Contact
12 programs
Explore the site

Zoho Sheet SKEW Function: Measure Asymmetry with Worked Examples

Statistical Intermediate Zoho Sheet

Calculate skewness in Zoho Sheet with an eight-week sales example, outlier comparison, practical dashboard formula, troubleshooting, and statistical interpretation.

Verification status: Official documentation referenced; example outcomes independently reasoned or arithmetically checked where stated. Not executed in Zoho Sheet. How formula examples are checked

Quick answer: Enter =SKEW(B2:B9) to measure the skewness of numerical observations in cells B2 through B9. For the eight values 10, 12, 14, 14, 16, 18, 20 and 50, the independently calculated sample-skewness expectation is about 2.489766088. This large positive number indicates a pronounced right-hand tail, driven by an unusually high value. The example is a mathematical expectation based on an adjusted sample-skewness convention; it has not been executed in Zoho Sheet.

SKEW describes asymmetry, not average size, profitability, or probability of a future event. Positive skewness generally means a longer tail toward high values; negative skewness means the opposite; a value near zero suggests approximate symmetry, but does not prove the distribution is normal. A few extremes can dominate a small sample. Always inspect the actual observations alongside the statistic.

Official syntax and arguments

Zoho’s current official statistical function reference gives this syntax:

SKEW(number; [number1]; ...)
Argument Required? Meaning Example
number Yes A number or reference/range containing observations to assess. B2:B9
[number1] and following No Additional observations or ranges, separated using the syntax shown in Zoho’s reference. D2:D5 or 25

The reference uses semicolons between function arguments and a colon for a contiguous cell range. Thus =SKEW(B2:B9) takes a range as one argument, whereas =SKEW(-13;5;27;-1.5) specifies four values. Zoho’s official example prints an answer of 0.859383956 for those four numbers; an independent adjusted-sample calculation gives 0.8593839555, agreeing at the shown precision. That mathematical agreement does not substitute for current native execution or establish every undocumented coercion rule.

Zoho explicitly says that text, logical values and empty cells are ignored for SKEW. A numeric zero is an observation and should not be confused with a blank. Zoho’s documented error list identifies #DIV/0! when fewer than three values are supplied. The help page does not establish what first product release introduced SKEW or guarantee identical behaviour in every old workbook, locale, mobile client, or desktop beta. These guides use the currently published English-language cloud documentation, checked October 12, 2026. Zoho has separately announced Windows/macOS desktop beta apps with many, but not necessarily all, web features; this is not a test of SKEW on either desktop platform.

Example 1: assess eight weeks of order counts

Enter these observations as ordinary numbers, not text with a leading apostrophe:

Row A: Week B: Orders
1 Week Orders
2 Week 1 10
3 Week 2 12
4 Week 3 14
5 Week 4 14
6 Week 5 16
7 Week 6 18
8 Week 7 20
9 Week 8 50
  1. Select a separate empty cell such as D2.
  2. Enter =SKEW(B2:B9).
  3. Confirm the formula. The independently calculated expected result under the adjusted sample formula is 2.489766088 (rounded to nine decimals); check the application result and its formula settings before reporting it as Zoho-tested.
  4. In D3, enter =AVERAGE(B2:B9); the independently calculated ordinary arithmetic average of the eight numeric inputs is 19.25. This companion metric measures a different property.
  5. Compare the maximum value, 50, with the cluster of 10–20. The extreme week is why a positively skewed distribution is plausible.

This guide intentionally contains no simulated screenshots.

Why the expected result is about 2.49

A commonly used adjusted sample skewness estimator uses the sample standard deviation s and the number of observations n:

adjusted sample skewness = n / ((n-1)(n-2)) * SUM(((x - mean) / s)^3)
s = SQRT(SUM((x - mean)^2) / (n - 1))

For the eight observations: n = 8, mean = 19.25, SUM((x-mean)^2) = 1151.5, s ≈ 12.825755338, and SUM((x-mean)^3) = 27578.25. Substitution gives 2.489766088. This is independent statistical arithmetic, not a claim that Zoho publishes the estimator formula, which its help page does not specify explicitly. The independent estimator agrees with Zoho’s separately published four-number SKEW example, but that still cannot establish all implementation details.

Example 2: quantify outlier sensitivity before making decisions

To find out how much the unusually high final week influences skewness, compare an explicitly labelled seven-week subset with the full eight weeks:

Formula Observations Expected adjusted sample skewness
=SKEW(B2:B9) 10, 12, 14, 14, 16, 18, 20, 50 2.489766088
=SKEW(B2:B8) 10, 12, 14, 14, 16, 18, 20 0.168964170
=SKEW(1;2;3;4;5) A symmetric five-number set 0

These are independently calculated expectations, not native test results. The contrast does not justify deleting real orders. It tells you to investigate whether Week 8 was a legitimate promotion, a one-time large order, a duplicated import, or an operational change. Report both the original series and a justified sensitivity analysis rather than silently trimming inconvenient data.

