If Trump Wins, It Likely Won’t Be Because There Are ‘Shy’ Trump Voters
Since 2016, a theory has circulated that “shy” Trump voters helped make it happen — and could do so again in 2020. That is, some unknown segment of Trump’s support is too “shy” to admit they back him, so the polls are underestimating him. Despite scant evidence to support the idea, we’ve heard it again and again, even in the closing days of this campaign.
But if “shy” Trump voters were a thing, you might expect to find a difference in how respondents reply to surveys conducted by telephone versus those anonymously submitted online — the idea being that social desirability bias is less likely to kick in when a respondent is dealing with a faceless computer instead of a real person. But a September study by Morning Consult showed that Trump performed about the same against Biden whether the pollster interviewed respondents by phone or online.
Support for Trump held steady online and by phone
Presidential support among likely voters, by whether respondents were polled via live-phone interviews or online
While this study is just the latest dismantling of this idea, that doesn’t mean the polls are perfect predictors of the future. As we saw in 2016, the polls can be off just enough for an underdog to win. In other words, some degree of polling error could happen, and while it would have to be much larger this time around for Trump to win, that’s part of the reason that our forecast gives Trump about a 1 in 10 shot of winning the election. We know one of the problems from 2016: Many state-level polls underrepresented the number of white voters without four-year college degrees, a group that overwhelmingly backed Trump in 2016. Although many pollsters have adjusted their methodologies to better account for the education divide among white voters, that doesn’t mean the problem has been solved entirely.
But we also can’t know what new problems may arise in 2020. For instance, the COVID-19 pandemic has led a much larger share of voters to cast ballots early. If pollsters’ models of likely voters haven’t been properly tuned to this new reality, that could create problems.
Yet should Trump win, it probably won’t be because voters who support him have tried to hide their feelings.