+ Reply to Thread
Results 1 to 1 of 1

Thread: MANOVA and MANCOVA - Help needed

  1. #1
    Points: 4, Level: 1
    Level completed: 7%, Points required for next Level: 46

    Thanked 0 Times in 0 Posts

    Post MANOVA and MANCOVA - Help needed

    I have collected data with one Independent Variable (two levels - nominal) and 4 dependent variables (all scale - continuous). I decided to do a MANOVA to test whether my IV had an impact upon the DVs (which can be grouped together under one related category of wellbeing). The results of the MANOVA showed that the effect of IV was statistically significant.

    Therefore based on theory I proceeded to complete a MANCOVA using 4 covariates (all scale - continuous), which can be grouped together under one related category of social support. However on doing this the data has shown that when controlling for social support the effect of the IV on the DV is non-significant (which is an unusual finding).

    Before I draw conclusions from this I want to make sure that it not due to some breach of assumptions or problems within the data.

    On exploration of the correlations matrix and tests of between subjects effects I have found that one of the covariates is highly correlated to one of the DVs - I feel as though I need to do more tests to establish whether this is the cause of the non-significant finding. I am not sure if this breaches any assumptions? Can anybody suggestion what tests I could do based on this information? Or would it be a case of removing that single covariate and seeing if the finding is still non-significant?

    Furthermore the other three covariates have a non-significant effect on the DV (identified on the multivariate tests output). Not sure if this is cause for concern? Should I remove these covariates?

    Any help you can offer would be greatly appreciated!
    Last edited by Em19133; 01-01-2017 at 06:42 PM.

+ Reply to Thread


Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts

Advertise on Talk Stats