KVI "GREATEST HITS" FORMAT NOT A HIT WITH LISTENERS

[davideduardo]

Can you get a decent idea of what white women between 29-44 years old with high income, heavy listening and a preference for contemporary hits listen to? Or an asian male 18-25 with moderate income, moderate listening who likes rap? Or a hispanic male 44-54 with low income, low listening who likes country? I am less sure about getting good estimates of those level of groups from the outside. Of course this is other people's field of expertise but it does seem challenging.

First, Asians are not a stratification variable in any market. Black, Hispanic, Other. That's it. And music taste is not, either, a stratification variable. You sample a market based on facts, not taste... the survey then returns as results an indication of tastes.

Facts: age, gender, ethnicity, income levels, family size, Language preference among Hispanics, geography by county and sampling unit. The purpose of using a panel with the PPM is to be able to mirror the demographics of a market, just as a blood sample mirrors the entirity of the blood in your body.

 

[davideduardo]

Okay, I'll stick my neck out here. My MBA was marketing with a minor in statistics - lots and lots and lots of statistics! It's been a few years - so go easy on me. If the question is how small a sample size can be to accurately represent a population, the answer is 'pretty damn small' - surprising small for that matter.

More important than size (in this case size doesn't matter much) is WHO the sample size is. There is random sampling (whoever is walking by), quote sampling (try and get a representative population) and variations. Who and where the meters are is much more significant than how many there are, in fact, having too many out there can actually DECREASE the accuracy of the reporting.

Those are two great points, and well stated, too.

The test of a sample is replication, or the ability to do the same sample size and get the same results within the acceptable / tolerable margin of error.

I've seen such studies done for music testing for stations that cume several million, and the result is that a well recruited sample of 70 to 80 persons will replicate with only a few percent variation... as you also said, it's not the size that matters but the sample design and implementation.

Since the PPM system is a panel, and panels are not subject to the wobbles in sample that a weekly diary system was, the sample can pretty accurately match the market (I'm avoiding using the term "universe" or someone will think I am talking about alien radio stations...)

 

[davideduardo]

I think my biggest ??? when it comes to the PPM technology is how does that little beeper sized meter determine if you are an active or passive listener?

It does not matter. Ratings are intended to be used to sell advertising. Advertisers want impressions.

I was on a joint radio / advertiser committee for Arbitron last year that analyzed ways of measuring engagement or commitment to a station; the fact that there is no station by station measure of this quality tells you that it is not particularly needed.

 

[davideduardo]

uh huh, and another big question is how them small scant percentage was picked to ppm. and what is the demographic makeup of them?

The panel is supposed to be fully proportional to the market characteristics on a whole bunch of measurements. If the market is 10% Hispanic, 10% of meters will be in Hispanic households. If 22% are 18-34, then 22% of the meters are with 18 to 34's. If 40% of the population is in one county, so will 40% of the meters...

 

[davideduardo]

Bill, a couple of weeks ago I believe I read on here that there are approximately 750-800 PPMs for Seattle-Tacoma. Based on the Arbitron web site there are 3,453,400 people 12+ in the Seattle-Tacoma Market. Based on my possibly flawed math, that's one PPM per around 4500 people. This translates to roughly 0.022% of the population (2.2 people out of 10,000). Is this a significant sample? How does this stack up to the number of diaries under the old Arbitron method?

The diary used a weekly poll, sending out lots of diaries and getting back much fewer... often not a balanced sample. The average diary represented around 1,100 persons, but they only participated for a week. So if a market the size of Seattle, Phoenix or San Diego had around 3,ooo diaries that was only about 280 a week.

In the PPM, we have around 1000 in tab daily meters... for 4 times the weekly sample of the diary. And a panel can be built with balance on all demographics, while a diary that may not be returned is random...

The real limit is cost.. the PPM increased the already hefty ratings expense by sixty percent... adding sample in a significant manner would make it unaffordable.

 

[davideduardo]

okay, so learn me up on this. who instigated the ppm rating system? who lobbyed for it, and how long did it take to make the change? was there any opposition at the time? and if so , who?

Ad agencies and time buyers pushed for a better, faster delivery system.

Arbitron began working on the PPM around 1995. The first PPM I saw that year was about the size of a brick.

The system was tested in England, then Philly for two years, followed by two years in Houston, at which point it went into market by market rollout. So there were 4 years of pre-currency tests, with a lot of station participation... I was on a committee for the Philadelphia PPM test, and we met frequently in NY or Columbia, and were given detailed updates. All in market stations got the data, too, for internal use.

We now have all 48 of the intended PPM markets running, some are around 4 years into currency measurement.

 

[semoochie]

At some point, will they increase the amount of markets participating? Is it just a matter of time? Do you have a ballpark guess when it might be and how many? Will it ever expand to include all current Arbitron markets? I would think that at some point, advertisers would balk at ratings in the non-PPM markets.

 

[davideduardo]

At some point, will they increase the amount of markets participating? Is it just a matter of time? Do you have a ballpark guess when it might be and how many? Will it ever expand to include all current Arbitron markets? I would think that at some point, advertisers would balk at ratings in the non-PPM markets.

For the moment I forsee no further PPM expansion.

First, Arbitron needs to get MRC accreditation in all 48 markets; at present just a couple are accredited.

Then there is the issue of cost. Outside the top 50 markets, and even between 50 and 100, there are quite a few markets that don't even have continuous diary measurement... due to cost. The PPM is much more expensive and likely the payback is much smaller because the smaller markets have much less agency / transactional business. There are individual stations in some of the top 10 markets that bill more than all stations combined in the markets in the 80-100 range.

