SKEW Function (OpenOffice Calc)

• Introduced: First release not established; documented by OpenOffice.org 3.3.
• Supported in: Apache OpenOffice 4.1.16; this is a verified release, not a final supported version.
• Scope: Desktop Calc, English function names and semicolon-separated formulas. Locale and file-format mappings may differ.
• Checked: October 14, 2026. Eight reference formulas and the listed boundary cases were recalculated in 4.1.16 on Linux x86-64, Build 9816, using XLSX import. Source and independent mathematics were also checked.

SKEW returns adjusted sample skewness based on the third standardized central moment. It describes asymmetry in the collected observations. It is not a population-moment formula, a p-value or a guarantee of a particular distribution. Release implementation.

=SKEW(Number1 [; Number2; …])

Supply numeric observations as arguments or ranges. At least three observations and nonzero sample variance are required for the stated statistic. Numeric referenced cells contribute to the sample. As with other collected statistics, blanks and text in ranges should not be confused with recorded numeric zeroes.

These eight formulas were recalculated in the stated 4.1.16 build using XLSX import. Saved numeric results agree with independent high-precision calculations; values below are rounded to twelve decimal places. The comparison concerns stored values, not a claim that every display format prints the same digits.

For the range formulas, enter these values in the indicated columns and rows. Keep labels outside the calculation ranges.

The symmetric values 1,2,3 give zero skewness. For 1,1,4, =SKEW(1;1;4) gives approximately 1.732050807569. The isolated high value drives positive skewness. Adding 10 to every value gives 11,11,14 and leaves that statistic unchanged.

The B worksheet column has a large high-end observation; the E column reflects it through zero. Their skewness values have equal magnitude and opposite signs. This makes the sign convention visible without relying on an ambiguous verbal description of the “long side” of a distribution.

With n observations, sample mean x̄ and sample standard deviation s based on n−1, the implemented statistic is nΣ((x−x̄)/s)^3/((n−1)(n−2)). A raw third central moment or a version using population variance is a different estimator.

A positive scale factor leaves skewness unchanged, while a negative scale factor reverses its sign. Translation leaves it unchanged. These properties explain why 1,1,4 and 2,2,8 have the same result, and why reflected columns differ only in sign.

An exact zero third moment does not prove a distribution is symmetric or normal. Positive and negative contributions can cancel in asymmetric data. Examine the underlying observations, sample size and an appropriate plot before interpreting the summary.

A constant sample has zero spread, so standardization is undefined. The inspected evaluator signals an invalid argument in that case. An undersized sample also cannot supply the adjusted formula’s denominator. Preserve the observed error instead of converting it to an apparently meaningful zero.

Skewness is sensitive to outliers and can vary substantially between small samples. Large offsets with tiny differences can also worsen floating-point cancellation when subtracting the mean. The ordinary examples do not establish universal precision or a distributional test.

The following boundary formulas were recalculated in the same named method. H5 and H6 are deliberately empty cells where referenced. An error display remains an error even if the underlying ODS numeric attribute stores zero.

• stdev
• average
• median

• Function-specific release entry
• Numeric helper or argument collection

The retained input and saved ODS preserve the tested formula identities. A stale SUM cache changed from 999999 to 3, and an uncached SUM returned 7 after hard recalculation. Native evidence is limited to the eight reference formulas and listed boundary cases in this exact release and method. Composite explanatory examples are independent calculations unless listed among those cases; other releases, locales and extreme-parameter accuracy were not established.