Sort of off topic, but since you brought it up Bill.. I know you've been very clear that you're not a fan of PPM, but I'm curious to hear more details on your impressions. When you say flawed PPM technology, what examples of the flaws and results are causing you to arrive at those conclusions?
I don't have a dog in the fight either way, but am curious, given you've been in this business as long, or maybe even longer than me.
Although diaries aren't always that accurate either (that $1 doesn't buy accuracy), at least the diary model sampled a much larger population base over a period of time. And while I've never paid much attention to a particular book, long-term averaging provided a much higher sample. PPMs provide a more accurate sample of the users, but there is no assurance that the particular panels represent the larger population base. Some panelists keep the PPM for up to two years, and the broadcaster is at the mercy of those panelists, until new ones are chosen. A comparison might be that if we were to arbitrarily select a small group of voters and follow their changing opinions on the Presidential sweepstakes.
One of the problems statisticians have with something like audience surveys is that there are way too many variables. Day-part listening, 50 radio stations, etc. is much more detailed than a simple "who are you voting for".
There is the more recent discovery that talk stations were being under-reported due to the absence of PPM encoding when there is no music to mask the encoding. Voltaire was able to demonstrate those flaws, and Nielsen modified their model to address them.
PPM was an expensive project, much more expensive than Arbitron bargained for. To keep the cost down, sample sizes were reduced, therefore compromising the accuracy of the measurements.
Diaries in areas outside the metro are blended in with PPMs in the metro, producing a combination of long-term panelists and random diary recipients. I don't know how this works out. In our rural market, we sometimes show up in Thurston County on PPMs, but in other counties, we show up in the diaries. I fail to understand what methodology can produce any meaningful results.
Of course I am often wrong about things. I pontificated in the early '70s that FM would never work in Seattle due to the hills.