R help - Jan 2011 - Inconsistencies in the rpart.object help file?

Hello all,

I'm was going through the help for
?rpart.object
And noticed some inconsistencies, Some might be a mistake in the help file
and some might be my misunderstanding.

The help in the section:
value -> frame (first paragraph), states that:
>  yval, the fitted value of the response at each node, *and splits, a two
> column matrix of left and right split labels for each node. *

But from looking at the object, for example

fit1 <- rpart(Kyphosis ~ Age + Number + Start, data=kyphosis)
fit1$frame

      var  n wt dev yval complexity ncompete nsurrogate    yval2.1
 yval2.2    yval2.3    yval2.4    yval2.5

1   Start 81 81  17    1 0.17647059        2          1  1.0000000
64.0000000 17.0000000  0.7901235  0.2098765

2   Start 62 62   6    1 0.01960784        2          2  1.0000000
56.0000000  6.0000000  0.9032258  0.0967742

4  <leaf> 29 29   0    1 0.01000000        0          0  1.0000000
29.0000000  0.0000000  1.0000000  0.0000000

I can't see any "splits" column.

I'm also not sure I understand what the yval2 columns signify (even that I
read what it says in the help).

p.s: I hope I sent it to the correct e-mail.

Best,
Tal





----------------Contact
Details:-------------------------------------------------------
Contact me: Tal.Galili@gmail.com |  972-52-7275845
Read me: www.talgalili.com (Hebrew) | www.biostatistics.co.il (Hebrew) |
www.r-statistics.com (English)
----------------------------------------------------------------------------------------------

	[[alternative HTML version deleted]]

> splits, a matrix of split labels
 This is a mistake in the help file.  
> I'm also not sure I understand what the yval2 columns signify (even
that I read what it says in the help).

  For the kyphosis data set y is a factor with two levels.  For the
first split yval2 contains (1, 64, 17, .7901235, .2098765); this node is
classified into group 1, it contains 64 group 1 and 17 group 2
observations, pr(group 1) = .79 and pr(group 2) = .21.
  Note that if the model specifies priors that the last two values will
not be 64/81 and 17/81.

Terry T.