Hi all,
I had a discussion with a customer on Sample size and they had a specific requirement:
Here is an example of a sample size estimation they require for most of our projects.
Part of their product requirements will be to achieve a sensitivity of 88% with 95% confidence intervals between 86% and 90%. So they want to be 95% confident that the “real” sensitivity is somewhere between 86-90% based on the calculation for the tested population.
What they want to know is, assuming they have met the requirement of sensitivity = 88%, how many samples do they need to have tested for our confidence intervals to be as stated?
Their desired final output would be: for an assumed sensitivity, the confidence intervals based on the number of samples.
They are using NCSS for those calculation - Can JMP do the same?
Thanks in advance,
Valerie
