Doug Schwartz, Quinnipiac University poll director, told the Y’s Men of Westport/Weston on Thursday that though few expected the presidential election’s outcome, “the polls were not as bad as is commonly assumed.”
The “quality” polls, those done by the major television networks, by Pew Research and Quinnipiac, projected a four point popular vote victory by Hillary Clinton. She won by two. “If you’re off by two or three points on the margin, that’s not bad,” he said.
The biggest surprises—Michigan, Wisconsin and Pennsylvania—were where the polls tightened at the end and where the pollsters missed. Fewer than 80,000 votes — less than one percent—decided these states. Had their 46 Electoral College votes gone to Clinton, she would be president-elect.
In Pennsylvania polls showed Clinton in the lead, but none by wide margins. Quinnipiac’s final poll, released one week before the election, showed her up by five points. She lost by less than one, essentially a tie. “Unfortunately, we were on the wrong side.”
Ohio was their best poll. They were on the right side and close to the margin. Their final poll showed Trump up by eight points, he won by five. “No surprise.”
In Florida, another critical swing state, their final poll had Hillary up by one point. Trump won by a point, so we were off by two. “You guys were wrong.” Quinnipiac’s release said the race was too close to call, but the media ran with their numbers, not their caveat.
Why were they wrong? Their sample size includes a margin of error of three points — a mathematical reality, not a pollsters’ choice — “so there no way we can predict who’s going to win when there’s only a two point gap.”
What did they miss? Schwartz’s answer: late deciders in the swing states — about ten percent of the voters — fell overwhelmingly for Trump. Had they voted more like the final outcome, again Clinton would be president elect. Was it Comey? Wikileaks? Or? “Polling data doesn’t answer the question.”
Asked during Q&A whether they missed a “certain kind of voter?” Schwartz noted that the white working class, a key to Trump’s victory, was not under sampled. Rather, they “went for Trump by a greater margin than our polls were showing.” “He just did better in that segment,” he said.
The reason? Schwartz called it the “social desirability bias.” Some people did not want to admit they were voting for Trump even though they had made up their minds.
And Hispanic voters went overwhelmingly for Clinton, but went for Trump more than they had for Romney.
Similarly, African-American voters were correctly sampled, but fewer voted than in the two previous elections.
“Polls aren’t perfect. You can’t expect pinpoint accuracy because polls only talk with a subset of voters.”
Photo by Larry Untermeyer