Hi all,
I’m having some trouble trying to do data analysis leveraging between two destructive tests on the same product.
Background: I have variable data from 500 samples out of a population of 40,000 from a destructive test that ends up in a non-normal distribution. I need to perform a second destructive test that is pass/fail, but I want to be able to prove that the samples being tested in the second test contain a non-zero amount of samples at the bottom 5% of the range observed in the first test. Due to the destructive nature of both tests, is there a way to make an argument that sample size “x” is large enough for the second test to ensure the amount of bottom 5% of samples I need is included in that population?
Thanks,
Dan
.
.