[R] possible problem with plot.lm

Benilton Carvalho bcarvalh at jhsph.edu
Thu Aug 20 16:53:29 CEST 2009


Here's one toy example that shows what I believe to be a problem with  
plot.lm. This was brought up by another user (Marcos Tadeu Andrade  
Cordeiro). I took a look at the source and the problem appears to be  
related to the fact that you need to reorder the data by the group  
means and the correct order of the labels is lost.

a <- rnorm(50,0,2)
x <- rep(seq(10, 50, by=10), each=10)
y <- x+a
f <- rep(c("e","a","c","b","d"), rep(10,5))
f <- as.factor(f)
y[1] <- 200
mod1 <- aov(y ~ f)

The last plot suggests that the outliers belongs to group D, when in  
fact it belongs to E.

Does this make any sense or are we missing something? I'm reporting  
this with R-rc, but it is reproducible with previous versions too.

Thanks a lot,



 > sessionInfo()
R version 2.9.2 RC (2009-08-17 r49309)


attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base

More information about the R-help mailing list