Cognitive Biases and Polling: Navigating the Illusion of Certainty

1. Understanding Our Psychological Vulnerabilities
   1. Aversion to Uncertainty
   2. The Problem with Probabilities
2. The Sword and Shield of Self-Righteousness, a Suite of Cognitive Biases
   1. Confirmation Bias
   2. Motivated Reasoning
   3. Cognitive Dissonance
3. Polling and Political Reality
4. Image Attribution

While polling offers a numerical snapshot of the electoral landscape, it exploits our psychological weaknesses—fueling uncertainty, distorting probabilities, and reinforcing biases. This manipulation obscures the true dynamics of the race, leading us to misinterpret the actual chances of candidates like Harris.

We’re particularly vulnerable to these pitfalls. For starters, we hate uncertainty; it makes us nervous. In our quest for predictability, we have turned to everything from magic and religion to myths and conspiracy theories. Then there’s our struggle with probabilities; a 57% chance of Harris winning doesn’t register as a 43% chance for Trump. Instead, we hear, “Harris is gonna win.” Additionally, we engage in cognitive gymnastics to justify our preferences—confirmation bias, motivated reasoning, and cognitive dissonance help us dismiss polls that oppose our favored candidates while elevating those that support them. Finally, polls create an anchoring effect, shaping our perceptions based on the narratives we prefer.

These psychological tendencies converge to make polling appear more significant than it truly is, leading us to misuse and misinterpret polling data when predicting election outcomes. While we can’t fully isolate their effects, as they combine to create one overarching polling error, we can examine each bias individually and then explore their collective impact on our perceptions.

Understanding Our Psychological Vulnerabilities

Aversion to Uncertainty

Uncertainty aversion, or ambiguity aversion, “is the tendency to favor the known over the unknown, including known risks over unknown risks.” Our aversion to uncertainty is so strong that it is a central driver to our decision-making as a species. One of the most robust findings in behavioral economics is that people hate taking risks unless it is in the face of certain loss.

By definition, taking a risk is dealing with an uncertainty. You don’t know whether the risk will pay out or not. The only thing that makes taking a risk tolerable is the alternative of a certain loss. The only thing we hate more than uncertainty is loss.

We look for any way out of the uncertainty trap, anything that will show us a more certain path to the future. Polling helps to set up the illusion of certainty because of the way we react to probabilities.

The Problem with Probabilities

Probability requires A LOT of thinking, which is hard. Long time readers know that we evolved to avoid thinking because our brains use 20% of our energy while only making up 2% of our mass.

Instead of sussing out what a 57% chance of winning actually means, we prefer to ask and answer the easier question: Who will win the election? And use the poll as the answer, the person with higher polling result.

It turns into a kind of rounding error. We round the highest polling result up to 100%! We can’t really look at it any other way without sinking a lot of cognitive resources into it, so we don’t. I know, I know, YOU, dear reader, you know how to interpret polls and probabilities and make yourself do the hard work of saying to yourself a 57% chance of winning means that if the election were held one hundred times, so-and-so would win 57 times, but on any given iteration, the such-and-such could win, too.

See? That’s a much less satisfying outcome when you’re really looking for certainty, which brings us to all of the cognitive gymnastics we’re willing to go through not only to get to certainty, but also, to the outcome that we prefer.

The Sword and Shield of Self-Righteousness, a Suite of Cognitive Biases

Because we hate both uncertainty and losses, we have evolved little cognitive tricks to help us sleep through the night, reassuring ourselves that all will be well, we’re right, and that the “other bad people” are both wrong and bad. We can go on living our blameless sinless lives in the certainty of our self-righteousness. We could call this suite of cognitive biases, The Sword and Shield of Self-Righteousness, as it defends us from any challenges to our self-delusions and obliterates any nay-sayers that might come along. It consists of confirmation bias, motivated reasoning, and cognitive dissonance.

Confirmation Bias

In case you hadn’t notices, we human beings are an argumentative bunch. Nothing feels better than proving yourself right, and no where do you prove the rightness of your thoughts, words, actions, and deeds more than in your own mind. It is the only place that matters.

The first step in this self-validation is confirmation bias, “the tendency to seek out and prefer information that supports our preexisting beliefs.” It’s easy! You just ignore that which doesn’t agree with you. If you don’t like wearing a mask during the worst worldwide pandemic in a century, then you simply cherry-pick the “evidence” that wearing a mask doesn’t help stop the spread of the airborne disease. It’s that easy! just ignore anything that doesn’t confirm that you are indeed right.

Motivated Reasoning

If your confirmation bias isn’t getting the job done, you can elevate your game to motivated reasoning. Motivated reasoning is a cognitive bias that leads people to favor information that supports their beliefs or goals. It’s a sort of ass-backwards reasoning because you start at your conclusion and use only the information that leads to your starting point.

You don’t ever have to even consider anything contradictory. You’re a good person, so everything you do is good, right? because good people can’t do bad things. Duh. With reasoning like that, you are always certain of yourself. You can’t go wrong, but just in case you do, we got one more item in the C-Suite.

In the context of polling, this means focusing on the numbers that support your candidate while ignoring those that suggest otherwise. If you believe your candidate is destined to win, you’ll latch onto any positive data, while dismissing unfavorable polls as “fake news.” You don’t even have to consider contradictory evidence. You’re a good person, so everything you do must be good, right? Because good people can’t be wrong!

With reasoning like that, you are always certain of yourself. You can’t go wrong, but just in case you do, we got one more item in the C-Suite.

Cognitive Dissonance

If motivated reasoning is the next level of confirmation bias, then cognitive dissonance is confirmation bias on steroids. Cognitive dissonance is the mental discomfort you feel when your actions or beliefs conflict with your values.

This discomfort drives you to ignore anything that makes you uneasy. For example, if your favored candidate is behind in the polls, you might convince yourself that the polling methodology is flawed or that the pollsters have an agenda. You tell yourself, “Those polls can’t be right; they’re just trying to discourage us!”

Lie to themselves and everyone who will listen. Masks were ineffective. If you need proof, start at your conclusion and work yourself back to the question using only information that supports your conclusion. Once you’re back at the question, ignore any contradicting evidence. You’ve got one flawed study on mask wearing that suggests it didn’t prevent the spread of #COVID19, so you beat that dead horse into hamburger meat.

In the current political landscape, this dissonance manifests in the refusal to acknowledge unfavorable polling results. If a poll shows your candidate trailing, the instinct is to dismiss it as biased or irrelevant. You take one flawed poll or a single outlier and use it to justify your belief, ignoring the broader consensus that contradicts your view.

Polling and Political Reality

We started with the desperate need to predict the election, trying to avoid the discomfort of uncertainty. Polling data provides a semblance of “informed” prediction, but probability doesn’t guarantee certainty. Thus, we wield the Sword and Shield of Self-Righteousness to maintain certainty about our favored candidate’s standing—whether they are ahead, tied, or behind.

Currently, polls indicate a race within the margin of error—it’s a toss-up. We’re constantly told the outcome is fifty-fifty. But how can that be? We’ve known Trump as a politician for nearly a decade, witnessing his chaotic response to the #COVID19 pandemic, the economic fallout from his tariffs, and the ballooning deficit due to tax cuts for the wealthy. Who in their right mind would want to experience that again?

Our need for certainty—knowing the outcome of a future event—works against us. All our cognitive instincts are hijacked by polling numbers. We torture ourselves with the belief that our worst-case scenario is imminent or comfort ourselves with the hope that our preferred outcome is on the horizon.

Image Attribution

This image was found on Valori using a DuckDuckGo Creative Commons License search