NetNews Usenet Archive 1993 #3/NN_1993_3.iso/spool/rec/autos/32755

Internet Message Format | 1993-01-27 | 3.7 KB

Path: sparky!uunet!ogicse!uwm.edu!zaphod.mps.ohio-state.edu!magnus.acs.ohio-state.edu!jnielsen
From: [e-mail] (John F Nielsen)
Newsgroups: rec.autos
Subject: Re: yet another consumer reports article (Re: Before Buying Japanese.)
Message-ID: <[e-mail]>
Date: 26 Jan 93 15:27:10 GMT
Article-I.D.: magnus.1993Jan26.152710.2253
References: <[e-mail]> <[e-mail]> <[e-mail]>
Sender: [e-mail]
Distribution: na
Organization: The Ohio State University
Lines: 68
Nntp-Posting-Host: magnusug.magnus.acs.ohio-state.edu

In article <[e-mail]> [e-mail] (Spiros Triantafyllopoulos) writes:


I think you are missing my point. I probably wasn't clear enough.
(And believe it or not I am even taking a 400 level stats course this
quarter)

I am not claiming that the results determine the rate of failure for
the population. I am claiming that the results are for comparing
autos. And, for that purpose, the rigourous sampeling methods are not
necessary.

In other words to take a look at your coffee example:
What you are doing is not comparing the coffee drinking community
to the normal population and attempting to derive a concusion.

What you are doing it taking the community of people who drink Folgers,
Maxwell House, and Hills Bros and comparing them to each other.
In other words, you are already looking at people with type A personality

In this case, you are already looking at people who want to respond
to the CU questionaire.

>(note) But the bottom line John is that it is a scientific fact that
>people satisfied with their cars will NOT respond at the same rate as
>people who are NOT. When was the last time you bragged about your
>flawless widget A and when was the last time you bitched about the
>flawed widget B? If I recall a study done people with complaints were
>something like 11 times more likely to respond than people without
>complaints...

But, it is not an issue because it is the same for every make of car.
That is a constant factor. Perhaps you can say the problem side is
more heavily weighed than the non-problem side, but it is more heavily
weighed for everyone.

>There is also the issue of sample size; how many CR readers drive K-cars
>vs. Accords? and out of these K-car-CR-readers, how many respond vs
>how many Accord-CR-readers? Out of how many? and in what proportions?

First off, most people do not buy cars because they think it is
going to break down. And, it is not that hard to know when your car
broke down. And, most people are not pleased that their car breaks
down.

An annoying problem is an annoying problem. I don't think it
signifiantly matters *what* car you are driving. If the car has
to be in the shop for a week, you'll be annoyed and remember it.
Unless you assume people are morons.

>It is simply for the same reason that call-in audience polls are NOT
>accurate. CR or anyone else has no way of controlling the response
>rate, or of mapping the responses back to a correct population.

It is not necessary to map it back to the population. In other words,
you don't look at it to determine what the failure rate of the
population is since it is probably skewed by the readership.  You look
at it to compare if the Ford Crown Vic is more reliable than the Chevy
Caprice.

So, as I said, for their purpose it is not necessary.

john
--
John Nielsen   MAGNUS Consultant            ______   ______   __  __
[e-mail]         /\  __ \ /\  ___\ /\ \/\ \
                                           \ \ \/\ \\ \___  \\ \ \_\ \
  Back off man, I'm a Scientist!            \ \_____\\/\_____\\ \_____\