[R] linear correlation?

Andrew Perrin aperrin at email.unc.edu
Sat Mar 9 20:44:03 CET 2002


All I can say is, please read the whole thread. He's not correlating two
variables across cases, as you assume in your message. He's correlating
two cases across variables, and getting meaningless (but attractively
high) numbers.

ap

---------------------------------------------------------
   Andrew J. Perrin - Assistant Professor of Sociology
        University of North Carolina, Chapel Hill
269 Hamilton Hall CB#3210, Chapel Hill, NC 27599-3210 USA
   andrew_perrin at unc.edu - http://www.unc.edu/~aperrin

On Fri, 8 Mar 2002 Liu.Chunhua at epamail.epa.gov wrote:

> 
> What about if we measure the height (x1 in cm) and weight (x2 in kg) of
> a sample of people from some population. It seems it makes sense to me
> to get the correlation between x1 and x2.
> 
> Charlie Liu,
> Intern at ECO/EPA.
> 
> 
> 
>                                                                                                          
>                     "Scott, Uriel"                                                                       
>                     <uriel.scott at mirant.        To:     "Scott, Uriel" <uriel.scott at mirant.com>, 'dechao 
>                     com>                        wang' <dechwang at yahoo.co.uk>, r-help at stat.math.ethz.ch   
>                     Sent by:                    cc:                                                      
>                     owner-r-help at stat.ma        Subject:     RE: [R] linear correlation?                 
>                     th.ethz.ch                                                                           
>                                                                                                          
>                                                                                                          
>                     03/07/02 12:28 PM                                                                    
>                                                                                                          
>                                                                                                          
> 
> 
> 
> 
> 
> Sorry, I also misread your original question and thought x1 was in cm
> and x2
> in kg.
> 
> I don't think it makes any sense for some values of x1 (or x2) to be in
> cm
> and others in kg.  How can they represent samples from the same
> population?
> It would be okay if, say, some were in cm and others in km as they are
> equivalent units, and you could simply convert to the same unit, but
> otherwise I don't see how some members of a population are in cm and
> others
> in kg.
> 
> > -----Original Message-----
> > From:         Scott, Uriel [SMTP:uriel.scott at mirant.com]
> > Sent:         Thursday, March 07, 2002 10:12 AM
> > To:           'dechao wang'; r-help at stat.math.ethz.ch
> > Subject:           RE: [R] linear correlation?
> >
> >
> > Whether the two variables have the same units does not matter.
> Moreover,
> > even if there were some way of converting cm to kg the correlation
> would
> > still be the same because the correlation is invariant under unit
> > conversion
> > as it is invariant under multiplication of its arguments by a
> constant.
> >
> > As for your second question, the correlation estimator is a continuous
> > function of each of the individual data points, so perturbing the
> values
> > of
> > any of them by a sufficiently small amount will only perturb the
> > correlation
> > by a small amount.
> >
> > > -----Original Message-----
> > > From:       dechao wang [SMTP:dechwang at yahoo.co.uk]
> > > Sent:       Thursday, March 07, 2002 5:34 AM
> > > To:         r-help at stat.math.ethz.ch
> > > Subject:         [R] linear correlation?
> > >
> > > Hi, I have checked statistic textbooks about
> > > correlations, but I am still not sure the correlation
> > > analysis with different units, for example,
> > >
> > > x1<-c(1,  2,  3,   100, 200, 300)
> > > x2<-c(1.1,2.8,3.3, 108, 209, 303)
> > > the unit of the first 3 numbers is cm
> > > the unit of the last 3 numbers is kg
> > >
> > > cor(x1,x2)=0.999655
> > >
> > > Can I explain the correlation coefficient as normal in
> > > which all numbers have the same unit?
> > >
> > > Secondly, if keep the three large numbers unchanged,
> > > just change the three small numbers, the coefficient
> > > changes little, this means that the variation of three
> > > small numbers is hidden by the three larger numbers.
> > > Is there any solution in R to solve this issue?
> > >
> > > Thanks,
> > >
> > > Dechao
> > >
> > > __________________________________________________
> > >
> > > Everything you'll ever need on one web page
> > > from News and Sport to Email and Music Charts
> > >
> > >
> > -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.
> -.-.
> > > -.-.-
> > > r-help mailing list -- Read
> > > http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
> > > Send "info", "help", or "[un]subscribe"
> > > (in the "body", not the subject !)  To:
> r-help-request at stat.math.ethz.ch
> > >
> > _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._.
> _._.
> > > _._._
> > -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.
> -.-.
> > -.-.-
> > r-help mailing list -- Read
> > http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
> > Send "info", "help", or "[un]subscribe"
> > (in the "body", not the subject !)  To:
> r-help-request at stat.math.ethz.ch
> > _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._.
> _._.
> > _._._
> -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.
> -.-.-.-
> r-help mailing list -- Read
> http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
> Send "info", "help", or "[un]subscribe"
> (in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
> _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._.
> _._._._
> 
> 
> 
> 
> -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
> r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
> Send "info", "help", or "[un]subscribe"
> (in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
> _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
> 

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._



More information about the R-help mailing list