-
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
Arbitron
- Thread starter pmack
- Start date
E
eggplantradio
Guest
Great question. One would think so but Arbitron's weighting process and unfair recruitment policies are not an attempt to get accurate results as they claim. They are pure social engineering, and a deliberate attempt to boost certain formats over others.
An example I've never gotten a straight answer about: McAllen-Brownsville market is 90% Hispanic, yet Arbitron still still employs these practices there, in order to reach the "under represented."
If you press your Arbitron rep hard enough about weighting, and bust all their talking points, they'll eventually say something like "There are much smarter people than me, with PHD's that say this is necessary to get an accurate count, maybe you should talk to them."
Good luck getting a straight answer.
> An article in Thursday's (Aug. 11) Dallas Morning News
> states that Whites are no longer the majority in Texas. So
> does that mean Arbitron will have to
> re-configure how they do ratings and start counting Whites'
> diairies as double entries and offer Whites more money to
> fill out diaries?
>
An example I've never gotten a straight answer about: McAllen-Brownsville market is 90% Hispanic, yet Arbitron still still employs these practices there, in order to reach the "under represented."
If you press your Arbitron rep hard enough about weighting, and bust all their talking points, they'll eventually say something like "There are much smarter people than me, with PHD's that say this is necessary to get an accurate count, maybe you should talk to them."
Good luck getting a straight answer.
> An article in Thursday's (Aug. 11) Dallas Morning News
> states that Whites are no longer the majority in Texas. So
> does that mean Arbitron will have to
> re-configure how they do ratings and start counting Whites'
> diairies as double entries and offer Whites more money to
> fill out diaries?
>
Very true. I had the same experience in a market that is 40% hispanic. In a certain demo where 70 diaries were returned, 60 of the 70 were from hispanic households. And there were many more examples like this. We even had a few demos where we couldn't even run a ranker for anglos, and these were your bread and butter key demos. We'd get an error message that said "not enough data" because they didn't collect enough diaries. And this is a market that's 60% Anglo according to Arbitron. When I asked why they didn't just weight the anglo diaries as they've done with black and hispanic diaries when this problem has come up (I knew what they'd say, but it was fun to watch them backpedal)...well, let's just say the spin machine was in full effect.
Arbitron is supposed to deliver a product where the percentages of the sample reflect the percentages of the market, but this has not been the case where I am for quite some time. If you have a low return for a certain group, which is not some new issue by the way, then you weight the diaries you do get so you can estimate the number. But for some strange reason that methodology doesn't seem to apply to anglos. Or if it does them I'm not aware of it. We're told basically the same thing when you press them.
> Great question. One would think so but Arbitron's weighting
> process and unfair recruitment policies are not an attempt
> to get accurate results as they claim. They are pure social
> engineering, and a deliberate attempt to boost certain
> formats over others.
>
> An example I've never gotten a straight answer about:
> McAllen-Brownsville market is 90% Hispanic, yet Arbitron
> still still employs these practices there, in order to reach
> the "under represented."
>
> If you press your Arbitron rep hard enough about weighting,
> and bust all their talking points, they'll eventually say
> something like "There are much smarter people than me, with
> PHD's that say this is necessary to get an accurate count,
> maybe you should talk to them."
>
> Good luck getting a straight answer.
>
>
>
> > An article in Thursday's (Aug. 11) Dallas Morning News
> > states that Whites are no longer the majority in Texas. So
>
> > does that mean Arbitron will have to
> > re-configure how they do ratings and start counting
> Whites'
> > diairies as double entries and offer Whites more money to
> > fill out diaries?
> >
>
Arbitron is supposed to deliver a product where the percentages of the sample reflect the percentages of the market, but this has not been the case where I am for quite some time. If you have a low return for a certain group, which is not some new issue by the way, then you weight the diaries you do get so you can estimate the number. But for some strange reason that methodology doesn't seem to apply to anglos. Or if it does them I'm not aware of it. We're told basically the same thing when you press them.
> Great question. One would think so but Arbitron's weighting
> process and unfair recruitment policies are not an attempt
> to get accurate results as they claim. They are pure social
> engineering, and a deliberate attempt to boost certain
> formats over others.
