Arbitron Market Boundaries

[tanner]

Here is a new subject to toss around instead of WIXE 1060. How does arbitron draw the boundaries for the markets. Lets compare for instance Orlando and Charlotte. Orlando is a Top 20 TV market yet it is only #34 radio. It is a larger TV market and is more populated than Charlotte. Charlotte has counties that are not part of their metro which pushes them to #24. Lake is part of the core Orlando metro and is not included. All of southern Volusia county is attached to Orlando not included, it is its own market with Daytona, dominated by Orlando. Same for Brevard. Polk splits between Tampa and Orlando pretty even. The way they have Charlotte drawn if you drew Orlando the same way that would put Orlando in the top 20. What a windfall for rates that would create for Orlando stations. If logic says Daytona and Brevard remain micro markets at least Lake, Sumter, and Citrus should be included in Orlando as it dominates those areas. they are not included anywhere. Those 3 unaccounted for counties would add approximately 400,000 people which would move it into the Top 25.

JM Tillery and David Eduardo any thoughts?

 

[davideduardo]

Here is a new subject to toss around instead of WIXE 1060. How does arbitron draw the boundaries for the markets. Lets compare for instance Orlando and Charlotte. Orlando is a Top 20 TV market yet it is only #34 radio. It is a larger TV market and is more populated than Charlotte. Charlotte has counties that are not part of their metro which pushes them to #24. Lake is part of the core Orlando metro and is not included. All of southern Volusia county is attached to Orlando not included, it is its own market with Daytona, dominated by Orlando. Same for Brevard. Polk splits between Tampa and Orlando pretty even. The way they have Charlotte drawn if you drew Orlando the same way that would put Orlando in the top 20. What a windfall for rates that would create for Orlando stations. If logic says Daytona and Brevard remain micro markets at least Lake, Sumter, and Citrus should be included in Orlando as it dominates those areas. they are not included anywhere. Those 3 unaccounted for counties would add approximately 400,000 people which would move it into the Top 25.

JM Tillery and David Eduardo any thoughts?

Arbitron has a rule that determines the counties (and even, in some cases, county splits) that make up a metro. There is a formula that uses the amount of commuting within the market and the specific county and the amount of listening to stations that are home to the metro which qualifies and disqualifies counties.

Quite a few counties are added or removed from metros each year. The changes are based on the listening and commuting standard.

TV markets are in part based on the broadcast TV stations and market, directly or via cable, that determined most viewing... and go back before satellite, etc.

In any case, in radio adding counties where there is no significant listening to the existing metro's home stations would not benefit the market... in fact, there would be a decline in total shares and rating because of it.

Another factor is that significantly expanding a market so that it "swallows" another one requires approval of the subscribers. In 1981, the managers of the stations in Dade and Broward counties voted on combining the two counties. The combining barely passed, as some of us voted for and others against the change; most AM managers voted against, as their rank and share would diminish.

Similarly, about 15 years or so ago it was proposed that the San Bernardino Riverside market be consolidated with LA. This was not approved. Many LA stations had bad or no coverage in that area, and no station from Riverside San Berdoo covers more than a slice of LA... the result would be lower shares and ratings. In fact, there is still a slice of San Bernardino County from Pomona to the East that is in neither book because neither marked wanted the diluted shares.

Keep in mind that to the OMB and Census, MSA means metropolitan statistical area, while in radio it means metro survey area.

 

[jmtillery]

Here is a new subject to toss around instead of WIXE 1060.

The WIXC thread had taken on a life of its own and, as far as I am concerned, has become quite interesting reading.

How does arbitron draw the boundaries for the markets.

Dave beat me in giving you a response, and he has also given you a much more detailed explanation regarding Arbitron than I had drafted. However, I'll give you what I have relating to radio and expound a bit regarding television.

Arbitron draws the boundary lines based mainly on the population in any given MSA (Metropolitan Statistical Area/Metro Survey Area) and TSA (Total Survey Area). The number of stations serving a MSA/TSA are then included as part of the ratings process.

With regards to television, keep in mind there are far fewer Television stations than there are radio stations. And for this reason, a TV market will almost always be larger than a radio market. As mentioned above, Arbitron measures radio listening in a rated MSA/TSA while Television viewership is measured in a rated Area of Dominant Influence (ADI). In states such as Florida, a television ADI will be much smaller than an ADI in most Midwestern states. In states such as Wyoming and Utah, an ADI will consist of the entire state where – using Wyoming as an example – Casper is the main television center. The Casper stations may have satellite TV stations in Laramie and other parts of the state re-broadcasting the main Casper station. However, unless it has changed within the past few years, the Casper ADI is the entire state.

