Hello, I am quite new to this program and attempting to assist in some research involving pre and post-course testing of patients. For context: 62 patients underwent two courses, both of which had tests for certain measures which were administered both pre and post course. Test 1 has five measures while Test 2 has four. We have found significant changes in scores on the factors in both tests from pre to post, as well as correlation in the changes between the tests; ie. positive change in a factor on one test appears to correlate with positive change in a factor on the other. What we wish to do, and what seems to be appropriate to us, is to run a repeated measures MANOVA with these nine factors as our dependent variables, which would tell us the interactions of pre and post scores on the factors. I cannot seem to get stata to do this, and to be honest all my reading on the subject has me a little lost.
We have, essentially, nine factors repeated two times which we would like to measure. If anyone can give me a dumbed down explanation on how I need to organize my data and which commands to run to do this, I would appreciate it. Additionally, if a repeated measures MANOVA is not the test we want to answer this question (is there significant interaction between the pre and post Test 1 and Test 2 data?), I would also like to hear advise on that.
I have been looking at this tutorial for help http://www.stata.com/manuals14/mvman...peatedmeasures but am unsure how to apply this method to my data, or if doing so is even reasonable.
Where this example states at the beginning "here is data on five subjects, each of whom took 3 tests" I assume mine would instead say "here is data on 62 subjects, each of whom took 18 tests". (the data is organized to fit that format, at least.)
Can anyone provide some guidance?
We have, essentially, nine factors repeated two times which we would like to measure. If anyone can give me a dumbed down explanation on how I need to organize my data and which commands to run to do this, I would appreciate it. Additionally, if a repeated measures MANOVA is not the test we want to answer this question (is there significant interaction between the pre and post Test 1 and Test 2 data?), I would also like to hear advise on that.
I have been looking at this tutorial for help http://www.stata.com/manuals14/mvman...peatedmeasures but am unsure how to apply this method to my data, or if doing so is even reasonable.
Where this example states at the beginning "here is data on five subjects, each of whom took 3 tests" I assume mine would instead say "here is data on 62 subjects, each of whom took 18 tests". (the data is organized to fit that format, at least.)
Can anyone provide some guidance?
Comment