The survey results among seniors are another symptom of a deeper failure of this year’s survey. Unlike in 2016, polls consistently showed that Mr Biden won with voters 65 and over with comfortable profit margins. The latest NBC / WSJ poll found that Mr Biden scored 23 points in the group. The latest Times / Siena poll found he ascended at 10 a.m. In the final accounts, there will be no reason to believe that any of this was real.
This is a deeper type of mistake than 2016. It suggests a profound miscalculation of the attitudes of a large demographic, not just an underestimation of their share of the electorate. In other words, the survey’s underlying raw data has deteriorated over the past four years, undoing the changes respondents made to address the bugs in 2016.
This explains why the national surveys were worse than in 2016; You gained weight through education four years ago and have made little to no changes since then. It also helps explain why the error is so closely correlated to what happened in 2016: it focuses on the same demographic group, even if the underlying source of error is very different within the group.
The survey clearly has some serious challenges. The industry has always relied on statistical adjustments to ensure that each group, like unqualified white voters, represents their fair share of the sample. However, this only helps if the respondents you reach are representative of those you don’t. In 2016 they seemed representative enough for many purposes. In 2020 they weren’t.
How have the polls deteriorated over the past four years? This is mostly speculation, but consider just a few possibilities:
The president (and the polls) violated the polls. There was no real evidence of a “hidden Trump” vote in 2016. But maybe there was one in 2020. For years, the president attacked the news media and polls, among other things. The polls themselves lost some of their credibility in 2016.
Not surprisingly, the president’s supporters were less likely to respond to polls as their skepticism about institutions grew and polls stayed in a worse place than they were four years ago.
“We must now take one version of the Shy Trump Hypothesis seriously,” said Patrick Ruffini, a Republican pollster for Echelon Insights. It would be a “problem that the polls just don’t reach large elements of the Trump coalition, which leads them to underestimate Republicans across the board if he takes part in the vote.”