Regarding Orlando, Lake county is considered a part of the Orlando ADI while Sumter and Citrus counties are part of the Tampa Bay ADI. At the same time Citrus county is part of the Gainesville – Ocala radio TSA, but not part of the MSA. Keep in mind that ADI and MSA are not the same thing. MSA is a direct measurement of the Metro while ADI is measuring a much larger area outside the metro. An easier way to remember is MSA/TSA = Radio – ADI = Television.

Lastly, using your Daytona Beach and Lakeland - Winter Haven example, at one time, and it should still be, Daytona Beach was included as part of the Orlando Television ADI although Daytona is independent with its own rated Top 100 radio market. Lakeland – Winter Haven also has its own Top 100 rated radio market while Lakeland – Winter Haven is a part of the Tampa Bay Television ADI.

Hopefully this helps.

 

[Deano]

If memory serves me correctly, back in the early 1990s Arbitron was considering the addition of Lake County to the Orlando Metro. The rub was with subscribers who owned AM facilities. None of Orlando's AM stations deliver a good nighttime footprint over much more than the southern part of Lake County. With directional patterns sending the signal either to the east (540, 740, 950, 990, 1270, 1440) or to the south (580, 660, 810, 1030) most are not very listenable once you get up into the Golden Triangle, Leesburg and The Villages.

 

[davideduardo]

Arbitron has a rule that determines the counties (and even, in some cases, county splits) that make up a metro. There is a formula that uses the amount of commuting within the market and the specific county and the amount of listening to stations that are home to the metro which qualifies and disqualifies counties.

Here, from Arbitron's most recent "Description of Methodology" is what is called the "55/15" rule...

Generally, a sampling unit will be added
to the Metro’s definition if each of the
following is true: 1) the sampling unit is
contiguous to a Metro County, 2) at least
55 percent of all unweighted quarterhours
of radio listening reported by
Persons 12+ in the Monday-Sunday,
6AM to Midnight daypart over the
previous Spring and Fall surveys was
credited to radio stations home to the
Metro, and 3) 15 percent of commuting
from the county is into the existing Metro
(according to the most recent census).
Additional criterion may also apply,
particularly when considering the
disposition of counties in adjacent
Metros. (Note: an alternate set of criteria
may apply in rare circumstances where
the area being evaluated for re-definition
includes a ‘dual city’, e.g., cities that are
within a Metro and adjacent to a non-
Metro area. These criteria will be
reviewed on an as-needed basis with
parties to whom they apply.)

 

[shoothoops]

OK, i was here when it happened. It was Orlando-Daytona as late as '88. the counties
covered were the 3 now, and included Lake and Volusia. Some stations had better coverage,
some staions couldn't sell them. Plus, it was market 23 ish. which meant, market 23 paychecks.
So, in the best Chicago dirty secret political way, the GM's managed to get lake and volusia whacked.
Daytona became it's own market, Lake county left blowing in the wind, and knocked the Orlando market
size back to 42. You are correct, 5 miles to the north, lies lake, and 15 miles to the east
there is volusia. All in the name of money. 106 stopped bitching because they couldn't reach the beach, while 105, 107 , K92, could, and the others who couldn't reach north enough, stopped
bitching and filing complaints, they didn't cover lake. And the county they agreed on Osceola, still
makes us laugh, because 90 percent of Orlando radio is gone halfway in and over. So, the final kicker is, those who reach, still sell those counties, including brevard. revenue up, salary's down
and there you have, the rest of the story. Anybody who disputes 1 word of this, was on the grassy knoll, and fired the kill shot. Or actually drank the kool-aide, given out at the staff meetings

 

[davideduardo]

OK, i was here when it happened. It was Orlando-Daytona as late as '88. the counties
covered were the 3 now, and included Lake and Volusia. Some stations had better coverage,
some staions couldn't sell them. Plus, it was market 23 ish. which meant, market 23 paychecks.
So, in the best Chicago dirty secret political way, the GM's managed to get lake and volusia whacked.
Daytona became it's own market, Lake county left blowing in the wind, and knocked the Orlando market
size back to 42.

I checked back to 1982 (Using the Arbitron market map in the Orlando section of the Inside Radio Directory at http://www.davidgleason.com/Archive R&R/Inside Radio Fall 1982.pdf ) and found that the definition of the Orlando market way back then was the same three counties as today.

The market, per Arbitron, in 1987 was 51st with 746,000 and it went into the top 50 in 1988 at 49th with 773,000 population. It's now 34th.

