By Michael Todd
What have we learned from the disastrous polling predictions that came from so-called experts? If anything, it taught us how remarkably unreliable polls are. The purpose of polling is to measure or gauge the electorate, which is obviously outside the current statistical models’ purview. Yet the seductive temptation of preemptively understanding a complex and varying set of conditions and replacing it with something that appears to be science has become commonplace.
As a culture, we are heavily influenced by media propaganda, which consistently uses subjective language to push political preferences. This is the case with terminology formulated to lull people into believing a statistical model is a fact, such as “95 percent confidence intervals,” “margin of error,” and “non-response bias.” There is no such process that provides entirely accurate predictions; that’s why they incorporate a margin of error. For this reason, polls should be used in conjunction with other tools to offer reasonable expectations.
Why does this matter? Because in its current use, it is being exploited as propaganda to create an aura of despondency among some voters that their particular candidate can not win when in reality, they are winning. This is psychological warfare in its most corporeal form. It works by paralyzing those who support candidates or policies the media and political elites disagree with while encouraging those whose ideology is reinforced by the numbers.
Suppose polling became one of many instruments used by politicians and not the primary tool to demonstrate their popularity. Without being able to sucker people with delectable polling data that cannot be trusted, politicians would be compelled to persuade voters that their ideology has substance and is the right path. They would have to take up ideas and argue them decisively and persuasively, pulling people over to their side on the merits of their platform.
Something to take into account is that the polls in 2020 were more wrong than in 2016. In 2018 the polls were also wrong, which showed a blue wave taking the Senate and the House of Representatives. The massive polling leads for Biden allowed him to rake in massive donations during the run-up to the election, producing an unfair advantage thanks to the media. This, in combination with the dejection of a potential loss for a favored candidate, can make a difference in tight races.
While these are significant indictments over the reliability and usefulness of political polling and its place in the electoral process, it is worth noting that polling does have a role to play, just not to the extent it’s being used now. These are not unfounded claims but serious questions that need to be addressed. If not taken care of immediately, why would anyone continue to rely on political polling ever again?