People meters not accurate?

[ercjncpr]

DRUDGE REPORT has posted the following this afternoon:

New people meters from ARBITRON show radiostations appealing to minorities have FEWER listeners... Developing...

 

[Shoot From Hip]

This story's been posted on radio industry websites since Friday. Check the home page of this site ("Abitron agrees to Independent Review") or www.radioandrecords.com or www.allaccess.com.

 

[DaleGriffin]

What's with this ridiculous headline?

 

[Stewy]

Wait, wait. Drudge and accurate together in the same thread? ;D

 

[hamNcheese]

well aint that just darn tootin'!

Id like to hear the positive on it. If somebody dropped off, im sure somebody else must have kicked up a few notches.

 

[ITC Delta]

Here's why certain groups are up in arms...the days of falsely writing down that you listened to a certain station for hours even when you didn't are over. The loyalties incurred from big prizes given away in the past, now mean nothing. While still not perfect, PPM's show things far more accurately.

 

[davideduardo]

Wait, wait. Drudge and accurate together in the same thread? ;D

Actually, the report is incomplete, not inaccurate. Listening levels in PPM measurement are down for all stations... the aveage weekly listeing is around 11 hours as opposed to 18 to 19 hours in the diary.

 

[InTIMadate]

DaveEduardo is correct.
Arbitron keeps singing about the cume increases while agencies and other clients want to know what happened to 30% of the audience. Here's the scary part. Some agencies in Houston and Philly (where the PPM is the only ratings) are spending their money on other media until radio can straighten out their story and drop their rates to reflect the ratings drop. Could that be one of the reasons for the big revenue drops in Sept? Imagine what will happen in first quarter 2008 with LA and NYC coming on board.

It's going to be ugly. Our empty suit leaders have given in and are trying to cash out and get out before this hits big. Watch for a lot of gray hair white guys retiring with big green parachutes in the next six months.....

Anybody remember John Hare of ABC? I'm sure he's happily on a golf course right now not thinking about radio.

 

[davideduardo]

Our empty suit leaders have given in and are trying to cash out and get out before this hits big. Watch for a lot of gray hair white guys retiring with big green parachutes in the next six months.....

Funny line!

 

[Hunter]

Here's the scary part. Some agencies in Houston and Philly (where the PPM is the only ratings) are spending their money on other media until radio can straighten out their story and drop their rates to reflect the ratings drop.

These must be agencies that don't buy TV, which went through a similar switch in rating methodology a few years ago. Local market TV ratings have switched from a diary method to Local People Meters (not to be confused with Arbitron's Personal People Meters) starting a few years ago. Agencies really pushed the stations to drop rates to reflect a similar seeming drop in ratings, however, supply and demand, in LA at least, helped most stations hold rate. Many did give extra added value, promotions and/or lower valued free spots to help the agencies work with the Cost Per Point increases.

The good news for radio is that the agencies have been through this and should work with stations, knowing that no station has really lost any audience, it's just the reporting of who's listening is more accurate. The bad news for radio is that the agencies have been through this....and have had practice negotiating through it. Fortunately for radio, next year is a political year which should help keep demand up....radio, which is more competitive than TV will need to be very creative to help the agencies keep them in the buy at the higher CPP levels.

Remember, agencies want PPM's. When properly deployed, they provide significantly more accurate ratings.

As a side note, TV national overnight ratings have been generated using a set top recording box for years. The local market ratings were generated using diaries (just like Arbitron) until the recent switch to LPM's - a set top box similar to that used for national ratings - and similar in scope to PPM's.

 

[Kyle D]

ITC Delta nailed it!

Anyone that makes the argument that PPM's aren't WAY more accurate, obviously doesn't know what they're talking about. 90-percent of PPM holders could forget to dock their units for a night and the information from the other 10-percent would still be more accurate than the diary on its BEST day.

As for the whining from the stations here in Philadelphia, as well as in Houston, about the difference in the "currency", if they'd just buck up and learn how to use the new data this issue would fade away. But, they're human and have that urge to complain, so the story continues to gain steam. Arbitron has plenty of materials and had numerous "training sessions" on how to interpret the info (just as Nielsen did when Mr. Murdoch launched the coalition against the LPM markets a few years back). That all faded over time, and if you didn't see it, Nielsen recently announced that it's taking LPM service to the top 55 or 56 markets, so I think TV buyers have learned how to use that data.

