Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: correlation


From   Maarten buis <maartenbuis@yahoo.co.uk>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: correlation
Date   Tue, 4 Dec 2007 09:57:18 +0000 (GMT)

--- fponzano@unipmn.it wrote:
> I want to check whether, for each individual (id), x1, x2 and x3 are
> correlated. My question is: is it correct to run a Spearman's 
> correlation test (or a Pearson's correlation test if I manage to
> collect more observations) even if observations within the same group
> are probably not independent?

No, but you knew that already. 

Such correlation tests have a very limited field of application: they
are applicable if the main point of your study is to test the
hypothesis that x1, x2, and x3 are univariately linearly associated
with one another. 

Two comments:
- often these are not the hypotheses of interest, but are part of some
sort of model selection / data exploration excersise. In both cases
test are not needed and not desirable. 

- A scatterplot matrix is much more informative, as this also shows
non-linear relations, see: -help graph matrix-.

-- Maarten

-----------------------------------------
Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands

visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434

+31 20 5986715

http://home.fsw.vu.nl/m.buis/
-----------------------------------------


      __________________________________________________________
Sent from Yahoo! - the World's favourite mail http://uk.mail.yahoo.com

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index