Z-Score Calculator
A z-score measures how many standard deviations a value lies from the mean. This calculator applies z = (x − μ) / σ and explains whether the value is above or below the mean and by how much.
A unit that travels between data sets
A raw measurement means nothing without a scale. Eighty-five is excellent on one test and mediocre on another, and there is no way to tell which from the number alone. A z-score fixes that by expressing the value as a distance from the mean, measured in standard deviations.
This calculator performs that conversion and states what the result means in words — how far above or below the mean the value sits.
Standardising strips the units out. Centimetres, marks and seconds all become the same dimensionless quantity, which is what allows a height and an exam result to be compared at all — something the raw values make impossible.
It is also the input most statistical tables want. Probability tables, control limits and outlier rules are all stated in standard deviations, so converting first is the step that makes them usable.
How to use this calculator
- Enter the value to standardise The observation of interest, in whatever units it was measured. It may lie above or below the mean.
- Enter the mean The centre of the distribution the value belongs to, in the same units. This is what the value is measured against.
- Enter the standard deviation Must be strictly positive. It sets the scale, and a value of zero is rejected because a distribution with no spread has no scale to measure against.
- Read the interpretation line It restates the result in words, naming the direction and the number of standard deviations, which is easier to check against intuition than the bare figure.
The formula, and where it comes from
z = (x − μ) / σ
The subtraction moves the centre of the distribution to zero, so what remains is a displacement rather than a position. That step alone makes values from different distributions comparable in direction, though not yet in size.
The division rescales that displacement into standard-deviation units. A distribution whose values naturally spread by fifty and one that spreads by half become directly comparable, because the divisor absorbs the difference in scale.
The result carries no units at all. The units of the numerator and denominator are identical and cancel, which is precisely what makes a z-score portable between quantities that have nothing physical in common.
The transformation is linear and reversible: multiplying a z-score by the standard deviation and adding the mean recovers the original value exactly. Nothing is lost in standardising, only re-expressed.
What each input means
- x Value — form field “Value x”
- The observation being standardised, in the units of the measurement.
- μ Mean — form field “Mean μ”
- The centre of the distribution the value is judged against, in the same units.
- σ Standard deviation — form field “Standard deviation σ”
- The spread, which must be strictly positive. It is the unit the answer is expressed in.
- z Standard score
- Unitless. Positive above the mean, negative below it, and zero exactly at it.
Worked examples
Every number below is produced by the same calculation engine the tool above runs. Nothing here is typed by hand, so the walkthrough cannot drift from what you get when you enter the same values yourself.
A value above the mean
An observation of 85 against a mean of 70 with a standard deviation of 10 — a value one and a half standard deviations high.
Inputs Value x = 85, Mean μ = 70, Standard deviation σ = 10
- Given x = 85, μ = 70, σ = 10
- Formula z = (x − μ) / σ
- Substitute z = (85 − 70) / 10 = 1.5
- Interpretation 1.5 standard deviations above the mean
Result z = 1.5
The displacement is 15 and the scale is 10, giving 1.5. That figure is more informative than the raw 85, because it says the value is unusual without needing to know anything else about the test.
Under a roughly bell-shaped distribution about 7% of values sit this high or higher. Notable, but not rare — a z-score of 1.5 is well short of the threshold at which most conventions call something an outlier.
A value below the mean
An observation of 58 against the same mean and spread. The displacement is now negative.
Inputs Value x = 58, Mean μ = 70, Standard deviation σ = 10
- Given x = 58, μ = 70, σ = 10
- Formula z = (x − μ) / σ
- Substitute z = (58 − 70) / 10 = -1.2
- Interpretation 1.2 standard deviations below the mean
Result z = -1.2
The z-score is −1.2, and the sign is doing real work: it says below the mean, which the magnitude alone would not. Dropping it is the most common error with standard scores.
The interpretation line spells the direction out rather than leaving it to the sign. That redundancy is deliberate, since a misread minus changes the meaning entirely.
Reading the result
What the magnitude conveys
Roughly 68% of a bell-shaped distribution lies within one standard deviation of the mean, 95% within two and 99.7% within three. Those proportions turn a z-score into an immediate sense of how unusual a value is.
The sign is half the answer
A z-score of −2 and one of +2 are equally far from the mean and mean opposite things. Quoting the magnitude without the sign discards the direction, which is usually the part that matters.
It says nothing about shape
The calculation itself assumes nothing about the distribution — it is arithmetic. The familiar percentages that make a z-score interpretable do assume approximate normality, and for a strongly skewed distribution they do not hold.
When you would use this
Comparing results across different scales
Two marks from tests with different means and spreads become comparable once both are standardised. That is the whole reason standardised scores exist in assessment.
Flagging outliers
A common convention treats values beyond two or three standard deviations as unusual. Converting to a z-score is what makes that rule applicable to any quantity.
Assumptions and limitations
What this calculator assumes
- The mean and standard deviation describe the distribution the value belongs to.
- The standard deviation is strictly positive; zero spread leaves no scale.
- All three inputs share the same units, which then cancel.
- No distributional shape is assumed by the calculation itself.
Where it stops being the right tool
- One value at a time; standardising a whole data set means repeating the conversion.
- No probability is reported — the normal distribution calculator converts a z-score into a tail probability.
- The mean and standard deviation must be known rather than estimated from data here.
- The rules of thumb about percentages assume approximate normality, which is not checked.
Common mistakes
Entering the variance instead of the standard deviation
Why it happens. Both describe spread and both appear in statistical output, but the variance is the square. Using it makes every z-score too small and understates how unusual the value is.
How to avoid it. Take the square root first if what you have is a variance. A z-score much closer to zero than expected is the symptom.
Dropping the sign
Why it happens. Distance feels like it should be positive, and the magnitude is what the outlier rules are stated in.
How to avoid it. Keep it. Above and below the mean are different results, and the interpretation line names the direction explicitly.
Reading a z-score as a percentage
Why it happens. It is a small number that measures unusualness, which invites reading 1.5 as fifteen per cent or similar.
How to avoid it. Treat it as a count of standard deviations. Converting it into a probability needs the normal distribution calculator.
Key terms
Frequently asked questions
What does a z-score tell you?
How many standard deviations a value lies from the mean, with the sign giving the direction. Because the units cancel, scores from completely different quantities can be compared directly.
What is a negative z-score?
A value below the mean. The magnitude is the distance and the sign is the direction, so −2 and +2 are equally far out and mean opposite things.
What counts as an unusual z-score?
Beyond about 2 is often treated as unusual, since roughly 95% of a bell-shaped distribution falls within two standard deviations, and beyond 3 as rare. Those thresholds are conventions resting on approximate normality rather than absolute rules.
Can I get back to the original value?
Yes — multiply the z-score by the standard deviation and add the mean. The transformation is linear and loses nothing, so it reverses exactly.