Holiday ratings

[radiopunk99]

Hey guys,
I was able to get my hands on the ranker for 2 demos (P25-54 and W25-54) from a friend who has access. But I would be really curious to see P18-34 if anyone else has it!

P25-54 (top 10)

KSNE (1)
KOMP (2)
KKLZ (3)
KISF (4)
KLUC (5)
KCYE/KWID/KMXB (6t)
KXPT (9)
KVEG (10)


W25-54 (top 10)

KSNE (1)
KCYE (2)
KMXB (3)
KOMP (4)
KKLZ/KLUC (5t)
KRGT (7)
KISF (8)
KWNR (9)
KXPT (10)


Notes from me:

-It’s no surprise Sunny is #1 with Christmas music and it’s always interesting to see who they pull from. More on that later.

-Another big book for the Country format with Coyote coming out on top in a big way. Coyote had their best book ever, especially with women. KWNR’s share was still pretty decent but they’ve gotta be worried about Coyote, losing by significant shares 2 books in a row.

-The music stations that took the biggest beating this month were MIX, KFRH, KXTE and KCEP (in that order). You have to wonder how much of that went to Christmas music. Mix alone was down over 20%.

-The music stations with the biggest gains were Sunny, KWID, KOMP, and KKLZ (in that order). Nice bumps for KOMP and KKLZ, despite Christmas music.

-We finally have a Spanish station in the top 5. I find it odd that this doesn’t happen more often considering the high population.

 

[davideduardo]

-We finally have a Spanish station in the top 5. I find it odd that this doesn’t happen more often considering the high population.

One of the reasons why the Spanish language stations have shown as they have is that Las Vegas has been the only major Hispanic market in the US that did not employ language proportionality to insure a balanced survey. The reason is that Nielsen did not provide that data (which Arbitron buys) but as of the January book, there will be language (Spanish dominant and English dominant) proportionality and it is very possible that this will benefit the Spanish language stations.

 

[Pinegroves]

Good to see KOMP and KKLZ making moves even against the Christmas music of Sunny. That's about all Clear Channel has going for them, they've even changed the name of their company to protect the innocent. Looks like Coyote is pulling ahead by long strides against KWNR, mainly due to cost cutting decisions of CC to eliminate long time successful radio people, both on and off the air. Coyote has live local personalities all day as opposed to out of town jocks voice tracking on KWNR. You think maybe a station with a strong signal will go against Sunny and grab some of those Holiday numbers next Xmas season. Whaddye' Think?

 

[holyradio]

David, you are 100% correct about the new methodology that Arbitron has just implemented in Las Vegas. However, I don't buy that it's "One of the reasons why the Spanish language stations have shown as they have". I think that's just an excuse that the Spanish companies have created to pass the blame. And this new methodology that Arbitron has created is their way of appeasing the Spanish Language radio companies. In fact, if you look back at the past PPM numbers for Las Vegas, Spanish stations have historically done very well. It is only lately that they have been out of the top tier.

Here are just the example of ranks 1 and 2:

Mar 2011, KWID was #1 (P6+)
Feb 2011, KWID was #1 (P6+)
Jan 2011, KWID was #2 (P6+), off by only .2 to KKLZ
Nov 2010, KWID was #2 (P6+), a share behind KKLZ
Sept 2010, KWID was #1 (P6+)
Aug 2010, KWID was #2 (P6+), off by a half share to KKLZ
July 2010, KWID was #1 (P6+)
June 2010, KWID was #1 (P6+)
May 2010, KWID was #1 (P6+)
April 2010, KWID was #2 (P6+), off by .1 to KKLZ
Mar 2010, KWID was #1 (P6+)
Feb 2010, KWID was #1 (P6+)
Jan 2010, KWID was #1 (P6+)

Again, this is only when they showed 1 or 2, but there’s many examples of when they were #3. And in almost every case, they also placed extremely well P25-54, if not #1. In 26 PPM currency books, KWID was either #1 or #2 thirteen times!!! I'm not very good at math, but isn't that exactly HALF??? In a 26 book rolling average, KWID is #3 behind Sunny and KKLZ.

I would say that’s pretty good! Wouldn’t you? Maybe these Spanish radio companies should spend more time fixing their product instead of blaming Arbitron.

 

[davideduardo]

David, you are 100% correct about the new methodology that Arbitron has just implemented in Las Vegas. However, I don't buy that it's "One of the reasons why the Spanish language stations have shown as they have". I think that's just an excuse that the Spanish companies have created to pass the blame. And this new methodology that Arbitron has created is their way of appeasing the Spanish Language radio companies. In fact, if you look back at the past PPM numbers for Las Vegas, Spanish stations have historically done very well. It is only lately that they have been out of the top tier.