>
> An example I've never gotten a straight answer about:
> McAllen-Brownsville market is 90% Hispanic, yet Arbitron
> still still employs these practices there, in order to reach
> the "under represented."
>
> If you press your Arbitron rep hard enough about weighting,
> and bust all their talking points, they'll eventually say
> something like "There are much smarter people than me, with
> PHD's that say this is necessary to get an accurate count,
> maybe you should talk to them."
>
> Good luck getting a straight answer.
>
>
>
> > An article in Thursday's (Aug. 11) Dallas Morning News
> > states that Whites are no longer the majority in Texas. So
>
> > does that mean Arbitron will have to
> > re-configure how they do ratings and start counting
> Whites'
> > diairies as double entries and offer Whites more money to
> > fill out diaries?
> >
>
Re: Arbitron misinterpretation and misunderstanding.
> Great question. One would think so but Arbitron's weighting
> process and unfair recruitment policies are not an attempt
> to get accurate results as they claim. They are pure social
> engineering, and a deliberate attempt to boost certain
> formats over others.
Weighting is a process by which you get true proportionality in a survey... any survey.
For example, if you want half women and half men, but end up with 50 women and 40 men, you tabulate the male responses by 1.25 so the men are represented at 50% of the sample.
This is exactly what Arbitron does.
The weighting is actually minimal in the Arbitron cells since they are constantly trying to get proportionality through all 12 weeks of a survey.
> An example I've never gotten a straight answer about:
> McAllen-Brownsville market is 90% Hispanic, yet Arbitron
> still still employs these practices there, in order to reach
> the "under represented."
Hispanics are not over weighted in the LGRV. Arbitron uses DST, differential survery treatment, to incentivize groups that do not return diaries as well as the norm. that means they give each person a larger monetary incentive and more phone follow up. DST participants are not weighted any differently than any other participant. DST is used to enhance participantion, not to overweight certain groups.
>
> If you press your Arbitron rep hard enough about weighting,
> and bust all their talking points, they'll eventually say
> something like "There are much smarter people than me, with
> PHD's that say this is necessary to get an accurate count,
> maybe you should talk to them."
That is not true. Every Arbitron book and every download of Arbitrends has a sample report where they show the percentage of each cell in the market and the dpercentage of diaries they got back from the cell. For example, if 18-24 men is 5.8% of the market, they might get 5.6% return, and weigh up the returned diaries so the total is equal to 5.8%, and is proporitonal to the market. It is really very simple.
>
> Good luck getting a straight answer.
Hey, it is in every printed book and in every download. You don't need luck... you just have to be a subscriber and have a minimal understanding of polling.
> Great question. One would think so but Arbitron's weighting
> process and unfair recruitment policies are not an attempt
> to get accurate results as they claim. They are pure social
> engineering, and a deliberate attempt to boost certain
> formats over others.
Weighting is a process by which you get true proportionality in a survey... any survey.
For example, if you want half women and half men, but end up with 50 women and 40 men, you tabulate the male responses by 1.25 so the men are represented at 50% of the sample.
This is exactly what Arbitron does.
The weighting is actually minimal in the Arbitron cells since they are constantly trying to get proportionality through all 12 weeks of a survey.
> An example I've never gotten a straight answer about:
> McAllen-Brownsville market is 90% Hispanic, yet Arbitron
> still still employs these practices there, in order to reach
> the "under represented."
Hispanics are not over weighted in the LGRV. Arbitron uses DST, differential survery treatment, to incentivize groups that do not return diaries as well as the norm. that means they give each person a larger monetary incentive and more phone follow up. DST participants are not weighted any differently than any other participant. DST is used to enhance participantion, not to overweight certain groups.
>
> If you press your Arbitron rep hard enough about weighting,
> and bust all their talking points, they'll eventually say
> something like "There are much smarter people than me, with
> PHD's that say this is necessary to get an accurate count,
> maybe you should talk to them."
That is not true. Every Arbitron book and every download of Arbitrends has a sample report where they show the percentage of each cell in the market and the dpercentage of diaries they got back from the cell. For example, if 18-24 men is 5.8% of the market, they might get 5.6% return, and weigh up the returned diaries so the total is equal to 5.8%, and is proporitonal to the market. It is really very simple.
