by Dave Zornow
CSP Magazine, March 2002

From time to time you read about one of those unfortunate far away countries with economies in a self-inflicted free fall. Consumers and companies scramble madly to keep up with hyper-inflation as buyers and sellers constantly change what they charge trying to predict what their currency will be worth, next month, next week and next year. And you my smugly think about how lucky we are, living with a stable currency and a relatively predictable economy. If you have ever felt that way, please now raise your hand.

All of you people who buy, sell, plan or support cable network, national broadcast or syndication, please put your hand down. You may be in for some big surprises come next September.

For the first time in the history of national TV ratings, Nielsen will be weighting the results of the national people meter panel. If the preliminary results of only four months of test data are any indicator, weighting will have some unpredicted, unstable and unsettling effects. For example, usually any change in a network's core demo is reflected with a similar directional change in its TV household rating. But Nielsen's proposed weighting scheme seems to have an opposite effect on household and demo ratings: networks with a 2-11 skew show higher TV HH ratings but lower demo ratings for kids and channels with an older skew have lower TV HH ratings while Persons 55+ ratings are for the most part flat. Cable researchers and Nielsen alike are scratching their heads to explain some of these impacts as Nielsen's announced implementation data of September 2002 threatens to turn the current upfront into a guessing game for both buyers and sellers.

Researcher's biggest gripe with Nielsen has been the ratings service's refusal to release a year worth of parallel trend data before implementation. The complaint isn't without merit: this is the biggest change in methodology since the introduction of people meters in the mid-80's and that change included at least a year of side-by-side data before implementation. This concern, along with cable networks' nagging doubts about how new procedures for creating cable universe estimates and Nielsen's reliance on un-audited sources for demo populations, led the CAB's CONCAM network research committee to ask Nielsen to delay in weighting until more was known about the how it would effect the credibility of the ratings currency.

Why is the long stable ratings system now in crisis? Up to now, Nielsen has claimed their strictly implemented methodology for selecting people meter homes has been able to select a properly apportioned, representative national sample. But as the U.S became more diverse, and we all became more wary of letting strangers into our homes to connect wires to our TV, the sample became increasingly out of step with the reality of who is watching TV in U.S..

The Media Ratings Council, the watchdog auditing group that reviews the policies and procedures of major research vendors, has asked Nielsen to weight their sample. That's because over time Nielsen has accumulated too few harder to recruit black and Hispanic homes and families with young children, and too many TV households with older Americans and homes that receive cable TV.

Although cable networks agree weighting is needed, they say the devil is in the details of how it is executed. Nielsen's source for universe estimates, Claritas, has so far refused to have its procedures audited by the MRC, an issue that raises both cable and broadcast researchers ire because Claritas is owned by the same company that owns Nielsen. There's also concern that the practical application of weighting will actually make things worse, because weights will increase the volatility of projections made by both buyers and sellers. Finally, some of the cable sources that stand to benefit most are concerned they will still be left with unusable data after weighting. Although weighting should theoretically increase ratings for ethnic audiences, Nielsen's threshold for sample size will stay the same, meaning that the ratings service still won't publish some of the numbers these channels say advertisers want to see.

If Nielsen's plan is executed as scheduled, most agree that it will effectively take GRPs out of the network TV marketplace in the 2002-2003 upfront. That combined with an already weak advertising economy could make it doubly tough for both broadcasters and cable networks to limp through this recession. ##

Dave Zornow is President/TNG Research, a media research consultancy and applications development company that works with media sellers and research providers.

Return to: Home • Published Articles