Can you explain a bit about PPM "weighting" and survey sample sizing"?
For a station like WABC to be so top heavy with 55 and even 65+ male listeners and in the bottom half of the $ demos, yet be within a share point of the overall Top 5, 6+, seems to illustrate a dearth of younger folks willing to wear meters and Nielsen having to over compensate the older demo #'s into the overall pool.
The PPM sample is very close to proportional on all the age, gender, ethnicity, language, income and education categories. When there is a shortage in a category, those that are there are slightly weighted to achieve proportionality; the opposite is done if a cell gets over-populated and it is weighted down.
"Weighting" is simple. If you have a quota of 10 in a statistical cell, and only get 9 participants, you weight each of the 9 up by about 10% so that the cell has proportional weight. Similarly, if you got 11 participants, you weight each down by the appropriate amount so that the cell is proportionally represented in the total sample.
There is no "dearth" in younger participants; Nielsen just works harder to get them to participate. There are considerable cash and premium incentives to everyone to get participation.
WABC has huge shares in 55+ and 65+, and those average out across the whole sample when you look at 6+, 12+ and 19+. That is normal.
WABC's "anemic" cume combined with their through the roof TSL also super illustrates how lower cuming/high TSL "passion" formats like talk, urban, Latin, and country can easily hold their own against the higher cumer AC, CHR, and Classic Hit stations depending on the market.
Spanish language formats in competitive markets don't have such a huge advantage in TSL as they did when there might have been one station in each format or just one in the whole market.
Talk formats have always much higher TSL than others. Because today they tend to only appeal to 45 and over, and principally 55 and older, they can not generate the broad cume that some other formats generate.