A consensus has formed around the proposition that polling, especially in multi-candidate presidential races, has become inaccurate and unreliable.This judgment, rendered by respected experts, exposes the absurdity and corruption of the recent decision by the
Commission on Presidential Debates to continue using polls to determine who will be on the stage in the fall of 2016.
Jill Lepore's excellent article in the Nov. 16 issue of The New Yorker is only the latest in a series of authoritative pieces providing evidence of how "horse-race" polling isn't working anymore - and explaining why. Dan Balz made similar points in a Washington Post article on Nov. 7. Michael Barone, the dean of election analysts, co-founder of The Almanac of American Politics, and pollster himself for nine years, made the case against polling on Nov. 11 in the Wall Street Journal. He also quoted his American Enterprise Institute colleague, Karlyn Bowman, saying that this may be "the end of polling as we know it."
The New York Times has run several pieces on the subject, including an op-ed in June by Cliff Zukin, a Rutgers scholar who is past president of the American Association of Public Opinion Research.
Dr. Zukin concluded:
"We are less sure how to conduct good survey research now than we were four years ago, and much less than eight years ago... In short, polls and pollsters are going to be less reliable. We may not even know when we're off base."
The evidence abounds. It includes the Israeli and British elections and, on Nov. 3, the Kentucky governor's race. The polls had the Democrat winning, but the Republican won in what was practically a landslide, 53% to 44%.
The reasons for the declining accuracy of polls are now well known:
- sharply falling response rates by voters called for their opinions,
- the difficulty and expense involved in reaching people who have abandoned land lines for cell phones (a law bans auto-dialing to mobile phones),
- and the near-impossibility of adjusting results to achieve the proper weighting of demographic groups and to determine which of the respondents will actually show up to vote.
As a result, Gallup, the premier U.S. polling organization, announced recently that it would not poll for the primaries this year and might not poll for the general election either.
But the main problem with polling is the uses to which it is being put - especially making it the main criterion for admission to debates. Lepore quotes Scott Keeter, director of survey research for Pew: "I don't think polling is really up to the task of deciding the field for the headliner debate."She also quotes Bill McInturff, co-founder of Public Opinion Strategies as saying, "I didn't think my job was to design polling so that Fox could pick people for a debate."
Even in the face of all this evidence, the Commission on Presidential Debates just announced it will rely on polling criteria to decide who participates in the final fall presidential debates. Their decision cannot possibly be justified - except as a way to keep someone other than a Republican and a Democrat off the stage.
62 percent of Americans say they would vote for an independent for president in 2016. But because of the CPD's partisan stranglehold on the process, we won't see an independent candidate in the debates. And everyone knows you can't be president if you're not in the debates.
Instead of polls, the Commission on Presidential Debates (CPD) could use direct voter endorsement as a criterion. Every state in the U.S., for instance, uses signature collection as a way to determine ballot access. The CPD could do the same. Or, as several members of the group Change the Rule - including former Democratic Gov. Bob Kerrey and former Republican Rep. Vin Weber - have advocated, the CPD could agree that a national online primary can pick a single independent candidate for the stage.
The deficiencies of polling, as used by CPD, were cited in July by Ann Ravel, chair of the Federal Election Commission:
said that the world may have a polling problemHe gave a number of examples
Ravel and one other commissioner favored Level the Playing Field's request to open a rulemaking. Ravel's side lost, 3-2, but the erosion of the duopoly's hold on power is evident - not just at the FEC but at the CPD itself.
On Oct. 29, a month later than it promised, the CPD announced its criteria for the 2016 debates, and they turned out to be the same as for previous debates. An independent or third-party candidate will have to receive 15% support in an average of five horse-race polls taken in September 2016. This unfair and biased rule forces independents to meet an impossible polling threshold only 7 weeks before the election. No candidate who did not compete in a primary has scaled this barrier since the televised debates began in 1960.
What was significant was this sentence in the CPD's press release: "The criteria for 2016 were adopted by a majority vote of the CPD board." Not a consensus, not a unanimous vote, but a "majority vote."
The CPD conducts its business entirely in secret and did not report to the public what the vote was, but it has an obligation to do so. We demand it here. Members of the CPD's board, including several very distinguished journalists, publicly call for transparency in public affairs. Let's see them practice what they preach.
Editor's note: This article originally published on Presidential Debate News, and may have been slightly modified for publication on IVN.