That has been the nonstop buzz at WSB Radio Group this year. "PPM is coming in September." Now it IS September! IS PPM here YET?
-
Get involved.
We want your input!
Apply for Membership and join the conversations about everything related to broadcasting.
After we receive your registration, a moderator will review it. After your registration is approved, you will be permitted to post.
If you use a disposable or false email address, your registration will be rejected.
After your membership is approved, please take a minute to tell us a little bit about yourself.
https://www.radiodiscussions.com/forums/introduce-yourself.1088/
Thanks in advance and have fun!
RadioDiscussions Administrators
Portable People Meter's Atlanta debut
- Thread starter MCMagicCracker
- Start date
MCMagicCracker said:
The panel is built, precurrency measurement happens in October and in November, with a currency report for Decemberreleased in early January.
Summer will be the last diary book.
Keep in mind PPM books are for 28 days, not three months. There are no rolling or trended periods of more than a monthy.
Given a copy of Excel, one can create one's own rolling averages and trends from the PPM data.
C
CONSultant
Guest
If I recall correctly, Arbitron used to get back about 3300 diaries in Atlanta area. How many PPM's will be used? What is the methodology for PPM distribution - anyone know the specifics?
J
JoshuaC
Guest
I don't recall exact numbers, but I know it is FAR less PPM units than diaries (somewhere in the vicinity of a thousand I think?). The argument I heard at the SBE meeting where they discussed it was that the extended length that any one person is part of the panel means less people need to be on the panel... I don't really get the math of that. I must have missed that day in Statistics back in college.
That being said... they said that the theory is that even though people sign up for 2-year tours of PPM duty, the reality is that the panel will constantly be changing as people move in/out of the city, eligibility changes, etc etc etc.
Again... not a mathematician, just an engineer.
That being said... they said that the theory is that even though people sign up for 2-year tours of PPM duty, the reality is that the panel will constantly be changing as people move in/out of the city, eligibility changes, etc etc etc.
Again... not a mathematician, just an engineer.
The numbers of diaries over 3 months is misleading, but here's what Arbitron presented at its recent breakfast:
Atlanta Market:
Diaries: 337 weekly, 1,347 monthly, 4,040 quarterly
PPM: 1,335 weekly, 1,335 monthly, 1,335 quarterly
As you can see, the PPM has a much larger in-tab sample each week.
The way Arbitron explains it, the focus will move from number of participants to number of days of observation. With the PPM, each week will have 9,345 days of observation (1,335 X 7) compared to 2,359 for the diary (337 X 7). After 12 weeks, the PPM will generate 112,140 days of observation versus 27,699 for the diary.
As said farther up, October and November will be defined as pre-currency. After December, the October-December period will be considered currency. As far as a rolling average, Arbitron clients will be able to pull up anything they want. Due to the far greater number of participants each week, agencies probably will be willing to buy just on the past 3 months, whereas currently, they usually buy on a 2-book average. But that could be any consecutive 3 months they want, January-March, February-April or whatever.
Atlanta Market:
Diaries: 337 weekly, 1,347 monthly, 4,040 quarterly
PPM: 1,335 weekly, 1,335 monthly, 1,335 quarterly
As you can see, the PPM has a much larger in-tab sample each week.
The way Arbitron explains it, the focus will move from number of participants to number of days of observation. With the PPM, each week will have 9,345 days of observation (1,335 X 7) compared to 2,359 for the diary (337 X 7). After 12 weeks, the PPM will generate 112,140 days of observation versus 27,699 for the diary.
As said farther up, October and November will be defined as pre-currency. After December, the October-December period will be considered currency. As far as a rolling average, Arbitron clients will be able to pull up anything they want. Due to the far greater number of participants each week, agencies probably will be willing to buy just on the past 3 months, whereas currently, they usually buy on a 2-book average. But that could be any consecutive 3 months they want, January-March, February-April or whatever.
C
CONSultant
Guest
Thanks for the clarification, Roddy and JoshuaC. Is Arbitron contacting potential PPMers via landlines or have they figured out a way to contact the cellphone only crowd?
CONSultant said:
As posted elsewhere, there are 1335 panelists in the market, ready to start the October pre-vurrency period.
