Does 1200 people qualify as a sampling for the entire population of the United States?

Question

This is one of those polls often quoted by proponents of GW.

It seemed restricted to political affiliation; excluding, age, race religion, gender, education, and nationality. Is this a true reflective poll?

Answer 1

Only if they agree with me.

Answer 2

If a truly random sample of 1200 people is taken from a population of 300 million people the result will have a error of margin of +/-3.7% with 99% confidence. Not bad.

Edit: Oops that should have been margin of error. If the poll results said that 50% believe in the tooth fairy, then we can say that with 99% confidence that the ACTUAL percentage of people who believe in the tooth fairy is between 46.3 and 53.7%.

I should mention that the margin of error I quoted was based on a "result" of 50%. As that value approaches 0 or 100% the margin of error goes down.

Edit2: If you are willing to do enough math. The calculator I linked only allows 95 and 99%. For a given sample size, the higher your confidence level the wider the margin of error and vice versa. Pollsters usually quote the margin of error for 95% confidence.

Answer 3

Depends on the pollster. It sounds like you are assuming it was a fraud. Why don't you go out and get the polling method and argue it, instead of make veiled accusations.

EDIT - no pretense, you're the one making the accusation. You must present evidence for your claim.