[R] graphics question

[Thomas Lumley]

On Wed, 15 May 2002, Paul Lemmens wrote:

> --On woensdag 15 mei 2002 8:59 -0700 Thomas Lumley
> <tlumley at u.washington.edu> wrote:
>
> > On Wed, 15 May 2002, Paul Lemmens wrote:
> >
> >> The graph I'm looking for has on the y-axis the RT values (in ms), so
> >> this should range from 350 - 450. On the x-axis I want two ticks for
> >> each level (sc and oc) of the 'cat' factor. In the plot, three lines
> >> should be displayed, for each level of the factor 'loc', pref. including
> >> a legend.
> >>
> >> How do I do this???
> >
> > Does example(interaction.plot) look like the sort of thing you want?
> >
> Yes!! This goes a great ways into what I want. However, if I do:
>
> data <- c(390,416,406,396,421,415)
> psc <- gl(2,3, labels=c("samecat","othercat"))
> loc <- gl(3,2, labels=c("corresp", "noncorresp", "noninf"))
> interaction.plot(psc, loc, data)
>
> I get only one line (for noncorresp) which looks as if it is the average of
> all data points (2 times the average of 3 points). This makes sense because
> of the fun argument of interaction.plot (which, to me, seems to default to
> mean). Can I override this with some function that just returns one point
> of data??
>

The problem is with your definition of psc and loc. You haven't defined
them the way you think you have:

> data.frame(data,psc,loc)
  data      psc        loc
1  390  samecat    corresp
2  416  samecat    corresp
3  406  samecat noncorresp
4  396 othercat noncorresp
5  421 othercat     noninf
6  415 othercat     noninf

You need something like
loc <- factor(rep(1:3,2), labels=c("corresp", "noncorresp", "noninf"))

and then it works


	-thomas


-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._