Jump to content
 

This sub-forum is for the 2019 series and any individual entry topics. Content from the 2018 series can be found in the Challenges Archive.

GMRC Series 2 - Episode 1 - 'The Restless Earth'


Recommended Posts

10 hours ago, john new said:

How do the figures include recorded and watched later? We did that due to two shows we wanted to see being on at the same time.

 

They don't. When we were at Fawley Hill the producers stressed to us that we should encourage people to watch live and not on catch up as it is the BARB live figures that C5 will use to decide whether to commission a series Three.

 

Mind you, if the viewing figures stay North of one million per heat the chances of another series are good.

  • Like 1
  • Informative/Useful 2
Link to post
Share on other sites

  • RMweb Premium
5 minutes ago, whart57 said:

 

They don't. When we were at Fawley Hill the producers stressed to us that we should encourage people to watch live and not on catch up as it is the BARB live figures that C5 will use to decide whether to commission a series Three.

 

Mind you, if the viewing figures stay North of one million per heat the chances of another series are good.

Hi

 

So if you record it when it airs but watch it later does that count as watching live as you are recording it while its being broadcast?

 

Cheers

 

Paul

  • Like 1
Link to post
Share on other sites

1 hour ago, PaulCheffus said:

So if you record it when it airs but watch it later does that count as watching live as you are recording it while its being broadcast?

 

No. It's the viewing time that matters, not the recording time. The reason for that is commercial: only live broadcast viewers are fully exposed to normal, broadcast adverts - catch-up viewers see different adverts, and time-shift viewers tend to skip over ad breaks. So it's the live broadcast viewing figures that are needed to sell a programme to advertisers.

 

For comparison purposes, what's known as VOSDAL - Viewing On Same Day As Live - time-shifting is included in the published figures, as a significant number people will do that to avoid a clash (eg, so they can watch both Bake Off and GMRC on the same evening) or because they don't happen to be at home at the time the programme is broadcast but do get home shortly afterwards (although this is more common for popular tea-time programmes, such as Pointless, than prime time evening shows). But the internal figures shown to advertisers will strip this out.

 

But, if the live ratings stay in C5's top twenty - and even better if they can break into the top ten - then the channel bosses will be happy with it. At around a million an episode, it won't be a "must do another series" type of show, but if the makers come back with a pitch for a third series then it's likely to get the go-ahead.

  • Like 2
  • Thanks 1
Link to post
Share on other sites

9 hours ago, PaulCheffus said:

Hi

 

So if you record it when it airs but watch it later does that count as watching live as you are recording it while its being broadcast?

 

Cheers

 

Paul

 

7 hours ago, MarkSG said:

 

No. It's the viewing time that matters, not the recording time. The reason for that is commercial: only live broadcast viewers are fully exposed to normal, broadcast adverts - catch-up viewers see different adverts, and time-shift viewers tend to skip over ad breaks. So it's the live broadcast viewing figures that are needed to sell a programme to advertisers.

 

 

But your recorder's watching the live transmission and so seeing the same adverts as the live viewers?

 

At one time, there used to be a panel of viewers who provided a subset of the figures which were then extrapolated to the wider populous, so was very approximate. Have things changed?

  • Like 1
Link to post
Share on other sites

  • RMweb Premium
4 hours ago, JohnDMJ said:

But your recorder's watching the live transmission and so seeing the same adverts as the live viewers?

This is true, but when I play back my recording I can (and sometimes do) fast forward through the adverts, so I am less valuable commercially as a viewer.

If I watch a programme on a "Catch-up" or "On-demand" service, the clever software they use prevents me from fast forwarding through the ads - so I have to let them run at normal speed - but they are different ads. I don't know why they use different ads but my commercial value as a viewer is again different. I presume this is why the production company want us to watch live if we can: we're premium customers if we do.

Edited by Keith Addenbrooke
  • Like 1
Link to post
Share on other sites

7 hours ago, Keith Addenbrooke said:

This is true, but when I play back my recording I can (and sometimes do) fast forward through the adverts, so I am less valuable commercially as a viewer.

 

 

But the ads have still been 'watched' in their original entirety by the recorder.

  • Agree 1
Link to post
Share on other sites

  • RMweb Premium

I can't watch it live... I'm at the MRC modelling that night, I'm betting a lot of MRCs have that as modelling night. (Although Broadland MRC is open 24/365 to members, as I work that's the most convienient club night)

  • Like 1
Link to post
Share on other sites

46 minutes ago, JohnDMJ said:

 

But the ads have still been 'watched' in their original entirety by the recorder.

Unfortunately, recorders don't buy teabags, or holidays, or whatever else they're trying to flog. I'm not sure how the market research people distinguish between people who watch live, and people who record to watch later, but they evidently do.

 

What I've noticed is that when you watch on demand through a Sky box, you can fast forward through the commercials. But the cannier channels have a reduced number of them in the on-demand version, sometimes only one or two, so that by the time you've picked up the remote, fast forwarded, missed the programme starting back, rewound, and so forth, it's easier just to watch them.