This is not a "new methodology." Arbitron started implementing language proportionality about a decade ago because it was seen from the week to week and trend to trend wobbles in the diary that the enormous instability in the Spanish language stations from New York to Miami to LA and Houston was caused by the total lack of proportionality in the sample subsets (weeks and months) for Spanish Dominant and English Dominant Hispanics.

We had a case in Los Angeles in around 2000 where one month the total share of all Spanish language stations combined changed by nearly 50%. The MRC and advertisers were as appalled as stations were.

Nielsen had been using language as a stratification variable for some time, and its Hispanic estimates had a far greater degree of stability than the radio estimates. The reason, of course, was the total lack of control of the sampling of Spanish Dominants and English Dominants despite this being a characteristic that is quite definable and does not change rapidly or spontaneously... unlike the radio ratings numbers.

Finally, Arbitron arranged with Nielsen to buy the Nielsen data, which is done by a separate enumeration process which has been MRC approved for decades now.

Unfortunately, Nielsen did not provide data for Las Vegas until this year. All the other major Hispanic markets have had language weighting going back to around 2002.

Proportionality on any stratification variable does not "appease" anyone, but it does provide a more accurate survey. And that is the goal of the MRC, which mostly represents advertisers, in seeking a correct and consistent methodology.

On the nit-picking side of things, "Spanish radio companies" are all in Spain, as are "Spanish radio stations."

 

[Jay F]

It's surprising (and impressive) to see KOMP #4 in women 25-54. They are beating many of the stations that actually target adult women.

So if the new methodology benefits Spanish language stations who is it going to hurt? Stations with large English dominant Hispanic audiences?

 

[holyradio]

David, you are 100% correct about the new methodology that Arbitron has just implemented in Las Vegas. However, I don't buy that it's "One of the reasons why the Spanish language stations have shown as they have". I think that's just an excuse that the Spanish companies have created to pass the blame. And this new methodology that Arbitron has created is their way of appeasing the Spanish Language radio companies. In fact, if you look back at the past PPM numbers for Las Vegas, Spanish stations have historically done very well. It is only lately that they have been out of the top tier.

This is not a "new methodology." Arbitron started implementing language proportionality about a decade ago because it was seen from the week to week and trend to trend wobbles in the diary that the enormous instability in the Spanish language stations from New York to Miami to LA and Houston was caused by the total lack of proportionality in the sample subsets (weeks and months) for Spanish Dominant and English Dominant Hispanics.

We had a case in Los Angeles in around 2000 where one month the total share of all Spanish language stations combined changed by nearly 50%. The MRC and advertisers were as appalled as stations were.

Nielsen had been using language as a stratification variable for some time, and its Hispanic estimates had a far greater degree of stability than the radio estimates. The reason, of course, was the total lack of control of the sampling of Spanish Dominants and English Dominants despite this being a characteristic that is quite definable and does not change rapidly or spontaneously... unlike the radio ratings numbers.

Finally, Arbitron arranged with Nielsen to buy the Nielsen data, which is done by a separate enumeration process which has been MRC approved for decades now.

Unfortunately, Nielsen did not provide data for Las Vegas until this year. All the other major Hispanic markets have had language weighting going back to around 2002.

Proportionality on any stratification variable does not "appease" anyone, but it does provide a more accurate survey. And that is the goal of the MRC, which mostly represents advertisers, in seeking a correct and consistent methodology.

On the nit-picking side of things, "Spanish radio companies" are all in Spain, as are "Spanish radio stations."

Hey David,


Very well thought out and well researched. However, you make it sound like you were in the room when Arbitron decided to implement language proportionality into their methodology. I happen to know that Spanish LANGUAGE companies (is that appropriate enough for you?) have been fighting with Arbitron for more than a decade and asking for programs like this to increase their shares. In fact, I have personally witnessed a high level Spanish Language Company executive berate an Arbitron executive while claiming Arbitron’s system was seriously flawed and unfair to Spanish Language stations (does that term work for you?). Remember a major Spanish Language Company Univision did not renew numerous specific market contracts when PPM was first released, literally fusing to encode their signals in many markets around the Country (including Las Vegas). It was only until Arbitron agreed to make changes that would benefit Spanish language radio that they resigned.

http://news.****************/cgi-bin/rol.exe/headline_id=b11762


Arbitron is a publicly traded company and Spanish Language Radio Companies are big clients. Do you really think Arbitron just decided to go buy this data from Nielsen because “it was the right thing to do”? No, they did it to “appease” and make more money! Come on man, this is a Capitalistic Society not Utopian.

