In the first 26 metered markets to be tabulated, the average response rate was a mere 30%, down substantially from May 1997; and, in Miami, only 18% of prospective respondents actually filled out and returned usable diaries.
How good are the ratings that Nielsen derives from such anemic levels of cooperation?
Very good, according to Nielsen, which "has concluded that the reported data are projectable, and that there is no increase in the bias" of the ratings.
Barely usable, according to the American Association of Advertising Agencies media policy committee, whose chairman characterized Nielsen's diary system during the sweeps as "almost bankrupt."
Most of the buyers and sellers of local air time, though stopping short of condemning Nielsen outright, nonetheless seem deeply concerned about what they perceive as serious flaws in Nielsen's diary operations.
Thus far, the outcry seems largely confined to the 39 metered markets in which diaries are used solely for audience composition. But what about the country's remaining 172 markets, in which Nielsen obtains all of its local audience data exclusively from diaries?
How much confidence can we have in published ratings derived from a diary sample with a cooperation rate of 30%? Or 20%? Or 18%?
Very little, we suspect, given a) today's highly fragmented television audience; and b) the known tendency of Nielsen's seven-day diary to understate viewing widely but irregularly, depending upon such factors as daypart, station, program and demo.
Does this mean that, in the 172 diary-only TV markets, we have no choice but to watch a bad situation grow progressively worse until some exotic new technology, affordable at the local level, comes along? Not at all!
MORE ACCURATE RATINGS
TN Media's local broadcast unit, working in consultation with Lawrence Friedman, a highly regarded media research specialist, believe that Nielsen, using existing technology and survey methods, can very easily arrive at far more accurate TV audience ratings in its diary markets than it now publishes -- and which we can use to negotiate the purchase of billions of dollars' worth of air time.
Here, in a nutshell, is what we have in mind.
In the nation's 172 diary markets, Nielsen now has about 2,000 metered households that belong to the Nielsen Television Index's national people meter panel. These homes are more than enough to provide stable, reliable electronic measurement for the diary markets as a group. So why not use them for this purpose?
All Nielsen has to do is multiply all diary-derived household audiences in these 172 markets by the appropriate constant so that they add back to the corresponding meter-derived number. If, for example, at 7 a.m. (ET) the diary markets' ABC affiliates have a combined meter audience of 1.2 million but a combined diary audience of only 1 million, each station's diary projection can be multiplied by 1.2.
This is hardly a radical idea; in fact, it is essentially what Nielsen does now in each individual metered market.
NARROWING THE GAP
How much more accurate would ratings in diary markets be if Nielsen did as we suggest? There's a very easy way to find out. Let's simply have Nielsen test the TN Media local broadcast unit proposal on a national scale using May 1998 sweep data. Then, in each of the 39 metered markets, we can compare meter-derived household ratings by station and daypart with diary-derived ratings:
* As is.
* Adjusted so that all local markets combined are synchronized with NTI's national people meter panel.
Is there any doubt this adjustment would dramatically narrow the huge gaps that exist between meter ratings and diary ratings locally? We think not. In any event, it would do no harm to find out.
We believe, in sum:
* The sharp falloff in diary response rates during the May 1998 sweep points up the weakness of using diaries exclusively to measure TV ratings in 172 markets.
* The problem arises chiefly from Nielsen's policy of ignoring the presence of 2,000 metered households in its diary markets.
* The metered households are there, so let's use them! All it would take on Nielsen's part is a little computer programming skill and a willingness to do the right thing. Is this asking too much?
This, at any rate, is our reaction to the recent drop in TV diary response rates.
Mr. Nass is senior VP-director of spot broadcasting, TN Media, New York.