Whatever your opinion of Nielsen Audio, they have a tough job. When Nielsen started measuring radio back in the 1930s (long before Arbitron existed), recruiting survey respondents was pretty easy. People were impressed that a company wanted to know what they listened to on the radio, and even into the ‘50s and ‘60s, response rates were high.
We’ve gone through a transition of declining response rates combined with the change from landlines to cell phones as a method of recruiting people. Door-to-door interviewing went away in the ‘70s with the demise of Pulse, and we’ve moved to an address-based sample frame because of the change from household phone numbers to individual phone numbers. The task keeps getting harder to accomplish.
I’ve spotlighted this issue before in this column. As costs go up, Nielsen, like any other company, will look for ways to keep their expenses in line. Some initiatives work well, and others, like offshoring their data crunching, haven’t been as successful, as I’ve noted here recently. South Bend was added to the list of screwups last week, putting the tally at 58 markets so far.
Back in my time at Arbitron, we were occasionally accused of reusing sample. I remember having to write a piece when someone in the industry called it the “windshield wiper effect,” a completely bogus accusation. In the pre-PPM days, a household could end up in the diary sample more than once in a short period of time—say, a year—but it was purely by chance.
In random samples, anything can happen. Baseball is full of those oddities; for example, the 8-6-2 double play by the Brewers against the Dodgers in the NLCS. Or one of my recent favorites: the New York Rangers set an NHL record by being shut out three times in a row at home to start the season. You would guess correctly that I’m not a Rangers fan. It’s the same long shot as someone winning the Powerball or Mega Millions jackpots.
Now, Nielsen has added “targeted reselects,” which means going back to households that have been in the diary service previously. The emphasis is on households with hard-to-reach individuals, which is people who aren’t in my demo—more specifically, young and minority. If you have been doing, as I suggested a while back, reviewing the Presurvey Bulletin (PPM every four weeks, diary service every quarter), you’d know that Nielsen is looking for 25% of the sample in diary markets and non-metro areas to be targeted reselects.
In general, there’s nothing wrong with this approach. Nielsen has said that no reselected individual will show up twice in a survey, which means an extended period for the smallest markets that use two-book averages. It seems likely this will increase PUR just a bit.
While we know that one diary can have a real effect on ratings if it includes a great deal of listening to one station, the overall effect is typically limited.
One positive side effect is that if Nielsen can use targeted reselects to better balance the sample, that can reduce the effects of weighting. Sometime in your diary market career, you will discover that a heavy-listening diary also carried a heavy weight due to shortages in a particular cell, whether demo (especially young males), race/ethnic, or geography. The closer Nielsen’s sample can get to matching the market’s population estimates, the less the effect of weighting.
In the past, I’ve strongly suggested that subscribers check the E-Book to understand the makeup of their markets and to look at how the sample was distributed. Per MRC disclosure requirements, Nielsen includes the targeted reselect sample results in the E-Book. The disclosure is divided into “Fresh Sample” and “Targeted Reselects.” You can easily see the number of initial households, the diaries mailed, the intab diaries, and the percentage of the sample that came from these households.
Nielsen has set the goal for targeted reselects at 25% of the sample as of Fall 2025, in both diary metros and non-metro areas—in other words, the entire diary service. Their view is that testing and practice have shown little difference in listening, but the gains in proportionality are positive. The unspoken part is that this change will help control sample costs as well.
It’s hard to knock Nielsen for keeping costs in line. We’ve read too many trade press articles of late about friends in the radio business who will have to pursue new jobs and potentially new careers as their former employers “right-size” or use whatever euphemism they choose. For those of you who are still working and responsible for ratings success, keep abreast of the changes.
Let’s meet again next week.
Barrett Media produces daily content on the music, news, and sports media industries. To stay updated, sign up for our newsletters and get the latest information delivered straight to your inbox.
