home *** CD-ROM | disk | FTP | other *** search
- Xref: sparky sci.math:16974 rec.puzzles:7946
- Newsgroups: sci.math,rec.puzzles
- Path: sparky!uunet!think.com!linus!linus.mitre.org!gauss!bs
- From: bs@gauss.mitre.org (Robert D. Silverman)
- Subject: Re: Marilyn Vos Savant's error?
- Message-ID: <1992Dec15.202207.27644@linus.mitre.org>
- Sender: news@linus.mitre.org (News Service)
- Nntp-Posting-Host: gauss.mitre.org
- Organization: Research Computer Facility, MITRE Corporation, Bedford, MA
- References: <1992Dec15.052211.24395@CSD-NewsHost.Stanford.EDU> <1992Dec15.160000.3714@cs.cornell.edu> <BzBE0J.A6F@watdragon.uwaterloo.ca>
- Date: Tue, 15 Dec 1992 20:22:07 GMT
- Lines: 65
-
- In article <BzBE0J.A6F@watdragon.uwaterloo.ca> deghare@daisy.uwaterloo.ca (Dave Hare) writes:
-
- stuff deleted for brevity....
-
- :>> 1) You meet a man on the street and ask him how many children he has.
- :>> He replies "two, and one is a boy." What is the probability that
- :>> his other child is also a boy?
- :>
- :>It depends entirely on how your probability space is constructed.
-
- etc.
-
- :This is entirely bogus. The question is clearly intended to be a particular
- :probability problem phrased in familiar terms. The intent of the problem is
- :the basic probability exercise, not a problem in the sociology of people
- :meeting on the street.
-
- If one really wants to be fussy about language, I can make a case that
- the answer to the question should be ZERO.
-
- The man replied "two, and one is a boy". He did NOT say:
-
- "two, and at least one is a boy"
-
- The first can be intrepreted as saying: number of boys = 1,
- the second: number of boys >= 1
-
- The problem with many of these problems lies not in the mathematics of
- condition probability, but rather in the sloppiness of English. Does
- the phrase "one is a boy" imply "exactly one is a boy", or does it imply
- "at least one is a boy"? It can be interpreted either way.
-
- For those who disagree, look at this another way: Suppose the man had said
- "seven, and 3 are girls". Would you not interpret this to mean that
- exactly 3 are girls? So why does the phrase "two, and one is a boy"
- imply that the number of boys is greater than or equal to 1 and not
- exactly one??
-
-
- I remember an old statistics course, in which the instructor had posed
- a similar question on an exam. I put down the answer ZERO, just as above
- and had it marked wrong. I went back and presented the above argument and
- the professor agreed that the sloppiness in the way the question was worded
- could lead to my intrepretation. Indeed, that a careful reading, rather than
- a colloquial one would lead ONLY to my interpretation.
-
- Let me give another example, taken from an article in Notices of the AMS
- from a while back. The discussion was about the GRE exam. One question
- gave an equation or equations and asked which of the following sets
- satisfied the equation(s). It then gave 4 answers plus (e) none of the
- above. The trouble is, while one of the answers was clearly intended as
- the correct one, none of the answers presented were in set notation.
- That is to say, they were not sets, but conditions satisfied by sets.
- [i.e. 1 < x < 3 is not a set, but: {x: 1 < x < 3} is]. The article went
- on to say that for students who were taught to be careful about language
- and careful about definitions, the only correct answer was (e), even though
- (b) was the intended answer.
-
- Too many of these problems are exercizes in clarifying the ambiguity in
- English phrases, rather than in probability.
- --
- Bob Silverman
- These are my opinions and not MITRE's.
- Mitre Corporation, Bedford, MA 01730
- "You can lead a horse's ass to knowledge, but you can't make him think"
-