survey results??

[mutantpoodle]

it must be me...but I cannot understand why companies advertise on TV and use obviously fake survey results

there are many examples around, but last night I watched one that boasted that 88% of 227 people ..agreed/liked or whatever

as with the others I refer to...thats not a whole number of people...so cannot be true

why do they not say xx% agree
or
xx people

surely that way us targets would be able to accept it might be true!

or have I fallen into the trap..............which still escapes me!

[Dod101]

it must be me...but I cannot understand why companies advertise on TV and use obviously fake survey results

there are many examples around, but last night I watched one that boasted that 88% of 227 people ..agreed/liked or whatever

as with the others I refer to...thats not a whole number of people...so cannot be true

why do they not say xx% agree
or
xx people

surely that way us targets would be able to accept it might be true!

or have I fallen into the trap..............which still escapes me!

The percentage is presumably rounded to make it simpler for people to understand, After all 88% of 227 is 199.76, making your point, but presumably the actual result is 200 in this case. If they had said that 88.1057% of the people agreed, that would probably have a smaller impact. Quoting the sample size is I suppose meant to convey that it was a decent sized sample and 88% is simply meant to convey that it is an overwhelming number. Whether any of it is actually helpful or not is another matter.

Dod

[DrFfybes]

Quoting the sample size is I suppose meant to convey that it was a decent sized sample and 88% is simply meant to convey that it is an overwhelming number. Whether any of it is actually helpful or not is another matter.

Dod

ISTR this becoming a requirement for adverts that quote statistics, so they couldn't start the survey and then stop if 4 of the first 5 agreed with them.

it must be me...but I cannot understand why companies advertise on TV and use obviously fake survey results

No, they have to provide the survey details if challenged.

I just checked, the first google hit for "how do adverts get survey results" explains it

https://www.asa.org.uk/news/a-quick-gui ... rveys.html

Paul

[gryffron]

Comedian Dave Gorman did a whole program on this. He pointed out sometimes the survey results showed less-than-half actually agreed the product worked.

[mutantpoodle]

thank you for the replies
but why not just say 200 out of 227

mixing numbers with % achieves nothing and simply makes a nonsense of common sense...
or do they assume that most if us are too stupid to calculate that 88% is not a whole person (in this example)

[servodude]

it must be me...but I cannot understand why companies advertise on TV and use obviously fake survey results

there are many examples around, but last night I watched one that boasted that 88% of 227 people ..agreed/liked or whatever

as with the others I refer to...thats not a whole number of people...so cannot be true

why do they not say xx% agree
or
xx people

surely that way us targets would be able to accept it might be true!

or have I fallen into the trap..............which still escapes me!

The percentage is presumably rounded to make it simpler for people to understand, After all 88% of 227 is 199.76, making your point, but presumably the actual result is 200 in this case. If they had said that 88.1057% of the people agreed, that would probably have a smaller impact. Quoting the sample size is I suppose meant to convey that it was a decent sized sample and 88% is simply meant to convey that it is an overwhelming number. Whether any of it is actually helpful or not is another matter.

Dod

The concept is called "social proof" in the book "Influence: The Psychology of Persuasion" by Robert Cialdini
- basically it boils down to "8 out of 10 cats do it so I don't need to bother thinking about it"
- and use precise numbers to make it look "science-y" so folk don't think you've just made it up (same reason you throw in facts)

Very good book
-sd

[modellingman]

thank you for the replies
but why not just say 200 out of 227

mixing numbers with % achieves nothing and simply makes a nonsense of common sense...
or do they assume that most if us are too stupid to calculate that 88% is not a whole person (in this example)

The whole point of surveys, which are a form of sampling, is to get insights into a population by measuring not the population as a whole but a much smaller sample. Much of the data published by, for example, the Office for National Statistics, is based on surveys such as the Labour Force Survey.

The 88% derived from the survey applies, within limits, to the broader population from which the sample survey has been drawn. The underlying message that the advertisement is trying to get across is that, if you buy this there is an 88% chance you are going to like it.

The problem with such surveys is not typically the sample size but the sample frame (ie how the sample has been selected) and whether the sample is representative of the target population as a whole. Opinion pollsters go to great lengths to try and get this right but don't always succeed.

[mutantpoodle]

thanks modellingman

I understand.....but I still find it annoying when it does not add up

why do they not suggest that 7.6 out of 10 cats......?.........because it sounds stupid!

[servodude]

thanks modellingman

I understand.....but I still find it annoying when it does not add up

why do they not suggest that 7.6 out of 10 cats......?.........because it sounds stupid!

More that that sounds impossible (without a chainsaw anyway)

[modellingman]

I understand.....but I still find it annoying when it does not add up

why do they not suggest that 7.6 out of 10 cats......?.........because it sounds stupid!

The marketing trick of using surveys to convince you to buy something, clearly isn't going to work with you. I wonder how many others share your annoyance - perhaps I could run a survey

[dionaeamuscipula]

thanks modellingman

I understand.....but I still find it annoying when it does not add up

why do they not suggest that 7.6 out of 10 cats......?.........because it sounds stupid!

Originally the 8 out of 10 cats claims was just that, a bald statement. It was backed up by surveys, but then it was discovered that in fact most cat owners said their cats didn't give a toss what brand their cat food was.

Therefore the tag line became "8 out of 10 cats whose owners expressed a preference"

The 8 of 10 tag is clearly so powerful a message that it has survived the need to explain its meaninglessness. It presumably speaks to cat owners partly because of its simplicity.

DM