[R] pie charts?

[Joshua Gramlich]

I did find ?pie and saw the explanation as to why not to use pie
charts.  I may end up using it anyway, because the comparison is
something like 94%,5%,1%, so the difficulty of the human eye to read
area as opposed to length(as in a bar chart) doesn't make much
difference in this case.

Thanks for the reply.


Joshua Gramlich



On Thu, 2003-02-13 at 16:33, Marc Schwartz wrote:
> >-----Original Message-----
> >From: r-help-admin at stat.math.ethz.ch 
> >[mailto:r-help-admin at stat.math.ethz.ch] On Behalf Of Joshua Gramlich
> >Sent: Thursday, February 13, 2003 4:17 PM
> >To: R-help at stat.math.ethz.ch
> >Subject: [R] pie charts?
> >
> >
> >I don't suppose it's possible to create a pie chart in R?  
> >I've got 1500 some odd elements in a frame that are valued at 
> >either -1, 0 or 1 and I'd like to find a reasonable way to 
> >represent the distribution graphically...any ideas?
> >
> >
> >Joshua Gramlich
> >Piocon Technologies
> >Chicago, IL
> 
> 
> You can. See ?pie.
> 
> However, I believe the general disposition would be to not do so.
> 
> You would probably be better off with either a bar chart (see ?barplot
> in base R or ?barplot2 in the gregmisc package) or perhaps a Cleveland
> dot plot (see ?dotchart in base R) depending upon what you might wish
> to show.  barplot2() has some additional features like plotting
> confidence intervals if you wish to include these in your graphic.
> 
> The reasoning behind this is covered in W.S. Cleveland's "Elements of
> Graphing Data" on pages 262 - 264 in the section dealing with what he
> calls "Pop Charts".
> 
> Hope that helps,
> 
> Marc Schwartz
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> http://www.stat.math.ethz.ch/mailman/listinfo/r-help