  • Like 1
Link to post
Share on other sites

14 hours ago, JohnDMJ said:

But your recorder's watching the live transmission and so seeing the same adverts as the live viewers?

 

Yes, but you don't see them. You fast forward through them when you watch the recording. It's you that matters, not your equipment.

 

Quote

At one time, there used to be a panel of viewers who provided a subset of the figures which were then extrapolated to the wider populous, so was very approximate. Have things changed?

 

It is done from a panel. The system records what the people are watching at the time they are watching it, not what their equipment is recording. So it is extrapolated, and therefore not precise. But a lot of work goes in to ensuring that it is a representative sample of the viewing public. 

  • Like 1
Link to post
Share on other sites

1 hour ago, RLBH said:

Unfortunately, recorders don't buy teabags, or holidays, or whatever else they're trying to flog. I'm not sure how the market research people distinguish between people who watch live, and people who record to watch later, but they evidently do.

 

There are a number of different methods of doing it. One is to use a device fitted to the TV which logs what is actually showing on the screen at any one time. That will obviously log a recorded programme at the time it is viewed, rather than being recorded. Another, more old-fashioned system is to get people to keep a diary of their viewing over a week. That, again, records what you are watching when you watch it, not when it is recorded for later viewing. There are also snapshot polls conducted via telephone or Internet asking people what they watched, live, that day and what, if anything, they watched on PVR or catch-up that day.

 

Those different methods will tend to give different results. But that's helpful, because they can then be combined into an overall figure that's more reliable than a single method.

  • Like 1
Link to post
Share on other sites

9 hours ago, MarkSG said:

 

There are a number of different methods of doing it. One is to use a device fitted to the TV which logs what is actually showing on the screen at any one time. That will obviously log a recorded programme at the time it is viewed, rather than being recorded. Another, more old-fashioned system is to get people to keep a diary of their viewing over a week. That, again, records what you are watching when you watch it, not when it is recorded for later viewing. There are also snapshot polls conducted via telephone or Internet asking people what they watched, live, that day and what, if anything, they watched on PVR or catch-up that day.

 

Those different methods will tend to give different results. But that's helpful, because they can then be combined into an overall figure that's more reliable than a single method.

 

So it is not an automatic, involuntarily placed, spy in the TV! Compiling these figures is thus an arbitrary process; potentially "Lies, D**n Lies and Statistics"?

 

11 hours ago, RLBH said:

Unfortunately, recorders don't buy teabags, or holidays, or whatever else they're trying to flog. I'm not sure how the market research people distinguish between people who watch live, and people who record to watch later, but they evidently do.

 

I don't buy teabags either! I have enough to last me my lifetime.

 

Unless the TV or recorder has been specifically modified to convey what is actually being watched (opening cries of Big Brother!) then the above is pure speculation.

 

Is there anyone on this thread actually involved in the gathering of this 'data' who can confirm or deny this?

  • Like 1
Link to post
Share on other sites

41 minutes ago, Captain Kernow said:

So the best thing to do is to have it on live in the house, in the background and also recording and then do something else more useful (eg. your own modelling) at the time and watch speed through the ads on the recording later?

 

 

Who knows? This seems to make the proverbial 'Black Arts' look relatively simple! :jester:

  • Like 1
Link to post
Share on other sites

 

But, unless you are actually one of the households that contributes to the viewing panels, then it doesn't really matter what you do....


Like the old days with the record charts, only certain retailers contributed their sales figures to the people compiling the charts, so any sales from other shops didn't actually make any difference!

 

  • Like 1
  • Agree 1
Link to post
Share on other sites

On 25/09/2019 at 19:12, JohnDMJ said:

 

So it is not an automatic, involuntarily placed, spy in the TV! Compiling these figures is thus an arbitrary process; potentially "Lies, D**n Lies and Statistics"?

 

It's not arbitrary. It's a very detailed, and well-researched, science. Working out how to get an accurate representaton of the whole, from a sample, is a matter of applied maths.

 

On 25/09/2019 at 19:12, JohnDMJ said:

Unless the TV or recorder has been specifically modified to convey what is actually being watched (opening cries of Big Brother!) then the above is pure speculation.

 

The TV or recorder itself isn't modified, at least not directly. As I said elsewhere, there are three sources of data: a long-term panel of people who have an device fitted to their TVs which logs what is being viewed, a short-term panel of people who are asked to keep a daily diary of their viewing, and random polling of people (eg, by phone or online) asking them what they watched yesterday. Those three different sources are combined to give a more representative picture of viewing numbers.

 

On 25/09/2019 at 19:12, JohnDMJ said:

Is there anyone on this thread actually involved in the gathering of this 'data' who can confirm or deny this?

 

Yes. Me. That's why I'm trying to be helpful by explaining how it works.

