0
$\begingroup$

so if I've done a MANOVA and followed that up with Bonferroni correction and found that only 2 out of 7 variables are significant (whereas before there were a few more) does that mean that only these 2 variables are significant overall then and I only talk about them in my study? Hopefully this makes sense! Thank you.

Improve this question
$\endgroup$
1
  • $\begingroup$ Welcome to the community. Can you please contextualise your question more? Why do you do this multiple comparison? Why are you doing MANOVA to begin with? Etc. (Some punctuation would be appreciated too.) You can edit the original post of your question by clicking the edit link below the tags. $\endgroup$
    – usεr11852
    Commented Apr 13, 2016 at 18:49

1 Answer 1

Reset to default
1
$\begingroup$

It's generally best to report all the tests you performed, whether they were significant or not.

Improve this answer
$\endgroup$
2
  • 1
    $\begingroup$ This may be all that can be said here, but it might be preferable if you could explain this a little more. Eg, why is it "generally best"? Can you elaborate? $\endgroup$
    – gung - Reinstate Monica
    Commented Oct 14, 2016 at 14:55
  • 1
    $\begingroup$ Because honesty is the best policy. $\endgroup$
    – Bonferroni
    Commented Mar 7, 2017 at 20:43

Your Answer

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