Nearly two weeks ago, The Washington Post and ABC released a poll which was the first to show Clinton gaining a majority of support (53 percent) among primary voters. In an article on the results, WaPo described Clinton as “surging,” and the media attention and continued increase in her numbers in the wake of the poll certainly backed the claim. Two weeks ago Clinton was the
embattled frontrunner, today her “inevitability” is openly discussed. CNN’s Bill Schneider put the already heavy significance of the WaPo/ABC poll in context, pointing out that every candidate since 1980 who has gotten majority support in polls taken the year before the election has gone on to win the nomination.
But while the stated significance of polls lies in their ability to detect public interest in a candidate, it is too often left unsaid that these polls also help to create that interest. As pollsters Mike Murphy and Mark Mellman wrote in the L.A. Times this summer:
The result is a cycle. Early national polling is used to declare winners and losers. Those declarations affect the flow of money and coverage, which is then reported as winners and losers, part two, thereby driving the next polls.
Pollster.com’s Mark Blumenthol furthered this point in a cautionary post:
The horse race nature of the coverage leads voters to more positive evaluations of the winners and more negative evaluations of the losers. Media accounts portray the winners and their campaigns as competent and able, while the losers look hapless and faltering. Which set of characteristics would you want in a president? Not surprisingly, voters readily make the connection between winning and competence.
But the cycle of polls “driving” other polls (and therefore driving the vote) is not so worrisome if the polls are accurate and thus the momentum they document actually exists, or if the public is informed enough to realize that it must take poll results with a grain of salt. Blumenthol, Murphy, and Mellman most recently (and many other political experts in the past), have argued that the first scenario is unlikely. Which leaves us at the mercy of the second scenario—a poll-savvy public—which is difficult to discern (without another poll!) and even harder to believe.
Those members of the public who have been following the election closely, are informed on political theory and the intricacies of polling - in other words, the segment of the population who plan to vote in the upcoming primary, are not likely to be caught up in any of the artificialities of the polling cycle. Those who are less informed, the readers who read the headlines but rarely make it to the op-ed section, the “impressionistic” citizens which a recent LA Times article claimed make up a significant portion of voters, are far more likely to be affected.
Or, as Frank Farley, a psychologist at Temple University and a former president of the American Psychological Association put it to me, “If [voters] get a mental perception that so and so is the winner at this stage that can often be an end game for them. They get an early conclusion and they often go with that… In a country of millions and millions of people, very busy people, you can’t go on editing and revising.”
Although these less-informed voters are not likely to participate in the primaries, their opinions are likely to have an impression on those voters who do make the trip to the booth. As Bill Schneider explained shortly after the release of the WaPo poll, “What’s behind the Clinton surge? Fifty-seven percent of Democrats think she’s the candidate with the best chance to win the White House.”
In other words, it is her perceived electability that is largely driving her surge in the polls. And the well-informed primary voters will gauge electability the same way everyone else does, by looking at the numbers reported in polls. As news outlets continue to flog their poll results as the quasi-scientific measure of voter sentiment, they would do well to include a bit of this context when divining great political truths in the numbers.