[R] comparision among multiple subgroups

[David Winsemius]

On Jul 4, 2009, at 11:36 AM, sdzhangping wrote:

> Dear R users:
>
>    In my recent works, I compared the cumulative incidences among  
> three different treatment groups. The cuminc function (cmprsk  
> package ) yielded a graph (refer to figure 1)

No figure attached. Perhaps it did not conform to the rules in the  
Posting Guide and so was scrubbed before distribution?

> and a p value (p = 0.0007). I don’t know how to interpret the  
> meaning of the p value ( one p value and three subgroups) ? Similar  
> puzzles exist in log-rank test of survival data (figure 2).
>
>   What dose the meaning of p value in these conditions?

Hard to tell. No data or output is visible to the list readers.

> What packages and functions should I apply?

To do what?

>
>   thanks
>
>  Ping Zhang



David Winsemius, MD
Heritage Laboratories
West Hartford, CT