts it. Democratic and independent women did not turn out in droves for Liddy Dole. If anything, research supports the notion that Clinton is judged more harshly by other women simply because she is female. (I've been meaning to blog about that w/ citations).

Also - I suspect choosing Sebelius, Napolitano etc. would be framed by the press as pure pandering, a negative for Obama.

Wow, I should be working right now.


Post a comment

Please be patient while your comment posts - sometimes it takes a minute or two. To check your comment, please wait 60 seconds and click your browser's refresh button. Note that comments with three or more hyperlinks will be held for approval.