Nielsen Releases New ‘LPM’ Data6/06/2005 9:51 AM Eastern
Nielsen Media Research Monday disclosed full-month demographic data from its “Local People Meters” in Washington, D.C., and Philadelphia, claiming that they represent more accurate viewing information for those two markets.
In a press release, Nielsen pointed out that while the May LPM data can’t be used for buying and selling TV ads right now, they “show that LPMs deliver more accurate and representative ratings than the meter-diaries they will replace.”
At least one Washington broadcaster, Allbritton Communications Co., immediately questioned Nielsen’s claims about the reliability of LPMs. Last week, after howls of complaints from 17 TV-station groups, Nielsen agreed to delay its launch of LPMs in D.C. and Philadelphia by about one month, to June 30.
On Monday, the ratings service pointed out that its LPM samples are larger and more representative than set-meter samples in the two DMAs in question. In Washington, the size of the African-American-household LPM sample was up 29% from set meters, from 113 to 146, while Latino and Asian households went up 122% and 133%, respectively.
In Philadelphia, the size of African-American homes in the LPM sample increased 59%, from 104 to 166, while the size of the Latino and Asian household samples increased 124% and 69%, respectively.
Nielsen also said that when measured by LPMs, viewing was up in key demographic groups. For example, among viewers 18-49 the number of people in Washington watching TV during the full day was up 22% with LPMs versus the meter-diary in the same period last year.
And the ratings service maintained that LPMs more accurately capture the diversity of viewing in African-American and Latino households. LPMs in Washington showed that African-American viewers age 18-49 watch 93% more viewing sources.
The broadcasters that called for a delay in the LPM rollouts were especially concerned about so-called fault rates with the LPMs.
On Monday, Nielsen claimed that fault rates were lower in the LPM sample than in the set-meter sample. In Washington, the LPM metered non-persons fault rate for African-American households was 11.6% versus 16.8% for the set-meter sample during the same period.
But Jerald Fritz, Allbritton’s senior vice president for legal and strategic affairs, challenged Nielsen’s assertions about the LPM fault rates based on what he claimed was past experience.
“They play fast and loose with what they call the fault rates,” Fritz said. “You have to be very careful that they’re comparing apples to apples.”
When comparing fault rates, Nielsen needs to look at not just “non-person” fault rates, which would reflect to technical problems with a meter, but also to compare personal fault rates, which have to do with people in a sample household not pressing the right buttons, according to Fritz.
“There is no data on the number of minorities not pushing the buttons,” Fritz said. “Those figures must necessarily be added to the set faults to get an accurate representation of how the LPM system is measuring the fault rates.”
He also pointed out that there is no mention of the relative faulting among minorities versus nonminorities. The broadcasters claimed that with LPMs, African Americans and Hispanics are faulting at higher rates than nonminorities.