As a side note, TV national overnight ratings have been generated using a set top recording box for years. The local market ratings were generated using diaries (just like Arbitron) until the recent switch to LPM's - a set top box similar to that used for national ratings - and similar in scope to PPM's.

As I understand it Hunter, the LPM equipment is the exact same as used in the National sample. In fact, the inclusion of the LPM markets has increased the size of Nielsen's national sample significantly, as those homes are counted in that sample as well as the local.

I definitely don't think we've heard the end of the Arbitron hoop-la with NYC about to go live and LA coming soon, but in the end I believe that this story will go away and PPM will survive. I'm sure the PPM will eventually be rolled into folks cell phones and the demo challenges that have been seen thus far will improve. Just like anything else, these things take time.

I haven't always been a supporter of Arbitron over the years, but I think they are finally headed in the right direction.

 

[davideduardo]

Anyone that makes the argument that PPM's aren't WAY more accurate, obviously doesn't know what they're talking about. 90-percent of PPM holders could forget to dock their units for a night and the information from the other 10-percent would still be more accurate than the diary on its BEST day.

That is not precisely true. While the PPM, being a panel, has an average weekly in-tab that exceeds that of the weekly diary sample, the number of panelists active on any single day is very small. Let's take LA: 3275 panelists (vs. 7700 diaries per 12 week diary survey) of which the average should be around 2000 daily in-tab (the rest did not carry the minimum time). If you take Black Males 18-24, a group that is about 0.5% of the population, there would be less than 10 meters active on a day and about 5 at any particular time of the day. Does that sound accurate?

Considering that the sample is not adequate in all of 18-34 in two of the three markets, after 5 years of "testing" and in New York is less than 65% of the proportional quota, we have real issues, as the joint letter to Arbitron today indicated.

[/quote] As for the whining from the stations here in Philadelphia, as well as in Houston, about the difference in the "currency", if they'd just buck up and learn how to use the new data this issue would fade away.[/quote]

We know how to use the data... those of us in Houston have been receiving it for 30 months. The fact is that agencies are taking the "new reality" of a PUR of around 11 to 12 in the PPM to try to get rates that are 30% or more below those attained with the diary method, without taking into account the fact that the measurement, whenever it is accurately done, is more precise. When Nielsen went into the LPM, it took lawsuits and the threat of intervention from Congress to get things more in line.

But, they're human and have that urge to complain, so the story continues to gain steam. Arbitron has plenty of materials and had numerous "training sessions" on how to interpret the info (just as Nielsen did when Mr. Murdoch launched the coalition against the LPM markets a few years back). That all faded over time, and if you didn't see it, Nielsen recently announced that it's taking LPM service to the top 55 or 56 markets, so I think TV buyers have learned how to use that data.

The Philly and NY markets are not even MRC accredited; that should give you a clue that the issue is sample and proportionality.

As a side note, TV national overnight ratings have been generated using a set top recording box for years. The local market ratings were generated using diaries (just like Arbitron) until the recent switch to LPM's - a set top box similar to that used for national ratings - and similar in scope to PPM's.

And it took legal action and pressure from the Hill to get things right with TV.

I definitely don't think we've heard the end of the Arbitron hoop-la with NYC about to go live and LA coming soon, but in the end I believe that this story will go away and PPM will survive.

Run over to All Access and read the joint letter from Cox, Cumulus, Clear and Radio One to Arbitron. It is labeled as an "ultimatum" and is based on sample and proportionality, not ability to sell. NY is already "live" but just not currency, by the way. And LA, Riverside and Chicago are slated to start in 6 weeks.

Heck, the PPM does not even give us a differentiation for in car and at work listening!

 

[Kyle D]

If you take Black Males 18-24, a group that is about 0.5% of the population, there would be less than 10 meters active on a day and about 5 at any particular time of the day. Does that sound accurate?

Sounds more accurate to me than those same 10 people filling out a diary.

The Philly and NY markets are not even MRC accredited; that should give you a clue that the issue is sample and proportionality.

Are you talking about Arbitron or Neilsen?

NY is already "live" but just not currency, by the way.

You really like to nit-pick, don't you?

It soon will be currency and the cries and screams will be very loud.

Heck, the PPM does not even give us a differentiation for in car and at work listening!

Maybe Arbitron should put a little button on the PPM that you press when you get to (and leave) work, and another button for the car.

Would it be nice to know that difference? Sure it would, but no method is going to be perfect.

I just can't believe anyone that has been in the business as long as yourself could defend the diary method with a straight face. I don't even think it's a discussion.

