Plotting points with color and shape based on data variables

Question

I am trying to make a scatter plot with the colors of each point corresponding to one variable and the shape of each point corresponding to another variable. Here is some example data and the code I used to make the second plot:

Example data:(of 3 points)
 X    Y    att1    att2

.5    .5    1       A
.24   .8    3       B
.6    .7    5       C

code:(for image2)
> plot(X,Y, col=statc[att2], pch = 15)
> legend("right", statv, fill=statc)

Where:
> statv
[1] "A"  "B" "C"  
> statc
[1] "red"    "blue"   "orange"

I have done this individually but dont know how to do both. Here is two plots:

1

2

For example: I want the colors to apply to the points with the same att1 and the shapes to apply to points with the same att2

Answer 1

One of the domain where ggplot2 excels , comparing to other R system plots, is mapping plot parameters to data variables.( via aesthetics mechanism)

library(ggplot2)
dat <- data.frame(X =runif(20),
                     Y =runif(20),
                     att1 = gl(5,20/5),
                     att2 =gl(3,20/3))
ggplot(dat,aes(x=X,y=Y,color=att1,shape=att2)) +
    geom_point(size=5) 

You can do it in the base plot also, but you should generate manually the legend ...

plot(dat$X,dat$Y,pch=as.integer(dat$att1),col=as.integer(dat$att1))

Answer 2

Is this what you want? [df is your data formatted as above.]

library(ggplot2)
ggplot(df) + geom_point(aes(x=X,y=Y,color=factor(att1),shape=att2),size=5)

Produces this with your data: