Let's say I have sampled all adult males (N=30) and females (N=25) in a group. There is a property (z) that emerges from pairing male A with female B. I run a Monte Carlo simulation of size 25 (every female is paired with one random male). For every iteration sampling 25 pairs, I calculate the average z [zmean = sum(z)/25]. I run this 100,000 times and draw the distribution of all 100,000 zmeans. I then plot the observed z mean (mean z in field observations = Ozmean) to test for a significant difference of Ozmean from the simulated data.

My question is: What is the best way to test the power of this analysis? Would the power of a Z-test be appropriate? If so, which sigma should I use (the simulated/population, or the observed)?

My goal is to determine if my sample size of observed "z" is good enough to reject the null hypothesis (zmean = Ozmean) in a two.tailed test.

I appreciate any help!


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.