Doug Darfield is senior vice president of Hispanic services for Nielsen Media Research. Darfield previously was vice president of research at Univision and director of network research at the Hispanic Broadcasting Corp. Darfield spoke recently with Multichannel News contributor Luis Clemens about the switch to the Nielsen Television Index and changes in incentives for Hispanics in the National People Meter sample as well as the ongoing debate over whether to include country of origin as a selection criteria. An edited transcript follows:
MCN: You’ve been including Spanish-language networks in the Nielsen Television Index for nine months now. Where do things stand at this point?
Doug Darfield: We’ve added TV Azteca. For a while, there was a rather large difference in the numbers that we really were never able to explain satisfactorily in terms of it being procedurally driven. That gap in the numbers, for a couple of networks, has closed up at least for the networks that we have been looking at for a long time. It is still too early to say anything about TV Azteca one way or another.
We just announced our latest series of initiatives in terms of some incentive test changes that we are going to be doing to drive some further change in terms of improvement in the National People Meter sample to get the representation a bit better before we get to the final single sample date.
It appears that most of the market is still using NHTI [Nielsen Hispanic Television Index] for currency and that is fine, because that is still the currency data that we are putting out there for the Hispanic market. We are beginning the next phase and we are going to be making some announcements and making a tour of agencies and advertisers, as well as the broadcasters, to acquaint them with our plans for reporting and how the data is going to be produced and distributed.
MCN: Tell me a bit more about the incentives.
DD: We don’t discuss publicly the quantity because different people get different amounts. We noticed when we looked at our overall profile of Hispanics we had developed a tendency to have too many Hispanics — I mean bigger than the universe estimate said we should have as a percentage total of the United States. And within the Hispanic market, we had too many English-dominant Hispanics who were receiving more money just because they were Hispanic, even though in a lot of respects their profile was more similar in terms of behavior to non-Hispanic persons.
So what we are going to do is try to get the in-Hispanic market balance right. We are going to pull back a little on incentives to English-dominant Hispanics and do another round of incentive increases for Spanish-dominant Hispanics, where we know from our universe estimates that we are still short.
MCN: Why is it difficult to get Spanish-dominant Hispanics to participate?
DD: A large chunk of it, you know … remember a certain percentage of them are, of course, um, undocumented and especially in this rather toxic climate are [reluctant] to say anything about themselves to anybody. There is also the question of lack of familiarity with the ratings process, lack of familiarity with the Nielsen name, both of which are major cooperation drivers in the U.S. in the population as a whole.
MCN: Earlier this year, the gap for Telefutura and Telemundo between the NHTI and NTI ratings was significant. What sort of numbers were we talking about then and what sort of numbers are we talking about now?
DD: There were 25% to 30% differences, and by the end of August those were closer to high single digits, in terms of percentages.
MCN: Any explanation for the mysterious close in the gap?
DD: Not really. Aside from the fact that these are smaller-rated networks and the numbers were never — how should I put this — the numbers were never something that wasn’t explainable by standard error. Because of the sample sizes and the size of the ratings of the networks involved, we would expect them to converge at some point and possibly swing the other way.
As long as you run the two samples that is going to happen, the numbers are going to revolve around a mean. It happens quickly for Telefutura, because they run movies in primetime, so it is kind of a different choice every night. It happens more slowly for Telemundo because of the fact that it is novelas and you kind of make your up or down vote on that for a longer period of time.
MCN: What about reporting?
DD: Well, seeing as we haven’t shared these with clients yet, it is not something I am really at liberty to talk about, except for the fact that I can say that all of the data that anybody gets now they are going to continue to get, and there will be more information that they haven’t gotten historically that they will be getting.
MCN: At a higher price?
DD: Oh, we don’t talk about prices, but we are not significantly changing for the agencies and the advertisers.
MCN: Why not ask country of origin?
DD: Why ask country of origin? We have a system we control on a variable that we know from countless years of studying it is very tightly tied to TV viewing, which is what we are trying to measure. If it ain’t broke, don’t fix it, except for the fact that you’ve got some people who don’t like the way the game is going so they’d rather turn over the game board.
MCN: When it comes to establishing language competence and dominance, it is a notoriously tricky area when you rely on self-reporting. Are you confident that self-reporting is working?
DD: If it wasn’t working, then we would see big chunks of people in the sample who report speaking one language and then we’d see them viewing television in the other language, and we don’t. There are lots of things that may happen as a marketer that would make it valuable to know that information. But having that data doesn’t make me do this job any better. That being said, we might add it just so that people don’t have anyplace to hide anymore. So just to finally finish the controversy one way or the other.
MCN: So you are actively considering including country of origin?
DD: We are thinking about it and if we decide it doesn’t hurt anything, we are thinking about doing it.