Statalist


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

Re: st: RE: easy command for correlation tables???


From   "Bastian Steingros" <Steingros@gmx.de>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: RE: easy command for correlation tables???
Date   Mon, 17 Nov 2008 17:01:45 +0100

Hi Nick,

thanks for your reply.

Do you know why many studies report the pearson and the spearman correlation coefficients?
I have looked in some statistic books but they did not explain well why one should both...

B


-------- Original-Nachricht --------
> Datum: Mon, 17 Nov 2008 15:33:12 -0000
> Von: "Nick Cox" <n.j.cox@durham.ac.uk>
> An: statalist@hsphsun2.harvard.edu
> Betreff: st: RE: easy command for correlation tables???

> Martin Weiss made some suggestions. 
> 
> I have three quite different comments: 
> 
> 1. -correlate- rather than -spearman- is more closely related to
> properties affecting multiple regression. 
> 
> 2. With that many variables it is easy to produce a scatter plot matrix
> using -graph matrix-. Mentioning the response (dependent variable in
> your terms) last means that the last row of the matrix has scatter plots
> with the response on the vertical axis: 
> 
> . graph matrix weight length disp turn mpg 
> 
> 3. -corrci- from Stata Journal 8(3) 2008 offers an alternative to
> -correlate-. This is the default display: 
> 
> . corrci mpg  weight length turn disp
> 
> (obs=74)
> 
>                            correlations and 95% limits
> mpg          weight           -0.807   -0.874   -0.710
> mpg          length           -0.796   -0.867   -0.693
> mpg          turn             -0.719   -0.814   -0.587
> mpg          displacement     -0.706   -0.804   -0.569
> weight       length            0.946    0.915    0.966
> weight       turn              0.857    0.782    0.908
> weight       displacement      0.895    0.838    0.933
> length       turn              0.864    0.792    0.913
> length       displacement      0.835    0.750    0.893
> turn         displacement      0.777    0.667    0.854
> 
> The -ci- stands for confidence interval. -corrci- is dedicated to the
> idea that a confidence interval is often superior to a P-value, just as
> a P-value is usually superior to stars. 
> 
> -corrci- thus does not support stars, least of all in a non-standard
> scheme like that requested by Bastian (although possibly he did not mean
> what he wrote). 
> 
> I have already vented my spleen against stars on this list, so I won't
> repeat that in detail here. 
> 
> <http://www.stata.com/statalist/archive/2007-03/msg00646.html> 
> 
> Nick 
> n.j.cox@durham.ac.uk 
>  
> Bastian Steingros
> 
> I want to make a correlation analysis before running some regressions.
> 
> I have 4 independent and 1 dependent variable.
> Using the command *spearman*, is it possible to use a further command
> help me to put the result directly into a table (inclusive the
> significance stars * 0.01 ** 0.05)??
> 
> 
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/

-- 
"Feel free" - 5 GB Mailbox, 50 FreeSMS/Monat ...
Jetzt GMX ProMail testen: http://www.gmx.net/de/go/promail
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   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