A

logical fallacy in

induction where the

sample used is not

representative of

the population as a whole. This reduces the

probability that the

proposition is

valid.

Example: "people are generally opposed to the new law, according to a sample of 1000 Republicans." What of Democrats, or Independents?

To prove the fallacy, show how the sample is relavently different from the population, and how this means the conclusion may be different. Statistics can help, but an inductive argument can work.