 

(Disclaimer: I haven't been involved, in any way, with the long-term panels of viewers with hardware logging. That's a very small proportion of the data; most of it is obtained from short-term "daily diary" panellists and random snapshot polling. But it's an important part of the mix as it helps to verify the data from the other sources).

Incidentally, if anyone wants to be a potential contributor to some of the online polling that makes up the random snapshot polls, then sign up to be a YouGov panellist. Signing up doesn't guarantee that you'll be asked what TV programmes you have watched (as that will depend on the demographic information you'll need to supply when registering), but there's a fair chance that you will get one every now and then. Just hope it's for one of the days when you watched GMRC :)

  • Informative/Useful 1
Link to post
Share on other sites

On 25/09/2019 at 21:38, Sarahagain said:

 

But, unless you are actually one of the households that contributes to the viewing panels, then it doesn't really matter what you do....


Like the old days with the record charts, only certain retailers contributed their sales figures to the people compiling the charts, so any sales from other shops didn't actually make any difference!

 

 

That's partly true. But you can't know, for certain, whether or not you will be randomly selected for a telephone poll the day after you watched it (or not, as the case may be). Unlike the record charts, it isn't just a single, long-term panel. Part of the data is random polling, in the same way as other opinion polling such as voting intentions. And you're as likely to be randomly telephoned by a polling organisation as anyone else.

  • Like 1
  • Informative/Useful 1
Link to post
Share on other sites

10 hours ago, MarkSG said:

 

It's not arbitrary. It's a very detailed, and well-researched, science. Working out how to get an accurate representaton of the whole, from a sample, is a matter of applied maths.

 

Yes. Me. That's why I'm trying to be helpful by explaining how it works.

 

Paul from Team Grantham also does this work, albeit for radio listening figures. He'll be instructed to visit a random location in the area, knock on doors to ask if folks would be happy to take part by filling in a listening diary then returns later to collect the diaries.

 

I do vaguely remember, from the dim and distant past of my A level maths, being shown the 'chi-squared' test (chi being the Greek letter that looks a little like a capital X). To do with the statistical significance of sample size versus total population. It's surprising how small a sample you need to get a good result when the total population is very large. The maths is overwhelmingly against the result ever being significantly 'wrong'.

  • Like 2
Link to post
Share on other sites

53 minutes ago, LNER4479 said:

I do vaguely remember, from the dim and distant past of my A level maths, being shown the 'chi-squared' test (chi being the Greek letter that looks a little like a capital X). To do with the statistical significance of sample size versus total population. It's surprising how small a sample you need to get a good result when the total population is very large. The maths is overwhelmingly against the result ever being significantly 'wrong'.

 

Yes, the science of sampling is, in many ways, counterintuitive, but nonetheless delivers reliable and consistent results. It isn't just used for sociographic things like TV viewing and voting intentions, but also things like determining wildlife populations, measuring pollution and archealogical research.

 

What tends to put people off it, and make them think it's unreliable, is that the most well-known use of sampling is opinion polling for voting intentions. But these are notoriously difficult to measure accurately, and therefore have a much higher margin of error. Partly, that's because anything related to future intentions can be affected by people genuinely changing their minds - they may say they will vote one way but then decide to vote another - and partly because people are often reluctant to be fully truthful - they will either give a deliberately incorrect answer, or say something like "I don't know" when they mean "I do know, but I'm not telling you".

 

But TV viewing is much less of an issue in this respect. For a start, it's questioning peple about past facts rather than future intentions - what you did watch rather than what you plan to watch - and it's also something where people have much less of an incentive to mislead. It is likely that polling does somewhat under-report viewing for some of the more, er, specialist channels in the high numbers of the Sky TV guide that tend to be watched late at night ;), but not for the mainstream channels - if someone says they watched GMRC then they almost certainly did watch it, and if they don't say they watched it, they almost certainly didn't. So the raw data is reliable, and all that's left is the maths.

  • Like 1
  • Informative/Useful 1
Link to post
Share on other sites

  • RMweb Premium
13 hours ago, LNER4479 said:

 

I do vaguely remember, from the dim and distant past of my A level maths, being shown the 'chi-squared' test (chi being the Greek letter that looks a little like a capital X). To do with the statistical significance of sample size versus total population. It's surprising how small a sample you need to get a good result when the total population is very large. The maths is overwhelmingly against the result ever being significantly 'wrong'.

I actually did statistics as part of my work 40+ years ago!

All about calculating the quality & probability of failure and such like from a sample of X number of products from a total of Y etc.

It was boring..........:(

I did pass my MIQA exams though:good:

 

  • Like 1
Link to post
Share on other sites

Just been catching up with the series due to being out of the country - the young scientists are obviously customers of the company I work for as I noticed bottles of our Gibco product in the short film about where they actually work.  Wonder if they know that these bottles are excellent for storing dilute PVA, paint, water, etc etc. (I have an almost endless supply of discards!)

Enjoying the series so far.

 

Jim

  • Like 2
Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share

×
×
  • Create New...