>
> Good luck getting a straight answer.
Hey, it is in every printed book and in every download. You don't need luck... you just have to be a subscriber and have a minimal understanding of polling.
Re: Arbitron misinterpretation and misunderstanding.
> Very true. I had the same experience in a market that is
> 40% hispanic. In a certain demo where 70 diaries were
> returned, 60 of the 70 were from hispanic households. And
> there were many more examples like this.
In every case, the diaries returned will be wieghted to achieve proportionality. The Hispanic diaries will be weighted down, and the "other" diaries weighted up.
> We even had a few
> demos where we couldn't even run a ranker for anglos, and
> these were your bread and butter key demos.
Of course, you want to blame Arbitron. Arbitron will increase sample size to avoid this kind of thing if the subscribers will pay for it. Since a reduction by half of the margin of error requires 4 times the sample, usually the stations will not pay.
You get what you pay for. If the sample is not big enough, it is not Arbitron's fault that this stuff happens. If a key cell only has 70 diaries, the sample is woefully small.
> We'd get an
> error message that said "not enough data" because they
> didn't collect enough diaries.
This is pretty common, even in the larger markets. It does not mean they did not "collect enough diaries." The collected the number of diaries the subscribers paid for. But in some cells, that sample does not allow very deep breaks... such as a narrow demo in one sex and one ethnicity in one daypart. If you look at LA, Hispanic females 55-64, Saturday, 7-mid, you get the message. LA has 7,500 diaries. You can not get certain data if you try to slice it too thin. It is not a defect of data collection... it is the cosequence of the agreed on sample size.
> And this is a market that's
> 60% Anglo according to Arbitron.
Arbitron does not use the term "anglo" anywhere. Non-Hispanics and non-Blacks are in the "other" category which includes Asians, non-Hiispanic whites and anything elese that does nto fit the first two.
> When I asked why they
> didn't just weight the anglo diaries as they've done with
> black and hispanic diaries when this problem has come up (I
> knew what they'd say, but it was fun to watch them
> backpedal)...well, let's just say the spin machine was in
> full effect.
The will not give an estimate if there is not a minimum number of diaries with data in the cell. That is done pre-weighting. If there are not enough, you get the error message. It is really enormously simple.
>
> Arbitron is supposed to deliver a product where the
> percentages of the sample reflect the percentages of the
> market, but this has not been the case where I am for quite
> some time.
What is you PPDV? Did you market accept the sample increas initiatives of abuout 5 years ago?
> If you have a low return for a certain group,
> which is not some new issue by the way, then you weight the
> diaries you do get so you can estimate the number. But for
> some strange reason that methodology doesn't seem to apply
> to anglos. Or if it does them I'm not aware of it. We're
> told basically the same thing when you press them.
It applies to every cell. And the data on sample and wighting is in the front of every book, and in every trend download if you are a continuous measurement market.
You might download and read the Pueple Book about methodology, as some of these issues are very clearly explained there.
> Very true. I had the same experience in a market that is
> 40% hispanic. In a certain demo where 70 diaries were
> returned, 60 of the 70 were from hispanic households. And
> there were many more examples like this.
In every case, the diaries returned will be wieghted to achieve proportionality. The Hispanic diaries will be weighted down, and the "other" diaries weighted up.
> We even had a few
> demos where we couldn't even run a ranker for anglos, and
> these were your bread and butter key demos.
Of course, you want to blame Arbitron. Arbitron will increase sample size to avoid this kind of thing if the subscribers will pay for it. Since a reduction by half of the margin of error requires 4 times the sample, usually the stations will not pay.
You get what you pay for. If the sample is not big enough, it is not Arbitron's fault that this stuff happens. If a key cell only has 70 diaries, the sample is woefully small.
> We'd get an
> error message that said "not enough data" because they
> didn't collect enough diaries.
This is pretty common, even in the larger markets. It does not mean they did not "collect enough diaries." The collected the number of diaries the subscribers paid for. But in some cells, that sample does not allow very deep breaks... such as a narrow demo in one sex and one ethnicity in one daypart. If you look at LA, Hispanic females 55-64, Saturday, 7-mid, you get the message. LA has 7,500 diaries. You can not get certain data if you try to slice it too thin. It is not a defect of data collection... it is the cosequence of the agreed on sample size.
> And this is a market that's
> 60% Anglo according to Arbitron.
Arbitron does not use the term "anglo" anywhere. Non-Hispanics and non-Blacks are in the "other" category which includes Asians, non-Hiispanic whites and anything elese that does nto fit the first two.
> When I asked why they
> didn't just weight the anglo diaries as they've done with
> black and hispanic diaries when this problem has come up (I
> knew what they'd say, but it was fun to watch them
> backpedal)...well, let's just say the spin machine was in
> full effect.
The will not give an estimate if there is not a minimum number of diaries with data in the cell. That is done pre-weighting. If there are not enough, you get the error message. It is really enormously simple.
>
> Arbitron is supposed to deliver a product where the
> percentages of the sample reflect the percentages of the
> market, but this has not been the case where I am for quite
> some time.
What is you PPDV? Did you market accept the sample increas initiatives of abuout 5 years ago?
> If you have a low return for a certain group,
> which is not some new issue by the way, then you weight the
> diaries you do get so you can estimate the number. But for
> some strange reason that methodology doesn't seem to apply
> to anglos. Or if it does them I'm not aware of it. We're
> told basically the same thing when you press them.
It applies to every cell. And the data on sample and wighting is in the front of every book, and in every trend download if you are a continuous measurement market.
You might download and read the Pueple Book about methodology, as some of these issues are very clearly explained there.
Re: Arbitron misinterpretation and misunderstanding.
> "Arbitron does not use the term "anglo" anywhere.
> Non-Hispanics and non-Blacks are in the "other" category
> which includes Asians, non-Hiispanic whites and anything
> elese that does nto fit the first two."
>
> Well, isn't that rather racist?
No, actually not. "Anglo" is not the description of all persons who are not Hispanic or Black, but "other" fits it well.
> "Arbitron does not use the term "anglo" anywhere.
> Non-Hispanics and non-Blacks are in the "other" category
> which includes Asians, non-Hiispanic whites and anything
> elese that does nto fit the first two."
>
> Well, isn't that rather racist?
No, actually not. "Anglo" is not the description of all persons who are not Hispanic or Black, but "other" fits it well.
E
eggplantradio
Guest
Re: Arbitron misinterpretation and misunderstanding.
David really speaks the Arbitron rep Spin very well. Too bad you've drunk the Kool Aid to the point that you cannot see how badly flawed the system is.
You are very right about their methodology, however it is a flawed methodology, and you cannot cannot escape the fact that there is significant social engineering going on when only special groups are given additional incentives to participate. Then once they do, they are counted in special ways that always enhance two formats. Black and Spanish.
Spin away.
> > Very tru
e. I had the same experience in a market that is
> > 40% hispanic. In a certain demo where 70 diaries were
> > returned, 60 of the 70 were from hispanic households. And
>
> > there were many more examples like this.
>
> In every case, the diaries returned will be wieghted to
> achieve proportionality. The Hispanic diaries will be
> weighted down, and the "other" diaries weighted up.
>
> > We even had a few
> > demos where we couldn't even run a ranker for anglos, and
> > these were your bread and butter key demos.
>
> Of course, you want to blame Arbitron. Arbitron will
> increase sample size to avoid this kind of thing if the
> subscribers will pay for it. Since a reduction by half of
> the margin of error requires 4 times the sample, usually the
> stations will not pay.
>
> You get what you pay for. If the sample is not big enough,
> it is not Arbitron's fault that this stuff happens. If a
> key cell only has 70 diaries, the sample is woefully small.
>
>
> > We'd get an
> > error message that said "not enough data" because they
> > didn't collect enough diaries.
>
> This is pretty common, even in the larger markets. It does
> not mean they did not "collect enough diaries." The
> collected the number of diaries the subscribers paid for.
> But in some cells, that sample does not allow very deep
> breaks... such as a narrow demo in one sex and one ethnicity
> in one daypart. If you look at LA, Hispanic females 55-64,
> Saturday, 7-mid, you get the message. LA has 7,500 diaries.
> You can not get certain data if you try to slice it too
> thin. It is not a defect of data collection... it is the
> cosequence of the agreed on sample size.
>
> > And this is a market that's
> > 60% Anglo according to Arbitron.
>
> Arbitron does not use the term "anglo" anywhere.
> Non-Hispanics and non-Blacks are in the "other" category
> which includes Asians, non-Hiispanic whites and anything
> elese that does nto fit the first two.
>
> > When I asked why they
> > didn't just weight the anglo diaries as they've done with
> > black and hispanic diaries when this problem has come up
> (I
> > knew what they'd say, but it was fun to watch them
> > backpedal)...well, let's just say the spin machine was in
> > full effect.
>
> The will not give an estimate if there is not a minimum
> number of diaries with data in the cell. That is done
> pre-weighting. If there are not enough, you get the error
> message. It is really enormously simple.
> >
> > Arbitron is supposed to deliver a product where the
> > percentages of the sample reflect the percentages of the
> > market, but this has not been the case where I am for
> quite
> > some time.
>
> What is you PPDV? Did you market accept the sample increas
> initiatives of abuout 5 years ago?
>
> > If you have a low return for a certain group,
> > which is not some new issue by the way, then you weight
> the
> > diaries you do get so you can estimate the number. But
> for
> > some strange reason that methodology doesn't seem to apply
>
> > to anglos. Or if it does them I'm not aware of it. We're
>
> > told basically the same thing when you press them.
>
> It applies to every cell. And the data on sample and
> wighting is in the front of every book, and in every trend
> download if you are a continuous measurement market.
>
> You might download and read the Pueple Book about
> methodology, as some of these issues are very clearly
> explained there.
>
David really speaks the Arbitron rep Spin very well. Too bad you've drunk the Kool Aid to the point that you cannot see how badly flawed the system is.
You are very right about their methodology, however it is a flawed methodology, and you cannot cannot escape the fact that there is significant social engineering going on when only special groups are given additional incentives to participate. Then once they do, they are counted in special ways that always enhance two formats. Black and Spanish.
Spin away.
> > Very tru
e. I had the same experience in a market that is
> > 40% hispanic. In a certain demo where 70 diaries were
> > returned, 60 of the 70 were from hispanic households. And
>
> > there were many more examples like this.
>
> In every case, the diaries returned will be wieghted to
> achieve proportionality. The Hispanic diaries will be
> weighted down, and the "other" diaries weighted up.
>
> > We even had a few
> > demos where we couldn't even run a ranker for anglos, and
> > these were your bread and butter key demos.
>
> Of course, you want to blame Arbitron. Arbitron will
> increase sample size to avoid this kind of thing if the
> subscribers will pay for it. Since a reduction by half of
> the margin of error requires 4 times the sample, usually the
> stations will not pay.
>
> You get what you pay for. If the sample is not big enough,
> it is not Arbitron's fault that this stuff happens. If a
> key cell only has 70 diaries, the sample is woefully small.
>
>
> > We'd get an
> > error message that said "not enough data" because they
> > didn't collect enough diaries.
>
> This is pretty common, even in the larger markets. It does
> not mean they did not "collect enough diaries." The
> collected the number of diaries the subscribers paid for.
> But in some cells, that sample does not allow very deep
> breaks... such as a narrow demo in one sex and one ethnicity
> in one daypart. If you look at LA, Hispanic females 55-64,
> Saturday, 7-mid, you get the message. LA has 7,500 diaries.
> You can not get certain data if you try to slice it too
> thin. It is not a defect of data collection... it is the
> cosequence of the agreed on sample size.
>
> > And this is a market that's
> > 60% Anglo according to Arbitron.
>
> Arbitron does not use the term "anglo" anywhere.
> Non-Hispanics and non-Blacks are in the "other" category
> which includes Asians, non-Hiispanic whites and anything
> elese that does nto fit the first two.
>
> > When I asked why they
> > didn't just weight the anglo diaries as they've done with
> > black and hispanic diaries when this problem has come up
> (I
> > knew what they'd say, but it was fun to watch them
> > backpedal)...well, let's just say the spin machine was in
> > full effect.
>
> The will not give an estimate if there is not a minimum
> number of diaries with data in the cell. That is done
> pre-weighting. If there are not enough, you get the error
> message. It is really enormously simple.
