hello, as ive written in other posts, im a major in media and cultural studies in australia,

and as far as i know, nielsen here is the same nielsen you have.

so here goes my theory, and whomever can benefit i hope does...

firstly, this information in its FULL form, not my edited down version is available here:

http://entertainment.howstuffworks.com/question433.htm

pretty much, the way its done (and its got a pretty decent rate of accuracy as its the same technique used by quantitative med studies) is nielsen will try to take a sample of the population and then poll how many shows are taken.

the samples are relative to the amounts of people in a community. for example:

in a village theres 300 old people over 60, 400 people over 30 and 600 under 30.

what nielsen would do in this simplistic example is randomly select 30 old, 40 30+'s and 60 under 30's

the point being, X100 and u get a rough estimate of your population.

the problem with this is, their random sample can easily be skewed and are therefore not really indicative of ANYTHING.

for example, out of the 40/400 under thirties, only 10 out of the forty might watch a program like jericho. so thats 100 out of 400 in extrapolated terms.

HOWEVER the reality might be that 250 of the under 30's watch jericho.

therefore the number of 100 is less than half the viewer ship of under 30's.

and so on. so in that one village the poll might be skewed completely and therefore show a programme to be niched programming when it isnt.

however, this is only the math behind the science

go to http://entertainment.howstuffworks.com/question433.htm

for information on the physical side of the testing to better understand here im talking just numbers

*CONTINUED IN REPLY AS COUNTER IS GOING DOWN*

4  out of 5 found this valuable. Do you find this valuable? Do you?    

• Post reply

Show Last Reply