3
$\begingroup$

I have a set of pre and post treatment scores on 8 measures. I'd like to test which of these show significant improvement after treatment. If I use within-subjects t tests it will mean carrying out 8 t tests, and risking type I error.

My textbooks all say that repeated subjects ANOVA is for more than two measures, but I only have 2 (pre & post). There is only one group - no control group or anything. What is the most appropriate test to use?

Improve this question
$\endgroup$

1 Answer 1

Reset to default
1
$\begingroup$

Why not just perform a MANOVA?

(Post-Test Variables of Interest) = (Group-Related Variables of Interst).

Alternatively, you could perform a MANCOVA, (Post-Test Variables of Interest) = (Pre-Test Variables of Interest AND Group-Related Variables of Interest).

Improve this answer
$\endgroup$
1
  • 2
    $\begingroup$ Welcome to the site, @M.Buchler. Your post is being automatically flagged as low quality, because it is a little short, but I think it is a real answer. Would you mind expanding a little on the implicit argument behind this answer? $\endgroup$
    – gung - Reinstate Monica
    Commented Jun 28, 2014 at 18:50

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.