Jonathan Jakubowski, executive chairman of the Wood County, Ohio, Republican Party, and Christos A. Makridis, a research professor with Arizona State University and senior adviser at Gallup, this week predicted in an op-ed for The Hill that “President Trump is going to win the 2020 presidential election — and win big.”
In their analysis of the polls which have been leaning heavily in the direction of Democratic candidate Joe Biden, Jakubowski and Makridis identify at least three key problems with those polls (Don’t believe the polls — Trump is winning):
1. The two experts suggest that behavioral psychology has shown that the way a question is framed determines the potential answers. Gallup has long ago discovered that the same respondents answer very differently on the very same issue depending on the wording. And there’s the issue of respondents feeling ashamed of their actual choice, as in the 2016 Republican candidate Donald Trump. In many instances, those respondents gave the answers they though the pollsters were looking for, instead of reflecting their real views. Jakubowski and Makridis cite a recent poll by the Cato Institute that suggests that almost two-thirds of Americans say they prefer not to share their genuine political opinions.
2. The other issue has to do with the definition of the sample of respondents, beginning with the medium – landline, cellphone, internet, email; location; sample size; and demographic factors. Then there’s the distinction between “registered voters” and “likely voters,” the first being fact-based, the other a much softer definition. The margin of error in these cases is not reliable, seeing as the entire population sample is skewed from the start. They cite a Rasmussen poll that showed Trump supporters are less likely to share their position with pollsters. Sample sizes matter, too – the article quotes Robert Cahaly of the Trafalgar Group whose service would not use fewer than 1,000 respondents in their samples, claiming that the polls featuring only a few hundreds respondents incur a margin of error that’s far too high.
3. Finally, the current news cycle can influence voter attitudes–especially in swing states, which means that the answer they give the pollster reflects what they think about this or that outrageous thing that Trump or Biden had said or done last night, but they won’t necessarily base their vote on that. For example, a Gallup survey recently found that 56% of respondents said they are better off than they were four years ago – while in the same poll criticizing the way the administration has managed the pandemic and the economy.
Jakubowski and Makridis conclude: “We don’t pretend to have all the answers, but we do have some of the right questions. How are people actually feeling? Surveys are great when respondents know what they’re answering and when the sample is representative, but surveys can be misleading otherwise.”