LO
So, the viewing figures for the top programmes on christmas day are out and if you haven't seen them, are below..
what i want to know is whether the underlying dataset gives an authentic extrapolation.
my understanding is that BARB put a special set top box in a selection of homes (a few thousand) and from this they are able to derive the viewing figures for the entire country.
i suppose, upon seeing the top ten programmes, I have to wonder whether that data is as robust as it is made out to be...
i can just about accept that about 12% of the entire country is watching the Queen's Christmas message, but is 10% of the population watching Mrs Brown's Boys? really?
With all the ways we can consume entertainment these days, those figures seem rather surprising.
I also wonder whether the sample size is so small that the extrapolated data is ultimately meaningless - i mean, look at how election pollsters do pretty much the same thing, and most of the time, they get it wrong. I know it's not the exact same science, but it is still magnifying a small dataset to get something many, many times larger... a process which will also magnifies errors in the approach.
Do you think viewing figures are a load of old cobblers?
what i want to know is whether the underlying dataset gives an authentic extrapolation.
my understanding is that BARB put a special set top box in a selection of homes (a few thousand) and from this they are able to derive the viewing figures for the entire country.
i suppose, upon seeing the top ten programmes, I have to wonder whether that data is as robust as it is made out to be...
i can just about accept that about 12% of the entire country is watching the Queen's Christmas message, but is 10% of the population watching Mrs Brown's Boys? really?
With all the ways we can consume entertainment these days, those figures seem rather surprising.
I also wonder whether the sample size is so small that the extrapolated data is ultimately meaningless - i mean, look at how election pollsters do pretty much the same thing, and most of the time, they get it wrong. I know it's not the exact same science, but it is still magnifying a small dataset to get something many, many times larger... a process which will also magnifies errors in the approach.
Do you think viewing figures are a load of old cobblers?
Quote:
1. The Queen - 7.6m (BBC One, ITV and Sky News)
2. Mrs Brown's Boys - 6.8m (BBC One)
3. Strictly Come Dancing - 6.5m (BBC One)
4. Call the Midwife - 6.3m (BBC One)
5. EastEnders - 6.3m (BBC One)
6. Doctor Who - 5.7m (BBC One)
7. Coronation Street - 4.8m (ITV) rising to 5.1m when ITV+1 is included
8. BBC Teatime News - 4.2m (BBC One)
9. The Highway Rat - 4.0m (BBC One)
10. Cinderella - 3.6m (BBC One)
2. Mrs Brown's Boys - 6.8m (BBC One)
3. Strictly Come Dancing - 6.5m (BBC One)
4. Call the Midwife - 6.3m (BBC One)
5. EastEnders - 6.3m (BBC One)
6. Doctor Who - 5.7m (BBC One)
7. Coronation Street - 4.8m (ITV) rising to 5.1m when ITV+1 is included
8. BBC Teatime News - 4.2m (BBC One)
9. The Highway Rat - 4.0m (BBC One)
10. Cinderella - 3.6m (BBC One)