'USA TODAY; 'JOURNAL' FUME AT MRI

By Published on .

Most Popular
USA Today and The Wall Street Journal are locking horns with Mediamark Research Inc. over the latest syndicated readership surveys.

The national dailies are incensed over MRI's decision to discard new results that would have shown a huge upsurge in their audience counts.

MRI claims the increases-the result of a new measurement method-were so dramatic as to be statistically unbalanced. Instead, MRI is substituting data from its fall 1995 report and releasing it, along with the latest results for national magazines, in a package due out April 1.

And MRI is returning to its old measurement method for future reports.

The Journal and USA Today are not happy with those decisions.

"Our readership skyrocketed," said Paul Atkinson, VP-advertising at the Journal. "Our argument is not to poke a hole in other people's numbers; it is just to be treated fairly."

He and Carolyn F. Vesper, senior VP at USA Today, urged MRI to reconsider and release the latest survey results with the April 1 report. The papers sent a joint letter to MRI on March 19 saying they were "alarmed" over its "arbitrary decision" not to release the new numbers.

PROTECTING MAGAZINES?

Mr. Atkinson charges MRI is protecting its main clients, magazine publishers. MRI executives deny that, claiming they will only present accurate results.

So far, the newspapers' arguments are falling on deaf ears.

"They are not credible data on their face," MRI President David Bender said of the newest numbers, while refusing to confirm a significant uptick.

`SCREEN-IN' METHOD

At the heart of the dispute is the "screen-in" method used to whittle down the 20,000 people surveyed by MRI every year.

For weekly and monthly magazines, participants are shown logos and asked if they think they've read an issue of the magazine in the past six months. For daily newspapers, the screen-in period was limited to seven days.

For years, the big dailies had felt that method unfairly reduced their pool of potential readers and artificially deflated results.

Simmons Market Research Bureau is slated to release results of its spring readership survey on March 27, but has steered clear of the controversy by maintaining a six-month screen for magazines and newspapers.

Simmons' fall results showed significantly higher audience levels than MRI's. The Journal's adult readership was reported at 4.8 million by Simmons, but at 29.2% less by MRI with 3.4 million. USA Today was tracked with 5.9 million adult readers by Simmons, and 34% less by MRI at 3.9 million.

For the spring 1996 report, MRI agreed to change its procedure and allow a six-month screen for newspaper logos.

CONSISTENT LEVELS

"We expected it would yield a higher screen-in rate but that the eventual readership levels would be somewhat consistent with their historical levels," said Julian Baim, senior VP-research at MRI.

He said MRI would be willing to release the higher numbers sometime after publishing the spring report, for discussion in the research community.

But the data will not be part of the information ad agencies use to evaluate media decisions.

"Some media decisions are made based on these data," Ms. Vesper said, "and that's especially true in smaller markets where we may not have offices."

The New York Times stayed out of the fray because it was evaluated as a local paper and accorded a seven-day screen-in by MRI in the latest results. Kathi Love, director of marketing at the Times, was annoyed by that move. "I think all the newspapers should be treated equally," she said.

ANOTHER AGGRAVATION

Mr. Atkinson at the Journal is further angered by MRI's decision to use fall '95 numbers because that report had produced particularly poor showings for the national dailies. "We both had drops of 25% from '94 to '95," he said. "How come MRI didn't consider that abnormal?"

MRI also said it is permanently returning to a seven-day screen for newspapers.

"The conclusion is that no matter how well intentioned, we believe the results were flawed," Mr. Baim said.

In this article: