2 months of L.A. radio data may be wrong, Nielsen says

In an unprecedented move, ratings giant Nielsen acknowledged problems with its audience estimates for Los Angeles radio stations.

Nielsen conceded on Wednesday that two months of ratings data might have been compromised because of undisclosed problems with members of the firm’s sample audience. The company this week delayed the release of its May radio ratings, saying it needed time to perform “quality control reviews.”

Nielsen also said it needed to recalculate April’s ratings.

The disclosure rocked the region’s radio stations, which rely heavily on Nielsen’s numbers to set their advertising rates. Los Angeles is the nation’s largest radio market, as measured by revenue, with an estimated $1 billion spent annually for advertisements on dozens of stations, according to industry executives.

“I can’t remember a time when they went back and revised an entire month of ratings,” said Bill Davis, chief executive of Southern California Public Radio, which operates KPCC-FM (89.3). “This is the first in my experience.”

Radio executives said they noticed April’s ratings seemed off-kilter. KPCC’s ratings were actually lower than the previous month. That seemed unusual because in March, the station was conducting an on-air fundraiser, which typically results in lower audience levels.

Meanwhile, ratings soared for Spanish-language radio station KSCA-FM (101.9), owned by Univision Communications. KSCA, which plays regional Mexican music, leapfrogged over four other stations in the ratings to win the key “morning drive” time period — weekdays from 6 to 10 a.m. — in terms of audience share in the L.A. market.

Univision credited the surge to the popularity of its recently introduced morning drive-time show, “El Bueno, la Mala y el Feo” (The Good, the Bad and the Ugly). Univision said in a statement that it was “pleased with the steps taken by Nielsen and are confident in their commitment to quality data, which we share.”

Nielsen declined to detail the problems it uncovered with its sample audience in Los Angeles. The company noted that it acted quickly to identify and resolve the problem that might have produced tainted ratings.

“After reviewing preliminary data for the Nielsen Audio Los Angeles PPM market, we discovered inconsistencies with the household ratings that led us to conduct a more in-depth analysis of the data,” Nielsen said in a statement distributed to radio stations on Wednesday.

The ratings giant has been trying to verify that individuals who participated in its sample audience panel were truly independent. They want to verify that none of the participants has ties to any radio stations or radio personalities.

Nielsen acknowledged that it needed to “remove a household from the panel for not meeting our quality standards.”

The issue is a thorny one for Nielsen, which acquired the Arbitron radio ratings company last year.

The company strives to recruit an audience pool that is representative of the region, which can be difficult in such a large and ethnically diverse market as Los Angeles. The recent problem, in addition to allegations of audience pool tampering last year, may expose larger problems with Nielsen’s methodology.

Some industry insiders said the firm has been relying on too small of a listener pool to accurately extrapolate audience levels for such a large city.

There are more than 2,700 families in the Los Angeles region that participate in Nielsen’s sample audience for radio ratings, the firm said. Nielsen plans to expand the pool with an additional 175 homes by the end of the year.

Historically, the company has grappled with the fair representation of Latino families, and other ethnic minorities.

Some immigrant families are suspicious of an organization that seeks to monitor its TV viewing and radio preferences.

Nielsen has used mathematical calculations to “weight” the sample to try to achieve a representative sample of Latinos.

The company said it would release the May ratings Tuesday. Revised ratings for April, which were released last month, are expected to be completed and distributed to stations at the end of next week.

“Nielsen is committed to upholding the highest standards of data integrity and acts swiftly to meet those standards,” the company said in a statement. It called the audience pool problem in Los Angeles an “isolated situation.”

Nielsen also experienced problems with its sample audience last summer and launched an investigation into possible audience pool tampering.

Last year, Spanish-language radio host Ricardo Sanchez, who then worked at Spanish Broadcasting System’s popular La Raza station KLAX-FM (97.9), rocketed to the top of the ratings for morning drive-time programs. Other radio stations complained, questioning the accuracy of the ratings.

Sanchez, who goes by the nickname El Mandril, left La Raza late last year. He returned to the airwaves this year, but at a new station, KXOS-FM (93.9). A person familiar with the Nielsen investigation said there was no evidence to connect Sanchez to the audience sample problem.

In fact, Sanchez’s new station didn’t crack the Top 10 in Los Angeles during the questionable April measurement period.

Twitter: @MegJamesLAT