SKEW.P in ONLYOFFICE: Population-Moment Skewness without Sample Adjustment

Create ShapeData. Put Duration in A1 and enter numeric values 1,2,3,4,10 in A2:A6, measured in minutes. These five observations are deliberately small enough to reconstruct the calculation; they are not a representative study or proof of a distributional model.

Their mean is 4. The deviations are −3,−2,−1,0,6. The sums of squared, cubed and fourth-power deviations are respectively 50,180,1394. In C1:E1 enter Squared, Cubed and Fourth. Put =(A2-AVERAGE($A$2:$A$6))^2, =(A2-AVERAGE($A$2:$A$6))^3 and =(A2-AVERAGE($A$2:$A$6))^4 in C2:E2 and fill through row 6. Summing each helper column reconstructs those totals.

Keep the original numeric values unchanged when adding controls. For typed-input tests, place the literal text note in H2, the logical value TRUE in H3, leave H4 genuinely empty, and put =1/0 in H5. These are different cell types, not four interchangeable missing-value markers. The expected source behavior of references can differ from literal arguments.

The official reference documents SKEW.P and notes that referenced text, logical values and empty cells are ignored while numeric zero values count. The tagged helper computes moments using divisor n.

Calling a formula a population statistic does not prove the selected rows are the complete population relevant to a question. It describes the formula’s convention. If the records are a sample, the unadjusted coefficient may still be descriptively useful, but it is not the same estimator as SKEW.

Let m2=Σ(x−x̄)²/n and m3=Σ(x−x̄)³/n. The result is m3/m2^(3/2). For the five values, m2=50/5=10 and m3=180/5=36. Thus 36/(10√10)≈1.13841995766062.

The adjusted SKEW result is approximately 1.69705627484771. Multiplying SKEW.P by √(n(n−1))/(n−2) reproduces it for these nondegenerate data. Do not compare results from different systems without checking whether they use the adjusted or population-moment convention.

The statistic is dimensionless. Positive rescaling and translation leave it unchanged; reflection changes the sign. A positive result is consistent with a long right-side observation in this tiny dataset, but cannot by itself identify a distribution family or quantify a specific tail probability.

If every observation in the main dataset is repeated once, the empirical distribution and its population moments are unchanged. =SKEW.P(1,2,3,4,10,1,2,3,4,10) therefore still targets 1.13841995766062. The sample-adjusted SKEW correction, in contrast, depends on n and changes when the count changes.

This is a definition check, not permission to duplicate observations to improve evidence. Repeated records may represent actual frequency counts or accidental duplicates; those situations have different data-quality implications. Keep the dataset’s provenance and selection rule separate from the numerical formula.

Zero skewness is not proof of symmetry or normality. The third central moment can vanish for more than one shape, and small samples have substantial variability. Pair the statistic with the data and a clear count rather than replacing a distribution assessment with a single threshold.

The tagged shared helper rejects n≤2 even for SKEW.P. Mathematically, a varying two-point population can have a defined third standardized moment, but =SKEW.P(1,2) is source-predicted #DIV/0! in this implementation. This is an application-domain restriction, not a universal mathematical impossibility.

Three identical numbers also produce #DIV/0! because their zero standard deviation makes standardization invalid. Do not replace that error with zero and interpret the result as symmetry; a constant dataset has undefined standardized skewness under this calculation.

The function includes numeric cells and skips text, logicals and blanks in ranges and arrays. Direct strings, booleans and empty arguments are skipped too, so =SKEW.P(A2:A6,TRUE) retains the main result. Numeric zero counts. Errors in ranges and arrays propagate, while the separate single-reference numeric filter can omit a referenced error. Direct errors propagate. These branch differences are reasons to inspect input types, not to rely on error omission.

The implementation is separate from SKEW but shares its helper with a population flag. That flag changes both the standard-deviation divisor and final scaling. Source array handling does not demonstrate any editor’s native array-entry or spill behavior.

Checked October 18, 2026 against current official help and SDKJS tag v10.0.0.169, commit 3679e0375a758d7dc59d31337a45461be3525ed8. Introduction dates and a final supported-version endpoint are not established. Desktop Editors, Docs, mobile releases and imported files require separate compatibility checks.

Neither the ONLYOFFICE application nor its calculation engine was executed for this guide. The numerical targets are independently calculated mathematical values; code-path and error observations are manual source traces. They are not screenshots, displayed workbook results or promises of identical final digits. Use English function names and comma separators as shown, adapting separators to the editor’s locale.

Keep inputs and intermediate probabilities numeric and unrounded. Format the display rather than replacing values with rounded strings. These moderate scalar examples do not certify extreme-tail accuracy, unusual type coercion, array entry or dynamic spill. Model assumptions must be checked separately from arithmetic correctness.

• Official SKEW.P help, checked October 18, 2026.
• Pinned implementation.
• Shared skewness helper.
• Numeric wrapper and error conversion.
• KURT.
• SKEW.
• TRIMMEAN.
• PROB.