Feet on the Street

[Zeb Norris]

From R&R

• Arbitron is expanding its "Feet on the Street" program, which is scheduled to have bilingual Arbitron representatives knocking on the doors of newly recruited Hispanics and African Americans 18-34 in the top 10 Portable People Meter markets by the end of April.

I'm assuming that this is an outgrowth of the New York/New Jersey settlement of issues with Arbitron and minority broadcasters.

My question; is this program ONLY targeting Hispanics and African Americans? And if so, is that fair? (Note to DE; this is a QUESTION. You misinterpreted me in the past on a similar issue.)

But if it's better to include this kind of recruiting, why exclude white folks? I mean non-Hispanic whites. I was going to say Euro-Americans, but that wouldn't work either as Spain in a European country.

Anyway, lots of non-Hispanic whites get missed by Arbitron's traditional recruiting methodology. So why not use this type of effort for everyone?

 

[davideduardo]

• Arbitron is expanding its "Feet on the Street" program, which is scheduled to have bilingual Arbitron representatives knocking on the doors of newly recruited Hispanics and African Americans 18-34 in the top 10 Portable People Meter markets by the end of April.

I'm assuming that this is an outgrowth of the New York/New Jersey settlement of issues with Arbitron and minority broadcasters.

I think it has to do with getting MRC Accreditation; Houston uses similar in-home techniques and got accredited.

My question; is this program ONLY targeting Hispanics and African Americans? And if so, is that fair? (Note to DE; this is a QUESTION. You misinterpreted me in the past on a similar issue.)

Of course it is "fair." The idea is to make sure that panelists that are hard to recruit, or hard to retain, or hard to get daily in-tab from are encouraged to participate by personal attention. All panelists get personal by-phone attention... this is just an effort to retain, train and maintain panelists. It's no different than DST, introduced in the 70's, to get better cooperation via increased incentives, from the hard to capture groups.

In your comment, you seemed to state that Hispanic or Black diaries were weighted out of proportion. They are not. Weighting is always to achieve proportionality, even if over or undersampling has occured.

But if it's better to include this kind of recruiting, why exclude white folks? I mean non-Hispanic whites.

... because, except for 18-34 males and cellural only homes, they have good cooperation rates. And cell onlies and 18-34 males are not easy to find at home.

Anyway, lots of non-Hispanic whites get missed by Arbitron's traditional recruiting methodology. So why not use this type of effort for everyone?

Because they get their quotas on panel households (theydon't recruit people, they recruit households) and there is no need. It is very expensive to do.

 

[Zeb Norris]

Hmmm...

In my previous post that you took exception to, I asked if Arbitron had trouble finding minorities in LA. Yes, I did imply that they might be weighted up regularly, but I didn't mean to imply that they were weighted up disproportionally.

In response you said that Hispanics tend to be MORE co-operative than non-Hispanic whites. What you're saying now seems to contradict that. I mean, if they're more co-operative, they aren't hard to "capture" are they? So why make an expensive extra effort, either with DST or with "Feet on the Street'?

Here in Vermont the population is mostly non-Hispanic whites. But in the latest book they had to weight up all of the men under 50 and all of the women under the age of 45. ALL of them. Every single cell. That tells me that Arbitron is NOT finding it all that easy to reach non-Hispanic white folks.

And of course, as it's a diary market, they don't reach ANY cellphone only folks.

Again, these are questions. I don't know that it's unfair (as an aside, I do support affirmative action, even though it has cost me one promotion that I know of). But it does seem to me that Arbitron should make an equal effort to reach their goals in all of their cells without weighting or racial preferences.

Setting aside fairness, if they actually hit all their targets in all of their cells their data would be more accurate and reliable. On that we can agree, yes?

 

[davideduardo]

In response you said that Hispanics tend to be MORE co-operative than non-Hispanic whites. What you're saying now seems to contradict that. I mean, if they're more co-operative, they aren't hard to "capture" are they? So why make an expensive extra effort, either with DST or with "Feet on the Street'?

The issue with PPM, as explained by Arbitron in their conference calls, is... and I paraphrase... that PPM panel homes are make or break all during the first 30 days or so. The training, encouragement and support given in that period determines if the panel household will continue on for up to the remaining 23 months.

Use of the phone works with "Other" but Hispanics respond better to the personal visit. As an anecdote to this, in the decades that I also sold in Latin America and Puerto Rico, I would never think of pitching a client on the phone... it is considered rude at the worst, and it's not effective at best.

Arbitron has overindexed on their "goals" for Hispanics in most PPM markets, but the issue is keeping the panelists here... the cost of finding a replacement matching household is quite high vs. the cost of retention. And that is where this sort of technique comes in.

