|
|
|
Portable People Meter
Wilmington test data way off from diary results
By Kevin Downey
Anyone hoping that Arbitron’s Portable People Meter system would quickly replace the much maligned diary systems now used to measure TV and radio audiences is in for a long wait.
It’ll be another year or more, in fact, before even a gradual rollout is to begin.
The reason: Early numbers from a test of the device in the Wilmington, Del., market differ, wildly in some cases, from numbers recorded by diary entrants.
One of two things is at work.
Either the sample audience for the PPM is too small and thus doesn't accurately reflect true media use in that market. If that is the case, broadening the study to include more households would, in theory, bring results more in line with the diaries.
The other, more ominous possibility is that the PPM results are indeed reflective of true media use and that the diaries are even less reliable than even their worst critics suspected.
Whichever the case, the launch of the PPM to all markets will be held off until the test expands to a wider sample audience, as Arbitron had originally planned.
"I think Arbitron would like to do it as soon as possible but all of us want to do it as accurately as possible," says Jack Loftus, a spokesman for Nielsen Media Research, which may partner with Arbitron in deploying the PPM system.
"You’re talking about using this as a tool in commerce, so people have to have a certain degree of confidence in the data."
The second wave of results from the Wilmington test of the multimedia measurement tool, which began last December, show ratings that are as much as 133 percent higher than ratings from the diary.
The results may confirm that the diary-based system, which serves as the basis for many of the decisions about when and where $35.1 billion in local ad time is placed, is as bad as media people have long complained.
Or the discrepancies may simply mean that Arbitron’s test of 300 people is too small to gauge the accuracy of the PPM, at least for television.
"You are looking at the Wilmington radio metro area and you are comparing that with Nielsen data for the entire designated market area," says Loftus. "So you would be quite amazed if the numbers were the same or even remotely similar."
Arbitron plans to expand the PPM test in 2002.
"We will move into phase-two, which means we expand the sample to 1,500 people throughout the Philadelphia DMA," says Thom Mocarsky, vice president of communications at Arbitron.
"We get through phase-two and we’ll be ready to expand to rolling out the system. Conceivably, it could be available in the second half of 2002. We’ll need more time to build the sample for radio, though, which would be available no sooner than 2003."
The second set of test results released from the first phase of the PPM test were collected in July and show that, overall, the rating for all broadcast media was 18 percent higher than the diary rating.
The broadcast TV rating was a 9.9, or 13 percent higher than the 8.8 rating measured by the current system, which is based on household meters and diaries.
The PPM results were especially striking for cable TV. The rating from the PPM test was 4.6, or more than twice the 2.1 measured by the current system.
"It’s a well-known fact that the diary does not report those niche cable networks that we spend time with," says Mocarsky.
"Otherwise there wouldn’t be this phenomenon called zero-cell, which is when the meter says viewing is happening but the diary can’t find anybody to attribute it to."
Radio ratings were the most similar between the two systems, both of which are based on the same geographic region.
The PPM rating of 10.3, in fact, was just 2 percent above the rating from the diary method.
Ratings for younger people, in general, were considerably more out of whack than for older people.
The rating from the PPM for broadcast and cable TV, in fact, was 81 percent higher than the diary rating for people 6-17 years old. That compares to a 16 percent difference for adults over the age of 55.
Those results do not come as a complete surprise.
Younger people are less inclined to fill out media-usage diaries than older folks.
One of the appeals of the PPM is that it is expected to be a more accurate measure for broadcast audiences because it is a passive system.
PPM participants only have to wear a pager-sized device, which picks up inaudible signals from TV and radio broadcasts, and put it in a docking station at night to download the data to Arbitron.
But because the PPM is passive, it presents another problem for comparing test results to ratings from the diary.
The PPM picks up signals anywhere that a participant is exposed to broadcast media, such as a TV in a bar. Arbitron is not yet able to determine how much viewing is done outside the home or how much of that viewing may have caused the discrepancies between the two systems.
"And there is a whole new definition of watching television," adds Nielsen’s Loftus.
"In one definition, people have to proactively say when they consider themselves to be in the audience. With the PPM system, it’s automatic, meaning simply that exposure translates into audience.
"That does not mean one is better than the other. What it does mean is that it’s different," he says.
|
ARBITRON'S
PORTABLE PEOPLE METER TEST |
|||
|
Media Type |
Portable People Meter |
Existing Methodologies* |
% Diff. |
|
Combined Media** |
24.8 |
21.0 |
18 |
|
Broadcast Television |
9.9 |
8.8 |
13 |
|
Cable TV |
4.6 |
2.1 |
119 |
|
Radio |
10.3 |
10.1 |
2 |
|
* Radio - Arbitron Wilmington metro diary; TV & cable - Nielsen Media Research, Philadelphia DMA, July 2001** 38 Radio stations, 7 broadcast television stations, 15 cable outlets Source: Arbitron Portable People Meter Test - Wilmington, Delaware, Sept. 2001 |
|||
|
ARBITRON'S
PORTABLE PEOPLE METER TEST |
|||
|
Media Type |
Portable People Meter |
Existing Methodologies* |
% Diff. |
|
Broadcast & Cable Television** |
|||
|
5:00 a.m.-10:00 a.m. |
9.5 |
7.0 |
36 |
|
10:00 a.m.-4:00 p.m. |
10.8 |
8.2 |
32 |
|
4:00 p.m.-8:00 p.m. |
19.5 |
16.3 |
20 |
|
8:00 p.m.-Midnight |
30.0 |
23.7 |
27 |
|
Midnight-5:00 a.m. |
7.7 |
3.3 |
133 |
|
Radio** |
|||
|
5:00 a.m.-10:00 a.m. |
10.0 |
12.2 |
-18 |
|
10:00 a.m.-4:00 p.m. |
15.7 |
18.0 |
-13 |
|
4:00 p.m.-8:00 p.m. |
14.0 |
11.1 |
26 |
|
8:00 p.m.-Midnight |
7.2 |
4.7 |
53 |
|
Midnight-5:00 a.m. |
3.6 |
2.0 |
80 |
|
* Radio - Arbitron Wilmington metro diary; TV & cable - Nielsen Media Research, Philadelphia DMA, July 2001** 38 Radio stations, 7 broadcast television stations, 15 cable outlets Source: Arbitron Portable People Meter Test - Wilmington, Delaware, Sept. 2001 |
|||
|
ARBITRON'S
PORTABLE PEOPLE METER TEST |
|||
|
Media Type |
Portable People Meter |
Existing Methodologies* |
% Diff. |
|
Broadcast & Cable Television** |
|||
|
People 6-17 Yrs. |
15.2 |
8.4 |
81 |
|
People 18-34 Yrs. |
13.3 |
7.6 |
75 |
|
People 35-54 Yrs. |
13.0 |
10.0 |
30 |
|
People 55+ Yrs. |
18.8 |
16.2 |
16 |
|
Radio** |
|||
|
People 12-17 Yrs. |
7.4 |
9.7 |
-24 |
|
People 18-34 Yrs. |
12.3 |
11.3 |
9 |
|
People 35-54 Yrs. |
10.3 |
10.7 |
-4 |
|
People 55+ Yrs. |
9.1 |
8.0 |
14 |
|
* Radio - Arbitron Wilmington metro diary; TV & cable -
Nielsen Media Research, Philadelphia DMA, July 2001 |
|||
October 2, 2001 © 2001 Media Life
-Kevin Downey is a staff writer for Media Life.
