Sandy Robertson

Director of ResearchCanvideo Television Sales, TorontoI think the biggest gap in television audience research lies between the views of various people in the industry. Some people still think a diary can do the job, some people think people meters can do...

Director of Research

Canvideo Television Sales, Toronto

I think the biggest gap in television audience research lies between the views of various people in the industry. Some people still think a diary can do the job, some people think people meters can do the job. The answer is a mixture of the two.

All of the methodologies have problems. But as long as the industry disagrees on which of those problematic methodologies to select, then the money is going to be divided. A good job won’t be affordable on any methology.

Sample size

The methodology is not so much the problem as the size of the sample. If you take a sample of 100, and you subdivide it into ever more finite pieces of demography, the number of people in each group becomes so small, the data becomes unreliable.

I think conventional television broadcasters can be adequately measured by the diary. Both our national networks seemed to think people meters were the answer. I don’t know whether they think that any more. Any research I’ve seen shows that the results are less desirable to them than were the results they were getting with the diary method.

Desirable

[Qualitative] data is desirable, but to say it’s needed is probably too strong. We’ve gone along for 20 years with not too much of that kind of data.

If you take a single respondent and ask them to tell you what they’ve viewed for the past week, and then you ask them, on top of that, what they’ve bought and what they’ve been doing, pretty soon they are going to say they’re not going to participate in your study. If you ask for too much, it sometimes destroys all the data.

That’s why the desire for electronic measurement is so great.

If you can measure what people are watching electronically, and then ask those whose homes are wired to answer some questions, you’re not going to affect your sample.

Unfortunately, we have not developed a passive electronic measure to the stage where anybody wants to bet on it.

Combination

Given today’s technology, I would think that a combination of the diary, for conventional broadcasters, and people meters, for nationally-based specialty channels, in particular, which probably don’t get enough viewing to be accurately remembered or recorded by the diary method, would be desirable.

If I ruled the world, I would stick with an improved diary until we had a more reliable electronic method.