:-(
A former member
so lets just say 2 out of 5000 wachted a show and it got poor rating BUT infant Over 7 million watched it
this is not a good system!
this is not a good system!
BA
I find it silly when people go ape over rating , They'll never be accurate Until they get fitted into all cable boxes .
RO
I used to be on their panel up until a few years ago, when AGB's equipment (at the time at least) couldn't handle digital cable, or our mess of SCART switch boxes and various recorders etc.
It was super fun. I used to be '6', until they changed to letters, then I became 'F'. That was less fun. Could this be the worst-constructed post this forum has ever seen?
It was super fun. I used to be '6', until they changed to letters, then I became 'F'. That was less fun. Could this be the worst-constructed post this forum has ever seen?
NB
How does that make sense? I'm not sure I follow.
If it got 7 million then it would have to be incredibly unlucky to get just 2/5000 people tuning in. There's going to be some difference in the figures but I'd imagine a sample of 5000 people would be a reasonable indicator.
The real difference is probably made in the methodology, things like the minimum amount of time it takes for a 'viewing' to occur, and the time intervals these values are taken which may give misleading results.
623058 posted:
so lets just say 2 out of 5000 wachted a show and it got poor rating BUT infant Over 7 million watched it
this is not a good system!
this is not a good system!
How does that make sense? I'm not sure I follow.
If it got 7 million then it would have to be incredibly unlucky to get just 2/5000 people tuning in. There's going to be some difference in the figures but I'd imagine a sample of 5000 people would be a reasonable indicator.
The real difference is probably made in the methodology, things like the minimum amount of time it takes for a 'viewing' to occur, and the time intervals these values are taken which may give misleading results.
CW
Although demographics are of course considered (it's not just 5000 homes taken at random 62334424whatever) I don't consider 5000 homes as much of a representative sample of the entire population of the country.
It's probably fine for primetime, promoted parts of the schedule, but I do seriously wonder about quieter parts of the schedule, such as late nights, where I think the sample just isn't wide enough to deliver a meanigful result.
When a niche programme on a niche channel has a small, but visible and evident following, but the given programme gets zero-rated by BARB, then they need a bigger sample.
Considering that BARB have technology at their core to calculate ratings, I do think they could respond more quickly to changes in technology.
Until comparitively recently a programme had to be watched 'live' in order to count in the ratings - recordings were not considered until *years* after domestic recording was commonplace.
Similarly, I do wonder how long we'll have to wait until VOD services get BARB ratings.
cwathen
Founding member
Quote:
If it got 7 million then it would have to be incredibly unlucky to get just 2/5000 people tuning in. There's going to be some difference in the figures but I'd imagine a sample of 5000 people would be a reasonable indicator.
Although demographics are of course considered (it's not just 5000 homes taken at random 62334424whatever) I don't consider 5000 homes as much of a representative sample of the entire population of the country.
It's probably fine for primetime, promoted parts of the schedule, but I do seriously wonder about quieter parts of the schedule, such as late nights, where I think the sample just isn't wide enough to deliver a meanigful result.
When a niche programme on a niche channel has a small, but visible and evident following, but the given programme gets zero-rated by BARB, then they need a bigger sample.
Quote:
The real difference is probably made in the methodology, things like the minimum amount of time it takes for a 'viewing' to occur, and the time intervals these values are taken which may give misleading results.
Considering that BARB have technology at their core to calculate ratings, I do think they could respond more quickly to changes in technology.
Until comparitively recently a programme had to be watched 'live' in order to count in the ratings - recordings were not considered until *years* after domestic recording was commonplace.
Similarly, I do wonder how long we'll have to wait until VOD services get BARB ratings.
JE
Jez
Founding member
Ive never been part of it or known of anyone who has. My mum used to fill in a questionnaire type form on a monthly basis many years ago which was basically tv schedules for a month and she had to tick which programmes she had watched and rate them. I think this was more to do with market research though.
I would think ratings are very inaccurate the way they do it but someone was saying they are infact very acurate?
