Two-group log-rank comparison
Compare right-censored event times using the pooled-risk-set log-rank score.
Your entries stay in this page’s memory and are not submitted to us. Downloads are generated locally. Reloading or leaving the page clears your entries. Do not enter patient information.
Use two nonempty groups, positive times and binary event indicators. All enter at time zero; no weights or delayed entry. Tied event counts use the hypergeometric variance.
The starting values are hypothetical. Results update as you edit. Use aggregate or synthetic inputs only.
Calculation results
- Group 1 observed minus expected
- -0.6
- Score variance
- 0.74
- Chi-square
- 0.48648649
- Asymptotic p
- 0.4854988
How to interpret the output
The ordinary log-rank statistic is (observed−expected)²/variance with one chi-square degree of freedom. It is sensitive to hazard differences but does not estimate a hazard ratio. Crossing curves, dependent censoring or few events can undermine its usefulness and asymptotic approximation.
Method referenceSource checked 2026-11-06. Read the reproducible calculator validation report.