 

[davideduardo]

Very well thought out and well researched.

I did not do any "research" on this. I simply lived through it and was a significant player in the process.

However, you make it sound like you were in the room when Arbitron decided to implement language proportionality into their methodology.

I was. Ask Dr. Ed at Arbitron.

I happen to know that Spanish LANGUAGE companies (is that appropriate enough for you?) have been fighting with Arbitron for more than a decade and asking for programs like this to increase their shares.

Actually, this question was asked and answered a dozen years ago. Around "the turn of the Century" and after increasing wobble in book to book and month to month results, all the Spanish-language station owners of significance, from HBC and SBS to Entravision and Z Spanish and Radio Unica demanded language proportionality, just as had been done in TV some considerable time before.

After going back and forth on how to do it, Arbitron finally presented a plan that would add language as a stratification variable just like age groups, gender, ethnicity, income, etc., already were. The problem in implementation was rather unusual... the very old database structure inherited from the Arbitron minicomputer Control Data days had no more unused fields in a fixed-field structure, so they had to do a major database redesign and conversion to do this.

At the same time, they arranged with Nielsen to buy the language preference data that Nielsen obtained from its own enumeration surveys in each major Hispanic market (but only on 12+ and later, for PPM, 6+... a major flaw that exists to this day).

Language preference does not increase shares. It had no average effect of increasing shares when implemented over a decade ago. But it did to much of what was desired, which was to eliminate the enormous wobbles in the numbers... such as one LA book about 13 years ago where the total shares in 18-34 for Spanish language stations went from 24 to 40 to 30 in three books. Today, the shares are within a percent +/- from book to book in that market.

Remember a major Spanish Language Company Univision did not renew numerous specific market contracts when PPM was first released, literally fusing to encode their signals in many markets around the Country (including Las Vegas). It was only until Arbitron agreed to make changes that would benefit Spanish language radio that they resigned.

The PPM problem was not a language preference issue. It was a sample design issue. In fact, it was a subset of the same issues that prevented the MRC from accrediting the PPM initially and which has kept 80% of the PPM markets from being accredited still. Here, Arbitron responded by slowly implementing address based recruiting (done in the 2004-2006 Houston PPM tests I was a participant in and, thus allowing Houston alone to be initially MRC accredited) in all markets, the addition of Cell Phone Only households to the sample, the introduction of personal recruiting and the creation of GeoZones within each County and Sampling Unit.

In other words, the issue was about the distribution and proportionality of the sample in a survey with about one-third the sample size of a traditional three-month diary survey and where each meter had to be perfectly placed; a panel survey assumes a perfectly proportional pre-designed panel, but that was not initially the case for the PPM due to recruitment problems of major proportion the solution to which still does not satisfy the MRC in most places.

Arbitron is a publicly traded company and Spanish Language Radio Companies are big clients. Do you really think Arbitron just decided to go buy this data from Nielsen because “it was the right thing to do”?

They bought it from Nielsen because it was cheaper to do that than to do their own enumeration... and go through the MRC process for this procedure. The did the overall process because it was necessary in order to achieve proportionality in the survey and because the MRC had focused its attention on the issue due to the rapidly growing Hispanic population in the US and there was a very real possibility of losing the diary MRC accreditation in the larger Hispanic markets.

No, they did it to “appease” and make more money! Come on man, this is a Capitalistic Society not Utopian.

In this case, they did it to prevent loss of accreditation and to avoid considerable loss of money should the MRC have withdrawn its accreditation of markets like NY, Chicago, LA, Dallas, Houston and San Francisco... five of the top 10 markets... where the huge increases in Hispanic population from 1990 to 2000 made it so obvious that new techniques were required to retain proportionality.

 

[holyradio]

Very well thought out and well researched.

I did not do any "research" on this. I simply lived through it and was a significant player in the process.

However, you make it sound like you were in the room when Arbitron decided to implement language proportionality into their methodology.

I was. Ask Dr. Ed at Arbitron.

I happen to know that Spanish LANGUAGE companies (is that appropriate enough for you?) have been fighting with Arbitron for more than a decade and asking for programs like this to increase their shares.

Actually, this question was asked and answered a dozen years ago. Around "the turn of the Century" and after increasing wobble in book to book and month to month results, all the Spanish-language station owners of significance, from HBC and SBS to Entravision and Z Spanish and Radio Unica demanded language proportionality, just as had been done in TV some considerable time before.

After going back and forth on how to do it, Arbitron finally presented a plan that would add language as a stratification variable just like age groups, gender, ethnicity, income, etc., already were. The problem in implementation was rather unusual... the very old database structure inherited from the Arbitron minicomputer Control Data days had no more unused fields in a fixed-field structure, so they had to do a major database redesign and conversion to do this.

