SPSS command: SPSS SURVIVAL
Actuarial life tables in your browser
A life table groups survival times into equal intervals and reports, for each interval, how many cases entered, withdrew, were exposed to risk and had the event.
Free · No account · Files stay on your device · Updated 18 October 2026
What it tells you
Crosstabs runs the SPSS SURVIVAL procedure on your device, with cumulative survival, probability density and hazard rates with standard errors, the life-table median, and the Wilcoxon (Gehan) comparison of groups.
- The proportion of cases surviving each interval and the cumulative proportion surviving to its end, with standard errors.
- The hazard rate per unit of time in each interval, which shows whether risk rises, falls or stays flat.
- The probability density of the event in each interval, i.e. where in time events concentrate.
- The life-table median survival time, or that it lies beyond the last interval.
- Whether survival differs across factor levels, by the Wilcoxon (Gehan) statistic, overall and pairwise.
When to use it
Use it when
- Large samples where a row per event time would be unreadable, such as thousands of customers or policyholders.
- Times recorded in coarse units (months, years) with many ties, where interval summaries are the natural presentation.
- Reporting hazard by period, for example to show that cancellation risk peaks in the second quarter.
Use something else when
- Small samples with precise times: Kaplan-Meier uses every event time and loses no information to grouping.
- You need to adjust for covariates: use Cox regression.
- You need a second, controlling factor: this panel takes one factor, so split the file by the control variable or use Kaplan-Meier with strata.
How to run it in Crosstabs
- 01
Open your data
Open a file with one row per case in the workspace and switch to Statistics mode.
- 02
Choose Life Tables
In the Survival group, pick Life Tables (SURVIVAL).
- 03
Set Time, Status and the event value
Choose the Time and Status variables and the Value indicating the event has occurred; other status values are treated as withdrawals (censored).
- 04
Define the intervals
Enter Display time intervals 0 through and By (interval width), e.g. 36 by 6. If the width does not divide the range evenly it is adjusted to the nearest whole number of intervals; times beyond the last limit fall in an open-ended final interval.
- 05
Add a factor
Optionally choose a Factor (optional) for one table per level and the Gehan test; tick Pairwise comparisons when it has three or more levels.
- 06
Choose a plot and run
Set Plot to Survival, One minus survival, Hazard or Density, declare any weight as a frequency weight, and press Run.
Worked example
Computed by the Crosstabs engine from the synthetic data below when this page was built — not typed in.
Data
90 synthetic patients, 45 per treatment arm. Time is months to relapse (one decimal); patients lost to follow-up or relapse-free at 36 months are censored. Times come from exponential distributions with a fixed seed. Intervals run from 0 through 36 by 6.
| Interval Start Time | Number Entering Interval | Number Withdrawing during Interval | Number Exposed to Risk | Number of Terminal Events | Proportion Terminating | Cumulative Proportion Surviving at End of Interval | Hazard Rate |
|---|---|---|---|---|---|---|---|
| 0 | 45 | 7 | 41.5 | 13 | 0.31 | 0.69 | 0.062 |
| 6 | 25 | 3 | 23.5 | 8 | 0.34 | 0.45 | 0.068 |
| 12 | 14 | 0 | 14.0 | 6 | 0.43 | 0.26 | 0.091 |
| 18 | 8 | 1 | 7.5 | 2 | 0.27 | 0.19 | 0.051 |
| 24 | 5 | 1 | 4.5 | 1 | 0.22 | 0.15 | 0.042 |
| 30 | 3 | 0 | 3.0 | 0 | 0.00 | 0.15 | 0.000 |
| 36 | 3 | 3 | 1.5 | 0 | 0.00 | 0.15 | — |
Median survival time: 10.79. The last interval is open-ended and collects every time beyond 36.
| Item | Wilcoxon (Gehan) Statistic | df | Sig. |
|---|---|---|---|
| Overall | 3.928 | 1 | .047 |
Mean scores: New 8.56; Standard −8.56.
How to read it
By the end of the 24–30 month interval, the cumulative proportion surviving relapse-free is 0.15 on the standard arm and 0.41 on the new treatment. The life-table median is 10.79 months for the standard arm and 19.74 for the new arm. The Wilcoxon (Gehan) statistic is 3.93 on 1 df, p = .047. A positive mean score means longer times than the other arm: the new arm scores 8.56.
Output tables explained
- Life Table
- Per interval: start time, number entering, withdrawing, exposed to risk (entering minus half the withdrawals), terminal events, proportions terminating and surviving, cumulative proportion surviving, probability density and hazard rate, each with a standard error. The footnote gives the median.
- Overall Comparisons
- The Wilcoxon (Gehan) statistic, df and significance for all groups, then one row per pair when pairwise comparisons are requested. Mean scores show which group survives longer: a positive mean score means longer times.
Assumptions and notes
- Withdrawals are spread evenly across each interval, which is why half of them count as exposed to risk.
- Censoring is non-informative and survival does not depend on calendar time of entry.
- The hazard is roughly constant within each interval; very wide intervals hide changes in risk.
- Weights must be declared as whole-number frequency weights; sampling weights stop the procedure.
How the results are validated
No mainstream library reproduces SPSS's actuarial table, so the engine is checked against an independent NumPy implementation of the SURVIVAL algorithm on a weighted 150-case reference file: entering, exposed, proportion terminating, cumulative surviving and its SE, density, hazard and their SEs, the median, and the Gehan (Lee-Desu) statistic and p value all agree to 1e-9. Validation evidence
Questions
How wide should the intervals be?
Wide enough that most intervals contain several events, narrow enough to show how the hazard changes. For monthly churn data over two years, 3- or 6-month intervals are typical.
Why does the median differ from the Kaplan-Meier median?
The life-table median interpolates within the interval where cumulative survival crosses 0.5, while Kaplan-Meier reads it off the exact event times. With narrow intervals the two converge.
Is the Gehan test the same as the Breslow test in Kaplan-Meier?
Both are generalized Wilcoxon tests that weight early differences heavily, but they are computed differently. Gehan's D here compares individual survival times (Lee-Desu), not the grouped intervals.
What is the open-ended last interval?
Every time beyond your last limit falls into one final interval with no width, so its density and hazard cannot be computed and are shown as —.
Related procedures
- Kaplan-MeierKM
- Cox RegressionCOXREG
- Complex Samples Cox RegressionCSCOXREG
- FrequenciesFREQUENCIES
- Free SPSS alternative: full coverage list
Run Life Tables on your own data
Open a .sav, CSV or Excel file. Everything runs in your browser; the file never leaves your device.
SPSS is a trademark of IBM Corp. Crosstabs is not affiliated with or endorsed by IBM.