As an update on Arbitron's "Better Measurement" initiatives - - an ongoing series of enhancements to the Arbitron service - - the company has completed its fourth study of cell phone consumers who don't have a landline phone at home. As part the ongoing effort to sample more young listeners, Arbitron VP/Domestic Radio Research Dr. Ed Cohen presented the findings of the study yesterday and discussed Arbitron's plans to recruit cell phone respondents.

---

Cohen said that based on the National Health Interview Survey, in the first half of 2005, 7.4% of households and 6.7% of Adults 18+ were cell phone only. The percentage is higher among Hispanics (9%), slightly higher among Blacks and a little lower among Whites. For Adults 18-24, 16.1%; Adults 25-54, 9%. The number was over 7% of every region in the country except the Northeast where it was 4%.

Cell phone respondents listen to radio differently - - they generally listen more and they prefer certain formats. After weighting the cell-only sample, there were slight differences overall estimates in format preferences from landline-only households. For 12+, CHR is up a tenth; Country is down a tenth of a point. The other formats showed no change. The effect is greater in 18-34. Also given a cell-only penetration of 10%, CHR was up a tenth. Spanish was down two tenths of a rating point.

"Even in a 10% cell-only household rate, the effects on the overall estimates are slight," said Cohen. "We believe the trend is going to be for continued growth, so the time is right to start the process of adding cell-only sample to the Arbitron sample frame. But adding it requires a lot of changes in our systems. And from a personnel standpoint alone [given that by law all calls must be hand-dialed], the calling volume increases drastically as you add markets. We need a lot more interviewers. Systems have to be redesigned for sampling, for weighting and for reporting. Universe estimates have to be devised; response rate calculations don't exist right now. We have to work on the economics of measuring cell phone only samples."

He also mentioned it will be expensive and not an easy fix, but it's something that should be done. Response rates will decline. More testing will be done between now and 2008 to see if it can be done in a more efficient, cost-effective manner without compromising quality.

"So our goal is to start calling cell phones in some diary markets beginning in 2008. As a reminder, PPM in Houston already includes cell-only households. Moving forward, PPM will include cell-only households in all the new PPM markets," Cohen concluded.