At the same time, they arranged with Nielsen to buy the language preference data that Nielsen obtained from its own enumeration surveys in each major Hispanic market (but only on 12+ and later, for PPM, 6+... a major flaw that exists to this day).

Language preference does not increase shares. It had no average effect of increasing shares when implemented over a decade ago. But it did to much of what was desired, which was to eliminate the enormous wobbles in the numbers... such as one LA book about 13 years ago where the total shares in 18-34 for Spanish language stations went from 24 to 40 to 30 in three books. Today, the shares are within a percent +/- from book to book in that market.

Remember a major Spanish Language Company Univision did not renew numerous specific market contracts when PPM was first released, literally fusing to encode their signals in many markets around the Country (including Las Vegas). It was only until Arbitron agreed to make changes that would benefit Spanish language radio that they resigned.

The PPM problem was not a language preference issue. It was a sample design issue. In fact, it was a subset of the same issues that prevented the MRC from accrediting the PPM initially and which has kept 80% of the PPM markets from being accredited still. Here, Arbitron responded by slowly implementing address based recruiting (done in the 2004-2006 Houston PPM tests I was a participant in and, thus allowing Houston alone to be initially MRC accredited) in all markets, the addition of Cell Phone Only households to the sample, the introduction of personal recruiting and the creation of GeoZones within each County and Sampling Unit.

In other words, the issue was about the distribution and proportionality of the sample in a survey with about one-third the sample size of a traditional three-month diary survey and where each meter had to be perfectly placed; a panel survey assumes a perfectly proportional pre-designed panel, but that was not initially the case for the PPM due to recruitment problems of major proportion the solution to which still does not satisfy the MRC in most places.

Arbitron is a publicly traded company and Spanish Language Radio Companies are big clients. Do you really think Arbitron just decided to go buy this data from Nielsen because “it was the right thing to do”?

They bought it from Nielsen because it was cheaper to do that than to do their own enumeration... and go through the MRC process for this procedure. The did the overall process because it was necessary in order to achieve proportionality in the survey and because the MRC had focused its attention on the issue due to the rapidly growing Hispanic population in the US and there was a very real possibility of losing the diary MRC accreditation in the larger Hispanic markets.

No, they did it to “appease” and make more money! Come on man, this is a Capitalistic Society not Utopian.

In this case, they did it to prevent loss of accreditation and to avoid considerable loss of money should the MRC have withdrawn its accreditation of markets like NY, Chicago, LA, Dallas, Houston and San Francisco... five of the top 10 markets... where the huge increases in Hispanic population from 1990 to 2000 made it so obvious that new techniques were required to retain proportionality.

Touché David,

Although, I wasn’t suggesting that the “PPM Issue” was language preference. I was only suggesting that these companies have been pressuring Arbitron to make changes that benefit them. I’m sure you’ll agree that is the case, even with this issue. Anyway, you obviously know more about what’s going on here than I do. So… I’ll shut up now.

 

[davideduardo]

Although, I wasn’t suggesting that the “PPM Issue” was language preference. I was only suggesting that these companies have been pressuring Arbitron to make changes that benefit them. I’m sure you’ll agree that is the case, even with this issue. Anyway, you obviously know more about what’s going on here than I do. So… I’ll shut up now.

You have a point. In this case, everyone involved felt that there was non-equal treatment, via methodology, of the listeners of Spanish language stations because there was no control over the sample for language preference. Advertisers, seeing the huge wobbles, did not know whether to believe the high or the low side and so often they did not buy or lowballed stations on costs. Stations in all other segments did not go through this, so there was an element of fairness involved.

Language weighting achieved the desired stability... until Arbitron messed up again, rushing PPM to the market before the sample was done right. Of course, Arbitron had never done a panel-based survey before, and in fairness I don't think they ever realized how difficult ongoing recruitment would be, and they had no idea that keeping panelists in compliance was such a task!

 

[radiopunk99]

Well holyradio and DavidEduardo,

I enjoyed reading your spirited debate and learned a lot. Here are the 1st results of your discussion. According to a PD friend of mine who has access to the weekly data, after many books of no Spanish Language station in the top 3, there are now 2 (KISF and KWID in Jan 2012 Week 1)! The last time there was even one Spanish outlet in the top 3 was May 2011 when KWID was tied for 3rd with KKLZ.
It looks like the Spanish outlets are going to benefit from this Arbitron change.

I wonder how this will affect the English speaking outlets? What does everyone think about this? Is this fair? Does this mean that all of the previous Arbitron PPM data was WRONG? Or is this new data wrong?