[R] How can I "declutter"/make a biplot less messy in R

[Sarah Goslee]

Hi Derek,

It's good form to provide an example with code so that we can see what
functions you're using and what you've tried.

I'm going to assume that you're using princomp(). Here's one way to
reduce the clutter:

pc.cr <- princomp(USArrests, cor = TRUE)
biplot(pc.cr, xlabs=rep("x", nrow(USArrests)))

You can also change the colors and the scalings: see
?biplot.princomp
and
?biplot.default
for details.

Outlier removal may not be a good idea.

You also didn't tell us what your intent was: a publication-quality
plot? Something simply for you to look at yourself? Without knowing
that, it's hard to "improve" it.

Sarah

On Mon, Jun 18, 2012 at 10:37 AM, dss <Derek.Schanze at srnl.doe.gov> wrote:
> I am doing a principle component analysis on a dataset with a lot of
> different variables and have constructed a biplot of the data.
> Unfortunately, as can be seen on the attached image, the biplot is very
> messy, cluttered, and hard to read.  I have performed a few modifications
> including outlier removal from a few of the variables, which has made the
> plot better, however it still is too cluttered to read the variable names.
> Is there any further improvment that can be made to the attached plot and if
> so what steps should I take?
>
> Thanks,
> Derek



-- 
Sarah Goslee
http://www.functionaldiversity.org