[R] mca-graphics: all elements overlapping in the help-example for multiple correspondence analysis

Michael Reinecke Reinecke at consultic.com
Thu Feb 1 16:08:30 CET 2007


Thank you very much, that works fine!

I now realize that I should have looked up not only the help pages for plot and mca but also for plot.mca, which I did not think possible (unfortunately I am a too sporadic user to know where to get the appropriate information).

Best regards,

Michael
 

> -----Ursprüngliche Nachricht-----
> Von: Prof Brian Ripley [mailto:ripley at stats.ox.ac.uk] 
> Gesendet: Mittwoch, 31. Januar 2007 10:19
> An: Michael Reinecke
> Cc: r-help at stat.math.ethz.ch
> Betreff: Re: [R] mca-graphics: all elements overlapping in 
> the help-example for multiple correspondence analysis
> 
> On Wed, 31 Jan 2007, Michael Reinecke wrote:
> 
> > Dear all,
> >
> > I tried out the example in the help document for mca (the 
> multiple correspondence analysis of the MASS package):
> >
> > farms.mca <- mca(farms, abbrev=TRUE)
> > farms.mca
> > plot(farms.mca)
> >
> > But the graphic that I get seems unfeasible to me: I cannot 
> recognize 
> > the numbers (printed in black) because they are all overlapping and 
> > concealing each other. I don ´t dare using my own data, 
> which consist 
> > of several hundred cases - I guess I won ´t see anything.
> >
> > How can I solve this? Thank you for any idea!
> 
> Some levels do overplot, as they are identical (this is an 
> unusual example).  But as you see in the book, not many, and 
> you can adjust pointsize of your device or 'cex' to mitigate 
> the problem.
> 
> Plotting the rows is optional: see the help page.  I would 
> not recommend plotting rows for several hundred cases.
> 
> -- 
> Brian D. Ripley,                  ripley at stats.ox.ac.uk
> Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
> University of Oxford,             Tel:  +44 1865 272861 (self)
> 1 South Parks Road,                     +44 1865 272866 (PA)
> Oxford OX1 3TG, UK                Fax:  +44 1865 272595
>



More information about the R-help mailing list