PPM ratings

[Mike Sheridan]

I am no expert when it comes to PPN ratings, all I have are questions.

It seems like there has been a search for a more accurate measurement of listening than the old diary method. PPM is supposed to be the answer. The results using PPM are different. This shakes everything up so the feeling is it can't be right so lets sue. Did anyone ever think the diary method was not accurate?

If all you wanted was the same results why not just keep the diary system? Isn't the goal accuracy?

David Eduardo feel free to chime in here. I know you have a lot of experience with Arbitron so your thoughts would be enlightening. Is PPM flawed just because the results are different?

 

[Groove1670]

I think the PPM's are correct. I have said for years a rhythmic AC would do well. Every office has WLYF on... and FLC is has adjusted their music like y-100 was in the 80's, and POW & EDR are the heritage leaders. Several companies will kick and scream because the status quo has been broken. The qestion is who will change or adjust to some of the shocking PPM changes of the future.

 

[Jay F]

WEDR did better with PPM than they did with the diary. That bucks the national trend of Urban doing poorly with PPM.

How come the Univision stations weren't listed?

 

[davideduardo]

WEDR did better with PPM than they did with the diary. That bucks the national trend of Urban doing poorly with PPM.

How come the Univision stations weren't listed?

They do not encode. They are not participating in a survey that even the MRC thinks is flawed enough to warrant witholding accreditation. This means that the shares are not shares of 100, but shares of the total less the univision share. Typicall, in 12% that was about 14 to 15 shares, so what we have is shares that are around 15% too high...

Of course, when the State of FLorida was filing an injunction to prevent the release of the data, Arbitron released the first Currency book two days early to beat the state to it... it makes you wonder.

As to WEDR, it had a .7 rating in Winter of this year, 12+. In the PPM, it had a .6 rating, so it was off by a little over 10% in the PPM. So it did not do better in the PPM. It did slightly worse.

 

[Mike Sheridan]

Wasn't there some weighting that favored the Urban stations because of poor diary return in the black neighborhoods?

What is the problem with PPM, is it a technical issue? I know there is some kind of encoding, is it flawed?

The whole ratings thing seems like such a crap shoot to me, depending on where the diaries are placed (now PPM units).

With all this going on isn't this a good time for someone to come in with a new ratings service?

 

[davideduardo]

Wasn't there some weighting that favored the Urban stations because of poor diary return in the black neighborhoods?

Weighting is used in any cell if there are too many or not enough diary returns for said returns to be in proportion to the standard for the whole project, which is, of course, to be in proportion for the whole market. If returns in a problem demo, like non-Hispanic white males 18-24, are lower than desired, each is increased in weight to be proportional. Same for ethnic groups, gender, geography, etc.

And some groups will return too many diaries, so each one has less value. There was never overweighting of any cell, just weighting up or down to achieve proportionality.

What is the problem with PPM, is it a technical issue? I know there is some kind of encoding, is it flawed?

The encoding is the most robust part of the process. Nobody has voiced a valid objection here.

The issues are related to sample design, cell phone sample, the proportionality of each subset, etc. There are also issues with recruiting and other areas. In this whole situation, I defer to the MRC which has not accredited most PPM markets... just Riverside and Houston. If they don't find the service accreditable after a year in many markets, there are pending issues.

With all this going on isn't this a good time for someone to come in with a new ratings service?

In truth, the biggest issue is cost. Some issues would be solved by a much larger sample... big samples tend to drown small problems in reality... but the cost of the PPM is already about 60% higher than the diary survey, and in these times, even that is too much.

 

[pjc1961]

WEDR did better with PPM than they did with the diary. That bucks the national trend of Urban doing poorly with PPM.

As to WEDR, it had a .7 rating in Winter of this year, 12+. In the PPM, it had a .6 rating, so it was off by a little over 10% in the PPM. So it did not do better in the PPM. It did slightly worse.

