Friday, November 16, 2007

Sample Sizes and PPM

Arbitron continues to face a lot of challenges in rolling out PPM. On one side we have the reality that some ethnic formats are seeing less listening in the PPM system, on the other side the weak samples, especially in the younger 18-34 demos, are also a big concern. It sort of came to a head this past week as the news hit that Clear Channel, Cox, Radio One and Cumulus crafted a letter that calls for a full review from Arbitron. It's also spread to the media with NYC newspapers sounding some alarm bells.

Sample sizes are really at the heart of the issue. With the big increases in cost for each participant in PPM Arbitron is looking for ways to keep the sample as tight as realistically possible. In the diary samples in the medium and smaller markets 18-34 returns have been a disaster for nearly 20 years. Arbitron has tried to pay more and recruit harder but we still end up with a lot of books out there where 18-24 and 25-34 males (and even females) are doubling the value of the diaries to make the demo samples match the population proportions.

The reality in PPM is that the samples are a lot bigger than the 30 response minimum Arbitron requires to print a demo in the big PPM markets. There is no planned PPM market where there is a danger of living with as few as 10 people as rumored. You have to keep in mind that with the PPM meter keepers holding the device for every day of every week for up to 18 months you get a lot more data than in a diary system.

Look at it on a monthly basis. Let's say in a big market you have 1000 diaries in an average month. In any given week you have an average of 250 people with a diary to fill out. Now let's say you have 500 people holding the meters, bet remember in any given week you have all those 500 meters gathering data. In the diary system you only had 250 people so the data collected is really on twice the sample and a lot more reliable.

The big issue here is the 18-34 sample. As long as we are using land line phones to reach them it's a losing battle. This group is mobile and ignoring land lines for cell phones at a rapid rate. Till Arbitron finds a way to reach them via Text, IM, Email or maybe even in person they will be challenged in these demos and have to weight them up a lot.

The other issue is costs. They are paying more to PPM households and in some cases paying a lot more to hold them in the sample. Arbitron hasn't really spelled out the compensation, but off the record comments from some insiders claim it's a lot more to insure a successful roll out. PPM is a different system than the diaries and comparing the two in sample is misleading.

We all know we need PPM to compete with the stronger metrics advertisers get from other media. No question PPM is a lot better and more accurate than the old recall diary system. Both sides (Arbitron and the stations) need to work together and listen to each other carefully.

No comments: