Thread: Equivalence test or non inferiority test for non parametric data using stata or spss

1. Equivalence test or non inferiority test for non parametric data using stata or spss

I am doing a retrospective study about a new medical device, comparing it with an older device. the first group has 18 cases and the other group has 83 cases. I performed Mann-Whitney test for the data as they are non-parametric and there was no statistical difference between the two groups. and now I want to perform non-inferiority test or equivalence test to prove that the newer device is almost the same or not inferior to the older device but I don't know how??? could anyone help me in that matter???

2. Re: Equivalence test or non inferiority test for non parametric data using stata or s

What is your reasoning for nonparametric? Do you have a measure of how different they can be and still similar?

3. Re: Equivalence test or non inferiority test for non parametric data using stata or s

I performed KS test and shapiro-Wilk test for the data

4. Re: Equivalence test or non inferiority test for non parametric data using stata or s

I performed KS test and shapiro-Wilk test for the data
These tests are not typically strong reasons to dictate using parametric or non-parametric methods. The formal tests can reject Ho for an immaterial departure from normality (even if the sample was drawn from an underlying distribution that's normal), or similarly, they can fail to reject Ho from insufficient data (even if the data were drawn from a very nonnormal population). I would also clarify a point of terminology: data are not "parametric" vs "nonparametric", these refer to kinds of analysis.

What is the variable being analyzed?
How is it measured?
Is it bounded?
If it is bounded, what are the bounds?

5. Re: Equivalence test or non inferiority test for non parametric data using stata or s

my research is a retrospective study with 101 patients underwent liver resection, I divided them into two groups (n1=83, n2=18) using two devices. I compared the operative time and blood loss between the two groups using Mann-whitney test, and the P-value was more than 0.05. So I want to prove that the newer device is not inferior to the older device.

 Tweet

Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts