Suppose you sampled 2,500 people for a marketing survey, but your ...

[maychende]

Suppose you sampled 2,500 people for a marketing survey, but your boss thinks that you should have sampled more, say 4,000 people, because he says the results would have been much more accurate, even under the tight budget and time constraints on the company. Using statistics, how would you convince your boss to agree with you?

Question 2

With a random sample of 2,500 people, we usually get an estimate that is accurate to within what percent of the truth (using the general rule)?

[uniquea123]

Answer to Question 1

FOR 4,000 PEOPLE, THE MARGIN OF ERROR IS ABOUT 1.6, COMPARED TO ABOUT A 2 MARGIN OF ERROR FOR 2,500 PEOPLE. ALMOST DOUBLING THE SAMPLE SIZE COSTS MUCH MORE IN TIME AND MONEY BUT WILL ONLY INCREASE ACCURACY SLIGHTLY.

Answer to Question 2

2

[maychende]

Thanks for the timely response, appreciate it

[ebonylittles]

TYSM