In diaries, the extra incentive was likely to be "guilt money" that increased diary returns, not the acceptance of participation. Many people accept, but only a fraction return the diaries.

Here in Vermont the population is mostly non-Hispanic whites. But in the latest book they had to weight up all of the men under 50 and all of the women under the age of 45. ALL of them. Every single cell. That tells me that Arbitron is NOT finding it all that easy to reach non-Hispanic white folks.

Or, the market is so small that they have trouble based on a high refusal / no answer rate. More Other households have answer machines and refuse than other groups. Generally, they pump out more diaries in the last week of a book if there is underrepresentation based on early returns, but it may not be enough.

We have seen horrible 18-34 undersamples in the last several books. Perhaps all the PPM attention is taking the interest and care away from the diary markets.

And of course, as it's a diary market, they don't reach ANY cellphone only folks.

They do now, but it is costly since RDD can not be used for cellphone calls.

Again, these are questions. I don't know that it's unfair (as an aside, I do support affirmative action, even though it has cost me one promotion that I know of). But it does seem to me that Arbitron should make an equal effort to reach their goals in all of their cells without weighting or racial preferences.

Weighting is a technique used in every poll known to man. It's nearly impossible to do a poll and get a perfect representative random probability sample. So the variations are weighted to achieve proportionality... even the Census does it with the long form questionnaire.

And there are no racial preferences (Hispanic, of course, is not a race anyway...) but there are techniques used for subsets to get the most proportional sample possible.

Setting aside fairness, if they actually hit all their targets in all of their cells their data would be more accurate and reliable. On that we can agree, yes?

In theory, that is a perfect sample. It does not exist in practice. They would have to oversample horrendously, and that is cost prohibitive as well as requiring downward weighting.

 

[djmimi]

Regarding minorities, I still think it pretty funny that they continuously ignore Asians...considering that in San Francisco, a third of the population is really pretty Asian lookin. In Los Angeles, Asians are more than a few percent, last I checked.

If you look at the college population, Asians are 30 percent of the students in the UC school system. And how much of that population is Arbitron really sampling??

HMPH! we're still considered "other".

Of course, white people are the minority in Los Angeles...so should they get the special treatment here??

 

[davideduardo]

Of course, white people are the minority in Los Angeles...so should they get the special treatment here??

The real issue is that groups, whether ethnic or age based, don't get special treatment unless they are notoriously bad returners of diaries.

All the Arbitron incentives are post-recruit. In other words, they come after acceptance, and are intended to get the person or household that accepts to participate to finish that participation.

Unless they are not English speaking, Asians should participate in a survey in rough proportion to their presence in the market. If they do not speak English, the cost of having survey staff proficient in a dozen or more languages would be prohibitive and advertisers likely would not need this degree of results. That's why there is no subset for Kreyol speakers in Miami, either.

So, if younger Asians are recruited, they get the same incentive 18-34 white, Black, or Hispanic males get.

 

[AM FM listener]

Even if Asians are good diary returners, or PPM wearers, I still think it would be interesting if they were a separate demographic, similar to Hispanic and African Americans, when it comes to breakouts along racial/ethnic lines. Are there just not enough Asian-Americans in the US to make such a breakout worthwhile?

 

[davideduardo]

Even if Asians are good diary returners, or PPM wearers, I still think it would be interesting if they were a separate demographic, similar to Hispanic and African Americans, when it comes to breakouts along racial/ethnic lines. Are there just not enough Asian-Americans in the US to make such a breakout worthwhile?

We need to remember that ratings are produced to provide advertisers, particularly agencies, with data that enables them to buy radio ads. The fragmentation of Asians into many national and linguistic groups makes the marketing efforts towards this segment of the population difficult.

Arbitron adds sample when there is agreement among subscribers to pay more. They might add an Asian ethnicity break were subscribers to ask for this and pay for it; until ad agencies require it, it's unlikely it will be done. And since significant (5% or more) Asian populations exist only in a small number of markets, it's unlikely larger advertisers will push the media to get these breakouts.

 

[PTBoardOp94]

If you look at the college population, Asians are 30 percent of the students in the UC school system. And how much of that population is Arbitron really sampling??

To the best of my knowledge, Arbitron does not recruit survey participants on college campuses.

 

[davideduardo]

If you look at the college population, Asians are 30 percent of the students in the UC school system. And how much of that population is Arbitron really sampling??

To the best of my knowledge, Arbitron does not recruit survey participants on college campuses.

While there is no recruitment physically on the campus (or, physically, anywhere else except in the Houston PPM), students with cellular phones or those in off-campus housing can be recruited.