Spring Ratings 2016

[tbolt909]

The public ratings are available in the usual places.
KB 1520 AM must be below a 0.5 as they didn't register.
Seems like a waste of electricity keeping that old dog alive.

ALT Buffalo still hasn't reached a 2 share during its existence.
JACK and MIX went up a little, while WGR and 97 slid backward some...

 

[Rusty Bridges]

Google search found them here.

 

[dand5780]

Pergy and friends at the rag must be upset with WBEN ratings.

WHAM did ok too.

Maybe they should publish the ratings for the "public" station.

 

[SirRoxalot]

WBEN went up, WGR went down. Significant drop by WBLK. My guess is that diary placement favored angry old white suburbanites. Let me know if the trend continues for another book or two.

 

[davideduardo]

WBEN went up, WGR went down. Significant drop by WBLK. My guess is that diary placement favored angry old white suburbanites. Let me know if the trend continues for another book or two.

I thought you'd know that the sample is proportional on Blacks, whites in the older demo cells and on areas of residence within the metro.

 

[SirRoxalot]

The sample is "proportional", which means that the diary return is "adjusted" based on how many diaries were returned from different zip codes and demographics. The actual in-tab sample is so small that a diary or two can skew numbers significantly - and has in markets like ours. Yeah, I know, statistics says that it shouldn't happen, but we've seen it before, and we'll see it again. If you've actually gone to look at the in-tab diaries, you can see that it happens. The "weighting" is far from perfect.

 

[davideduardo]

The sample is "proportional", which means that the diary return is "adjusted" based on how many diaries were returned from different zip codes and demographics.

No, the sample is weighted for proportionality. That is not the same thing. The sample can be very disproportional, but by weighting it is made proportional.

As Orwell said, "all animals are equal but some animals are more equal than others." And that is the effect of weighting.

The actual in-tab sample is so small that a diary or two can skew numbers significantly - and has in markets like ours.

The in-tab sample is as large as the local broadcasters will pay for. If they want a larger sample, they can buy it. Their choice.

Yeah, I know, statistics says that it shouldn't happen, but we've seen it before, and we'll see it again. If you've actually gone to look at the in-tab diaries, you can see that it happens. The "weighting" is far from perfect.

The weighting is perfect. The whole purpose of weighting is to make an imperfect sample "perfectly" proportional.

I've looked at millions of diaries since 1970, and never saw seen sample issue by reviewing individual diaries. Sample issues are seen by looking at tabulations of the in-tab for each stratification variable and seeing what the diary return was for each and the amount of up or down weighting applied to each diary.

 

[flybynight]

I've looked at millions of diaries since 1970

Holy cow!

Even if by "Millions" you meant only 2 million, that would mean you've looked at roughly 120 diaries every day since 1970!

Now if you took weekends off, that would be roughly 175 diaries every day since 1970.

Geez Louise!

 

[SirRoxalot]

Now, fly, he didn't say he read them. He just "looked at them". In stacks. Bundled.

David has essentially staked his career on statistics. In his mind, they're perfect. Most statisticians are somewhat less sure of themselves, which is why they look at longer-term trends. We certainly have all seen books that were an anomaly. I'm sure that the "weighting" was perfect, but the underlying data was not. If you've only got a few books representing a particular zip code or demographic, one skewed book can carry immense weight - far more than it would have in a zip code or demographic with many books. That's the fallacy of "proportional weighting".

But, David will assert that he's right, and chastise the rest of us for being mere local peons who simply aren't smart enough to understand the grand design of statistical analysis and the beauty of their results. After all, if we were either smart or talented we would have moved on to larger markets, right?

 

[TheBigA]

We certainly have all seen books that were an anomaly.

Of course. That's why Arbitron and now Nielsen allows representatives to READ the actual diaries at their office in Maryland. And if you see an anomaly, you bring it to their attention. An anomaly can cost money. The goal is to prevent that. So we'll see what happens in the next book.

Look, everyone argues everything, including weighting. Unfortunately we have a monopoly running the ratings system, and we don't have much choice.

 

[SirRoxalot]

Hey, there's always Eastlan.

 

[tbolt909]

WBLK only went down 1 share. Nothing else looks too dramatic.
WGR always falls in the offseason. WBENs feeble minded seniors & tea party
loonies remembered to mail in their diaries this time...

 

[davideduardo]

Now, fly, he didn't say he read them. He just "looked at them". In stacks. Bundled.

Nice try. It's "saw" as in looked at them, page by page, on the computer system used for diary review for the last two decades.