I would think ratings are very inaccurate the way they do it but someone was saying they are infact very acurate?
CW
BARB themselves argue that it's very accurate - but then they would.
What I think would be better is if they had two samples of 5000 homes distributed amongst the same demographics but each independently generating their own ratings. Ratings between the two would have to be more or less the same in order for it to be considered a valid ratings, if they varied greatly then both would be disregarded and they should report that they can't sample an accurate rating for that programme. If the process is as accurate as they claim then this will never happen.
cwathen
Founding member
Quote:
I would think ratings are very inaccurate the way they do it but someone was saying they are infact very acurate?
BARB themselves argue that it's very accurate - but then they would.
What I think would be better is if they had two samples of 5000 homes distributed amongst the same demographics but each independently generating their own ratings. Ratings between the two would have to be more or less the same in order for it to be considered a valid ratings, if they varied greatly then both would be disregarded and they should report that they can't sample an accurate rating for that programme. If the process is as accurate as they claim then this will never happen.
BR
Have you ever read this forum before?
I'm sure it's all been statistically assessed etc. It can never be 100% accurate, but I imagine they've chosen a sample which is at least 95% accurate, which is often the bench mark for these type of things.
Very roughly, but with around 11,000 in the sample - if we take the population to be 55 million, each person in the sample would represent 10,000 viewers.
In terms of the big ratings winners, it should be pretty accurate for shows attracting figures in the millions, but those at the lower end of the scale such as the news channels and smaller digital channels, it's not going to be too accurate when a rating of 50,000 is based on just 5 people - possibly all in one house!
As we go through the digital revolution, has it ever been proposed the all Freeview/Sky/Cable boxes send back viewing data - I've never seen anything suggesting it.
It's possible Sky and Cable could do it already, but I'd assume they'd need the subscribers consent.
Barney Boo posted:
I used to be on their panel up until a few years ago, when AGB's equipment (at the time at least) couldn't handle digital cable, or our mess of SCART switch boxes and various recorders etc.
It was super fun. I used to be '6', until they changed to letters, then I became 'F'. That was less fun. Could this be the worst-constructed post this forum has ever seen?
It was super fun. I used to be '6', until they changed to letters, then I became 'F'. That was less fun. Could this be the worst-constructed post this forum has ever seen?
Have you ever read this forum before?
I'm sure it's all been statistically assessed etc. It can never be 100% accurate, but I imagine they've chosen a sample which is at least 95% accurate, which is often the bench mark for these type of things.
Very roughly, but with around 11,000 in the sample - if we take the population to be 55 million, each person in the sample would represent 10,000 viewers.
In terms of the big ratings winners, it should be pretty accurate for shows attracting figures in the millions, but those at the lower end of the scale such as the news channels and smaller digital channels, it's not going to be too accurate when a rating of 50,000 is based on just 5 people - possibly all in one house!
As we go through the digital revolution, has it ever been proposed the all Freeview/Sky/Cable boxes send back viewing data - I've never seen anything suggesting it.
It's possible Sky and Cable could do it already, but I'd assume they'd need the subscribers consent.
RO
As we go through the digital revolution, has it ever been proposed the all Freeview/Sky/Cable boxes send back viewing data - I've never seen anything suggesting it.
It's possible Sky and Cable could do it already, but I'd assume they'd need the subscribers consent.
There's another problem in that the data contains a lot of demographics as well - each person watching TV registers that they're watching on the meter, and that kind of pro-active interaction would be hard to force on people - if you make it optional, you end up having to make sure its demographically representative, and then you're right back to where we are now.
Brekkie Boy posted:
As we go through the digital revolution, has it ever been proposed the all Freeview/Sky/Cable boxes send back viewing data - I've never seen anything suggesting it.
It's possible Sky and Cable could do it already, but I'd assume they'd need the subscribers consent.
There's another problem in that the data contains a lot of demographics as well - each person watching TV registers that they're watching on the meter, and that kind of pro-active interaction would be hard to force on people - if you make it optional, you end up having to make sure its demographically representative, and then you're right back to where we are now.