Questions about PPM ...

Arbitron's new metered ratings system, now in use in STL for the past few months, replaces the older format of paper diaries.  The older system asked respondents to write down their best memories of their radio useage at some time before the expiration of their rating period, usually a week.

The handwritten diaries were returned by mail to Arbitron and the info garnered from them was transcribed to the Arbitron database by employees who may or may not be English speakers.  That's from my own personal experience, by the way.

(These ideas were discussed with a few radio programmers and engineers and all have agreed that technical issues exist with PPM.)

Now we use the Personal People Meter, a pager-sized device carried by respondents, that supposedly listens to what you listen to and reports it back to Arbitron via a wireless connection after each day of listening.  The data is incorporated into the Arbitron database and from this amalgam of info, the monthly listening date is derived, from ages 6 and up, with all former, relevant, demos included.

There may be serious problems with this system.

First is a very simple definition: 

Listening vs. hearing --

The PPM's hear what they're exposed to, rather than listening to them.  That's a very big difference.  If you walk through a shopping mall, you'll hear a lot of different audio sources: in-store audio from radio stations and satellite channels, loud music from other shoppers.  None of this music could be ascribed to you as intentional listening. But it's a station that is ascribed to you nonetheless.  It's music that you hear, not what you listen to.

Your personal listening preference has been discounted in favor of whatever noise is playing in the region around you, whichever is louder and more heard by the PPM.  Whatever your PPM hears, intentionally or not, means more than what you intended to listen to.

Next are some technical issues:

Spectral density --

Music is a complex waveform, much more complex than just voice. The Arbitron PPM injects into the station's audio an encoded signal for their PPM device to "hear it"...the more complex the waveform is, the more difficult it is to decode it.  Mistakes can be made.  A talk radio signal may or may not be recorded by PPM in a way that records the same way as a music signal.

Pauses --

PPM has a "pause button" built in, to allow for dead air, in the event a station drops audio or goes off the air.  I do not know what the time-out length is, but TalkRadio is primed to use pauses for effect.  This may account for the lower numbers for TalkRadio, which might use more of these pauses or may use them for PPM  reasons.

Signal insertion --

PPM inserts an encoding into each subscribing radio station's transmitted signal.  Presumably, the PPM signal is at a level within the broadcast signal that can be caught by the PPM reciever worn on the hip; it may or may not be loud enough.

There are lots of reasons that PPM is a mistake...

Discuss on the STL Media Message Board. (Registration required)