> >
> > Arbitron is supposed to deliver a product where the
> > percentages of the sample reflect the percentages of the
> > market, but this has not been the case where I am for
> quite
> > some time.
>
> What is you PPDV? Did you market accept the sample increas
> initiatives of abuout 5 years ago?
>
> > If you have a low return for a certain group,
> > which is not some new issue by the way, then you weight
> the
> > diaries you do get so you can estimate the number. But
> for
> > some strange reason that methodology doesn't seem to apply
>
> > to anglos. Or if it does them I'm not aware of it. We're
>
> > told basically the same thing when you press them.
>
> It applies to every cell. And the data on sample and
> wighting is in the front of every book, and in every trend
> download if you are a continuous measurement market.
>
> You might download and read the Pueple Book about
> methodology, as some of these issues are very clearly
> explained there.
>
Re: Arbitron misinterpretation and misunderstanding.
> No, actually not. "Anglo" is not the description of all
> persons who are not Hispanic or Black, but "other" fits it
> well.
>
You don't honestly believe what you just said, do you? You are WAY too far into the forest to see the trees! I would not put myself into a Hispanic or Black category, so am I "other"? Gee, where do those of "other" descent hail from? Any system that uses race to make determinations is RACIST. Period. I bet you would be singing a different tune if Arbitron gave more incentives to Whites than to those of "other" races.
> No, actually not. "Anglo" is not the description of all
> persons who are not Hispanic or Black, but "other" fits it
> well.
>
You don't honestly believe what you just said, do you? You are WAY too far into the forest to see the trees! I would not put myself into a Hispanic or Black category, so am I "other"? Gee, where do those of "other" descent hail from? Any system that uses race to make determinations is RACIST. Period. I bet you would be singing a different tune if Arbitron gave more incentives to Whites than to those of "other" races.
K
KPLEXCOMPLEX
Guest
Re: Arbitron misinterpretation and misunderstanding.
> > No, actually not. "Anglo" is not the description of all
> > persons who are not Hispanic or Black, but "other" fits it
>
> > well.
> >
> You don't honestly believe what you just said, do you? You
> are WAY too far into the forest to see the trees! I would
> not put myself into a Hispanic or Black category, so am I
> "other"? Gee, where do those of "other" descent hail from?
> Any system that uses race to make determinations is RACIST.
> Period. I bet you would be singing a different tune if
> Arbitron gave more incentives to Whites than to those of
> "other" races.
>
You make some very good points. Now the answer will probably be,"Whites" are "other" because there are more "white" races from Europe,and parts of Asia, than any other "culture".Too many in fact, so it's Other". Still simply being called "other" is dismissive in tone.
> > No, actually not. "Anglo" is not the description of all
> > persons who are not Hispanic or Black, but "other" fits it
>
> > well.
> >
> You don't honestly believe what you just said, do you? You
> are WAY too far into the forest to see the trees! I would
> not put myself into a Hispanic or Black category, so am I
> "other"? Gee, where do those of "other" descent hail from?
> Any system that uses race to make determinations is RACIST.
> Period. I bet you would be singing a different tune if
> Arbitron gave more incentives to Whites than to those of
> "other" races.
>
You make some very good points. Now the answer will probably be,"Whites" are "other" because there are more "white" races from Europe,and parts of Asia, than any other "culture".Too many in fact, so it's Other". Still simply being called "other" is dismissive in tone.
Re: Arbitron misinterpretation and misunderstanding.
> David really speaks the Arbitron rep Spin very well. Too bad
> you've drunk the Kool Aid to the point that you cannot see
> how badly flawed the system is.
The system is not badly flawed. The diary method is the accepted radio measurement tool worldwide in any nation where there is adequate literacy for it to work. The sampling technique and the tabulations are also impecable.
The only improvement would be electronic measurement. The reason this has not moved faster is the same reason Arbitron has small samples in some markets: cost. Clients determine whether the PPM will be implemented and whether samples will increase.
Interestingly, the people meter test in Philly sowed that the diary was an excellent tool, and highly accurate.
>
> You are very right about their methodology, however it is a
> flawed methodology,
There is nothing flawed about the concept of using a random probability sample to recruit, a diary to collect and the tabulationmethods Arbitron uses to present. All are used every day in every field of research into consumer, voter and specialized opinion and behaviour.
