Friday, January 22, 2016

Two Problems with The Way We Think: Confirmation Bias and Statistical Outliers


In my recent post on “The Seekers Who Aren’t” (that is, that people outside church walls aren’t really seeking the church to find answers), I wrote, 
“As proof we [church leaders] offer isolated anecdotes of those who were searching and found our churches.” 
Here I offer two reasons that anecdotes don’t prove what they promise.

There are two problems with anecdotes as proof. The first is “confirmation bias,” which the thoughtful and bestselling psychologist and author Daniel Kahneman takes on in Thinking Fast and Slow. We tend to look for evidence to confirm what we want to be true. For example, we tend be afraid of natural disasters over health risks, Kahneman writes
“Strokes cause almost twice as many deaths as all accidents combined, but 80% of respondents judged accidental death to be more likely.”
Many people look for the first sign to confirm their fears and visual representations—as well as media reports—inflame our fear. How often do we read about death by strokes versus natural disasters on the NBC Nightly News or Facebook news?
More specifically, as Kahneman writes, 
“Tsunamis are very rare even in Japan, but the image is so vivid and compelling that tourists are bound to overestimate their probability” (333).

For this reason, we remember and retell stories that tell reinforce confirming, and therefore comforting, preconceptions.

Second problem is misunderstanding outliers. Here’s the issue: I tell you that smoking, drinking, and being obese are all factors that tend to decrease life span. And then someone responds, “Ya, right. My uncle smoked two packs a day, had six beers every night, and weighed 350 pounds—he lived to be 95.” Of course, it’s possible, but it’s not counterproof. We’re not talking about one person, but a population sample of millions. Smoking, drinking, and being obese are all factors that tend to decrease life span for people as a rule. On every statistical distribution, which can be mapped on a “Bell Curve,” there’s the percentage of people on the outside ends or “outliers.” And that's where your miraculously long-lived uncle fits.


Bottom line: We might tell an anecdote or two of “real seekers,” as an antidote to evidence points in an unpleasant direction. That, however, is not good proof. We want our conclusions to be true so we seek whatever supports our convictions (confirmation bias), but exceptional stories don’t disprove the rule (misunderstanding outliers). I leave this as a question, Do we telling ourselves stories to make us feel better when feeling bad might be what’s needed?

No comments: