Z-Score Calculator
Standardize a measurement by its distance from a mean in standard-deviation units.
How the result changes
Only observed value changes; all other inputs stay at your current values. The line illustrates the formula across the shown range. This is not a forecast of real-world outcomes.
About this calculation
Z-Score Calculator: Standardize a measurement by its distance from a mean in standard-deviation units. Enter observed value, mean, standard deviation; the reported quantities are z-score (standard deviations), difference from mean (units). The equations and examples below show how those inputs lead to the result and where the model's assumptions matter.
Formula
z = (value - mean) / standard deviation
Worked example
With Observed value: 85 units, Mean: 70 units, Standard deviation: 10 units, z-score is 1.5 standard deviations.
Method and inputs
Subtract the relevant mean from the observed value, then divide by the positive standard deviation. The resulting z-score is dimensionless because numerator and denominator use the same measurement unit.
Worked example
For an observation of 85, mean 70 and standard deviation 10, the difference is 15 and z = 1.5. The observation lies one and a half standard deviations above the selected mean.
A score of 85 against mean 70 and standard deviation 10 gives z = 1.5. If the reference group’s standard deviation were 20 instead, the same raw difference would give z = 0.75. Standardization depends on the chosen comparison group.
How to read the result
Positive scores are above the mean; negative scores are below it. A z-score by itself does not give a percentile unless a distributional model, often normality, is also justified.
The z-score can help compare measures with different units only when each has a meaningful mean and standard deviation. It does not erase poor data quality. A small standard deviation also magnifies tiny measurement errors, so precision of the original observation matters.
Assumptions and checks
The mean and standard deviation must describe the same reference population or sample scale as the observation. A zero standard deviation makes standardization undefined, so the input must be positive.
Do not silently substitute a sample standard deviation for a known population value without noting the change. For strongly skewed distributions, a given z-score need not have the familiar normal-tail interpretation.
If the value equals the mean exactly, the numerator is zero and the z-score is zero for any positive standard deviation. If the value is one standard deviation below the mean, the z-score is minus one. These anchor cases make hand validation straightforward.
Common questions
What does the z-score calculator calculate?
It returns the signed standardized distance from the mean and the raw difference in original units.
When should I recalculate or check another method?
Recalculate when the reference group changes. For outlier decisions, inspect the distribution and sample size rather than applying a fixed z threshold without context.
Sources and references
Inputs
- Observed value (units)
- Mean (units)
- Standard deviation (units)