TV Viewing figures

  n4165si 14:34 28 Feb 09

Does anyone know how viewing figures that are compiled and published for a particular programme that has been shown,or are they guess work

  Si_L 14:45 28 Feb 09

Its not guess work. It works by having a cross sections of about 5,000 homes which a part of a panel. Their programs are tracked, and put into context of the wider population.

  n4165si 14:53 28 Feb 09

Thanks for that ,but when you see that millions have tuned into a particular programme ,against a panel of 5000 people tracked ,the figures produced have to be taken with a bit of licence i fear

  Si_L 16:42 28 Feb 09

I think 5000 is a fair sample to take, the figures are of course estimated based on this cross section but I doubt the real life figures will be far off. The main purpose is I think for advertising, and so it doesn't need to be 100% accurate, as it is just a means for companies to gauge how much exposure their adverts will get.

  interzone55 16:55 28 Feb 09

5000 is a good sample size, a lot of the surveys you read in the papers like "3m people drink to excess" are based on samples as low as a couple of hundred.

click here'_Audience_Research_Board

The RAJAR panel for radio listening figures uses a massive sample of around 130,000 people. These panellists fill in a listening diary every 1/4 hour for a week, with figures compiled once a quarter.
click here

  interzone55 16:56 28 Feb 09
  canarieslover 17:05 28 Feb 09

Does every survey use the same 5000 people?? Having reached the ripe old age of 66 I have never been asked for my opinion in any survey. Mind you I haven't even been called for jury service though I know someone who has sat on four juries. Am I lucky or not?

  Si_L 17:17 28 Feb 09

I would love the chance to sit on a jury so I guess unlucky.

  Si_L 18:03 28 Feb 09

"The problem comes with small audiences for the minor channels where the margin of error in the survey can be significant."

I have just read up on this and apparently, with just 5000 people, and so many channels, occasionally a channel is being watched by nobody, which causes a dilemma as to the viewing figures. Obviously there are people watching the channel, but if none of the 5000 are, what do you quote the viewing figure as?

Maybe I was wrong, and the sample size is too small.

  canarieslover 19:40 28 Feb 09

I'm with you on the like to sit on a jury thing. I feel that I have missed out on something.

  interzone55 19:58 28 Feb 09

It's not a survey as such.

The panellists have an adapted TV which reports back on which channel is being watched. Every 1/4 hour the viewers are prompted to tell the TV how many people are in the room...

This thread is now locked and can not be replied to.

Nintendo Switch review: Hands-on with the intuitive modular console and its disappointing games…

1995-2015: How technology has changed the world in 20 years

Prehistoric Britain is laid out in these Royal Mail stamp illustrations

Best running headphones | Best sport & fitness headphones: 4 brilliant pairs of wireless…