PPM: Obviously better...

As people concerned about accurate listing data, and not just ad sales, we need to stop repeating, as though somebody had proved it, that the people meter is “obviously better” than the diary.

They have not. It is not.

It isn’t obvious when the people meter is docked. Look at the data on how many hours a day these things are being carried, and at the low number of hours of use that Arbitron considers “compliance”.

It isn’t obvious if you can’t fix a mistake, like forgetting to carry it. If you forget to record a listening event in real time, with the diary you do have a chance to make it up 5 minutes or 5 hours later. There is zero ability to fix a PPM mistake. Yes, it is a “failure” of a sort when you have to use that feature of a diary. But at least it exists.

It isn’t obvious if the PPM is left in a purse, or in a desk drawer. This is happening. Nobody has an answer.

Studies have found that the diary methodology loses 15% of actual listening. For PPM, listening losses are far higher. How high? Arbitron should tell us. Will they?

It isn’t obvious if the PPM records sound, rather than listening. And that’s what it does. We have a technology to automatically record a special sound encoded into the broadcast signal. Wow! Aren’t we a clever species?

But that isn’t what we’re trying to measure. We want to measure listening. Cognitive reality, not background noise. Darn, but we don’t have the technology to do that. So we’ll just pretend we do, and measure the wrong thing. And evidently, we’re doing even that badly.

I’ve worked with tens of thousands of mechanical diaries and made too many trips to Arbitron to stare at the real ones to not have a good sense of the limits of the dairy methodology. The more I’ve worked with it, the better I’ve felt about it. it is an 85% solution.

I’m all for pushing that up to 95% or better. But PPM isn’t going to do that for you.

My biggest concern is for those who have to make programming decisions. With the PPM numbers, you don’t really know how many radio stations the listeners enjoy, only how many they were exposed to. TSL calculations are meaningless. Cume figures are somewhat LESS than meaningless. If you want to believe that the average listener carries around an interest in a huge number of radio stations, you go right ahead. Or you can trust decades of telephone, diary, in-person interviews, and simple common sense that suggest otherwise.