767d36b0   
   XPost: rec.arts.tv, alt.tv.scifi.channel   
   From: your.name@isp.com   
      
   In article , cloud dreamer   
    wrote:   
      
   > On 16/11/2010 4:27 PM, Your Name wrote:   
   > > "cloud dreamer"  wrote in message   
   > > news:7fCdnfKUqZsd43_RnZ2dnUVZ_vWdnZ2d@supernews.com...   
   > >> On 16/11/2010 9:09 AM, Neil Rieck wrote:   
   > >>> [...snip...]   
   > >>>>   
   > >>>> That's not how the rating system works. Neilsen has a set number of   
   > >>>> meters out there in selected homes. They base their numbers on them and   
   > >>>> extrapolate total viewership from the sample set.   
   > >>>   
   > >>> The measurement boxes are expensive so they (Nielson) still support a   
   > >>> larger paper diary system which allows people to lie (network bias).   
   > >>   
   > >> They would statistically lie in equal numbers so the difference would be   
   > >> negligible.   
   > >>   
   > >> So, tell me. What else are they supposed to do???   
   > >>   
   > >> Until they can start collecting data live from DVRs, what other system   
   > >> are they supposed to use???   
   > >>   
   > >> If they can't rely on people to tell the truth in the diary portion of   
   > >> their polling, they have to rely on the meters. If they're that   
   > >> expensive, they have to rely on extrapolated data and conclusions.   
   > >   
   > > Meters are no better than diaries. You would have to have a meter attached   
   > > to every TV in the house, as well as recording devices and download sites.   
   > > You would also have to determine whether anyone is actually watching the TV   
   > > or simply tuning in to some "news" show because they think thye should, but   
   > > in reality are in another room eating dinner or whatever.   
   > >   
   > >> Your Name seems to think they need to go out there and poll all 115   
   > >> million households in the US.   
   > >   
   > > Nope. Try the few billion households, worldwide. Despite what most   
   Americans   
   > > think, there are other countries on this planet.   
   >   
   > Get your head out of your ass.   
   >   
   > What you're advocating is simply IMPOSSIBLE.   
      
   I never said it was possible, but it is the only way to get near-accurate   
   results ... unfortunately you'll still get morons who lie or otherwise   
   circumvent the system.   
      
      
      
   > Extrapolating data works. Just because you can't understand it doesn't   
   > make it useless.   
      
   It doesn't "work" ... it's inaccurate "guesswork". It may well be   
   considered accurate "enough" for those in charge of the networks, but then   
   they ARE idiot management types so it's no surprise that they base their   
   decision on flawed reports rather than actual facts.   
      
   Try the coin test yourself and you'll easily see it is useless, flawed   
   methodology, and the smaller the sample size is compared to the total   
   population, the less accurate those results are.   
      
   END OF STORY.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)