STDEVPA in WPS Spreadsheet: Population Spread with Explicit Text and Boolean Encoding

• Introduced: documented by 2019-06-19 in the official WPS lesson; the first release is not established.
• Availability: the lesson identifies WPS Office for Windows Personal, without a numeric release range.
• Scope: exact current supported builds and other WPS clients remain unverified. Historical naming does not establish retirement.
• Checked: October 14, 2026. Documentation and independent calculations; no native WPS test was performed.

=STDEVPA(value1, [value2], ...)

The June 19, 2019 Windows Personal lesson describes 1–30 arguments, including arrays or references. That argument limit is historical; one range argument can contain several observations. The current build’s maximum remains unverified.

The source explicitly uses the population n denominator and contrasts STDEVPA with STDEVP, which ignores text and logical values. Its formula note loosely calls the center AVERAGE of the supplied values. For a mixed reference, do not simply apply AVERAGE to the unencoded cells: reproduce the center from the values after the documented encoding.

Examples use English names and comma separators. Empty-cell, formula-empty, error-input and direct numeric-text conversion details are not expanded beyond the source’s stated rules.

Set A2:A6 to numeric 2, numeric 4, =TRUE(), =FALSE(), and the text pending. The documented encoding produces 2, 4, 1, 0 and 0.

=STDEVPA(A2:A6)

The independently expected result is SQRT(56/25), approximately 1.496662955. Here is the full reconstruction:

The encoded numbers total 7, so their mean is 1.4 across five observations. Squared distances total 11.2; divide by five to obtain population variance 2.24, then take the square root.

Enter the five encoded numbers explicitly in B2:B6. Put =AVERAGE(B2:B6) in E2. In C2 enter =(B2-$E$2)^2 and fill through C6:

=SQRT(SUM(C2:C6)/5)

Expected result: approximately 1.496662955. This helper calculation makes the center and denominator inspectable without asking an averaging function to infer STDEVPA’s inclusion policy.

For contrast, STDEVP(A2:A6) uses only numeric 2 and 4 under its separately documented ignore rule. Its expected population standard deviation is 1, with mean 3. Both the mean and n differ from the STDEVPA calculation.

On a separate sheet, enter numeric 1, 2, 2 and =FALSE() in A2:A5. STDEVPA’s encoded set is 1, 2, 2, 0.

Its mean is 1.25 and squared-distance total is 2.75, giving population variance 0.6875 and standard deviation approximately 0.829156198.

If the intended analysis excludes the fourth record, explicitly select only A2:A4. The numeric set 1, 2, 2 has mean 5/3, population variance 2/9, and standard deviation approximately 0.471404521. Excluding a record and encoding it as zero are different statistical decisions.

This example does not certify what an actually blank referenced cell does. The omission is made by changing the selected range, so the comparison remains reproducible within the documented numeric/Boolean scope.

Copy the primary five observations twice into another range. The encoded mean remains 1.4, the squared-distance total doubles to 22.4, and the count doubles to 10. The expected population variance remains 2.24, so the expected STDEVPA remains approximately 1.496662955.

Duplicating the data does not create genuinely new information, even when the numerical population spread stays unchanged. Record duplication and a larger independent sample are different situations.

For multiple identical TRUE values, the encoded numbers are all 1, so their expected spread is 0. That multiple-observation control does not establish singleton or empty-input native behavior.

A text label such as pending can mean “not measured,” while the documented function encodes it as zero. Decide whether that encoding reflects the intended measurement before choosing STDEVPA. A mathematically correct formula can still describe the wrong population if the inclusion policy is inappropriate.

Keep raw inputs beside the encoded audit column. AVERAGE and SQRT reproduce the numeric helper steps but do not independently establish STDEVPA’s native edge cases. Exact handling of imported errors and all current client/build combinations remains unverified.

• WPS Academy: STDEVPA, 2019-06-19 (Chinese).
• Editorial testing standards.

Worked outputs are independently calculated expectations, not captured native results. Function-specific source gaps and product/client boundaries remain explicitly scoped.