VAR.S in Collabora Office: Sample Variance, Squared Units and Duplicate Records

• 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.

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

Collabora lists up to 255 arguments and requires at least two values. A range may supply many observations within one argument. These examples use numeric inputs only; the short dedicated section does not settle all logical, text, empty or error-valued combinations.

Sample variance describes squared deviations around the arithmetic mean. It does not itself take a square root. STDEV.S describes the related original-unit measure.

Enter 2, 6 and 10 in A2:A4:

=VAR.S(A2:A4)

Expected result: 16. The mean is 6, deviations are -4,0,4 and squared deviations total 32. Dividing by 3-1 gives 16.

If the original readings are in minutes, report 16 squared minutes. The square root is 4 minutes, which is a standard deviation rather than the requested variance.

For values 102,106,110, the mean is 106 but the same deviations -4,0,4 remain. =VAR.S(102;106;110) therefore has independent expected result 16.

A calibration offset can move every reported measurement while leaving this spread statistic unchanged. A correct variance does not prove that the measurement origin or mean is correct.

Multiply the original readings by -3 to obtain -6,-18,-30. Their mean is -18, and squared deviations total 288.

=VAR.S(-6;-18;-30)

Expected result: 144, nine times 16. The negative factor reverses ordering, but variance remains nonnegative. Treating it as -48 would confuse signed scaling with squared deviations.

Enter 2,6,10,2,6,10 in C2:C7:

=VAR.S(C2:C7)

Expected result: 12.8. The mean is still 6; squared deviations double to 64, while the sample denominator is five. It does not double from two to four.

This can expose a duplicated join or copied rows. Repeated records do not automatically add independent statistical information, even though the formula sees a longer range.

For the original A2:A4, put its mean in E2. Enter =(A2-$E$2)^2 in B2 and fill down. B2:B4 should contain 16,0,16. Then =SUM(B2:B4)/2 gives expected 16.

The explicit denominator is justified only for these three included numbers. If the range changes, recalculate the participating count under the chosen data policy rather than keeping a convenient fixed denominator.

The source names COM.MICROSOFT.VAR.S and states that it is outside the cited OpenFormula 1.3 standard. This format note does not establish a first Collabora release or a successful round-trip in every other application.

Exact error codes, direct text coercion and mixed-data placement rules remain unverified. A sample with fewer than two usable observations lacks a positive n minus one denominator; do not silently turn an unavailable estimate into a reported zero.

• Collabora Office 26.04: VAR.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.