Households are recruited by a random probability sample of phone homes and cellular-only homes, too. The entire houshold must be recruited, or the family is not used. Similarly, if anyone in a household drops out and can't be convinced to participate, the whole household is dropped and a new one put in.
All recruiting is on the phone, except in Houston, where some address based recruiting is used.
Randy, what was the reason for that?
Kabrich said:
How can a CPO sample be address based?
In any case, I think we have a simple difference in interpretation of terms. While a sample may be drawn from and address-based database, the recruiting in Houston is done on a combination of in-home (address based), phone and CPO recruit procedures.
The whole sample is DRAWN from addresses, but part of the recruitingiting may be in home and part by phone and part from a non-address drawn cell phone sample in proportion to the CDC percentage of cell phone only households.
One of the issues is that most of us who remember Personal Placement & Retrieval or who have worked in markets with in-home or intercept based radio ratings call "in home" the process of doing the interview or placement of the survey materials "at the front door." While Arbitron may use a geography based sample, the telephone is still a huge part of the actual survey technique.
In a recent conference call, Houston was described as being "partially address based" in recruit procedures. I'm asking clarification if the non-address based recruits are the cell phone sample or some of the non-cell sample. But from the July description of methodology, there is a mixed sample made up those identified from address based data who were recruited by phone and who were recruited in person.
In truth, even the SSI (or before it, Metromail) prefix based phone based samples were really address based since each phone prefix represents a geographic area if not a spcific set of overlapping addresses. The issue is whether the recruiting, not the database, is in person or on the phone. I think you are just nitpicking in that the Houston sample is not recruited 100% by address based recruiting as defined by doing the recruit at the home and not on the phone.
"Selection of Basics and Alternates
Addresses within the Houston-Galveston DMA are selected at random from the stratified frame using a systematic random sampling procedure. The required sampling interval (1:n) is calculated, and a random starting point is selected within the first interval. Each “nth” address is then selected to fulfill the required starting sample size. This first group of selected addresses serves as the predesignated Basic sample. Each Basic address constitutes a random geographic sampling point. Following the selection of Basic sampling points, a predetermined number of supplemental addresses are selected within the same zip+4 area as the Basic address. Three of these are designated as reserve Basics to serve as replacements in the event the first Basic proves to be ineligible to join the PPM panel (e.g., the address is a business, the address is demolished, the household is media-affiliated, etc.). The remaining addresses in each sampling point serve as Alternates in the event that an eligible Basic cannot be recruited or retained in the panel.
Alternates are, thus, always within close geographic proximity to the Basic. If the Basic household cannot be recruited or retained in the panel, it is replaced with an Alternate household. To be eligible to replace a Basic household, Alternate households must match the Basic on preassigned matching criteria. The matching criteria may include if the household subscribes to a cable television or satellite television service; the presence of children in the household under age 18; and/or householder’s employment status.
Alternate households must match the Basic on two of these criteria, as determined in advance by Arbitron. Arbitron makes extensive efforts to recruit the Basic household in each sampling point using mail, telephone and in-person methods, as described below. Alternates are recruited only after all
required steps to recruit or retain the Basic address in each sampling point have been exhausted.
Sample Preparation
Two months prior to recruitment, Arbitron mails Basic households the sample preparation questionnaire with a cash incentive, as well as a promised incentive for completion of the questionnaire. The
one-page questionnaire contains questions about media usage and asks for contact information, including the household’s phone number.
Telephone Recruitment Calls
Arbitron first attempts to recruit Basic households by telephone; all households are called for which Arbitron has secured a phone number (either from the questionnaire or provided by the vendor).
Every household that returns the sample prep questionnaire and includes their telephone number (as well as households for whom the vendor provided the phone number) is mailed a recruitment brochure with another cash incentive.
Prerecruitment Calls
Prerecruitment calls begin when recruitment brochures are mailed to the households. The purpose of
prerecruitment calling is to make contact with households in an effort to alert them that a recruitment brochure has been mailed to the household. During this call, Arbitron also secures a commitment from
the respondent to respond to the mailing when it is received and notifies the household that it will be recontacted in a few days. Shortly after the first prerecruitment call is made, the household is recontacted to verify the household’s address, secure a strong commitment on behalf of all
eligible household members to participate in the panel and arrange for the delivery of the household’s
equipment.
