Articles and Analysis


Measuring Nader and Barr

Topics: Barack Obama , Bob Barr , David Moore , John McCain , Measurement , National Journal , Ralph Nader

My Nationaljournal.com column, on the challenge posed to pollsters by third party candidates like Ralph Nader and Bob Barr, is now online.

The topic is timely since yesterday we also put up a new chart showing the results of national poll questions that include Nader and Barr as choices. One issue posed in the column was which form of the vote choice question is a better measure of support for the third party candidates: Those that include Nader and Barr as choices along with McCain and Obama or those that offer only the two major candidates (but typically record the preferences of those who volunteer a choice for another candidate).

I explore the reasons for skepticism toward the four-way vote question in the column. We have historical evidence that summer polls are poor predictors of November support for third party candidates (short version: the summer polls overstate their ultimate support). A more difficult question is which type of question -- the two-way or four-way choice -- provides a better measure of true preferences right now?

My hunch is that reality of Nader and Barr's current support lies somewhere in between the numbers that volunteer their support on questions offering only Obama and McCain as choices and those that offer all four. I believe that many voters are telling pollsters they support Barr or Nader now, especially when offered the four-way choice, because they are not entirely sold on Obama or McCain and would rather grab for the "independent"** as a way to give the interviewer a satisfactory answer (rather than trying to think through their final decision right there on the phone -- a tendency \survey methodologists sometimes call "satisficing").

This mushiness of support is something that David Moore touched on in his post here yesterday. You can also see evidence of the conflict in the fascinating cluster analysis of last month's USA Today/Gallup poll written up today.

**Or Libertarian or Green Party candidate. Incidentally, at least one commenter has wondered whether pollsters typically identify the candidate party on these vote questions, and the answer is almost always yes. You can usually get the actual language of the poll questions by clicking through the links in the tables below our charts or in the always invaluable, PollingReport.


Mark Lindeman:

Mark, great column. One twist: you note that in the late polls, Nader's support averaged 1.0%, but he received just 0.38% of the national popular vote. One confound you don't mention is that some folks might have voted for Nader had he appeared on the ballot in their states. (Here in New York, Nader got 1.35% of the vote on two different lines. I doubt he would have done nearly so well if he didn't have the Independence line.)



Do George Wallace and Ross Perot ('92) help prove the point? You note that Wallace polled at 16% and Perot at 22%, but that is not far off from their actual results (14% and 19%, respectively), with drops easily explained by particular events (Wallace's running mate Lemay musing about nuclear weapons, Perot's dropping out of the race, then back in), rather than an inherent tendency for people to identify with independent candidates.


Lorne Guyland:

correction/clarification: Bob Barr is indeed the "Libertarian candidate"; but Nader is not the "Green Party candidate", nor will he be. He's specifically said he will not seek their nomination.

Nader was only the official Green Party nominee once, in 2000. It's not a trivial distinction; much of his support that year came from progressives in non-competitive states who were hoping a respectable showing would help build a viable third party for the future. Certainly that was true of his volunteer base -- I was there and no one I knew was under any illusion that Nader could win.

At any rate, Nader's personal relationship with the Green Party was always a tenuous one, and in '04 he chose to run a purely vanity campaign, with predictable results. He won't be a factor this year either; but the Libertarians do have some party structure (albeit small), which gives Barr a better chance to make waves.

Hopefully pollsters will be aware of all this. Indeed, if they're going to list "all four" names, they should go all the way and list all candidates on the ballot in a given state. Since that's an unknown at the moment, and neither Barr nor Nader is guaranteed to be on all 51 ballots, the two-choice question is actually the less arbitrary choice for now.



I think if you only look at National polls you might get a skewed look. There is little disincentive for a voter to stick with a third party candidate all the way through the election IF THEY LIVE IN A NON-COMPETITIVE STATE. In other words, a voter who lives in New York or Utah might as well vote for a third party candidate because their vote for a major candidate won't matter. Contrast this with a voter in Ohio, Florida or Michigan. While they might state a preference for a third party candidate, they have a strong incentive to cast their actual vote for their preferred major party candidate, since that vote may make an actual difference in the outcome of the election.

If you really want to see the effect of third party candidates on early polls vs. late polls vs. the actual votes, I think you need to stick to the "battleground states" where the votes actually matter.



What's interesting to me is the new poll from Rassmussen of ND-


This is interesting in relation to the recent NJ poll. So are we to believe that obama gained in ND but lost ground in NJ??



2004 Election
Nader final RCP poll average vs. actual
from http://www.realclearpolitics.com/bush_vs_kerry_sbys.html

State / Poll / Actual
Arkansas / - / 0.6
Colorado / 1 / 0.6
Florida / 1 / 0.4
Iowa / 1 / 0.4
Maine / 1 / 1.1
Michigan / 1.5 / 0.5
Minnesota / 5 / 0.7
Nevada / - / 0.6
New Hampshire / 1.3 / 0.7
New Mexico / 1.3 / 0.5
West Virginia / - / 0.5
Wisconsin / 1.7 / 0.5

Minnesota appears to be an anomaly because the "5" is based on one poll from "Humphrey Inst".