 

[Bongwater]

The way I see it is a DECENT sample of a market like the Puget Sound would consist of MANY, MANY times of PPM holders than present. Were the paper diaries any BETTER? WHO KNOWS?!

And YES. I AM questioning every damn thing about it.

Would KXRX have survived with PPM?

Would KJET still be blasting on 1590 kHz?

Who knows?

All I know is for every diary cheat, there is a friendly dog willing to carry your PPM. And will as faithfully as in a scene only Francis Barraud could paint, listen to your radio.

(The overall digestive problems your dog may have upon consuming such a tasty electronic delicacy are those you should consult with a friendly professional veterinarian.....)

 

[dunno]

David, (by way of Guru)

Thanks for giving the Northwest outpost a PPM seminar. I'm able
to skip the meeting gobbledy goup and and get the meat that matters!! I've never read PPM explained so well.

Care to shed light on how you make you living?

 

[scott salvatori]

i think you nailed it. what normal business/proffesional person, wants to walk around with a geek box hooked up to them.

What normal person posts on radio message boards?

david, what normal person has almost 20,000 posts at radio-info.com? at that rate mr bongwater will be age 90 or so before he catches up from his scant 2778 posts, if you quit posting. ha ha....

at any rate, thats some good info, thanks for the learn up! . you should change your handle to "ppm guru". for some reason ppm still means, parts per million to me.

 

[davideduardo]

Care to shed light on how you make you living?

Click the link below, and hit the "My Biography" menu item and pick "Brief Resume" from the drop-down.

 

[davideduardo]

The way I see it is a DECENT sample of a market like the Puget Sound would consist of MANY, MANY times of PPM holders than present. Were the paper diaries any BETTER? WHO KNOWS?!

And YES. I AM questioning every damn thing about it.

And your background in research and statistics and radio will be forthcoming, correct?

Would KXRX have survived with PPM?

Would KJET still be blasting on 1590 kHz?

The two major differences betwen the diary and the meter can be stated in two words: "rounding" and "interruptions."

First, in the diary a person might really have listened from 2:10 to 2:50, but when they later filled in a diary they would likely have put 2:00 to 3:00. So,in the station got 4 quarter hours of credit for only 40 minutes of listening. The PPM immediately cuts out that rounding, which almost always causes lower TSL.

Second, a person may right that they listend to the same station from the time they got up to the time they got to work. In reality, they hit "snooze" twice, cutting off 20 minutes. They were in the shower... 10 minutes. They took the trash out... anotheer 10 minutes. And they took the kids to the school bus stop, 15 minutes. And when they got in the car, they put on the station with more frequent traffic for a few minutes... at the end, they really only listened to that station that got 3 hours in the diary for 47 minutes.

So, low cume stations with huge TSL tend to get whacked by the reality of the PPM.

The other thing the PPM does is register listening to stations that are a listener's second or third choice. Those may not have been prominent enough to be remembered at the time the diary was filled in but the PPM registers them. And the PPM picks up on stations a person hears that someone else turned on... in the home, in the car, at work, etc.

As to sample, I already posted that the test of a sample is a replication study. When you find a sample size where identically constructed samples all yield the same results, you have the right sample size; when you find that adding sample does not change the results you have too much sample.

Houston, with about 5 million persons 12+, has a daily in-tab of around 1,100 meters. And it has been accredited by the MRC, a group which consists of some of the advertising and media industry's best researchers. Remember, ratings are of no use to a station if the advertisers who use them do not trust them.

 

[davideduardo]

Second, a person may right that they listend to the same station...

And then again, they might wrong that they listened... ;D

That brain f--t should have been "write" of course.

 

[AQH]

That brain f--t should have been "write" of course.

Don't sweat the responses from some here, I'm sure you've gotten a good picture already about their stances.

Enjoy your weekend.

 

[peetah]

Yes, thanks David for the primer on PPM.
You mentioned Houston being accredited by the MRC.
Do you know how many markets are now accredited?
The last I heard it was Houston, Minneapolis and somewhere in North Carolina?
And if only a handful of markets are indeed accredited, what's the hold-up for other areas?
And should this be of concern to broadcastes?

Thanks.

 

[davideduardo]

Yes, thanks David for the primer on PPM.
You mentioned Houston being accredited by the MRC.
Do you know how many markets are now accredited?
The last I heard it was Houston, Minneapolis and somewhere in North Carolina?
And if only a handful of markets are indeed accredited, what's the hold-up for other areas?
And should this be of concern to broadcastes?

Thanks.

Minneapolis, Riverside / San Bernardino and Houston are accredited.

The rest are constantly being resubmitted. The issues are not totally public, as the MRC does not deal with the public any more than a company's outside auditor does. However, from what we see from Arbitron, the issues involve making the panel as representative as possible of each market. That is why use of some address based recruiting is being employed for ethnic recruiting, and why a 10% sample increase was implemented.

i'd say that the issue is not the system... that is, the metered panel... but the way the sample is recruited and, possibly, maintained. Since Arbitron never did panel based ratings, and thus this is new to them, that is understandable. Arbitron has a lot of incentives to be accredited as do all ratings and related electronic media measurement companies.

The concerns for broadcasters are that the ratings be accurate and that agencies trust them for buys. That's always been the case, though.

 

[dunno]

Quite the resume!

BTW

Reality bites!!!

 

[Jay Peninger]

Great information David and put out there in terms it can be easily digested... Thank you for taking the time to share your knowledge of PPM with everyone here...

I for one appreciate any opportunity to learn something new....

 

[semoochie]

That's an amazing site you have there! If I ever have another lifetime, I'd love to go through it.