📸 Log-Rank test in GraphPad Prism: Survival analysis menu
What Is the Log-Rank Test?
The log-rank test (Mantel-Cox test) compares two or more Kaplan-Meier survival curves to determine whether the groups have statistically different survival distributions. It is non-parametric and considers the entire follow-up period, weighting all time points equally.
Use log-rank when you want to test for an overall group difference. Use Cox regression when you need to adjust for covariates or estimate hazard ratios.
Running Log-Rank in GraphPad
Step 1: Enter data in survival table format: Column A = Time, additional columns = number at risk/events per group.
Step 2: Analyze → Survival analysis → Log-rank test (Mantel-Cox).
Step 3: Select the grouping columns → OK. GraphPad plots KM curves and reports the log-rank p-value, chi-square, and median survival times automatically.
📸 Log-rank test results — significant difference between survival curves
APA Reporting
A log-rank test indicated that survival curves differed significantly between treatment groups, χ²(1)=8.42, p=.004. Median survival was 18.4 months (95% CI [15.2, 21.6]) for Treatment A vs. 11.2 months (95% CI [8.4, 14.0]) for Treatment B.