How to Measure Skewness with SKEW

SKEW tells you whether your data has a long tail to the right or left — and whether the average can be trusted.

You will learn to

• Measure right skew in resolution times
• Recognise left skew in satisfaction scores
• Use a log transform to reduce skew

How it works

SKEW returns the sample skewness of a data set. Around 0 means roughly symmetric; positive means a long tail to the right (a few very large values, as with incomes, house prices or wait times); negative means a long tail to the left (a few very low values, as with satisfaction scores near a ceiling).

As a rule of thumb, skewness beyond about ±1 is strong — report the median rather than the mean. Taking the logarithm of right-skewed data (=SKEW(LN(B2:B13))) often makes it much more symmetric, which many statistical methods assume.

=SKEW(B2:B13)