If they’re right from now on, it will it only be by accident.
This column went up at PJ Media Monday evening and was teased here at BizzyBlog Tuesday morning.
Despite all of their supposed science, improved methodologies, and sophisticated turnout models, nation’s pollsters have just suffered through their worst midterm elections drubbing in 20 years. The last time they were off this badly was when they woefully underestimated Republican gains in the Newt Gingrich “Contract with America” midterms of 1994.
In this year’s U.S. Senate races, preelection “tossup” predictions really meant “comfortable Republican wins” in three instances — Georgia, Iowa and Kansas, where Republican victory margins were eight, nine, and 11 points, respectively. Four of the others — Alaska, Colorado, Louisiana, and North Carolina — went the GOP’s way, or appear destined to. The Democrats sole tossup triumph was in New Hampshire. Additionally, soon-to-be Senate Majority Leader Mitch McConnell’s race in Kentucky and the Arkansas Senate contest were both supposed to be fairly close. Instead, they were 16-point and 17-point blowouts, respectively.
Additionally, an expected easy ten-point Democrat win in the Virginia Senate race turned into a one-point nailbiter. I think it’s entirely possible that a comprehensive audit of the Old Dominion State’s results throwing out the ballots of foreign nationals and all others who illegally voted would have turned Ed Gillespie’s 17,000-vote loss into a win. Even if going through the exercise didn’t make up the margin, it would have been an ideal opportuinty to make an important point about election integrity. Unfortunately, Republicans all too often concede prematurely.
The polling fails in governors’ races were in some respects even worse, especially since two ordinarily solid blue states went red. Republicans’ five-point win in Maryland came after almost every post-Labor Day poll showed Larry Hogan’s Democratic opponent with a double-digit lead. The Democratic incumbent’s situation in Illinois was tougher, but he still was expected to beat back a challenge from Republican Bruce Rauner. Instead, Rauner won by five points. Finally, despite all-out attacks by Paul Krugman and others at the New York Times, the GOP’s Sam Brownback outperformed the polls by six points in defeating his Democratic challenger in Kansas. A Times preelection editorial claimed that Brownback’s tax cuts we “on trial.” Well, they won with the voters.
This time, pollsters’ usual excuses won’t cut it. Their models overestimated black and Hispanic turnout, even though, based on history, everyone should have expected significant apathy going in. And sure, there were last-minute Democrat gaffes which the polls had no way to pick up. Michelle Obama’s flat-out lie about job growth in Illinois and Joe Biden’s tagging of incumbent Republican Senator Pat Roberts’ opponent in Kansas as a de facto Dem on Election Day certainly helped Republicans, but they don’t fully explain the unpredicted shellacking Democrats took in those two states.
There are two far more fundamental problems pollsters face. I don’t see how they can overcome either.
As succinctly stated by Pew Research’s Center for the People & the Press in 2012: “It has become increasingly difficult to contact potential respondents and to persuade them to participate.”
A chart at Pew’s related report based on its own experience lays it out:
Framing the chart’s results in negative language:
- Between 1997 and 2012, the failure rate in contacting a potential survey participant almost quadrupled, rising from 10 percent to 38 percent. That’s not at all surprising. Caller ID was less common 17 years ago. Today, many targeted participants won’t pick up their phones if they don’t recognize a call’s origin phone number, or if the call comes in as “private” or “restricted.” Many families, whose members already have their own individual wireless phones, never bother to check messages left at their households’ landline phones, which many pollsters still usually treat as their main number.
- During the 15-year period studied, the failed-to-cooperate rate among those who were contacted increased from 57 percent to 86 percent. In other words, by 2012, six out of seven people contacted were refusing to go ahead with an interview, begging off before its completion, or getting terminated by the pollster for incoherence.
Combining these two problems caused Pew’s survey completion rate to plummet by 27 points, or 75 percent, in 15 years. In 2012, fewer than one in ten contacts ultimately resulted in a completed interview. It would be quite a surprise to see any kind of improvement in that figure in the past two years, while it’s quite easy to imagine that the completion rate is even lower now. There’s also no reason to believe that Pew’s experience is materially different from what other pollsters are facing.
Even in the 1990s, it was already shaky to contend that one could credibly generalize about an entire population based on the willing participation of barely one-third of the desired respondent group. Now we’re supposed to believe such generalizations can still be done based on participation rates as low as 9 percent — or even lower, given that this was Pew’s 2012 average.
For all of this not to matter, one would have to believe that there’s no significant difference between the views of those who can and can’t be reached, or between participants and refusers. That premise falls flat on its face once one recognizes that Republicans are twice as likely to distrust the media as Democrats, and that pollsters are usually and properly seen as representatives of the media.
There’s one more wild card which belongs in the mix, namely whether participants are even willing to tell pollsters the truth. Though unrelated (or so we certainly hope), revelations since 2012 about government spying on journalists and its massive data collection efforts relating to citizens’ phone, Internet and other activities might be expected to lead to lower willingness to participate in polls. Those who still participate may be less inclined to express non-”mainstream” opinions in surveys while still doing so in the voting booth on Election Day.
Other than the obvious political lessons everyone but President Obama and many Democrats have learned from this year’s midterms, there’s are two important strategic lessons for poiitical candidates. First, never, ever place full faith in internal or external polls. Second, don’t let up for a single minute until the polls close on Election Day.