Empirical Rule Calculator

The empirical rule, also called the 68-95-99.7 rule or the three-sigma rule, states that for a normally distributed dataset, approximately 68% of data falls within 1 standard deviation of the mean, 95% within 2 standard deviations, and 99.7% within 3 standard deviations. It provides a quick way to understand the spread of data in a bell-shaped distribution.

To apply the empirical rule, you need two values: the mean (μ) and the standard deviation (σ). The three intervals are: 68% interval = μ ± σ, 95% interval = μ ± 2σ, and 99.7% interval = μ ± 3σ. Simply subtract and add the respective multiple of σ from the mean to get the lower and upper bounds for each range.

With a mean of 100 and standard deviation of 15 (like IQ scores), 68% of values fall between 85 and 115, 95% fall between 70 and 130, and 99.7% fall between 55 and 145. Only about 0.3% of IQ scores would fall outside that last range.

The empirical rule is widely used in quality control (to set process limits), finance (to assess investment risk and normal return ranges), education (to interpret standardized test scores), healthcare (to evaluate normal ranges for clinical measurements), and any field where data follows a normal distribution.

No — the empirical rule only applies to data that is approximately normally distributed (bell-shaped). If your data is skewed, bimodal, or otherwise non-normal, the 68-95-99.7 percentages will not hold. In those cases, Chebyshev's theorem provides a more general (though less precise) alternative.

A value beyond three standard deviations from the mean is considered an outlier, since the empirical rule says only 0.3% of data in a normal distribution lies that far out. In practice, this could signal a data entry error, a rare but real event, or evidence that your data may not be normally distributed.

The empirical rule gives specific percentages (68%, 95%, 99.7%) that apply only to normally distributed data. Chebyshev's theorem applies to any distribution and guarantees that at least (1 − 1/k²) of data falls within k standard deviations — for example, at least 75% within 2σ. The empirical rule is more precise but more restrictive in when it can be applied.

For the empirical rule itself, the formula is the same regardless. However, if you are estimating the standard deviation from a sample, use the sample standard deviation (dividing by n−1) rather than the population standard deviation (dividing by n) to get a less biased estimate, especially for small samples.