[Rd] survdiff Mac binary in Survival pckg (PR#9302)

ulrich.beutner at kssg.ch ulrich.beutner at kssg.ch
Wed Oct 18 12:21:54 CEST 2006


Full_Name: Ulrich Beutner
Version: 2.4.0
OS: MacOS 10.4 (PPC G4)
Submission from: (NULL) (193.134.32.99)


It appears that there is bug in the Macintosh binary of survdiff (part of
survival library)
Following the output of the example for survdiff under Windows XP and
Macintosh.
While under WinXP correct Chisquare and P values are reported, in the Mac
Version Chisquare is 0 (also with other datasets) and as a consquence P is
always 1.

(all programs downloaded and installed on Oct18, 2006, thus I guess I used the
most current version in all cases)


Windows XP

> survdiff(Surv(futime, fustat) ~ rx,data=ovarian)
Call:
survdiff(formula = Surv(futime, fustat) ~ rx, data = ovarian)

      N Observed Expected (O-E)^2/E (O-E)^2/V
rx=1 13        7     5.23     0.596      1.06
rx=2 13        5     6.77     0.461      1.06

 Chisq= 1.1  on 1 degrees of freedom, p= 0.303


        ^^^^ what it should be       ^^^^^^^^  


Macintosh

> survdiff(Surv(futime, fustat) ~ rx,data=ovarian)
Call:
survdiff(formula = Surv(futime, fustat) ~ rx, data = ovarian)

      N Observed Expected (O-E)^2/E (O-E)^2/V
rx=1 13        7     5.23     0.596      1.06
rx=2 13        5     6.77     0.461      1.06

 Chisq= 0  on 1 degrees of freedom, p= 1 

   ^^^^^^      Incorrect data      ^^^^^^




More information about the R-devel mailing list