> and you cannot cannot escape the fact
> that there is significant social engineering going on when
> only special groups are given additional incentives to
> participate.
The main group incentivized is white 18/24 males, who otherwise get undersampled. The ONLY reason for higher incentives is to get true proportional diary returns within acceptable margins. This technique is approved by the MRC, which audits Arbitron and Nielsen on behalf of the advertisers who will spend about $22 billion on radio this year. And they endorse the methodology from start to finish.
> Then once they do, they are counted in special
> ways that always enhance two formats. Black and Spanish.
>
> Spin away.
> David really speaks the Arbitron rep Spin very well. Too bad
> you've drunk the Kool Aid to the point that you cannot see
> how badly flawed the system is.
The system is not badly flawed. The diary method is the accepted radio measurement tool worldwide in any nation where there is adequate literacy for it to work. The sampling technique and the tabulations are also impecable.
The only improvement would be electronic measurement. The reason this has not moved faster is the same reason Arbitron has small samples in some markets: cost. Clients determine whether the PPM will be implemented and whether samples will increase.
Interestingly, the people meter test in Philly sowed that the diary was an excellent tool, and highly accurate.
>
> You are very right about their methodology, however it is a
> flawed methodology,
There is nothing flawed about the concept of using a random probability sample to recruit, a diary to collect and the tabulationmethods Arbitron uses to present. All are used every day in every field of research into consumer, voter and specialized opinion and behaviour.
> and you cannot cannot escape the fact
> that there is significant social engineering going on when
> only special groups are given additional incentives to
> participate.
The main group incentivized is white 18/24 males, who otherwise get undersampled. The ONLY reason for higher incentives is to get true proportional diary returns within acceptable margins. This technique is approved by the MRC, which audits Arbitron and Nielsen on behalf of the advertisers who will spend about $22 billion on radio this year. And they endorse the methodology from start to finish.
> Then once they do, they are counted in special
> ways that always enhance two formats. Black and Spanish.
>
> Spin away.
Re: Arbitron misinterpretation and misunderstanding.
> >
> You make some very good points. Now the answer will probably
> be,"Whites" are "other" because there are more "white" races
> from Europe,and parts of Asia, than any other "culture".Too
> many in fact, so it's Other". Still simply being called
> "other" is dismissive in tone.
>
1. Not all whites are in the "other" category. Hispanic whites are in the Hispanic category.
2. Other inculdes non-Hispanic whites, Asians, native Americans, etc. In other words, everone but Blacks and Hispanics.
3. The reason it is that way is, again, that is what the clients of Arbitron want.
> >
> You make some very good points. Now the answer will probably
> be,"Whites" are "other" because there are more "white" races
> from Europe,and parts of Asia, than any other "culture".Too
> many in fact, so it's Other". Still simply being called
> "other" is dismissive in tone.
>
1. Not all whites are in the "other" category. Hispanic whites are in the Hispanic category.
2. Other inculdes non-Hispanic whites, Asians, native Americans, etc. In other words, everone but Blacks and Hispanics.
3. The reason it is that way is, again, that is what the clients of Arbitron want.
Re: Arbitron misinterpretation and misunderstanding.
> Still simply being called
> > "other" is dismissive in tone.
> >
> AND discriminatory - AND racist.
Why would you say it is discriminatory? There is no exclusion or different weighting for any of the groups, so I don't understand the use of the term. And "other" is just a catch all for everyone who is not either Hispanic or Black... probably the best would be "all other" but the field probably is not long enough in the software.
> Still simply being called
> > "other" is dismissive in tone.
> >
> AND discriminatory - AND racist.
Why would you say it is discriminatory? There is no exclusion or different weighting for any of the groups, so I don't understand the use of the term. And "other" is just a catch all for everyone who is not either Hispanic or Black... probably the best would be "all other" but the field probably is not long enough in the software.
Re: Arbitron misinterpretation and misunderstanding.
> > 3. The reason it is that way is, again, that is what the
> > clients of Arbitron want.
> >
> Doesn't make it any less racist and discriminatory.
