Sample Size Determination to Compare Three Independent Proportions

[abou55]

Dear All: good morning

Re: Sample Size Determination to Compare Three Independent Proportions

Situation:

Three Binary variables (Yes, No)

Three independent populations with fixed sizes (say: N1 = 1500, N2 = 900, N3 = 1350).

Power = 0.80

How to choose the sample sizes to compare the three proportions of “Yes” among the three variables.

If you know a reference to this topic, it will be very helpful too.

with many thanks in advance

abou

[StatDave · Accepted solution]

I assume you mean that you have one binary response that is measured on each subject in three independent populations and you want to compare the response proportions from those three populations. If you have the response for every subject in the full populations, whose population sizes you stated, then there is no testing or estimation to be done since you know everything in the full populations. Simply compute the proportions and you are done. There is no p-value to compute. However, if the populations are not finite and you do need to take samples from them, then see this note for determining sample size. See Example 3 in the Results tab which shows how to use either the macro or PROC POWER to do this.

[abou55]

Thank you very much for your help in this matter

Abou