Q1: This is definitely very interesting, will use this to compare two sets of repeat measurements for a DOE. Hope this use of paired sample T-test is alright
A1 : My advise is in order to use paired sample t-test, if you are interested in knowing whether there is significant difference between before and after a corrective action taken on your process, the same subject must be used for comparison of the difference between two time points, before and after.
Q2: What is the difference between the conclusion of using matched pair analysis vs. simple t test?
A2 : The answer is both methods concluded that the difference is statistically significant regarding the tutorial example of driving errors. That is not always the case though. Take a look of the attached sample files. There are from the same dataset. Unstacked Potency.jmp is using two sample t-test and Stacked Potency.jmp is using paired t-test. Click on the scripts to run the analysis and you will find that both have different conclusion about the mean difference.
(.jmp files at links below)
There are no labels assigned to this post.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.