David, maybe you were looking at/thinking of the wrong station or possibly at WEDR's (99 Jamz) ratings/shares in a particular demo/different market, but just going by the overall numbers for Miami/Ft. Lauderdale/Hollywood, FL, the original poster seems to be correct. For Winter 2009 (Jan.-Feb.-Mar.), which was the last "diary" 3-month book, WEDR had a 4.5 share. In the PPM pre-currency report for April 2009, WEDR had a 6.6 share and in PPM pre-currency for May 2009, WEDR had a 7.1 share. For June 2009 PPM, WEDR had a 7.3 share. (Data courtesy allaccess.com.)

Link to radio-info.com's ratings grid page for the Miami metro is here:

http://radio-info.com/site/markets/grid/miami-ft-lauderdale-hollywood

 

[davideduardo]

WEDR did better with PPM than they did with the diary. That bucks the national trend of Urban doing poorly with PPM.

As to WEDR, it had a .7 rating in Winter of this year, 12+. In the PPM, it had a .6 rating, so it was off by a little over 10% in the PPM. So it did not do better in the PPM. It did slightly worse.

David, maybe you were looking at/thinking of the wrong station or possibly at WEDR's (99 Jamz) ratings/shares in a particular demo/different market, but just going by the overall numbers for Miami/Ft. Lauderdale/Hollywood, FL, the original poster seems to be correct. For Winter 2009 (Jan.-Feb.-Mar.), which was the last "diary" 3-month book, WEDR had a 4.5 share. In the PPM pre-currency report for April 2009, WEDR had a 6.6 share and in PPM pre-currency for May 2009, WEDR had a 7.1 share. For June 2009 PPM, WEDR had a 7.3 share. (Data courtesy allaccess.com.)

WEDR had a 0.7 rating 12+ in Winter 2009 in the diary based 12 week survey. In the June PPM survey in 12+ they had a 0.6 rating.

WEDR went down from the diary based methodology to the PPM based survey.

Share to share comparisons are totally invalid in Miami, as there are 4 major stations which accounted for around 15 12+ diary shares that are not included in the PPM reports because they did not encode during the survey period. Comparisons have to be made on rating or AQH persons, which are derived from the universe, not the agregate total radio listening.

 

[pjc1961]

WEDR had a 0.7 rating 12+ in Winter 2009 in the diary based 12 week survey. In the June PPM survey in 12+ they had a 0.6 rating.

WEDR went down from the diary based methodology to the PPM based survey.

Share to share comparisons are totally invalid in Miami, as there are 4 major stations which accounted for around 15 12+ diary shares that are not included in the PPM reports because they did not encode during the survey period. Comparisons have to be made on rating or AQH persons, which are derived from the universe, not the aggregate total radio listening.

I will definitely agree that the audience shares for the April/May/June PPM Reports are wildly flawed due to the 4 major stations not being encoded & being placed at the bottom of the list with 0.0's. You could probably take away approximately 2.0-3.0 shares from each of the top 5-10 stations to be fair & bring them back closer to pre-PPM "diary" levels (just an estimation on my part; there's a more scientific, statistical formula to arrive at a fair number). It would be fair to say that the PPM share numbers for the Miami metro are at the very least inaccurate; you could even say they could be "thrown out".

But for WEDR-FM-99.1 in particular, which has been a top-performing station in the Miami metro for years (I'm guessing that it usually places in the Top 5 or 6 overall), I find it hard to believe that it would get overall shares of 0.7 & 0.6, unless it's for West Palm Beach or another nearby (possibly embedded) market; or for a particular demo in one of these markets. For example, WEDR-FM-99.1 in Miami Spring '08=5.1; Summer '08=5.3; Fall '08=4.4; Winter '09=4.5 (courtesy arbitron.com; overall 12+ Diary results). In West Palm Beach/Boca Raton: Spring '08=2.4; Summer '08=2.9; Fall '08=2.1; Winter '09=2.2; Feb/Mar/Apr '09 Trend=2.6; Mar/Apr/May '09 Trend=2.1 (courtesy arbitron.com & radio-info.com; overall 12+ Diary results). For Miami May 2009 PPM Cume=1,055,700; Miami June 2009 PPM Cume: 1,053,800 (courtesy allaccess.com; overall 6+ PPM results). So even if we take away some shares & even allow for the usual "PPM ratings drop" that Urban Contemporary-based stations face, WEDR would still have "some" ratings/shares.

So I wonder if you're quoting figures from a report that is not available to the general public (like me, a radio fan!), or possibly looking at something else. I only say this because the numbers don't seem to agree with the general, overall performance of WEDR over the years.

 

[davideduardo]

I will definitely agree that the audience shares for the April/May/June PPM Reports are wildly flawed due to the 4 major stations not being encoded

The shares for WAQI, WAMR, WRTO and WQBA in 12+ were "around" 15. That means that the shares in the new PPM data are percentages of 85 total shares, not percentages of 100 total shares. So every number, at least in 12+, is too high by the resultant lack of a 100 point base.

But for WEDR-FM-99.1 in particular, which has been a top-performing station in the Miami metro for years (I'm guessing that it usually places in the Top 5 or 6 overall), I find it hard to believe that it would get overall shares of 0.7 & 0.6,

I did not use shares because I stated that using shares was absurd for a comparison. That is why I compared ratings.

unless it's for West Palm Beach or another nearby (possibly embedded) market;

I don't think there is an embedded market in the entire state... there sure is not one in the coverage area of WEDR.

. So even if we take away some shares & even allow for the usual "PPM ratings drop" that Urban Contemporary-based stations face, WEDR would still have "some" ratings/shares.

"Rating" is not share. Rating is a percent of the universe, while share is the percent of only those listening to radio.

 

[pjc1961]

David, thank you & I now understand. You're using a figure termed as "rating" which is (apparently) not available to the general public (unless you'd like to include a link to such a report, but I'm sure it's proprietary information only available to Arbitron subscribers, thus an understandable reluctance to post it anywhere).

This makes sense when reminded of Nielsen TV rankings which include "rating" (universe of sets) & "share" (% of sets in use), both of which are made readily available if one searches online, checks trade magazines or looks in the entertainment section of some newspapers once a week for the "Top TV Shows Of The Week" that print both figures.

Please forgive us "radio fans" who sometimes use colloquial terms such as "ratings" loosely; some articles written about Arbitron reports might be headlined: "New Radio Ratings Put Station ---- at #1", when in actuality they're formally known as "Arbitron Share Ranking Reports" without actual "rating" info, just shortened to something easily understood by the general public. And for some of us, we get used to using the term "rating" rather than "share", so I understand your frustration when we're not technically correct.

Plus, forgive us for not having access to all the data (such as "rating") that's available out there - I wish I did, then I could make more informed, accurate posts based on fact. Having said that, I guess that's why you're here, trying to help us neophytes to better understand the radio business. For that I thank you.

 

[davideduardo]

David, thank you & I now understand. You're using a figure termed as "rating" which is (apparently) not available to the general public (unless you'd like to include a link to such a report, but I'm sure it's proprietary information only available to Arbitron subscribers, thus an understandable reluctance to post it anywhere).

"Rating" is an audience expressed as a percentage of the universe. Rating (or AQH Persons) is what advertisers use, not share.

What is released and has been released to the public for decades is share. But in this specific case, where some stations have decided not to participate in an encoded survey method, share is an inaccurate and dangerous metric to use.

In the case of WEDR, using share without compensating for the "missing" approximately 15 shares, gives an increase. But looking at reality, based on the universe, WEDR is slightly down.

Plus, forgive us for not having access to all the data (such as "rating") that's available out there - I wish I did, then I could make more informed, accurate posts based on fact. Having said that, I guess that's why you're here, trying to help us neophytes to better understand the radio business. For that I thank you.

Yep. That's why I tried to understand what you were referring to before being able to clarify that share, in Miami, should not be considered real or precise for the moment.