Consensus reached on radio audience measurement; next up? sample size

Some dilemmas require the wisdom of Job. Unfortunately, my name is David. When I took on my role as chair of the BBM Continuous/Extended Measurement Committee, I knew that this issue would prompt a vigorous discussion in all sectors of the radio and advertising communities. Little did I know that "extended" would come to refer to the lifespan of the committee as we sought a compromise position. After more than a year, a series of formal and informal meetings and various presentations, the moment has arrived. Now that we have finalized this matter, we can move on to something simpler like dividing up the Gaza strip.

We set out to come up with a determination amidst a flurry of position papers, press releases and significant amounts of lobbying on the most senior levels. The truth is that all of this activity says something very positive. Even though the topic may at first seem somewhat dry, it is clear that BBM audience measurement is the currency that is deemed vital by both buyers and sellers.

And the people at BBM are both conscientious and serious about their job. There was a consensus on all sides we do something positive for the industry and that we act in unison. Thus the need for a compromise. Most importantly, a genuine passion for radio was evident throughout the discussions. As we move toward an uncertain technological future, it is that passion which will allow the medium to survive and thrive.

BBM is jointly owned and controlled by broadcasters, advertisers and agencies. In fairness, it is important to point out that the broadcasters contribute the large majority of operational costs. It was clear from the outset that advertisers/agencies prefer continuous year-long measurement. Functionally, this would mean 48 weeks since weeks surrounding Christmas, for example, are impractical for enlisting diary keepers.

Among broadcasters, there are a wide variety of opinions concerning the optimal number of weeks and timing for surveys. Valid points were made on all sides. As part of the equation, the significant cost of an increased number of surveys (not to mention increased sample size) has to be factored in.

In the final analysis, the following is an overview of the extended measurement which will proceed in two phases over the next two fiscals. I should add that this is seen as the beginning, not the end. This is a transitional process by which we seek to establish an ever improving and more reliable approach to radio audience measurement.

Extending BBM Radio Measurement
4 Books: 28 Weeks 04/05

Extending BBM Radio Measurement
4 Books: 30 Weeks 05/06

• This increase in the number of radio audience surveys and weeks of measurement in the Top 9 markets should be viewed in the context of other significant improvements that BBM has introduced in the past year. These include:
• 30% sample increase for 42 large and Metro markets.
• Special $5 incentive for respondent households with Male 18-24 resulting in significant improvement in survey return rate for this demo.
• Guaranteed measurement of all radio markets with a 12+ population of 60,000 or more.
• Member access to audience data for all significant Canadian stations i.e. greater than a 1.0% share in Top 9 markets, 5.0% share or greater in other markets.
• Introduction of Random Digit Dialing for respondent enumeration resulting in more representative samples.

Now that peace has returned to the kingdom, perhaps we can all pack up our position papers and leave lobbying to the full-time politicians. Or perhaps not. We are currently examining the other issues deemed to be of critical importance, among them...sample size. A few of the broadcasters feel strongly that they would like to increase sample size and thereby lessen any statistical bounce. This, of course, comes with a hefty price tag. It remains to be seen if those same broadcasters will dig deep and the funding will be forthcoming.

On the not-too-distant horizon are a few developments that could significantly affect survey methodology. With the almost certain advent of subscription/satellite radio, the number of stations to be measured will increase exponentially. While this will certainly generate listener excitement and a further revitalization of the medium, it presents some measurement challenges that we are currently studying.

And that brings us to the next point. Portable people meters. These combined television/radio meters have been thoroughly tested by BBM in Quebec. They are currently being used for television measurement. The radio implications have yet to be fully fleshed out. Since the system is passive and picks up the encrypted station signals, reporting is very simple and reliable. No relying on a diary keeper's memory. Next, use of the meters certainly implies continuous measurement. Even more important is how the use of meters might have an effect on reported tuning. Early experience in tests show that while the average diary keeper reports listening to three stations, the meter reports show that same person listening to six stations. The use of panels involves far fewer respondents than the diary methodology (with an ever changing base of diary keepers), but monitors them much more comprehensively. This could produce significantly different results.

The next steps for the survey committee involve education. It is critical that buyers and sellers be equipped to take advantage of the significant resources afforded them by BBM. After all, research is only as good as the analysts using it. So break out the bugles. After that, perhaps the committee can take a holiday in the middle east.

(Published in "Media in Canada")