Yet another reason to hate early voting: It has eroded the journalistic exit-poll armistice, the agreement to embargo information about how Americans are voting until after the polls have closed. Without notice or discussion, mainstream news sources such as NBC and the New York Times have taken to giving out running tallies on how the vote is going so far. Will all bets be off when it comes to Election Day coverage of exit polls? Or will there be a one-day effort to pretend that, at least on November 8, the armistice still holds?
This week, Brian Williams was one of the cable hosts quick to report a Florida poll by TargetSmart and the College of William and Mary. The poll had the eye-popping takeaway that 28 percent of early voting Republicans in the Sunshine State had cast their ballots for Hillary Clinton. The poll contained detailed information about the choices early voters had made: “Among those early voters (who were asked which candidate they had voted for), Clinton outpaces Trump by a 17-point margin, 55 to 38 percent.” No one actually believes that this represents what the final vote margin will be in Florida, but the question remains: What are we doing, the week before the election, talking about vote tallies?
It’s no aberration or outlier. The New York Times also reported on early voting results this week. Noting that in North Carolina “the election is well underway,” the Times said that some 40 percent of voters “have already cast ballots, and the data from early voting suggests that [Hillary Clinton] has banked a considerable lead.”
Or how about Cleveland, where Baldwin Wallace University has been polling Cleveland.com readers: Mrs. Clinton’s allegiance to the Chicago Cubs, nemesis of the home team, notwithstanding, she has jumped to an staggeringly large lead among early voters in Democrat-heavy Cuyahoga County. “Clinton’s lead in the early voting—73.6 percent to 24 percent,” Cleveland.com reported, “is wider than Barack Obama’s victory margin over Mitt Romney” there.
Now, these are not the official exit polls, which are done by Edison Research for the National Election Pool consortium. Early access to those results comes with an agreement (one sometimes violated in this Internet age) not to publish the information from a given state until polls there are closed. Given the temptations to get the info out, NEP data is carefully quarantined.
But now that early voting is commonplace, everyday political polls become, in large part, exit polls. The pollster just adds in a question about whether one has already voted, and the data from those that say Yes become early exit-poll results.
And these new, everyday exit polls, unlike the National Election Pool, eagerly ask to be reported well ahead of the closing of the polls—indeed, even before the opening of the polls on Election Day. That makes for lousy public opinion research, as early voting tends to differ from that on Election Day. (Serious pollsters note that even among those who vote on election day, “Different types of voters turn out at different times of the day.”
But it also defeats the fairness principle that underpins the embargo: The voter who is told her candidate has already been beaten before she’s had a chance to vote is discouraged from going to the trouble of casting her own ballot, and thus effectively disenfranchised.
Discouraging that other guy’s partisans couldn’t possibly be the goal of all this exit-poll reportage on the early triumphs of the Clinton campaign. Could it?
