I'm officially growing sick of "This Week's Top TV Shows" Reports, and the dependence upon which the television industry places upon them (for advertising purposes, which then influences program decision making). While the Nielsen ratings are often taken as de Facto proof, they are, in fact, a very loose approximation of the viewing population.
According to Nielsen Media Research, they pull from a representative population:
"Our samples include homes from all 50 states, from cities to towns, from suburbs to rural areas. We have homeowners and apartment dwellers — some with children and some without — across a broad range of demographic categories. We include people of all ages, income groups, geographic areas, ethnicities and educational levels — all in proportion to their presence in the population at large."
While this will give you a nice demographic of the population in general, it leaves out a major criteria in TV viewing: viewing patterns. For example, I am a sports nut, and am more likely to watch a random sporting event than I am a half-hour sitcom. However, there are people who wouldn't watch sports if they could help it. If one type of person is selected over the other, it will skew the accuracy of the results.
Now, lets take that example and expound upon it. Let's take 20 people who all fall into the same demographic category and group them by their viewing habits:
- 2 who prefer historical content
- 5 who prefer sports and competitions
- 6 who prefer comedies and sitcoms
- 7 who prefer various types of dramas
Let's assume that Nielsen Media Research uses a 10% sample to calculate their viewer approximations. Regardless of who is chosen to be in the sample, the results will be skewed to different degrees (for example, if both history buffs are chosen to be in the sample, the viewer approximations for the larger group will be very inaccurate in comparison to actual viewership).
Even as the general population is increases and the likelihood of a more representative sample group increases, there is no guarantee that every viewing habit will be accurately represented. This is the inherent flaw in the system.
It might also be why "Dancing With The Stars" always has such high ratings, but the only person I know who watches it is my mother.
With the advances in television communications technology, they may soon be able to collect more data to give more realistic approximations, but until that happens, or Nielsen works to get better representations of the population, we're stuck with a flawed system of approximations.
So the next time your favorite TV show gets canceled, look to Nielsen. Their methods may have been the inadvertent cause of the cancellation.