Here's another example: suppose you wanted to know how many people did or did not (i.e., a yes/no question) watch the Doctor Who Christmas Special. You only need a sample size of 16583* to be 99% sure of being within 1% of the true answer (mathematical fact!).
(*assuming a population of 60m)
Where does the 16,583 figure come from? If BARB only uses 11,000, then its still quite a fair bit below that.
Just to put this into perspective, putting aside your personal preferences do you genuinely believe that last year 40m people in the UK chose not to watch X Factor?
Can I believe that two thirds of the UK population didn't choose to watch the X Factor [final (I'm assuming)]*?
Absolutely.
Once you take out those too young to watch, those too old to be able to decide to watch, those who are working, those who are out, those who are shopping, those who don't have a TV and the myriad of other reasons for people to be unable to watch, you're not going to be left with another 20m people to double your rating. And that's before you even get to those who don't like the X Factor so found something else to do / watch.
* That's a different question to "chose not to watch X Factor", as I doubt hardly any of the 40m deliberately chose not to watch it. Most people don't decide to not do something so trivial.
The consolidated figure was 16.6m viewers- ie factoring in people who were unable to watch at initial broadcast. If there are supposedly 62m people in the UK, thats over 45m who didn't watch by choice (although in reality that figure will be slightly lower due to the peak).
A better example is the premiership matches. The BARB figures suggest that anything from 1m to 3m people watch those. Last I heard BARB didn't take into account mass viewings (ie pubs), it might be different now but assuming its still the same, do you believe the BARB figures to be accurate and football matches genuinely attract that amount?
The consolidated figure was 16.6m viewers- ie factoring in people who were unable to watch at initial broadcast. If there are supposedly 62m people in the UK, thats over 45m who didn't watch by choice.
No it's not. It's 45m people who were either busy, working, out or otherwise unavailable or who were doing something else. I'm confident that only a tiny percentage of those who didn't watch didn't because they didn't want to watch the X Factor. I know at least 100 people who didn't watch it because they were at work with me. None of them made a concious choice not to watch.
A better example is the premiership matches. The BARB figures suggest that anything from 1m to 3m people watch those. Last I heard BARB didn't take into account mass viewings (ie pubs), it might be different now but assuming its still the same, do you believe the BARB figures to be accurate and football matches genuinely attract that amount?
Is that a good or bad amount? I'd be impressed if 1m is the lowest figure.
No it's not. It's 45m people who were either busy, working, out or otherwise unavailable or who were doing something else. I'm confident that only a tiny percentage of those who didn't watch didn't because they didn't want to watch the X Factor. I know at least 100 people who didn't watch it because they were at work with me. None of them made a concious choice not to watch.
The figure I quoted was the official rating including people who recorded and watched it within 7 days, not just those who watched it on the night. Just researched properly and had my stats slightly off (didn't include HD). 17.2m watched the final live whilst the official number of people who watched was 17.7m, still leaving 42m in total who didn't watch.
Is that a good or bad amount? I'd be impressed if 1m is the lowest figure.
Good amount for Sky Sports 1, in fact any channel except BBC1 at that particular time. The figures released are more or less above 1m, sometimes a bit lower. But that's on the assumption that only those at home are watching which isn't true.
Where does the 16,583 figure come from? If BARB only uses 11,000, then its still quite a fair bit below that.
It's from an online statistical calculator (two, in fact). 11,000 isn't a "a fair bit below" in statistical terms - with such a sample size you can still be 99% sure of being within 1.23%.
Quote:
If there are
supposedly
62m people in the UK,
Are you doubting the population figures as well?
Quote:
thats over 45m who didn't watch by choice
You still haven't explained why that figure is so problematic to you - unless you've conducted a survey with a larger sample size than BARB and got a different result?
Thanks David for backing up what I was saying - although of course now I'll discard any theory any reputable statistician (as hard to type as it is to say!) because Badger states it can't possibly be true!
You still haven't explained why that figure is so problematic to you - unless you've conducted a survey with a larger sample size than BARB and got a different result?
David
Nice to see you've skipped right over my point about Premiership matches though .
Maths isn't one of my strong points, thanks for the stats info, very interesting (genuinely).
I believe Sky have their own internal system which I assume collects data straight from subscribers digiboxes - though whether that's a sample of customers or all customers I don't know. I'm not sure what the legalities would be about collecting data about what people are watching if they don't know the data is being collected.
That doesn't sound right to me. I've never heard of Sky doing that and how they'd go about it I have no idea.
As you say I'd imagine that there would be legal problems but also in practice it would be useless data for ratings purposes: Firstly it wouldn't be representative - how many of their boxes are actually connected to a phone line? Secondly how would thy know who is watching? BARB boxes register how many people are watching - Sky wouldn't know if you're watching Sky News all day or whether you've just left the box on. Lastly BARB is an industry standard that all the broadcasters (including Sky) agree on and subscribe to, a separate set of statistics wouldn't be credible next to them.
Virgin almost certainly can tell what station your box is tuned to but most of the above points apply to them too
Just to put this into perspective, putting aside your personal preferences do you genuinely believe that last year 40m people in the UK chose not to watch X Factor?
Can I believe that two thirds of the UK population didn't choose to watch the X Factor [final (I'm assuming)]*?
Absolutely.
Once you take out those too young to watch, those too old to be able to decide to watch, those who are working, those who are out, those who are shopping, those who don't have a TV and the myriad of other reasons for people to be unable to watch,
That doesn't sound right to me. I've never heard of Sky doing that and how they'd go about it I have no idea.
As you say I'd imagine that there would be legal problems but also in practice it would be useless data for ratings purposes: Firstly it wouldn't be representative - how many of their boxes are actually connected to a phone line? Secondly how would thy know who is watching? BARB boxes register how many people are watching - Sky wouldn't know if you're watching Sky News all day or whether you've just left the box on. Lastly BARB is an industry standard that all the broadcasters (including Sky) agree on and subscribe to, a separate set of statistics wouldn't be credible next to them.
How many of their boxes are actually connected to a phone line?
I'm sure Sky sort of make it a condition that boxes have to be connected to a phoneline for 12 months IIRC, don't they?
Secondly how would they know who is watching? BARB boxes register how many people are watching
Maybe the count customers rather than people and work it out with some clever statistics?
Sky wouldn't know if you're watching Sky News all day or whether you've just left the box on.
Don't Sky boxes have an automatic ‘Auto Standby’ feature?
Lastly BARB is an industry standard that all the broadcasters (including Sky) agree on and subscribe to, a separate set of statistics wouldn't be credible next to them
I'm sure it's just so Sky can "big up"some of their own channel's say Sky 1 that don't necessirly get great 1st viewing figures, but better timeshift figures.
Sky DO have an internal ratings system using their Digiboxes, but it is of limited value as it reports only on what channels a box is tuned to and what shows are recorded and replayed. (Tivo also had a similar system)
What it doesn't tell you is the far more important information - WHO is watching...
The BARB ratings tell you about the audience - not the set-top box. (audience gender, demography etc.)
