Skewness Calculator

Skewness is a statistical measure that quantifies the asymmetry of a data distribution around its mean. A skewness of zero indicates a perfectly symmetrical distribution. Positive skewness means the right tail is longer (data leans left), while negative skewness means the left tail is longer (data leans right).

The moment coefficient of skewness (g1) is calculated using the third standardized moment: the average of the cubed deviations from the mean, divided by the cube of the standard deviation. Specifically, g1 = [( 1/n ) × Σ(xi − x̄)³] / [( 1/n ) × Σ(xi − x̄)²]^(3/2).

Pearson's SK2 is an alternative measure of skewness defined as SK2 = 3(mean − median) / standard deviation. It provides a simpler, robust estimate of skewness especially useful when the distribution is unimodal. Values typically range between −3 and +3.

A skewness between −0.5 and +0.5 is generally considered approximately symmetric. Values between ±0.5 and ±1 indicate moderate skewness, while values beyond ±1 indicate high skewness. Highly skewed data may require transformation (e.g., log transformation) before applying certain statistical tests.

Positive (right) skewness means a long tail extends to the right — most values cluster on the left with a few large outliers pulling the mean above the median. Negative (left) skewness means a long tail extends to the left, with a few very small outliers pulling the mean below the median.

Skewness can be calculated with as few as 3 data points, but results are unreliable with small samples. Most statisticians recommend at least 20–30 data points for meaningful skewness estimates. Very small samples tend to produce unstable, highly variable skewness values.

You can enter your numbers separated by commas, spaces, or a combination of both. For example: '2, 4, 6, 8, 10' or '2 4 6 8 10'. Non-numeric entries are automatically ignored. You can enter between 3 and 500 values.

Yes. Many parametric tests (like t-tests and ANOVA) assume normally distributed data with near-zero skewness. If your data is significantly skewed, you may need to apply data transformations or choose non-parametric alternatives such as the Mann-Whitney U test or Kruskal-Wallis test.