When I was living in California in 1982, I could not help but follow that year’s governor’s race between Los Angeles Mayor Tom Bradley and George Deukmejian. By election day, most polls showed Bradley with a significant lead, but after all the votes had been counted, Deukmejian had emerged the winner. Why were the polls so far off?
There is one other fact you need to know — Bradley was African-American, Deukmejian was Armenian (i.e., white). So one explanation was that whites lied to pollsters so as not to sound racially prejudiced (what we have come to call today as a “politically correct” answer). A more scientific term is social desirability bias, or responding with answers that are considered more favorable. So many people accepted this explanation that they began calling it the “Bradley Effect.” ( https://en.wikipedia.org/wiki/Bradley_effect)
The first question is does the Bradley Effect really exist, or were the polls simply flawed or ill-timed? For example, how do you account for absentee ballots? There were a bunch of them in that election.
Assuming there is such an effect, could this be an explanation for the generally inaccurate polling in our election this year? Some think so. As described in an article entitled ‘Shy Voters’ Can Help Explain Why Everyone Got the Election Wrong (http://nymag.com/scienceofus/2016/11/shy-voters-help-explain-why-so-few-predicted-a-trump-win.html?), reporters Katelyn Fossett and Steven Shepard organized a forum of people who got it right, or nearly right. One of the participants, Robert Cahaly, a senior strategist at the Republican polling firm the Trafalgar Group, who is a proponent of the “shy voter” explanation — those people who supported Trump but felt socially awkward about saying so. He said in part —
“I grew up in the South and everybody is very polite down here, and if you want to find out the truth on a hot topic, you can’t just ask the question directly. So, the neighbor is part of the mechanism to get that real answer. In the 11 battleground states, and 3 non-battleground, there was a significant drop-off between the ballot test question [which candidate you support] and the neighbors’ question [which candidate you believe most of your neighbors support]. The neighbors question result showed a similar result in each state: Hillary dropped [relative to the ballot test question] and Trump comes up across every demographic, every geography. Hillary’s drop was between 3 and 11 percent while Trump’s increase was between 3 and 7 percent. This pattern existed everywhere from Pennsylvania to Nevada to Utah to Georgia, and it was a constant.”
Which brings us back to social desirability bias — people answering in a way they think will make them look good — apparently as a variant of the Bradley Effect.
This election will be studied in every way imaginable for a long time. But I prefer a simpler explanation — “that’s why they play the game.”