These results show that not many people expressed a preference for Nader in the battleground states and even fewer voted for him (perhaps they learned from 2000 election).

Its possible that more voters may express a preference for Barr in this election and then go ahead and vote for Barr, but its my guess they wouldn't vote for McCain or Obama no matter what, so they are essentially irrelevant to the outcome of the election (in that respect they are no different than non-voters).

If the question is "are the poll results more accurate if they include the third party candidates", I think the answer may be no. I think the vast majority of the respondents who express a preference for a third party candidate are either casting votes that don't matter or will shift to one of the major candidates when they cast their actual votes. I think polling without asking about the third party candidates is more likely to capture how the voter will actually cast their vote. The most accurate might be to ask "McCain, Obama, neither or will not vote", but that is speculation on my part.



> So are we to believe that obama gained
> in ND but lost ground in NJ?

Looking only at Rasmussen, I see sample noise + a slight upward trend for Obama. I see no trend in ND.


Lorne makes a very important point above about Nader, Greens, and voter interest in either.

The Greens, by the way, formally nominate their candidate this weekend. It is almost certain to be Cynthia McKinney (who shares with Bob Barr having been a former congress member from Georgia).

I hope pollsters will be asking about all five.


The problem of casting a wasted vote is so important that I think questions should be asked about it in these polls. I'm not sure how they would be worded. Maybe follow up with those who say they support Obama (McCain) by asking, "suppose you could vote for anyone you wanted without helping McCain (Obama) get elected, whom would you support?".

Or do an instant runoff poll, in which each respondent ranks of all the candidates in order of preference. Without an elaborate explanation, that wouldn't completely reveal the extent to which voters believe their choice is constrained, but it would be a start.

In most polls that ask about "third" parties in general, over half of us say we want one. That tells me that people want more than two meaningful choices.


The numbers for all candidates vary considerably over the course of several months. Factors like the amount of news coverage and participation in debates have a dramatic effect on candidate performance. The inclusion of significant 3rd party and independent candidates helps to identify trends that may rear up over several months (such as perot or ventura) or over several election cycles (such as the green party).



I don't understand why polls don't force rank the top three choices, or why more don't include a measure of strength of support.

The 2 vs multi-candidate polls effectively leave important information unknown, particularly when uncoupled to a measure of 'enthusiasm' for the preferred candidate.

In my own case, as I told a pollster (2 candidate poll, with degree of 'enthusiasm' added) the other day, "nothing on god's green earth could persuade me" to vote for McCain. My first choice is Barr and my second albeit lukewarm choice Obama.

The terminology 'spoiler' and the pat phrase 'wasted vote' are annoying because they reflect perhaps unconscious bias or sour grapes on the part of the speaker, and more to the point those terms are highly misleading.

One sends a powerful message by voting for a third party candidate or independent, especially because nowadays virtually everyone doing so understands perfectly well that this will affect the relative totals of the inferior but leading contenders.

In my case the message is not just that the GOP has jumped the tracks or that its candidate is the lesser of two poor choices. That is the most one could infer from someone who might have been expected to be, say, a McCain supporter but votes for Obama or stays home.

In my case it is to state in bright and shining lights that to gain my vote, the Democrats and GOP need to make directional changes congruent with the positions of the third party candidate I voted for.

We may be stuck with one or the other of the major party candidates in this election cycle, but my vote is to refocus the major parties on the future. The leadership of the major parties may not like what they are being told, but they cannot fail to understand it.

Just as the Democrats fretted about Nader, and the GOP lost the Presidential '88 election because of Perot, the GOP has been beside itself since Libertarian votes began costing them congressional seats and a few years ago, the control of the US Senate.

There is MUCH more crisp information content in a vote for an independent or third party candidate.

Whether the major parties make changes to regain support they have driven away is a separate question, but at least they clearly know what they must do or the tradeoffs they must make to do so.

In a very real sense voting for one or the other of two unappealing 'major' candidates just because one of those two almost certainly will win, is the real 'wasted vote'.

A vote for a 'minor' candidate is anything but wasted.



Just realized, and a friend simultaneously hastened to point out to me, that I was referring to the Bush-Clinton-Perot election as being in '88, but of course it was '92.

It would bother me more to be mistaken, but it happens too often to worry about...



Maybe you can explain why the flip side of your arguement would apply, ie that people think that there's only John McCain and Barack Obama to vote for when the pollster asks and then in the voting booth on Nov. 4th notice the names of Cynthia McKinney, Ralph Nader and Bob Barr on their ballot and decide to vote for them.

I think a poll should mention all the names on the ballot, right? The poll reflect the ballot, right? Please enlighten me on this.


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.