[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: RE: statalist-digest V4 #3228

From   "Allan Reese (Cefas)" <>
From   Jochen Späth <>
To   <>
Subject   st: RE: statalist-digest V4 #3228
Date   Wed, 29 Oct 2008 10:00:59 -0000

Jochen Späth asks, Date: Tue, 28 Oct 2008 13:36:17 +0100: 
> I want to do a Manova (14 different dependent variables, 2 main factors) 
> and am stuck with the problem that most of my fourteen variables are not 
> normally distributed (and I do not want to transform them in order to get 
> them normal since I have only remote access to the data which complicates 
> things a lot). So, my question is: is there a way to do such a MANOVA in
> STATA using non-parametric techniques ...

Even if the data were normal, I'd ask what Jochen expects to get out of this analysis.  The chance that all 14 variables have constant means over all levels of two factors seems pretty unlikely, and I guess it's the post-hoc analysis that will provide interpretation.  The non-parametric equivalent might be a test that all levels have the same median.  So what, if the underlying distributions differ?  Not-normal in what ways?  Multi-modal?  Zero-inflated?  Skewed?  Outliers that may be mistakes?  I suggest more EDA and less assumption there must be a black-box silver bullet. 


This email and any attachments are intended for the named recipient only.  Its unauthorised use, distribution, disclosure, storage or copying is not permitted.  If you have received it in error, please destroy all copies and notify the sender.  In messages of a non-business nature, the views and opinions expressed are the author's own and do not necessarily reflect those of the organisation from which it is sent.  All emails may be subject to monitoring.

*   For searches and help try:

© Copyright 1996–2023 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index