> > This is just another indication of how inaccurate the
> > Arbitron numbers can be. How can you take single surveys
> > seriously when the Spring Arbitron books show a HUGE
> > INcrease for the Air America affiliate in Portland, Oregon
>
> > and a HUGE DEcrease for the Air America affiliate in
> > Madison, Wisconsin? Both are liberal markets and the only
>
> > explanation is that Arbitron may be somewhat useful for
> > spotting trends over several books, but a drop or increase
>
> > from one book to another is virtually meaningless.
> >
> > Some other Air America affiliates also had increases in
> the
> > Spring books, while others had decreases, which says to me
>
> > that the audiences probably stayed pretty much the same.
> >
>
>
> Hard to tell...many AA affiliates have crappy signals that
> may or may not cover the whole market. Diary distribution
> could cause quite large fluctuations from book to book in
> cases like this. Also, AA seems to be making a point of
> taking rather extreme positions...I'm sure their message
> wears thin in some markets while it resonates in others.
> Time will tell.
>
But the affiliates in both Portland OR and Madison have good signals and both markets are liberal. In fact, Madison is even more liberal than Portland, and it makes no sense that, all of a sudden, the message wore thin there and resonated even more strongly in Portland OR.