The log-rank test is used to find the difference between two curves. It is a nonparametric test. It is also known as the Mantel-Cox test. It compares the estimates of hazard functions at each observed time. It will test two hypotheses: there is no difference in the survival function based on gender, and there is a difference in survival function based on gender.
Perform the following step:
> survdiff(Surv(cancer$time, cancer$status)~cancer$sex) Output Call: survdiff(formula = Surv(cancer$time, cancer$status) ~ cancer$sex) N Observed Expected (O-E)^2/E (O-E)^2/V cancer$sex=1 138 112 91.5817 4.55228 10.3267 cancer$sex=2 90 53 73.4183 5.67850 10.3267 Chisq= 10.3 on 1 degrees of freedom, p= 0.00131116