Just a question I've been wondering about for a while ...
How do ratings work? I mean, when I watch a program on TV, I can check out the ratings the next day, but how on earth do those work?
I mean, how do they know how many people watched their show? How could they possible know that!!!
If anyone has any answers, please tell!
Pah! Isn't it obvious? They look through the reverse camera in the television, and count how many people are watching.
Thats a scary thought Babar...
In Hong Kong here, statistic companies installed devices in selected (well, randomly, they claimed) in a very small sample subset of voluntary familaries, which recorded the viewed channel informations, and then they'll use stupid (without doubt) but precise (they claimed) statistic methods to play with the data.
I don't know how it's done in other places, but I think it's just similar stupid tricks.
That's pretty much how it works in the U.S. as well. Check out the Wikipedia article on Nielsen ratings for more info: http://en.wikipedia.org/wiki/Nielsen_Ratings
Makes me think about when on the news they'll say "*80% of Leicester agrees with the decision to eat cheese" and I'll say "they never asked me!"
okay that example never happened but you get my point, right?
Heh I think Nielson is responsible for the research here too, so, no wonder. ;D
Of course with IPTV, when it comes in, they'll know exactly what everybody watched. They already know exactly where everybody carrying a mobile (turned on) has been ever...
Orwell was only a few years out: and the viewscreens are ostensibly voluntary and play downloadable ringtones!
But of course there could be multiple people watching the same set, and sometimes people just leave their sets on in the background during the day!
It'll never be accurate!
Also, how do the ratings for songs work? You'll hear "So and so was #1 in the charts this month!" How do they figure those out?
Wikipedia is your friend: http://en.wikipedia.org/wiki/Music_charts
QuoteRecord charts are compiled using a variety of criteria. These commonly include sales of records, cassettes and compact discs; amount of radio airplay; requests to radio disk jockeys; and, recently, number of downloads.
Another company started producing rival ratings to ACNielson here called OzTam or something [these businesses exist for the purpose of advertisers incidently, so they can determine the price of broadcast time in a certain show], and every 15 minutes or so they required the viewer to press a button to show they were actually there watching.
Also, you might note how many newspapers like to give away free copies at festivals or sporting events, or give subsriptions to schools for $5 for the year and the like. Even if 90% of these papers aren't read, they go into the circulation figures, which determine advertising prices.
Bottom line is that alot of these ratings mainly screw over advertisers who don't deserve sympathy, so don't worry, unless you're a fanboy to a show which isn't doing well or soemthing.
Well, started several months ago, a cheap crap newspaper (it's set up by a large newspaper press who wanted to battle their rival, so they published this paper which was a bit cheaper and with much worse content, not to mention severe proofreading errors) dropped its price to half of normal papers, AND included coupons (usually for supermarkets, so they're quite useful) everyday, which worth at least the price of the paper (sometimes the coupons could worth a lot, like $15 coupon for a $3 paper). So suddenly their sales raised exponentially (some people even bought several copies just for the coupons), and so they can now announce on TV, that they had become the 2nd top selling paper (with their big brother being the 1st) ...
I hate stupid statistics.
I had a student's job for one of those statistics companies once - we would call people and ask them in meticulous detail about every newspaper they read in the past year - the interviews tended to become rather long and confusing ("So you read this newspaper 2-3 times during the past six months - is there any chance you have read it at least once within the last week? The last two days, yesterday? today? Do you still have it? Could you find it and look at page 5? Do you remember having seen this commercial before? What did it make you think? Did you find it informative? humorous? rude?" etc - and this would go on for each newspaper they had ever touched).
On the TV side, families were chosen to represent the demographics of the country on a small scale, and they had to report their viewing activity by keeping a kind of TV-diary. This was some years ago, by now they probably have automatic sensors in the tv-sets as well. Having worked there, I'm pretty amazed how accurate the statistics can be for some things - the result of elections are usually remarkably precisely predicted.
I guess with sky, which is connected to a phoneline, data could be reported... i'm not sure really how they do it...
A few years ago we, my household was one of the selected for the Nielson rating. They paid us some money, I don't think it was a lot, but it was free money. They plugged something into the tv's, to record what we watched. Although, I don't think I represent anyone but myself.
-MillsJROSS
Quote from: MillsJROSS on Fri 26/05/2006 05:19:59
They plugged something into the tv's, to record what we watched.
They plugged something into the tv's, to watch you, recorded.
Can't workload on broadcasting devices be measured? Especially on cable?
Broadcasting no, since an television picking up the signal doesn't reduce the amount of signal around in the air.
And cable isn't broadcasting, since it uses a cable, but since most have the machines to determine which channels you have and haven't paid for, they could put monitoring equipment in there. But then, since you pay for cable, the ratings aren't as important since they're not making as much revenue off advertising.