Observers Rock Research

Mar 1, 2004  •  Post A Comment

A study utilizing a rarely conducted but highly regarded form of research-direct observation of media consumers-is raising some big issues for media planners and buyers.
The study, released last week by Ball State University’s Center For Media Design, had academic researchers “shadow” 101 Midwest consumers throughout their day-from the moment they woke to the moment they went to sleep-to record their actual media usage. The conclusion: Consumers utilize media-all forms of it-far more than some conventional forms of media research indicate.
The study, which observed consumers’ use of television, computers, the Internet, radio, newspapers, magazines and books, found that consumers spend twice as much time each day with those media than telephone surveys indicate, and about 15 percent more time than written diaries indicate. The one exception was radio, in which the academic researchers observed slightly less listening than was recorded in written diaries.
That last point is significant because radio ratings-the basis of radio advertising sales-currently are based on written diaries. Conversely, diaries, which are currently the basis of audience data in all but one television market, tend to underreport TV viewing by about 15 percent.
The findings indicate the distortions in media consumption aren’t uniform by medium, which would suggest that planners who rely on conventional forms of media research are not allocating advertising budgets to media based on an accurate share measurement.
“What this said is that people need to be a whole lot more conscientious about drawing conclusions based on media research,” said Bob Papper, a telecommunications professor at Ball State and a co-author of the study.
He said that while current forms of media research do have some value, they are not the empirical measures of media usage that many in the industry take them for, whether it is planning media mixes, allocating budgets to specific media or negotiating or posting media buys.
The problem, Mr. Papper said, is that static methods of audience measurement simply do not reflect the dynamic nature of actual media usage, something he said is growing more dynamic with the emergence of more media options-especially ones that take place outside the home.
“We found that something on the order of 8.5 percent of TV viewing took place out of the home in what is today an unmeasured environment. That’s 24 minutes a day, which is a fairly substantial amount,” Mr. Papper said.
That’s not the first time new research has shown that static audience measurement-whether it is from telephones, diaries or meters-undercounts media usage, particularly for television. Radio researcher Arbitron found similar patterns when it conducted tests of its portable people meter (PPM). Those studies shows some higher levels of usage for some radio dayparts, and some dramatically higher levels of usage for some forms of TV, particularly cable TV networks, than were reported by the conventional ratings sources for those media (Arbitron for radio and Nielsen Media Research for television).
“We’ve always wondered, `How are Nielsen panelists using their meters?’ or, `How are Arbitron panelists using their diaries?’ We invest billions of dollars for our clients based on these data. We know it’s not perfect. Maybe this can help,” said Tony Jarvis, senior VP and director of the Strategic Insights Group at media agency MediaCom Worldwide.
Mr. Jarvis, whose agency has long been a proponent of the PPM, said the Ball State study has its limitations, but nonetheless is providing new insights about how people really consume media. He said it is limited by its size and scope, but nonetheless he was impressed by the ability to directly observe consumer behavior.
“This is not a quantitative study. It’s a qualitative study. It should be dealt with like a focus group. It gives us avenues to explore and things to think about,” said Mr. Jarvis, who suggested that key bodies of the advertising and media industries, including the Advertising Research Foundation, the American Association of Advertising Agencies and the major media advertising bureaus, come together to back an expansion of the Ball State study. “It would be a wonderfully rich study to, in fact, expand this to a large quantitative study, to see if the findings hold up,” he said.
The Ball State researchers would like nothing better. In fact, Mr. Papper said the team already is looking at replicating the study with double the sample size and expanding it from the relatively small market of Muncie, Ind., where the initial study was conducted, to the bigger market of Indianapolis.
But even at a 200-person sample, he said, it still would not live up to the projectability of commercial research studies. To do that, he said, would require a 400-person sample.
“What we say is this is really a qualitative study. But it is such a large qualitative study that we think you can draw some quantitative conclusions based on it. It’s 101 people. Can you extrapolate from that? Yes and no,” said Mr. Papper. “It’s also 1,600 hours of direct observation, and that’s a huge number.”
One area the researchers have been extrapolating has profound implications for media planners: multitasking, or the simultaneous usage of multiple media. The researchers are still analyzing some of that data, but the initial findings suggest that multitasking is far more prevalent than previous studies have indicated and they are yielding some distinctive patterns among media. Among other things, Mr. Papper said, the data reveal that the dominance of television remains intact in a multitasking media environment.
“Multitasking poses a huge challenge for the advertising world-not just how to measure it, but how to interpret and evaluate it,” he said.