I may have mentioned this topic in the past- two sample dependent T test versus paired T test. This is not a shaming session. But people think they can take a paired -t test structure and perform a two sample mean comparison. It’s not translatable for several reasons. One of the assumptions of a two-sample t-test is that your groupings do not have the same individuals . Another words, you have two totally different people in each group. Therefore, you can not take a paired-t-test summarize before and after as if you have two distinct groupings.
Your interpretation of the paired -t test speak to whether or not there is a difference between the two time points. Another words, did the patients see an overall improvement or decline in the dependent variable. The interpretation of the two-sample independent test determines if the two sample populations are coming from two different populations or not. One statistic has two dependent factors per subject, and the other has one. Know the difference.
-Moore to follow Amy
