Arbitron's sampling methodologyisFARmoreimportantthanMrorMrsDePetro's misdeeds

[David Tomm]

Right now Arbitron has no plans to roll out the PPM beyond the top 50  
markets, so Hartford will be the smallest market using this  
methodology.  There are many statistical criteria they have to meet  
when placing PPM meters, including representation of various age  
demos, minority groups, geographic placement around the market etc.   
The biggest complaint with the PPM thus far is that certain groups  
are under-represented in the results, particularly persons 18-34 and  
minorities.  This was the case in Philadelphia and in New York, where  
ratings for AC and oldies stations soared in the PPM while Urbans,  
CHR's and Spanish stations plunged in relation to the diary.  As time  
has gone on and the troublesome cell targets have been properly  
populated, the data looks much better, albeit not perfect.

Despite all the flaws, PPM will eventually be far superior to the  
diary.   Most people will only write in a couple of stations each day  
or week, but in reality they are exposed to many more.  Plus, keep in  
mind that eventually PPM ratings will include non-comms and satellite  
radio stations.  Think about it.  You go out to do errands on a  
Saturday.  You head to the auto parts store to pick up oil for the  
car.  You're in there for 10-20 minutes and they're playing WAAF.   
That gets reported.  Then you go to Trader Joe's to pick up something  
for dinner and they're playing Sirius 18 The Spectrum (AAA).  That's  
recorded too.  Then, maybe a trip to the bookstore and they have on  
WBUR.  Yup, that's credited.  Compare that to the average diarykeeper  
who would probably write down that they listened to WMJX the entire  
time of the trip since that's what was playing in the car.  That  
makes up for occasionally leaving the meter idle somewhere and it  
shutting itself off.  PPM ratings in Philly have already shown that  
average listenership goes through the roof on Eagles flagship WYSP  
during games and drops back down to normal soon after.  This is very  
useful data that can be turned around quickly, something that can't  
be done with diaries.

It is Arbitron and it will never be perfect, but eventually PPM will  
surpass the diary.  They've just got to work through a few more bugs  
in the system.

-Dave Tomm
"Mike Thomas"


On Aug 23, 2008, at 2:58 PM, Don A wrote:

> Not at all.
>
> (I didn't realize you were an expert in this field Dan.)
>
> The statement was in response to Dan Billings comment about  
> "cheating".
>
> As Garrett pointed out, PPM combats the problem of "conscious  
> recall" and
> "actual listening".
>
> Given your scenario, you would have to GET 6 PP's...which will be  
> much harder to do than to get 6 paper diaries.  Not to mention that  
> at the end of the day, they have to report back to the  
> mothership...would all 6 report back from the same phone line?
>
> While no methodology is perfect, PPM should be better at situations  
> like Depetro's.
>