The BIG Theorem on Sampling

Assume there is this population of objects (fish, people, voters) for which each object is either “yes” or “no” to a certain condition.

Assume we take a random sample of sample size n of these objects.

Then p_pop_yes is within

of p_samp_yes …………

at a CONFIDENCE LEVEL OF 95%.

For Real!!

That is, there is a 5% chance that p_pop_yes will NOT be within the above margin of error of p_samp_yes. [And this is the little secret that you never hear about on TV.]

Previous slide

Back to first slide

View graphic version