Nate Cohn of the New York Times’ the Upshot has made a remarkable discovery about the USC Dornsife/Los Angeles Times poll, which has been unusually Trump-friendly throughout the campaign. According to Cohn, the poll for months has been skewed in a Trump-ward direction by the presence of a lone 19-year-old black Trump supporter in Illinois.
He’s a panelist on the U.S.C. Dornsife/Los Angeles Times Daybreak poll, which has emerged as the biggest polling outlier of the presidential campaign. Despite falling behind by double digits in some national surveys, Mr. Trump has generally led in the U.S.C./LAT poll. He held the lead for a full month until Wednesday, when Hillary Clinton took a nominal lead.
Our Trump-supporting friend in Illinois is a surprisingly big part of the reason. In some polls, he’s weighted as much as 30 times more than the average respondent, and as much as 300 times more than the least-weighted respondent.
Most polls weight responses from broad demographic groups in their sample of voters—men, women, 18-to-21-year-olds, etc.—to account for the actual sizes of those groups in the actual population of voters. But, as Cohn writes, the L.A. Times’ poll is unusual in that it applies weights for highly specific subgroups. In this case, the L.A. Times has elected to dramatically weight its limited sample of 18-to-21-year-old men to account for the fact that they’re thought to constitute 3.3 percent of the actual adult population. That makes individual voters in that category highly influential on the poll’s results:
A run of the U.S.C./LAT poll, for instance, might have only 15 or so 18-to-21-year-old men. But for those voters to make up 3.3 percent of the weighted sample, these 15 voters have to count as much as 86 people—an average weight of 5.7.
When you start considering the competing demands across multiple categories, it can quickly become necessary to give an astonishing amount of extra weight to particularly underrepresented voters—like 18-to-21-year-old black men.
The Illinois Trump supporter in question has been heavily weighted for both his age and his race, making him the most critical respondent on the L.A. Times’ panel. And it’s precisely because each iteration of the poll is based on the same re-contacted panel, rather than the new samples of voters gathered by most polls, that his pro-Trump influence on the results has been sustained. In fact, Clinton took the lead in Wednesday’s poll results purely because he hasn’t responded to the poll since last week! Maybe he’s been busy. Or maybe he’s decided to take some time off for reflection.
Cohn’s analysis of this quirk and the poll’s other flaws is well worth reading in full.