Pretend there is a virus that infects 1 in 1,000 people. (This is a hypothetical virus, not COVID). The test for this virus is 95% accurate in that only 5% of those testing positive are false positives. If you take the test and it says you have the virus, what is the probability that you do actually have it?
Most people say 95%. Most people are wrong. The actual probability is 2%. That's a crazy big difference.
Here's the math: Of 1000 people tested, on average 5% will be false positives (about 50 people) but only 1 of the sample is actually positive. Since 51 people test positive and only 1 has the virus, that equals about 2% of those testing positive have the virus.
This is an example of “base rate neglect,” sometimes called the base rate fallacy. It is extremely common and robust and the source of many - if not most - of the bad decisions we make.
“Base rate” refers to objective, statistical information, usually from a large relevant sample size. It is a great benchmark to use when we are assessing probabilities and we have little other information. The problem is that when we do have a “little other information” we tend to ignore the base rate.
Since most important decisions depend on assessing probabilities, ignoring the base rate can be costly and sometimes dangerous.
In the above example, most people completely ignore the base rate, which is the fact that only 1 in 1000 get the virus. Instead they focus only on the test accuracy.
Let’s look at another example. You are considering buying a new car and the Trekrunner appeals to you. Unfortunately, its reliability record is not very good and reliability is important to you. However, your friend, Jasper, has a Trekrunner and he says it’s been great - no major problems at all. In fact, he gushes over it.
What do you do? Most people weigh the opinion of someone they know and trust over faceless statistics. Add the emotional element - Jasper’s enthusiasm - and that tips the scale even further in the Trekrunner’s favor.
The same thing happens when we see a lottery winner with a big check, or a heavy smoker who lives to be 100. We subconsciously place more emphasis on a single case than in the overall rate especially when that case confirms what we want to be true - that the lottery is a good bet or that smoking isn’t that dangerous.
We are all suckers for this illusion. The term “lived experience” is in vogue, causing people to overvalue a single experience over the experience of a much larger population - the base rate. Since we often don’t see the base rate, we ignore the fact that it even exists!
Investors are extremely prone to this as well. Many years ago I bought stock in a company because I had a good experience with their products. Turns out, they were in an industry that had historically poor returns. I lost money on that bet because I ignored the base rate and went with anecdotal evidence.
Another cause of base rate neglect is when we value a category or stereotype over a base rate. In his book, Thinking Fast and Slow, Daniel Kahneman gave this example: You see a person on the subway reading the New York Times. Is it more likely that 1) she has a PhD or 2) she has no college degree. Most feel that she probably has a PhD because reading the newspaper seems like something a well-educated person would be more likely to do. That ignores the fact that a much greater number of nongraduates ride the subway than PhDs so she probably has no degree.
What strategies can we use to overcome the powerful impulse to ignore the base rate?
- Seek out a base rate. In this information age, there is no reason not to look for data that represents a large number of similar cases when you are faced with an important decision.
- “Confront the lizard” - this is my term for recognizing subconscious cognitive biases and outsmarting them. Before you make a decision, consider if it is based on the experience of one person (even if it is you) or may be based on a stereotype. If so, look deeper for a base rate.
- Get smart on how to calculate probability. It’s not a skill most people learn in school, so get a book or take a short course.
Remember, your performance depends on the quality of your decisions, and that quality is directly related to how accurately you can assess probabilities. Ignore the base rate at your peril!
Think well and be well.
- Steve Haffner
Want to learn more about improving your decision performance?
Click here for my free book, 7 Strategies for Making Better Decisions