Kabrich Disputes Arbitron's Explanation Of PPM In-Tab Drops
April 3, 2008 at 5:19 PM (PT)
YESTERDAY (NET NEWS 4/2), consultant RANDY KABRICH alerted ALL ACCESS to a dramatic drop of almost 20% in-tab in the PHILADELPHIA PPM during the first week in MARCH. ALL ACCESS contacted ARBITRON SVP/Press & Investor Relations THOM MOCARSKY, who explained from ARBITRON's viewpoint, what happened to cause this issue.
KABRICH is unhappy with both the PPM report and MOCARSKY's response, and wrote this to ALL ACCESS:
"ARBITRON tells us that the sharp loss in in-tab in the Week 1 PHILADELPHIA PPM estimates was due to the fact that they used the same initiative they've used to boost 18-24, implementing this now on the 25-34 cell -- a long anticipated improvement. This initiative included replacing non-25-34 households with those which contained a 25-34 year old. Unfortunately, the numbers don't bear that out and the impact on the rest of the demographics' in-tabs is far greater than any increase in 25-34 sample."
Quite simply, the Arbitron explanation falls far short of explaining the jaw-dropping loss in in-tab in the Philadelphia PPM for the first week in March.
"On the daily in-tabs, 25-34 has increased vs. the previous week, but it is currently the same in-tab that it was 4 weeks ago, while all other age cells have dropped, some significantly. Further, this appears to have erased several weeks of solid growth in the 18-24 in-tab."
"There is a gain of 6 Women 25-34 and 4 Men 25-34 in-tab. Even if these 10 25-34 in-tabs were from 10 different households, they should only take 10 in-tabs from the rest of the sample. They are supposedly 'robbing PETER to pay PAUL,' substituting non-25-34 homes with 25-34 homes. Even if ARBITRON de-installed the non-25-34 homes before they found enough 25-34 homes to substitute, they should not have lost as much in-tab as they did. So a gain of 10 25-34 in-tab shouldn't result in a loss of 56 P6+ in-tab, producing the lowest in-tab since the holiday week. Following the Daily in-tab chart below are the weekly in-tab figures, where the impact is even more significant."
"The impact on the weekly in-tab is more significant -- the weekly in-tab is down 15%, losing 215 P6+ in-tabs and producing the lowest weekly in-tab since last September, erasing six months of in-tab growth. ARBITRON tells us this is because the de-installation procedure removes the non-25-34 HHs and installs the 25-34 HHs in mid-week (so they can't be in the 'listened 6 out of 7 days' group that creates the weekly in-tab). This simply doesn't make sense -- there is actually a LOSS of 4 25-34's in the weekly number so not only do none of the 10 new 25-34 daily in-tabs have a full week of listening, hundreds of others in other demos have dropped out as well?? How is it possible that there is not only NOT a gain in 25-34 here, but there is a significant loss in every other age cell? It's does not seem possible that every age cell was impacted to this degree, just to harvest the 10 new 25-34 in-tabs that are in the daily in-tab sample."
Concludes KABRICH, "Quite simply, the ARBITRON explanation falls far short of explaining the jaw-dropping loss in in-tab in the PHILADELPHIA PPM for the first week in MARCH."
ARBITRON declined to comment on KABRICH's analysis.
Cornerstone Weighs In
Meanwhile, CORNERSTONE RESEARCH consultant BRAD RIEGEL did some work with XTrends that shows the percent change in PHILADELPHIA from FEBRUARY Week 1 to MARCH Week 1, and there are some big drops across the board. For the info, just click here.
RIEGEL added, "To be fair, while there are some significant changes, ARBITRON is still exceeding their targets."
For comparison, RIEGEL ran the same dates in HOUSTON, where the sample was a lot more consistent. Check it out here.
KABRICH read RIEGEL's info, and added more observations:
"While it's true ARBITRON did meet their 6+ goal, bottom line, they ARE NOT meeting their goal in core 18-54 demos (which is why they need the 18-24 and 25-34 initiative in the first place). As I stated, what they did in the first week of MARCH actually set them back in the gains we had seen in the past.
"Furthermore, PIERRE BOUEVARD noted on last FRIDAY's conference call that Q1 was the high level before we go into the SUMMER -- and now we are at the lowest point since the usual low points of CHRISTMAS and NEW YEAR'S week. What happens to the sample when we get to this SUMMER, which ARBITRON admits is weaker the what we see in Q1?
"As NEW YORK and its embedded markets, LOS ANGELES, RIVERSIDE, SAN FRANCISCO and SAN JOSE all use the same methods as PHILADELPHIA, not HOUSTON, we can only guess that the advances made in those panels (and revealed in last week's PPM Conference Call) have been wiped out as well -- but we have no weekly data at this point to compare it to. Perhaps this is why ARBITRON has been reluctant to release this data to stations thus far?"