It seems that one thing the human mind isn’t very good at is distinguishing between randomness and probability. Psychological studies* have shown that we tend to think that the randomness that appears in large samples will also show up in small samples while simultaneously ignoring probability.
For example, when a coined is flipped, it has an equal chance of coming up heads or tails. So when flipped many times, the number of times it comes up heads will be very close to the number of times it comes up tails. Yet if heads comes up seven times out of ten flips, most people would think something was wrong with the coin when in reality, if it was flipped 10,000 times, it would be very unlikely that a streak of seven in a row would not happen.
Another example of a poor assessment most people make is in guessing the number of people required so that there is a 50-50 chance that two of them will share the same birthday. Most people’s guess is three figures when in reality only 23 people are required. Only 75 people are required for there to be a 99% chance that two of them share the same birthday.
One of the reasons most people are poor at predicting the probability of an uncertain event is the representativeness heuristic is that they base their prediction on past experiences – their experiences. And obviously, their past experiences may not be representative of the total population.
I have found this to be true in some people’s decision to purchase life insurance. Early in my career, a client’s 39 year-old spouse died of a brain aneurysm. Three men in his circle of friends called me to write insurance on their wives. This was obviously the representativeness heuristic at work, as those women’s life expectancy was not altered one iota by my client’s misfortune.
Another time I received a phone call from a gentleman whose neighbor had recently died in a car crash. Again, that accident had no impact whatsoever on his life expectancy, but he used it as motivation to purchase life insurance on himself.
Now people buying needed coverage is never a bad thing, but it seems that many people use the representativeness heuristic as justification for not buying life insurance. If no one in their tight circle has ever experienced premature death, then they might be drawing the wrong conclusion from that lack of tragedy.
So we need to educate ourselves to guard against what appears to be the mind’s natural inclinations. Two books that I have found very helpful in the regard are Misbehaving by Richard Thayler and The Undoing Project by Michael Lewis.
*Much of what’s in this essay is from Daniel Kahneman’s and Amos Tversky’s work.