> From: "Hart, James" <jdh295@soton.ac.uk>
> Date: Sun, 26 May 1996 21:59:40 GMT
>
> The Base Rate Fallacy states that people routinely ignore
> base rate frequencies and that it is an error to do so
> and base your conclusions instead on the similarity between
> an individuals personality and the prototypes of the
> categories under consideration.
>
> You will find in most experiments that base rates will be
> equated with prior probabilities. Subjects' judgements and
> their deviations between the Bayesian posterior probability
> are used to measure the extent to which the base rate
> fallacy has been committed.
Kid-sib doesn't know what Bayesian priors are -- not even what base
rates are...
> For example if there was a disease which takes two forms,
> both of which are fatal and require two different medicines,
> only one of which can be taken at any time. Form A occurs
> 10% of the time, form B 90% of the time. There is a test to
> see which type of the disease a patient has, this test is
> 80% reliable and it says that the patient has form A of the
> disease. The patient is likely to take the treatment for
> form A of the disease even though there is only a 10% chance
> that he has form A, and a 20% chance that the test was wrong
> and that he has form B. The patient is ignoring the base
> rate.
Still don't know what base rates are. To see what's needed for an A, see
the other replies on the thread for this topic.
This archive was generated by hypermail 2b30 : Tue Feb 13 2001 - 16:23:45 GMT