Rational Consumers Test - Answers
Willaim Thurber - PhD Student
thurber at FMGMT.MGMT.UTORONTO.CA
Fri Sep 22 14:34:01 MDT 1995
Before you all scroll through to the answers let me say again that this
is not a reliable method for validating your existence nor in my
opinion is total rationality something to be admired or worse even copied.
It is important to note that rationality is not doing what others say
you should do, voting for Dole, Clinton, David Duke, or Jesse Jackson are
all equally "rational". Because I was unable to determine the motive
behind the voting patterns of some people is not proof of
"irrationality". The fact that my survey missed the motive that brought
these people to the polls, a fault that resides with the researcher not the
Neither is rationality a fundamental understanding of the laws of
statistics and probability, although I did include these types of
questions to make a point about health care, the discussion that spawned
this original post. That is if most people cannot evaluate the correct
probability of heads in a coin flip how can we expect these people to
make effective decisions about the probability of cancer, heart disease etc.
For the purposes of the test I will define irrationality as a reversal
of choice depending upon the frame of the situation. Many feel that they
make consistent choices and that nothing is relative, I argue that
everything is relative and universal truths are hard to come by.
Thanks to those of you who played along, since you all (I think) made
"irrational" choices, you have disproven that perception that some on the
left might hold: that all dittoheads are Rush drones hanging on his every
word to decide what to whine about, complain to, products to buy,
products not to buy, etc, etc, etc. I hope you have read between the
lines of my posts to understand that I have not and do not subscribe to
that point of view. [It is clear to me that you make up your own minds about
what to whine about... :) ]
Now (finally) the results....
Question 1 and question 11 are actually the same valued choices, to verify
this take $2400 with prob .66 out of both a and b in question 1. If you
chose a in Q1, you should have chosen a in question 11. If you did
not, you were not alone. In the original research 18% chose a in Q 1, but 83%
chose a in Q 11. This is the certainty effect, given a certain outcome most
people are risk adverse.
Question 2 and 12 are also the same. In question 2 we are gambling in
question 12 we are buying insurance. Somebody once said that insurance
is just respectable gambling. If you bought insurance, you also should
have picked a in Q2. All that changes is the frame of reference, 1 being
insurance one not.
Similarly in Q 3 and 13, if you valued option a in Q3 at more than $55
million you should have picked b in question 13.
Question 4 is a matter of perceived randomness, each (HHHTTT and HHTHTT)
is equally as likely, but HHTHTT LOOKS more random.
Question 5 and 15 are interesting, most people will guess b on Q5 but
realize their mistake on Q15. Since all bank teller feminists are bank
tellers, the probability that Linda is a bank teller must be larger, In
other words Linda cannot be a feminist without being a bank teller
first. It was the way I described her that made you (if you chose b)
think of her as a representative of feminists, sort of stereotyping.
Q6 to 10, given my lack of success in describing the test itself, I'll try
to post this explanation some other day.
q14, pure statistics, the prob of heads is 1/2. Coins don't have memories.
Q16: This was pure marketing. There are two pairs of responses that you
should look at. A dress shirt / business shirt and silk suit / italian
silk suit. Are the items in each of these pairs really different? What
is Italian Silk? There are 100's even 1000's of examples of how by
labeling things our judgements are affected. In politics, call it
liberal and it is bad, conservative and it is good, often without any
regard to the content. To some people a strawman is valid, usually until
they stop to think about it, but even then they will often claim that the
irrelevant attribute is important. This can be quite amusing when the
attribute is all in the mind of the ad agency or researchers mind.
So, what does this prove, maybe nothing, maybe that while you and I are
able to make the right choices about our own health care plans, many
people are not. The choice that we face is whether we want to pay for
these people now, later or make them suffer for their lack of foresight
by denying them any medical care that they cannot pay for themselves.
More information about the Rushtalk