[R] Appropriateness of survdiff {survival} for non-censored data

Frostygoat frostygoat at gmail.com
Wed Jul 7 17:25:52 CEST 2010


I read through Harrington and Fleming (1982) but it is beyond my
statistical comprehension.  I have survival data for insects that have
a very finite expiration date.  I'm trying to test for differences in
survival distributions between different groups.  I understand that
the medical field is most often dealing with censored data and that
survival analysis, at least in the package survival, is largely built
around these conventions and differs from a classical biological
perspective.  For example, for lifetable analysis of insects there is
often no need to estimate survival using a Kaplan-Meir estimate
because it is relatively easy to follow a cohort of individuals
through the entire course of life.  Thus I question the
appropriateness of using survdiff in my analysis; I have exact data
yet I would be testing on the Kaplan-Meir estimate of these data in
survdiff.  Thanks for any help.



More information about the R-help mailing list