Contact

STDEVA in Collabora Office: Sample Spread When Text Counts as Zero

Statistical Intermediate Collabora Office Calc

Audit Collabora STDEVA text-as-zero samples, denominator changes and explicit missing-data policies with reproducible values.

Verification status: See the evidence and test scope described in this guide. How formula examples are checked

Quick answer: STDEVA estimates sample standard deviation while assigning text a value of zero. This can be useful for an explicitly chosen encoding policy, but a word such as Pending is not evidence that a zero was measured.

Availability and version history

  • 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.

Syntax and the text rule

=STDEVA(argument1;[argument2];...)

Collabora lists up to 255 arguments and says at least two values are needed. Arguments may include numeric values, text and references. The source explicitly states that text has value zero; these controls use ordinary nonnumeric text rather than assuming how numeric-looking strings are parsed.

For each included value, use its encoded number, calculate the mean, divide squared deviations by n minus one and take the square root. This independently models the documented sample/text policy.

Example 1: inspect every encoded observation

Enter 4, 8 and Pending in A2:A4:

=STDEVA(A2:A4)

Expected result: 4. The encoded values are 4,8,0. Their mean is 4 and squared deviations are 0,16,16. Sample variance is 32/2=16, so standard deviation is 4.

The explicit scalar control =STDEVA(4;8;"Pending") has the same independently expected result under the stated text-as-zero rule. This does not certify arbitrary string coercion or a complete error matrix.

Example 2: exclusion asks a different question

For just the numeric values 4 and 8, STDEV has independent expected result 2.828427, from squared-deviation total 8 divided by one.

STDEVA’s result of 4 is not a rounding difference. Including an encoded zero changes both the mean and the number of observations. Decide whether text is a zero-valued category or an unavailable measurement before choosing the function.

Advanced example: another text value can reduce the result

Put text Review in A5 and extend the range:

=STDEVA(A2:A5)

Expected result: 3.829708. Encoded values 4,8,0,0 have mean 3, squared-deviation total 44 and sample denominator 3. The result decreases from 4 even though another nonnumeric status has been included.

There is no general rule that adding a zero or text entry must increase standard deviation. The new center and denominator both matter.

Advanced example: replace a status with a real measurement

Restore the three-cell fixture and replace Pending with numeric 6. Values 4,8,6 have mean 6 and squared deviations totalling 8, so expected sample standard deviation is 2.

The result now describes three actual numeric readings. Keep the original status or an audit note when a report changes from an encoded placeholder to a measurement; the new answer represents a change in data quality as well as arithmetic.

Advanced check: a literal zero makes the policy visible

=STDEVA(4;8;0)

Expected result: 4, matching the first text-encoded control. This equality shows the calculation’s chosen encoding. It does not establish that Pending and measured zero have the same real-world meaning.

Empty cells and other boundaries

The dedicated source’s text rule is not a complete specification for physically empty cells, formula-empty strings, logical values, numeric text or errors. Those cases remain unverified rather than borrowed from another spreadsheet application.

STDEVPA uses a population interpretation with a different denominator. Do not mix sample and population measures in one comparison without recording the choice. Zero variability in supplied values is also different from insufficient observations.

References and verification

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.