The table provides sample size (per group) for powers of 0.8, 0.9, and 0.95, Probability of Type I Error (α) of 0.1, 0.05, and 0.01, and for the one and two tail model. The survival rates are from 0.02 (2%) to 0.98 (98%) at 0.02 (2%) intervals

When there is more than two groups, the sample size should be the same as for two groups, but the two tail model should be used. The two survival rates should be the narrowest anticipated ot to be detected between any of the two groups.

Abbreviations are :

• α = Probability of Type I Error
• Power = 1=β
• s1 and d2 = Survival Rate of two groups being compared
• 1Tail and 2Tail are sample size (per group, assuming equal sample size groups) for the one and two tail model

• In the table, the sample size is that for each of two groups, assuming that the sample size in the two groups are same or at least similar. This allows a quick reference for planning purposes.
• The sample size calculated from the program in Sample Size for Survival (Kaplan Meier Log Rank Test) Program Page is that for both groups combined, allowing for a large difference between the sample size between the two groups. This flexibility is needed when the index cases are more difficult to come by, so that an index case is often matched by 2 to 3 background cases (as is in the case of cancer).
• Where the sample size are the same in the two groups, the sample size calculated using the algorithm in Sample Size for Survival (Kaplan Meier Log Rank Test) Program Page is for both groups, and therefore twice that in the table of this page.