Field Recruitment
Field recruitment of Basic households may be attempted by an at-home visit from an Arbitron recruitment associate forBasic households that were called during telephone recruitment but did not agree to participate, Basic households with hometelephone contact was not established, and qualifying Alternate households.
story@11 said:
I agree with you that those probably will be the 2 stations most affected. Most likely, they get diary mentions because of their high awareness in the marketplace. Also, I have a feeling V-103's TSL is somewhat exaggerated with the diary. But will the differences be enough to affect their ranks? Time will tell.
Even though Star 94's ratings have fallen, I still wonder if Star gets diary mentions because of its high awareness.
Roddy, I gotta agree with you on that. Incumbents have a great advantage because of their high awareness in the market.
But is that only when filling out a diary?
I think an incumbent has an advantage even when PPM is used because of that awareness, especially when you realize that most radios still have presets.
TSL, however, will no longer be exaggerated. And, incorrect entries won't happen, either. I'm convinced that when diaries are used, a listener (especially a passive listener) might absent-mindedly enter the established station instead of the other station that they were actually listening to. After all, listeners don't care or pay attention to this stuff nearly as much as we do. lol
But is that only when filling out a diary?
I think an incumbent has an advantage even when PPM is used because of that awareness, especially when you realize that most radios still have presets.
TSL, however, will no longer be exaggerated. And, incorrect entries won't happen, either. I'm convinced that when diaries are used, a listener (especially a passive listener) might absent-mindedly enter the established station instead of the other station that they were actually listening to. After all, listeners don't care or pay attention to this stuff nearly as much as we do. lol
tshanks said:
Houston sample comes from a geographic based sample, and recruiting is done by a mix of in home and by phone contacts, while CPO homes are contacted on the cellphone.
If you are in engineering, you have to get the PowerPoint presentation about the edit rule for quarter hour credits. Have your Arbitron rep get it for you. It's interesting how the tags received by the meter are processed.
RoddyFreeman said:
Sorry for the delay - I don't live on this site like others do daily.
Regardless of what others will tell you, the Houston sample is 100% address based. This was done because at the time Nielsen was involved in the Houston test and Nielsen is address based with their sample. Bottom line is it leads a better geographical sample and avoids issues such as Cell Phone Only Households, VoIP households etc.
The sample is drawn from addresses. Contact is then made with that address. If they can find a telephone crossreference with the address, they will call and try to place the PPM. Otherwise, they will mail, fed-ex or even in some cases, show up at the front door to try and recruit the household. The Vast majority of addresses can be crossreferenced with a telelphone number and are contacted as such - thus people, including Pierre Bouvard, have mistakenly stated that the vast majority of Houston is recruited by telephone, which is where others on this board have confused themselves.
Anyone that says anything to the contrary is out of the loop, but of course Pierre also said that a Spanish station was in the Top 10 in NYC in PPM when in fact it wasn't.
However, this is very expensive and Arbitron does not want to do this in other markets. In fact, they want to reduce Houston to phone based sample. However, they cannot do that as a phone based sample is not MRC accrediated - and Houston address based sample is the only MRC Approved PPM Service they have. So they cannot just switch Houston as then they would have NO PPM Service MRC accrediated.
Kabrich said:
No, that is not so. The CPO sample is not address based as there is really no way of knowing where a cellular phone "lives."
You are making false conclusions. Nielsen was involved back to the Philadelphia tests, and they did not use address based sample or recruiting in person in that market. Much of the reason for the address based sample and mixed recruiting including in-home has to do with the fact that Philly is less than 5% Hispanic while Houston is closer to 30% and have a different national origin.
Some of it.
In Houston, a lot of homes, particulaly in HDxA areas, were visited for on site recruiting.
There is a very significant percentage of in-home recruiting in Houston. Ask Pierre. Depending on which spin script they are all on this week, they may tell you the actual percentage.
Kabrich said:
In his testimony in NY last week Bob Patchen said the Houston sample was close to 50/50 recruitment phone/in person. While I'm not sure how those percents have shifted over time, this may be because we have never seen data on that.
Half and half does not leave a majority either way, and that is excluding CPO's.
- Status