Subject: Re: Unfair TV Ratings - Ron Reagan show (TP/30Something)
From: phil@eecs.nwu.edu (William LeFebvre)
Date: 1991-08-29, 14:43
Newsgroups: rec.arts.tv,alt.tv.twin-peaks

In article <1991Aug29.181319.15006@rodan.acs.syr.edu>, retants@rodan.acs.syr.edu writes:
|> In article <1991Aug28.222140.18463@casbah.acns.nwu.edu> phil@eecs.nwu.edu (William LeFebvre) writes:
|> >In article <56693@apple.Apple.COM>, unknown@Apple.COM (Matt Ackeret) writes:
|> >
|> >|> I'm not saying that the Nielsen families ARE a statistically valid
|> >|> sample.. they probably aren't.
|> >
|> >I don't want to get in to the position of defending A.C. Nielsen, but
|> >for what reason do you blithely assume that the Nielsen families
|> >"probably aren't" a statistically valid sample?  Do you have any idea
|> >at all what method the company uses to select those homes?
|> 
|> Well, let's start with the pitifully small sample size, and then continue
|> on the the fact that , last time i heard, all the Nielsen Families were
|> mom, dad, and 2 kids (preferably a dog and a white picket fence as well).
|> That is a DEFINATE minority in this age of 50% divorce rates and people
|> staying single until well into thier 30's. 

"...Last time I heard..."   Well that's certainly enough evidence to
convince me!  And where did you hear this great tidbit from, the
National Enquirer???  C'mon people, I want FACTS, not unsubstantiated
RUMOR!

I feel very confident in stating that the Nielsen family pool is
very diverse.  One of their goals in picking people to survey is
diversity.  The people (at Nielsen's, that is) that do this stuff
aren't *stupid*!  They know you need a diverse and statistically
representative sample for the results to be worth anything at all.
How do *you* think they pick these people, via friends, relatives
and neighbors????

And something most people probably don't realize:  there are several
different methods used to do the polling.  One method involves putting
a monitor in the house (this is the method that most people are familiar
with).  Naturally they don't want to be moving these boxes around alot,
so families with monitors tend to stay part of the sample for awhile.
But an OTHER method that they employ is the two-week diary.  The family
is sent a diary and in it they log everything they watch for two weeks,
then mail it back.  This method is not as reliable, but this group can
change each time a sample is taken, since there is no equipment to
install and remove.

So would someone who really knows what he's talking about like to comment
on the effectiveness of Nielsen's techniques?

William LeFebvre
Computing Facilities Manager and Analyst
Department of Electrical Engineering and Computer Science
Northwestern University
<phil@eecs.nwu.edu>