Trimmed and winsorized summary
Inspect robust location and spread under an explicit symmetric trimming fraction.
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 a fraction from zero to less than 0.5 and 3–500 values. The number removed at each end is floor(fraction×n); at least two observations must remain.
The starting values are hypothetical. Results update as you edit. Use aggregate or synthetic inputs only.
Calculation results
- Removed from each tail
- 2
- Retained observations
- 6
- Ordinary mean
- 9.5
- Trimmed mean
- 5.5
- Winsorized sample SD
- 2.1730675
How to interpret the output
The trimmed mean excludes extreme observations; winsorization replaces them with the nearest retained values before computing sample SD. These are different operations. No observations should be deleted from the original dataset based on this summary alone; the chosen rule changes the estimand and should be prespecified.
Method referenceSource checked 2026-11-06. Read the reproducible calculator validation report.