What is the relevant output of Equivalence TOST Test pairwise comparison?

[#1]

Hi,

I want to conduct an analysis for comparing two groups with paired samples for similarities (they should be similar).

My choice was the TOST Test with pairwise comparisons. My boundaries for no difference were 0.4. The output of the analysis is this:

BonferrroniMean_0-1659599123651.png

Does this mean that there is no difference between the groups as it says: mean difference ≤ 0.4 with p < 0.05 ?

Sorry this is the first time I work with this kind of test and the question might be trivial...

[#2]

If you are testing the difference between paired samples I suggest you use a paired t-test.

If you have the time I can also recommend the Decision Making with Data module in JMP's STIPS course. (Personally, I found the experience of doing this course highly informative.)

With regards to the TOST, if I remember correctly this is more adequate for a situation where you have one mean value and want to compare it to an upper as well as lower limit. This can be useful in situations where you want to compare, e.g., the mean of a process output to a minimum and maximum acceptable limit. These limits are than tested pairwise against the mean, to see if any of the limits are not statistically different from the mean.

[#3]

Assuming that your null hypothesis is that the mean pairwise difference is not zero and your alternative hypothesis is that this difference is practically equivalent to 0 (within 0.4 of 0), then your test (TOST) results indicate that you do not have sufficient evidence to reject your null. That is to day, because both tests in TOST are not significant, you should not reject the null and claim equivalence.