[R] Fill pattern for Boxplots?

[David L Carlson]

legend() supports density= and angle= arguments. For example, 

plot(1, 1, type="n")
legend("topright", c("Slant Left", "Slant Right"), density=c(20, 20),
     angle=c(135, 45))

You may want to use larger density values than I used in my earlier post
since the legend boxes are pretty small.

-------
David

> -----Original Message-----
> From: meyfarth at uni-potsdam.de [mailto:meyfarth at uni-potsdam.de]
> Sent: Thursday, August 09, 2012 6:55 PM
> To: dcarlson at tamu.edu
> Cc: r-help at r-project.org
> Subject: Re: [R] Fill pattern for Boxplots?
> 
> Thank you! This sounds good and easy to do (if I can as well make
> stripes to the legend)
> 
> --------Original Message----------
> From: David L Carlson <dcarlson at tamu.edu>
> To: <meyfarth at uni-potsdam.de>, 'John Kane' <jrkrideau at inbox.com>
> Date: 09.08.2012 22:10
> Subject: Re: [R] Fill pattern for Boxplots?
> > It's not too hard to use rect() to add shading to the boxplots. The
> boxes
> > are centered on consecutive integers and the width is +/- .4. The
> boxplot()
> > function returns the quartiles of each box.
> >
> > set.seed(42)
> > DF<- data.frame(val=rnorm(150), grp=rep(letters[1:6], 25))
> > outstat<- boxplot(val~grp, DF)           # save the boxplot stats
> > ngroups<- length(levels(DF$grp))         # get number of groups
> > dval<- c(6, 6, 6, 12, 12, 12)            # density vals for each
> group
> > aval<- c(0, 30, 60, 90, 120, 150)        # angle vals for each group
> > rect((1:ngroups)-.4, outstat$stats[2,], (1:ngroups)+.4,
> outstat$stats[4,],
> >       density=dval, angle=aval)            # draw the rectangles
> >
> > You can also play with lty= and lwd= but that will affect the box
> border as
> > well.
> >
> > ----------------------------------------------
> > David L Carlson
> > Associate Professor of Anthropology
> > Texas A&M University
> > College Station, TX 77843-4352
> >
> >> -----Original Message-----
> >> From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-
> >> project.org] On Behalf Of Susanne Meyfarth
> >> Sent: Thursday, August 09, 2012 12:08 PM
> >> To: John Kane
> >> Cc: r-help at r-project.org
> >> Subject: Re: [R] Fill pattern for Boxplots?
> >>
> >> Thank you. I saw these postings, but I don't want to learn lattice
> for
> >> this reason (was afraid to have to change then everything else in my
> >> graph). Anyway, I now tried with different shades of greyscale (4
> >> shades). I'm not fully satisfied with it, but it's ok. It's for a
> >> publication and depending on whether I have to change the graph, I
> >> decide to either put texture in some boxes manually or still look
> for a
> >> solution in R.
> >>
> >>
> >> --------Original Message----------
> >> From: John Kane<jrkrideau at inbox.com>
> >> To: suse<meyfarth at uni-potsdam.de>, r-help at r-project.org
> >> Date: 09.08.2012 16:52
> >> Subject: Re: [R] Fill pattern for Boxplots?
> >>> Probably not. See http://r.789695.n4.nabble.com/Boxplot-Fill-
> Pattern-
> >> td4457209.html on this topic.
> >>> What exactly are you doing?  There may be a workaround or
> >> alternative.
> >>> John Kane
> >>> Kingston ON Canada
> >>>
> >>>
> >> ______________________________________________
> >> R-help at r-project.org mailing list
> >> https://stat.ethz.ch/mailman/listinfo/r-help
> >> PLEASE do read the posting guide http://www.R-project.org/posting-
> >> guide.html
> >> and provide commented, minimal, self-contained, reproducible code.
> 
> --
> Susanne Meyfarth
> Plant Ecology&  Nature Conservation
> University of Potsdam
> Maulbeerallee 2
> D - 14469 Potsdam
> Germany
> 
> Phone: +49-331-977 1910
> Fax: +49-331-977 1930