In July I did a diary showing how AirAmerica had done up through Spring 2005. Today the last of the Summer 2005 ratings were made public, so I can once again, in the spirit of a reality-based community, present the most recent information.
Some of you may remember there was some discussion over what numbers I was using and how I arrived at them last time. Also, several questions were repeated asked last times, so please be sure to read the body of the diary before responding this time.
- I do not have access to the "real" Arbitron ratings. All I have is the most general public numbers. For the type of data I'm gathering, this is close enough to roughly accurate that I'm OK.
- Rush Limbaugh was chosen as a baseline because he is widespread and it would be easy to find a station carrying him in any given city. Also, he is an established personality and can be relied upon to show any widespread conservative radio trend.
- I have no way of tracking streaming or satellite listeners, so those numbers are not included.
The short story is that AirAmerica didn't do that well this last quarter. The longer story is that the Rush Limbaugh baseline numbers didn't either. In July, some predicted I would see a "summer slump" across the board and that seems to be the case. Both slumps were quite small, but AirAmerica's was larger, no doubt largely attributable to the fact that both Sacramento, CA and Providence, RI dropped out of the lineup. (Sacramento has a new station, but no ratings containing AAR exist yet.)
Based on the discussion from last time, the first graph shows absolute "pseudo-listeners". Again, since I don't have access to the real Arbitron ratings, I have to take the public numbers and use general knowledge to extract what I can. For instance, the market size of Portland, OR is 1,937,600. Generally speaking, about 14.5% of the public listens to the radio, so Portland has 280952 "pseudo-listeners". If AirAmerica pulls a 4.4 in Portland like they did in Summer 2005, they get 4.4% of 280952, or 12362 pseudo-listeners. The sum of these listener numbers across all AirAmerica-carrying cities is what is graphed below.
The vertical scale here is basically useless--the graph is only intended to show the general trend. In this case, a probably-mostly-noise 2.7% decline for AirAmerica and a definitely-noise .75% increase for Rush Limbaugh.
As I mentioned last time, my whole purpose in including Rush Limbaugh was to check AirAmerica's performance against a baseline. It should be obvious that liberal radio's numbers will be lower than conservative radio's--we are about 15 years behind. The question is, is AAR beating, matching or falling behind the general trend? The best way to check that is to remove the absolute numbers from the situation and only check relative growth. Whatever its size, has AAR been growing as well as conservative radio?
(Mathematical aside: The way to do that is to "normalize" the numbers. Divide every AAR number by some base AAR number--I chose their first quarter. Divide every Limbaugh number by some base Limbaugh number--again I chose the quarter AAR was first on the air. Say AAR got 10 listeners, then 15, then 20 and Rush had 1000, 1200 and then 1400. Rush is way ahead in absolute numbers, but now normalize them. AAR would normalize to 1, 1.5 and 2. Rush normalizes to 1, 1.2 and 1.4. AAR would be behind but growing better.)
The actual graph:
Some features:
- AirAmerica seems to follow the general trend of talk radio, including the bump during Election 2004, up until Spring 2005.
- Spring 2005 was incredibly successful for AAR relative to conservative radio.
- Both sides of the radio aisle were stagnant during the summer, with a slight decline for liberal radio.
Conclusion: While I'm a little disappointed that AAR didn't roar up the charts again this quarter, I'm not surprised they didn't. Vacation (both for listeners and for legislators) and nice weather draw people away from the radio just they do the TV. We will likely see more action, for better or worse, in the next quarter's numbers.