STDEV.S in Collabora Office: Sample Spread, Duplication and Measurement Scale

• Introduced: original Collabora release and launch date are not established by the reviewed sources.
• Availability: currently documented in Collabora Office 26.04 desktop help.
• Platform scope: online and mobile deployments are not inferred from this desktop-help release.
• Checked: October 14, 2026. Function-specific documentation and independent calculations; no native Collabora execution.

=STDEV.S(argument1;[argument2];...)

Collabora lists up to 255 arguments and says the parameters should provide at least two values. The examples supply ordinary numeric cells. Although the syntax allows text and references, the dedicated STDEV.S section does not fully explain their conversions, so those cases are not inferred from STDEV or Excel.

Enter 2, 5, 5 and 8 in A2:A5:

=STDEV.S(A2:A5)

Expected result: 2.449490, displayed to six decimals. The mean is 5; centered differences are -3, 0, 0 and 3. Their squares total 18, and dividing by three gives sample variance 6. Taking its square root returns standard deviation.

Use STDEV.P for the separately defined whole-population question. On these same four numeric values its independent population model gives 2.121320, using denominator four.

In C2:C9 enter 2, 5, 5, 8, 2, 5, 5 and 8:

=STDEV.S(C2:C9)

Expected result: 2.267787. The mean remains 5 and the squared-deviation total doubles to 36, but the sample denominator becomes 7. It is not twice the old denominator of 3.

This is a mathematical control, not a reason to duplicate records. If a join accidentally repeats every row, the changed sample result can expose the problem even though the list’s relative frequencies look unchanged.

For the original four observations, consider these independent datasets:

Adding a constant changes the mean but not deviations. Multiplication by -2 changes the standard deviation by a factor of two; the result stays nonnegative. Equal numeric readings have zero spread under this model, which is different from having no readings.

For the original four-value sample, the standard-deviation model gives sqrt(6). Dividing that by sqrt(4) gives 1.224745. This latter calculation is a different statistical quantity under additional sampling assumptions; it is not the output of STDEV.S.

Keep the measure’s name, sample count and units next to the result. This guide does not establish independence, a confidence interval or any particular inferential method for an actual dataset.

The dedicated section identifies the formula namespace COM.MICROSOFT.STDEV.S and says it is outside the cited OpenFormula 1.3 standard. That is a documented format detail, not evidence that the function is unavailable in Collabora or that every export/import succeeds.

The first Collabora release, Boolean/text conversion, mixed-error behavior and exact insufficient-data error display remain unverified. The older STDEV spelling has its own current section; do not use another function’s type rules to fill gaps in this one.

• Collabora Office 26.04: STDEV.S.
• Editorial testing standards.

Complete function-specific documentation reviewed. Worked outputs are independent mathematical or explicitly documented type-rule controls, not application screenshots or native test results. The stated unresolved boundaries retain their scope.