oldies76 said:
A "library size" or a playlist size.
Songs in regular rotation and exclusive of specialty shows. I have never heard anyone not call the regular playlist the library, but maybe you have. We have come a long way from the old room with shelves where records were stored. Many of us do not have any of our music on CD, disk, etc. It's all on a hard drive RAID Array and backed up to more hard drives.
Listeners do not choose the libraries..
sure they do. The stations that do not do this get killed by the ones that do.
Please tell me that listeners chose all 3000+ songs on CBS-FM's recent special. Didn't think so.
I have no idea one way or another. However, since they likely did extensive research prior to going to classic hits, and had research from Jack and from the old oldies format, they could have combined all that data.
"What the listener wants to hear today" is a very unproven statement. 100-200 people conduct a music test..There are 200,000 potential listeners..What do the other 199,800 people want to hear??
The exact same thing. A music test is usually 100 people or less, though. And we know from replication studies that we can test another 100 people, and then another, and the results will be the same. So it is not necessary to test any more than 100 persons (with that number actually being overkill).
If you look at New York ratings (and a music test is intended to make a station do well in the ratings) you find that there are 3800 meters in NY which has 16,000,000 12+ population. Of any station's cume, about 50% provides 92% of the quarter hours. Do the math on the cume, and you will find that a 100 person music test "talks to" more listeners than the "useful" meters track. So an AMT is more reliable, for this and many other reasons, than Arbitron.
Like I've said numerous times before, you do not know what other people's tastes are and what are their favorites are.
We know totally, and within a percent or two of the actual score on each song, too. A properly constructed and recruited sample is more than adequate to know with considerable precision what songs are the ones that have a consensus positive score and can be played.
Some of them are within the playlist but not all. Would it seem logical that maybe, just maybe a song that KRTH does not play (playlist of only 500), might be someone else's favorite?
Radio is not an iPod. we can not cater to individuals, only to the largest possible consensus tastes. If a song is loved by 3 people and hated by 7, then it is unplayable because we would lose those 7 listeners every time we played the tune in the PPM era. We have to play songs that are at worst neutral, but mostly positive with nearly everyone at a music test.
[/quote]