Base rate fallacy
From Wikipedia, the free encyclopedia
The base rate fallacy, also called base rate neglect, is an error that occurs when the conditional probability of some hypothesis H given some evidence E is assessed without taking sufficient account of the "base rate" or "prior probability" of H.
For example, suppose you know that an urn contains either 80 black balls and 20 white ones or 50 black and 50 white. Let these two hypotheses be H and H*, respectively. You randomly draw from the urn and get a black ball; call this evidence E. One might think that the probability of H given E is high simply because the probability of E given H is high (0.8). But suppose that you also know the prior probability of H is very low, 0.1. Maybe it was difficult for the urn-filler to get more than 50 black balls. In this case, the probability of H given E can be low as well (depending also on the prior of E). So, given that you draw a black ball, the probability that the urn had the 80/20 mix might still be very low while the probability it had the 50/50 mix might yet be much higher. To overlook or ignore the bearing of the low prior probability of H on the conditional probability of H given E is to commit the base rate fallacy.
For an intuitive example, suppose that Sue has won the town raffle and you hypothesize that she has won the raffle because she bribed the judges to print multiple copies of her ticket and to discard many other tickets. After all, given that this hypothesis holds, the probability of Sue winning the raffle is high. But to think that this consideration alone plus the fact that Sue won makes it probable that Sue successfully bribed the judges in such a manner is to commit the base rate fallacy. For, such reasoning ignores the prior probability that Sue pulls off such a bribe. Unless you have independent reasons for thinking the prior is high (or that successful bribes of this kind occur at a significantly high rate), the fact that Sue won is not good evidence that this bribing scenario has occurred.
Contents |
[edit] Findings in Psychology
In some experiments, students were asked to estimate the Grade Point Averages of hypothetical students. When given relevant statistics about GPA distribution, students tended to ignore them if given descriptive information about the particular student, even if the new descriptive information was obviously of little or no relevance to school performance.
This finding has been used to argue that interviews are an unnecessary part of the college admissions process because empirical evidence shows that interviewers are unable to pick successful candidates better than basic statistics[who?].
Psychologists Daniel Kahneman and Amos Tversky attempted to explain this finding in terms of the representativeness heuristic.
Richard Nisbett has argued that some attributional biases like the fundamental attribution error are instances of the base rate fallacy: people underutilize "consensus information" (the "base rate") about how others behaved in similar situations and instead prefer simpler dispositional attributions.
[edit] See also
[edit] References
- Bar-Hillel, M. (1980). The base-rate fallacy in probability judgments. Acta Psychologica, 44, 211-233.
- Kahneman, D., & Tversky, A. (1973). On the psychology of prediction. Psychological Review, 80, 237-251. (summary here)
- Nisbett, R. E., Borgida, E., Crandall, R., & Reed, H. (1976). Popular induction: Information is not always informative. In J. S. Carroll & J. W. Payne (Eds.), Cognition and social behavior, 2, 227-236.