The original Arbitron market definitions were based on a start point of the OMB's MSA definitons (Metropolitan Statistical Area), the application of the 55/15 rule or its equivalent, and the proximity of other rated radio markets to determine the Metro Survey Area. As Arbitron began in the mid-60's and achieved critical mass in the early 70's, we are talking about an era when AM radio coverage determined radio markets. Since very few Orlando stations had significant coverage outside of the three mentioned counties, the Daytona, Lakeland and other markets were not part of the Orlando MSA.

In general, as FM grew, and given the overall coverage superiority of most FMs over the bulk of AMs in each market, MSAs tended to grow... Miami became Miami-Ft Lauderdale in 1981 much to the displeasure of AM managers in the two former markets. So if anything, Orlando might have grown, but it is likely that the criteria of 55/15 were not met and the other rated markets did not want to be absorbed by Orlando.

Anybody who disputes 1 word of this, was on the grassy knoll, and fired the kill shot. Or actually drank the kool-aide, given out at the staff meetings

Daytona Beach was first measured by Arbitron in 1977. Lakeland started before 1975 (where I first have collections of data). They were separate, and not part of the Orlando MSA.

 

[shoothoops]

sorry, it was Orlando-Daytona, and thats how it was presented in R and R and arbitron.
not to be a board butthead, but thats how it was sold and marketed. diary's in volusia and lake counted. now what your reading? isn't how it was, but then again, many things weren't as they seem now. Thats all i got, when you explain Oswalds jammed rifle, from behind, hitting Kennedy in the side of the head, then you have the real story. plus the other 3 shots never explained. David, i belive you are reading what you are, but that smoke filled room changed everything, oh, of course orange, seminole and osceola are the 3 counties of record, you are correct sir. it's the bullet to the side of the head, supposedly shot from behind, that is as big a mystery how 2 counties, one 5 the other 15 miles away from DOWNTOWN, suddenly didn't count. SOMEBODY dig up an old R and R from 1988 or before, everytime the ratings came out, you will see. Now before, somebody with nothing else to do starts up on me, cause i see that alot around here, i am out. unless you can give me the bullet, from the wrong angle, nothing from a book means anything. our history books say Lee harvey Oswald acted alone, that must have been one of those cartoon rifles, with a huge loop for a barrel. I-100, BJ 105 went head to head. Y106 cried foul. Now i have to get back to watching baseball. David, always enjoy your post's though, your on so many boards, and it's like you know everyone, where are you from?

 

[davideduardo]

sorry, it was Orlando-Daytona, and thats how it was presented in R and R and arbitron.

Sorry to disappoint you, but Daytona was a separate market, Lakeland was a separate market. If you can get access to a copy of Duncan's "American Radio" 16th anniversary edition with tracking for all markets from 1975 to 1992 or the fuller 1975-2002 edition, you can see that the market was the specified three county metro survey area all along.

not to be a board butthead, but thats how it was sold and marketed. diary's in volusia and lake counted.

I don't have a history of the Orlando TSA, as nearly nobody used or uses TSA... but perhaps you are confusing the metro with the TSA.

David, always enjoy your post's though, your on so many boards, and it's like you know everyone, where are you from?

I have to admit I have never seen the Kennedy assasination connected to Arbitron (which did not even do radio ratings when that event occured), but I suppose that this concept is no further off-track than some of the explanations for PPM behaviour I have read recently.

As to your question, follow the link.

 

[shoothoops]

hyphens aside, and lakeland will have always been it's own market, neither tampa or Orlando
would even lay claim or market there. Now, PPM, thats both Kennedy's and Dr King together.
And you can throw in who really blew the BP rig, and for good measure the 2000 presidential
election. Your a smart guy, maybe you can explain this, how does a system, that is supposed to
end all confusion with tones from each station, making it clear, and removing human error or laziness, look so different , establish more confusion than the health care bill, and take radio backwards to 1980, instead of up to speed and real? i would respect an answer from a guy who
see's the same thing

 

[davideduardo]

...maybe you can explain this, how does a system, that is supposed to
end all confusion with tones from each station, making it clear, and removing human error or laziness, look so different , establish more confusion than the health care bill, and take radio backwards to 1980, instead of up to speed and real? i would respect an answer from a guy who
see's the same thing

What I have seen is every manner of explanation for the PPM results and all kinds of theories on how to program to win in the PPM.

In the last few days, a well known person in the industry tried to make conclusions based on the fact that there are many very short intervals of listening... just a few minutes long. The fact is that the PPM picks up exposure, not just listening and "exposures" are often short times when people were near another radio controlled by someone else. The fact that some stations have seen cumes double in the PPM illustrates that as much as half those cumers are "accidental" listeners or those who tune in just for a brief traffic reports and such.

