R help - May 2005 - Accounting for clustered data in Rpart or Mvpart

I am working on data that is in clusters (eg events from persons in
the family).So I have a clustered data set up and would like to build
a tree(regression /classification tree) taking into account this
clustered nature of the data.

Do anybody know how to do this or maybe the code to take into account
the clustered data in rpart

I have a dataframe in which I have used a cerain procedure to select
the response a subset of the response and obtained a vector y.Now I
need to fit a glm of the response y given x(the covariates) but now I
get an error that the variables are not of equal length.How can I come
up with a way to pick the x values corresponding to only the subset of
the response I am interested in.

Regards,

Luwis