Sample Size Calculations with JMP

#1
Hello,

I have used the sample size calculator of JMP for a single proportion. I need to add to my report the formula "I used" (JMP did). I couldn't find it in the JMP manual. Does anyone know how the JMP calculates a sample size for a single proportion, given the real known proportion, the hypothesized proportion, the significance level and the power ?

thank you
 

BGM

TS Contributor
#2
You can derive it.

If you know the hypothesized proportion and the significance level, you know the rejection region in terms of the sample size \( n \)

Under this rejection rule, if you know the power and the real known proportion, you can solve the \( n \) out.
 
#3
thanks for the hint, but I am afraid I have no clue how to do this...was looking for a link for the equation, but nothing came up.

what confuses me is how the power comes in. I know how to derive n from a CI formula for example, but not like this...
 
#4
I found a formula, the delta doesn't fit, it belongs to CI...if I remove it I get similar results to JMP, or is it just a coincidence...

I have no idea how to derive it like you suggested.
 
Last edited:

BGM

TS Contributor
#5
Ok do you know how to do the proportion test with normal approximation? with significance level \( \alpha \), hypothesized proportion \( p_0 \) and sample size \( n \)