 

[davideduardo]

Sounds more accurate to me than those same 10 people filling out a diary.

The issue is NOT what the behaviour is, but what the sample is. At this point, the sample is not proportional or representative.

The diary is subject to rounding. The meter is minute to minute precise, so instead of rounded hours and half hours, we have less time per tune in, and lots of very tertiary and beyond short cume incidents the diary did not pick up... for advertisers, listeners who listen 15 minutes or a half hour a week are useless, as there is no efficient reach and frequency ratio to hit those very light cumers.

Are you talking about Arbitron or Neilsen?

Arbitron.

You really like to nit-pick, don't you?

It soon will be currency and the cries and screams will be very loud.

The subscribers already have the data, one monthly and one weekly... and the sample sucks big time.

"Heck, the PPM does not even give us a differentiation for in car and at work listening"
Maybe Arbitron should put a little button on the PPM that you press when you get to (and leave) work, and another button for the car.

If as many as 40% don't even carry the meter every day, they are not going to flick buttons. Arbitron has a solution, and it is as practical as the headphone adapter....

I just can't believe anyone that has been in the business as long as yourself could defend the diary method with a straight face. I don't even think it's a discussion.

It is reasonably accurate, as proven by the PPM in cells where the sample is good. The problem is that the PPM sample is not overall correct, proportional or representative in any cell. Everyone who has analyzed the data on the subscriber side agrees.

 

[Francis]

Forgive me if this seems too simplistic but is it possible that the PPM is actually accurately measuring the listening habbits of 18-34 year olds and they just AREN'T listening to radio?

Can we not conceive of the idea that 18-34 year olds actually aren't listening to the radio on a daily basis and the PPM is reflecting that, where with diaries people would just fill in whatever at the end of the week?

 

[davideduardo]

Forgive me if this seems too simplistic but is it possible that the PPM is actually accurately measuring the listening habbits of 18-34 year olds and they just AREN'T listening to radio?

There are not enough meters placed in 18-34. Those that are placed show normal listening patterns.

Of course, if there are not enough meters placed, anything could be true.

 

[InTIMadate]

This issue is not about radio usage. It's about Arbitron selling junk research at a 65% rate jump that is costing talented people their jobs. If you want to bellyache about how bad radio is, go join one of the dozen other postings where you can whine and do nothing.

There's a good reason why the PPM system has not been accredited by the Media Ratings Council in Philly, NYC, Chicago, LA or any other market since Houston a year ago. The math doesn't work. You can't have a few hundred people representing the daily radio usage of millions.

I applaud the leaders of the three companies that reacted and sent the stern letter to Arbitron. Although I'm no fan, I respect Bob Neil for keeping his word to hold Arbitron's feet to the fire. Now I suggest holding the checks until they respond accordingly.

This is good news. Finally the suits are getting angry.

 

[Francis]

I think it's great that Arbitron is finally catching up with the 21st Century and untimately when the bugs get worked out it will be a very good thing for the industry.
Can anybody tell me if the PPM is able to register if people are listening throught the commercials? It seems as though commercial elements and songs could be encoded in a way that will tell us exactly when the button is being pushed, and that could really shake things up.

 

[InTIMadate]

PPM will tell programmers when the sample is tuning out. Due to the small demo samples, it cannot accurately reflect what the audience is doing. I hope that specific information is not released to advertisers until a true representative sample size is determined, targeted, measured and accredited.

Currently, less that 10 individuals can represent an entire demo in NYC reflecting the habits of hundreds of thousands of listeners.

If you're truly interested, go to the Arbitron site and read their propaganda. Then visit the Houston and Philly sites on this board and read about the results. It's eye opening.

 

[davideduardo]

I think it's great that Arbitron is finally catching up with the 21st Century and untimately when the bugs get worked out it will be a very good thing for the industry.
Can anybody tell me if the PPM is able to register if people are listening throught the commercials? It seems as though commercial elements and songs could be encoded in a way that will tell us exactly when the button is being pushed, and that could really shake things up.

Only the station is encoded (separate encoding for web streams, HD-2, etc) but not content. The way to look at moment to moment data is to match exact times of events to the minute by minute data to see audience behaviour. This has been done by Arbitron together with, I think, Edison Research. You might look at arbitron.com as I think there is a report there. In any case, a bad song is far more damaging than commercials... I think it is that listeners expect commercials, but are let down by a bad song.