Kaplan–Meier Survival Analysis in Gnumeric

The Kaplan–Meier estimator represents the fraction of subjects who have not yet experienced an event at each observed time. Gnumeric documents a Kaplan–Meier analysis tool with fields for time, optional censoring indicators, and optional group labels.

This is a teaching dataset only—it contains no patient information and is not a clinical analysis or medical recommendation.

Use the synthetic example

Download synthetic survival observations (CSV). It contains Time, Group, and Censored columns. In this example, 0 means an observed event and 1 means censored, matching the example described in Gnumeric’s manual.

Consider Group A in the synthetic CSV: it has five imaginary participants and its first event occurs at time 2. If all five are at risk immediately before that time and one event happens, the Kaplan–Meier estimate just after the event is:

S(2) = (5 - 1) / 5 = 0.8

A censored observation changes the number remaining at risk for later event times, but does not itself cause a downward survival step.

Set up Gnumeric’s analysis tool

1. Import the CSV and confirm that the Time values are numeric and the censor flags are coded consistently.
2. Open the Kaplan–Meier analysis tool.
3. Select the time column. When using censorship, enable the censoring option and select its indicator column.
4. Explicitly verify the meaning of the censor flag in the currently installed version—never assume every application uses the same 0/1 convention.
5. If comparing groups, enable the multiple-group option and map group labels to the correct column.
6. Review the resulting survival table, graph, at-risk counts and any group-comparison statistics.

Avoid misleading results

• Reversing event and censor codes changes the analysis entirely.
• Subjects entering at different times or competing events may need methods not represented by this simple example.
• Group differences and a log-rank p-value require attention to sample size, assumptions and censoring patterns.
• This tool is not a substitute for a reviewed research analysis plan.

Verification note: This guide follows the official Gnumeric Kaplan–Meier workflow. The synthetic data and arithmetic are independent illustrations; no Gnumeric native execution has been performed.

Official source: Application handbook or manual. This guide is an editorial draft prepared for staging; exact menus and outputs must be checked in the target software/version before public release.