For a negative-skewness sanity check, change each observation’s sign while retaining its magnitude: =SKEW(-10;-12;-14;-14;-16;-18;-20;-50) has an independently expected result of −2.489766088. Reflection around zero reverses the sign of skewness, while merely reordering the numbers should not change it.

Advanced application: a guardrail for a reporting dashboard

Skewness is undefined when too few numerical observations exist. A useful dashboard guard is:

=IF(COUNT(B2:B9)<3;"Need at least 3 numbers";SKEW(B2:B9))

COUNT supplies the numeric sample size. IF displays a readable message rather than requesting a skewness from one or two numbers. Because a constant-valued dataset has zero standard deviation, a stronger proposed guard is:

=IF(COUNT(B2:B9)<3;"Need at least 3 numbers";IF(STDEV(B2:B9)=0;"No variation";SKEW(B2:B9)))

These combined formulas are intended as practical extensions of individually documented functions; they have not been executed together in Zoho Sheet. Test their separators, evaluation and error paths in the target worksheet before relying on native behavior. COUNT and STDEV offer additional sanity checks: a zero standard deviation makes a standardized third moment mathematically undefined even if the nominal sample size is adequate.

For a real sales dashboard, include the sample size, mean, median, standard deviation and a distribution chart beside SKEW. A large positive value can be perfectly legitimate for infrequent wholesale orders, while the same number might expose an incorrect duplicate in a normally stable daily-sales series. SKEW is diagnostic information, not a pass/fail grade or a prediction.

Common errors and troubleshooting

Symptom Explanation and next step
#DIV/0! with one or two numeric cells Zoho officially lists fewer than three values as an error. Use COUNT(range) to verify the numeric sample size.
Undefined calculation for identical numbers Zero dispersion makes the standardised third moment mathematically undefined. Zoho’s exact native error display in this case was not tested; inspect STDEV(range) rather than inventing an error result.
Result appears unexpectedly small or has a different sign Confirm that a genuine high or low outlier was not omitted; verify the exact included range and the signs of imported values.
#NAME! Correct SKEW spelling, quotation marks in any surrounding formula, and any custom named range. Zoho lists invalid function/defined names as causes.
#REF! A row/column/range reference may have been deleted or changed; reselect the intended numeric cells.
#VALUE! Zoho lists invalid argument types generally. Inspect literal parameters, conversion errors and nested formulas; don’t assume how every direct text literal is coerced.
Blank, text or Boolean cells in the referenced range The Zoho SKEW reference says these are ignored. Check that numeric-looking text has actually been converted to numbers.
Spreadsheet from another locale rejects semicolons Verify the workbook’s formula language/argument separator rather than replacing delimiters indiscriminately.

Zoho’s generic error table also mentions #N/A!, usually a lookup-related error; it does not prove that SKEW normally generates a lookup error. An error in a referenced cell or nested formula needs independent tracing.

Standard deviation measures overall spread, regardless of direction. SKEW measures asymmetry, retaining information about which side has the longer tail. KURT measures a different aspect of distribution shape and is sensitive to extreme observations; it is not interchangeable with SKEW. Median and AVERAGE measure typical location. Always compare statistics on the same explicitly defined set of observations.

Zoho’s current catalogue identifies SKEW, but this guide does not establish support for a separate SKEW.P function. Microsoft Excel documents SKEW as sample skewness and SKEW.P separately as population skewness; do not paste an Excel SKEW.P formula into Zoho and claim support without checking Zoho’s live function picker. Google Sheets documents SKEW with comma-separated examples. Excel and Google Sheets support similarly named functions, but their argument handling, available variants and minimum-sample statements must be checked separately. In particular, Google’s public SKEW help text describes a two-value minimum-error condition, while Zoho explicitly names fewer than three values. A difference in documentation is not proof of an engine-level difference without native tests.

Related topics: KURT, STDEV, VAR, DEVSQ, AVERAGE, MEDIAN and COUNT.

Sources and verification (checked 2026-10-12)

Verification status: Official documentation checked October 12, 2026. Numerical examples and estimator arithmetic checked independently using reproducible non-Zoho calculations. Not executed in Zoho Sheet. No genuine Zoho Sheet screenshots are included, and no mock application images are used. Native edge-case validation remains pending.