Type: Biased Sample

Thought Experiment:

Let us suppose that you wish to buy a new car and have decided that on grounds of economy and longevity you want to purchase one of those solid, stalwart, middle-class Swedish cars—either a Volvo or a Saab. As a prudent and sensible buyer, you go to Consumer Reports, which informs you that the consensus of their experts is that the Volvo is mechanically superior, and the consensus of the readership is that the Volvo has the better repair record. Armed with this information, you decide to go and strike a bargain with the Volvo dealer before the week is out. In the interim, however, you go to a cocktail party where you announce this intention to an acquaintance. He reacts with disbelief and alarm: "A Volvo! You've got to be kidding. My brother-in-law had a Volvo. First, that fancy fuel injection computer thing went out. Had to replace it. Then the transmission and the clutch. Finally sold it in three years for junk."

Would you still buy the Volvo?

Source: Nisbett, R.E., et al., "Popular Induction: Information is Not Always Informative", in J.S. Carroll & J.W. Payne (Editors), Cognition and Social Behavior, Halsted, 1976

Exposition:

People tend to judge the probabilities of types of event by using what is called the "availability", or "ease of representation", heuristic.

The Availability Heuristic: The easier it is to remember, or to imagine, a type of event, the more likely it is that an event of that type will occur.

Like all rules of thumb, the ease of remembering or imagining—"representing"—a type of event is usually good evidence of degree of likelihood in ordinary circumstances. Instances of a type of event which we frequently experience will be easily remembered, so that that type will be correctly judged to be likely by the heuristic. Similarly, if there are many ways that a kind of event can come about, then it will be easy to imagine and also likely to happen. Moreover, if one has a hard time remembering an event of a given sort, then it is probably rare and unlikely. Also, if we cannot even imagine it, then there is probably almost no way for it to occur.

However, as with all rules of thumb, there are circumstances in which the Availability Heuristic leads to false results. Unusual events do happen, and if they happen to us then we tend to overestimate their likeliness when using the Availability Heuristic. You may have had the experience of seeing an accident on the road, then slowing down and driving more carefully afterwards. Of course, it's a good idea to slow down in the immediate vicinity of an accident scene, since there may be wreckage on the road. Also, it's possible that the accident took place where it did because the area is an unusually dangerous one. However, the vivid memory of the accident and your heightened caution may have lasted after you were away from the accident scene. The experience of seeing one makes accidents more available to your memory and imagination, thus making them seem more probable. However, the probability of getting in an accident in one place is not increased by seeing one in another. But when an unusual event is presented vividly to our minds, it becomes more available; and becoming more available, it seems more likely.

The Anecdotal Fallacy occurs when a recent memory, an unusual event, or a striking anecdote leads one to overestimate the probability of events of that type occurring―especially if one has access to better evidence of the frequency of such events. For instance, in the Thought Experiment, if the vividness of your acquaintance's anecdote about his brother-in-law's experience is enough to change your decision to buy the Volvo, you have committed the fallacy.

Sources:

• Daniel Kahneman, Paul Slovic & Amos Tversky (editors), Judgment Under Uncertainty: Heuristics and Biases (Cambridge University Press, 1982), Part IV.
• Massimo Piattelli-Palmarini, Inevitable Illusions: How Mistakes of Reason Rule Our Minds (John Wiley & Sons, 1994), p. 190.

Michael Shermer, "How Anecdotal Evidence Can Undermine Scientific Results", Scientific American, 7/2008

Acknowledgment:

Thanks to Stephen Rowe for a criticism of the auto accident example which led me to revise it.