1
$\begingroup$

I am trying to perform a multivariate multiple regression in SPSS. I have 5 independent variables and 4 dependent variables. I know it can be done in SPSS using the GLM-multivariate option. My questions are:

  1. What are the assumptions for a multivariate multiple regression?
  2. How do I interpret the findings (how are these findings different from the findings of a regular multiple regression)?

I am using this link as a reference (however, this is MPlus and I need the information for SPSS).

Any ideas, code, or references would be greatly appreciated.

Improve this question
$\endgroup$
3
  • $\begingroup$ Welcome to CV! I've edited your title to reflect your question content, please feel free to correct it if I've erred. $\endgroup$
    – Sean Easter
    Commented Nov 16, 2015 at 18:18
  • 4
    $\begingroup$ The terminology multiple regression is fine but increasingly it seems unnecessary to stress multiple as it's the same idea really and having multiple predictors is utterly routine. Conversely, the terminology multivariate regression seems, if not absolutely needed, then at least helpful as flagging a less common variant. I've not previously seen the composite form multivariate multiple regression and (although that composite puts all the cards on the table) I would commend just multivariate regression as the term for what you are doing. $\endgroup$
    – Nick Cox
    Commented Nov 16, 2015 at 18:35
  • 1
    $\begingroup$ There's a nice article in the AJPH that says the word multivariate should not be synonymous with multivariable. (I think that for Americans it is, more than Europeans). $\endgroup$
    – Jeremy Miles
    Commented Dec 30, 2016 at 18:05

1 Answer 1

Reset to default
1
$\begingroup$

Multivariate multiple regression tests multiple IV's on Multiple DV's simultaneously, where multiple linear regression can test multiple IV's on a single DV. This is why multivariate is coupled with multiple regression. The assumptions are the same for multiple regression as multivariate multiple regression.

Improve this answer
$\endgroup$
1
  • 1
    $\begingroup$ This is fine as a start but says absolutely nothing about what the assumptions are. $\endgroup$
    – Nick Cox
    Commented Aug 17, 2017 at 8:24

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.