Re: Carrying a big stick #2 Arb

[Dave Faneuf <tklaundry@juno.com>]

On Thu, 21 Feb 2002 15:16:05 -0500 "Dan Billings" <dib9@gwi.net> writes:
> With occasional exceptions, I think there overall numbers in the 
> Portland market are fairly accurate.  They usually reflect what I am 
> observing and hearing from people.

IIRC when Birch was still around their numbers were very different from
those of Arbitron because I believe they relied more heavily on telephone
interviews.  Birch gave station owners and advertisers a snapshot of what
the radio market is doing now versus Arbitron that averages numbers over
a period of three monts, so if you had an off April book because of
quesionable methodology or small sample size you had to carry those
numbers until the April book fell off the screen in August even if you
happend to have tremendous numbers later.  I personally feel the monthly
trends are better at giving management a picture of what is happening
than looking at the quarterly books.  

> They do use questionable methodology and a small sample size.  The 
> sample size could be larger, but I do not know of alternative methods
that 
> would be more accurate.

As you are aware, the smaller the sample size the greater the risk of
error so given that the small sample size that Arbitron takes in any
given county the numbers are suspect from the get go.  I know of
Providence broadcasters who complained to Arbitron when they dropped more
books in Northern Bristol County vs. Southern Bristol County, that had a
huge impact on how well Boston stations did in the Providence book.


> One thing that is crazy is how the numbers are broken down 
> demographically and by daypart.  As I said, I think the overall numbers
are 
> generally accurate, but when you start talking about Women 25-54 for
the 10am 
> to 3 pm daypart, the sample size is so small that the numbers mean
little.

See above response

> I also think it is crazy that stations put so much emphasis on any 
> one book. It is smarter to look at trends over time and not to put too
much 
> emphasis on any one survey.

I am a little confused by your use of the word trends.  I understood the
monthly trends to be just that, as much of a snapshot view of the market
as Arbitron can accomplish in one book.
 
> My view is that Arbitron is like democracy: it's the worst system 
> going, except for all the others.

Since Arbitron is really the only rating service major agencies subscribe
to, there are no others good or bad.

Nothing democratic about Arbitron, just ask any station manger who has
unsucessfully argued that while the person writing the diary loved their
morning show, naming names but put down the wrong calls that the book is
counted for the competitor and not the station the person was actually
listening to,  see the thread discussing why Oldies 103 changed to Oldies
103.3.  
df