Of course, prior to that in Beltsville and Laurel, we got plastic storage trays in specific order.

In either case, a review (and for decades I did reviews of multiple markets four times a year) consists of looking at your station's usage, analyzing sharing with competitors, searching for any errors and similar things. I personally got three books reissued based on diary reviews and was on Pat Duggan's "Hall of Fame" list (inside humor).

David has essentially staked his career on statistics.

No, I have staked my career on winning programming. The statistics help sell the dominant positions that I have tried to achieve for stations I have owned, managed or programmed.

In his mind, they're perfect.

If I thought that, I would not have gotten those books reissued. I believe I am one of only two or three who have gotten an Arbitrend reissued, too.

Once I am sure ratings reflect fairly the competitive environment, I have used them to generate sales. And that is the main purpose of having ratings.

Most statisticians are somewhat less sure of themselves, which is why they look at longer-term trends. We certainly have all seen books that were an anomaly. I'm sure that the "weighting" was perfect, but the underlying data was not. If you've only got a few books representing a particular zip code or demographic, one skewed book can carry immense weight - far more than it would have in a zip code or demographic with many books. That's the fallacy of "proportional weighting".

Actually, weighting is used to make a non-proportional sample adhere to proportionality.

And I am just one of many who pushed Arbitron over the years to improve sampling techniques, in situations that were, to say it mildly, adversarial.

But once I was in a top market that had no for several years. Once company pulled out, and over the two years, average rates and market revenue declined by nearly 50%. Agencies would say, "I don't know how much listening you have so I don't know what it is worth. Here is what I will give you. Take it or leave it." Revenues increased rapidly once we had ratings again. So I am a real believer in ratings as a way to get the highest possible rates.

But, David will assert that he's right, and chastise the rest of us for being mere local peons who simply aren't smart enough to understand the grand design of statistical analysis and the beauty of their results. After all, if we were either smart or talented we would have moved on to larger markets, right?

I am not saying you are wrong, although your use of inaccurate terminology and misinterpretation of data makes me believe you don't understand ratings very well. My point is that it is much easier to get good rates with good ratings. That means insisting that the ratings company constantly improve its product while spending most of the time creating a good product. I have always found it is a lot easier to sell a top rated station than one that is an also-ran.

 

[davideduardo]

Holy cow!

Even if by "Millions" you meant only 2 million, that would mean you've looked at roughly 120 diaries every day since 1970!

Now if you took weekends off, that would be roughly 175 diaries every day since 1970.

Geez Louise!

On a single diary review, I might have looked through 25,000 diaries or more in a couple of days. One does not read every one, of course. You look for listening to stations you care about, but you skim them all in the process. It gets to a point where you can "speed read" them if you know what you are looking for.

 

[Element9]

it's Persons 12+ ... That's why Nielsen releases them free of charge. It makes good fodder for speculation and debate on message boards. Surprising that 97 and 103 didn't do better. Then again, it's 12+. Wait... What's the #1 station? WYRK. Again.

 

[TheBigA]

What's the #1 station? WYRK. Again.

How many in a row? They must be doing something right.

I always think it's interesting how well country music does upstate, in Rochester, Syracuse, Albany, Binghamton, even Poughkeepsie. Obviously not just for southerners anymore.

 

[tbolt909]

How many in a row? They must be doing something right.

I always think it's interesting how well country music does upstate, in Rochester, Syracuse, Albany, Binghamton, even Poughkeepsie. Obviously not just for southerners anymore.

WYRK is the only Country format in the market.
If Country is so popular, where is their competition?

 

[TheBigA]

If Country is so popular, where is their competition?

Most of the popular formats in Buffalo have no competition.

 

[davideduardo]

WYRK is the only Country format in the market.
If Country is so popular, where is their competition?

It's the simple realization that splitting the audience would result in a second station that likely would, at least initially, have lower 25-54 numbers and, for a long time, much lower revenues. There is no full signal FM that would want to sacrifice 6-12 months worth of existing revenue during a conversion period in order to bill at a very much lower level than they probably have now.

The only full signal FMs at a low enough 25-54 level to even consider such a move are both owned by Townsquare. The rest are higher in billings and a change would be very high risk for very limited gain potential.

 

[TheBigA]

It's the simple realization that splitting the audience would result in a second station that likely would, at least initially, have lower 25-54 numbers and, for a long time, much lower revenues.

For comparison, in Syracuse, where country WBBS has been #1 for years, a competitor popped up a few years ago. It really hasn't taken audience away from BBS, and it gets about a quarter of the audience.