As I said, theories abound on how to meet the new challenge of PPM, and this makes me recall the 70's when stations used new theories of quarter hour maintenance, time warping and such to try to "beat" the then increasingly important diary method instead of the Hooper and Pulse methods. Most of the initial theories were seen as being wrong or ineffectual, though.

The PPM system is definitely state of the art, and far more accurate than the diary system. The PPM system encodes every station with a unique data pulse that can give an ID and time stamp as often as 12 times a minute. There is nearly no objection to the system.... just to the recruit methodologies and sample proportionality.

The results, however, have a major advantage. The diary measured cume, TSL and the diarykeeper's memory. The PPM just listens and identifies. The result is that the diary entry for listening to a morning show "6 AM to 9 AM" is found to be a bunch of separate incidents averaging 11 or 12 minutes, and totalling maybe 50 minutes, not 180. If we examine reality, the PPM is more accurate in measuring each person's listening and exposure. If the sample is truly proportional, as a panel should be, the results will be better, even far better, than the diary.

 

[shoothoops]

So walking through a mall, daily, will credit radio stations playing, you would never listen to. The more time in a store, the more time PPM picks up station 'Whatever" . i agree, diary sucked, but i also noticed, with these strange swings in listening ( you said it yourself) alot of guys lost there gigs after maybe round 2. Very talented people, who have been told PPM showing massive music listening. Doesn't it realy depend on where they go and hang, as opposed what they ( the carrier) prefer in reality. it seem's, that everytime a 3 month perid is looked at, the format " da jouir" flips. Or am i feeling this the wrong way. good gawd, you could spend a weekend walking through a flea market, and depending on what part of town, could cause huge swings

 

[davideduardo]

So walking through a mall, daily, will credit radio stations playing, you would never listen to. The more time in a store, the more time PPM picks up station 'Whatever" . i agree, diary sucked, but i also noticed, with these strange swings in listening ( you said it yourself) alot of guys lost there gigs after maybe round 2. Very talented people, who have been told PPM showing massive music listening. Doesn't it realy depend on where they go and hang, as opposed what they ( the carrier) prefer in reality. it seem's, that everytime a 3 month perid is looked at, the format " da jouir" flips. Or am i feeling this the wrong way. good gawd, you could spend a weekend walking through a flea market, and depending on what part of town, could cause huge swings

Those are very valid questions.

The PPM does not employ the term "listening" but, rather, exposure. Advertisers don't care, to a great degree, how you heard the ad... just that you were exposed to it; the PPM is largely a response to advertiser demands on radio.

The key elements in the ratings game are using a station at all (cume) and how long you use it (TSL in the diary, Average Time Exposed in the PPM). A station that gets exposed to a lot of people but not for long will not get TSL and thus not get a high share. One that gets fewer people, but who listen... or are exposed... for a longer amount of time can get a much higher share.

Since share, rating and AQH persons are all the same thing expressed in different fashions, and they are the thing advertisers look at, getting high shares or ratings is what produces sales results. Same as the diary, in other words.

Except... the PPM picks up exposure if for a minimum of five minutes in a quarter hour and gives credit. So if you are in a store and there long enough to register, you are cuming a station. But, since most of us don't go to the same store daily or for much time, we don't give much listening to a station via this incidental exposure, so it is not significant and only shows what is reality... we hear stations we did not pick regularly, but generally not for very long.

TV has used meters for several decades, and decisions are made very fast in the TV world... with 13 books a year, and results within under three weeks for any month, radio decisions will be done much faster. The PPM measures reality, not perception, and some changes are natural.

 

[shoothoops]

Makes better sense,you should speak on it, to these PD's and consultants that are taking 2 samples, and declaring it... the way it will be. Or to the "experts " that say, urbans are going to kill, or country will be back, or talk radio will tank, not because they are better or worse, because of management being to quick to judge. Rock will dominate, no talking, no matter the format. well the fact is, whatever your format, if your office friendly, and they have there meters, it's more important than ever. Thank you for your insight. I do welcome this change, cause i know how diary's were filled out, last second, and real listening was hardly ever charted correctly. This way, any station heard , shows. That is what the system is supposed to be about. And the confusion, when a personality changes stations, ends. Or 2 stations, same music, and you have people who listen to churbans / CHR's or Hot AC's where there are more than 1, just taking a stab at what they listened to most, to get the diary in on time, my biggest gripe, ends. And all these damn PD's and jocks scared of, accuracy. hmm. Hey bring on Mr Roboto, Styx had it right in 1981