>
The clients want Black and Hispanic broken out separately. And they want all the other groups together. I don't, as I said in the other post on this matter, see how that discriminates as it neither takes nor adds to any one group's proportionality in the survey process.
> > 3. The reason it is that way is, again, that is what the
> > clients of Arbitron want.
> >
> Doesn't make it any less racist and discriminatory.
>
The clients want Black and Hispanic broken out separately. And they want all the other groups together. I don't, as I said in the other post on this matter, see how that discriminates as it neither takes nor adds to any one group's proportionality in the survey process.
G
Gus
Guest
Re: Arbitron misinterpretation and misunderstanding.
> > 3. The reason it is that way is, again, that is what the
> > clients of Arbitron want.
> >
> Doesn't make it any less racist and discriminatory.
Shut the pie hole Pmack.
>
> > 3. The reason it is that way is, again, that is what the
> > clients of Arbitron want.
> >
> Doesn't make it any less racist and discriminatory.
Shut the pie hole Pmack.
>
Re: Arbitron misinterpretation and misunderstanding.
And "other" is just a
> catch all for everyone who is not either Hispanic or
> Black... probably the best would be "all other" but the
> field probably is not long enough in the software.
>
Geez, my friend, are you that much of a racist? So, you're okay putting me into an ethnic category of "OTHER"? The problem is using race to determine anything - PERIOD. To use race makes it a racist system. How would you feel being put into an "OTHER" category because you didn't fit into the "WHITE" category? It's a slap in the face to the 60's civil rights movement - you've got your "WHITES-only" water fountain and then one for "OTHER" - is that OKAY?
And "other" is just a
> catch all for everyone who is not either Hispanic or
> Black... probably the best would be "all other" but the
> field probably is not long enough in the software.
>
Geez, my friend, are you that much of a racist? So, you're okay putting me into an ethnic category of "OTHER"? The problem is using race to determine anything - PERIOD. To use race makes it a racist system. How would you feel being put into an "OTHER" category because you didn't fit into the "WHITE" category? It's a slap in the face to the 60's civil rights movement - you've got your "WHITES-only" water fountain and then one for "OTHER" - is that OKAY?
Re: Arbitron misinterpretation and misunderstanding.
> And "other" is just a
> > catch all for everyone who is not either Hispanic or
> > Black... probably the best would be "all other" but the
> > field probably is not long enough in the software.
> >
> Geez, my friend, are you that much of a racist? So, you're
> okay putting me into an ethnic category of "OTHER"? The
> problem is using race to determine anything - PERIOD. To use
> race makes it a racist system. How would you feel being put
> into an "OTHER" category because you didn't fit into the
> "WHITE" category? It's a slap in the face to the 60's civil
> rights movement - you've got your "WHITES-only" water
> fountain and then one for "OTHER" - is that OKAY?
>
Black is a race. Hispanic is not, as there are Black, white, Asian and Indigenous Hispanics. Other is all other persons, irrespective of race, who do not fit in either of the first two categories first.
There is no "white" category since that would have to include a huge percentage of Hispanics. This is why the US Census does not give "Hispanic" as an option on the "race question" on the Census form... it is a separate question separate from race. And this is why the correct term for what you describe is "non-Hispanic white."
> And "other" is just a
> > catch all for everyone who is not either Hispanic or
> > Black... probably the best would be "all other" but the
> > field probably is not long enough in the software.
> >
> Geez, my friend, are you that much of a racist? So, you're
> okay putting me into an ethnic category of "OTHER"? The
> problem is using race to determine anything - PERIOD. To use
> race makes it a racist system. How would you feel being put
> into an "OTHER" category because you didn't fit into the
> "WHITE" category? It's a slap in the face to the 60's civil
> rights movement - you've got your "WHITES-only" water
> fountain and then one for "OTHER" - is that OKAY?
>
Black is a race. Hispanic is not, as there are Black, white, Asian and Indigenous Hispanics. Other is all other persons, irrespective of race, who do not fit in either of the first two categories first.
There is no "white" category since that would have to include a huge percentage of Hispanics. This is why the US Census does not give "Hispanic" as an option on the "race question" on the Census form... it is a separate question separate from race. And this is why the correct term for what you describe is "non-Hispanic white."
- Status