Moving Target

Mark Ramsey scores some points about Arbitron’s Purple People Meter in this blog post. As always, the question remains, if the ratings vary wildly — whether between the diary system and PPM or just from quarter to quarter — which are accurate and which are bogus? Or are they all bogus? Mark’s comment is response to Arbitron’s attempt to smooth out the changes, and wildly varying results, when one group of panelists leaves and another arrives:

Put another way, smoothness creates the impression of validity while volatility creates the impression of non-validity. None of this has anything to do with validity, of course, but it has everything to do with impression.

When the ratings “bounce” and create “wide ratings swings” because the sample changes does that mean the new sample is wrong, the old one was wrong, or the system was wrong?  Or all of the above?

Or does it just mean that ratings – like all statistical estimates – are built to “bounce,” no matter how much we pretend they wouldn’t?

This issue throws light on a system where accuracy is ultimately less important than stubborn consistency masquerading as accuracy.

This is so because the samples are only panel-sized.

When we worship predictable numbers rather than true ones, what favors are we doing our clients and ourselves? Like the ocean itself, sometimes a shiny and smooth surface can hide a tumult below.

Maybe Arbitron – and radio itself – should focus less on the surface and more on the depths.


Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: