Z-Score Calculator

The z-score — the "standard score" — answers "how far is this one reading from the pack, measured in spread?" It converts any value into units of standard deviation from the mean, and because it is a pure ratio it makes very different datasets directly comparable. A single value then carries a z-score, a percentile, and a plain-language interpretation — all three printed by this calculator.

The formula

Z = (x − μ) ÷ σ

• x — the data point being placed
• μ — the mean (average) of the dataset
• σ — the standard deviation, the dataset's typical spread

Worked example

A class with mean 70 and standard deviation 10; where does a score of 85 sit?

Z = (85 − 70) ÷ 10 = 1.5

85 is 1.5 standard deviations above the mean. The sign of the result tells the direction: positive is above the mean, negative below, and a z near zero is right at the middle of the pack.

The percentile reading

Under the normal ("bell curve") distribution, a z of 1.5 corresponds to approximately the 93.3rd percentile — about 93% of scores sit at or below 85 and only about 7% exceed it. The percentile output turns an abstract "standard deviations" count into an immediately understood rank, which is why the same page shows both.

Rules of thumb

Why "standard" matters

Z-scores standardize everything: a student compared across different exam papers, a measurement tracked across different instruments, a financial return compared across different asset classes. Once every value lives in "units of its own dataset's spread", comparing them is legitimate — this is the same trick behind test-score normalization and many scaling schemes in machine learning and statistics.

Where z-scores are used

• Standardized tests — putting scores from different test versions or subject papers on a common scale.
• Quality control — flagging parts or measurements that drift beyond, say, ±3 standard deviations of a spec.
• Finance — detecting unusual price moves against a rolling mean and volatility estimate.
• Education and psychology — reporting how a result compares with a norm group.

The normal distribution assumption

The exact percentile conversion assumes the data is reasonably bell-shaped. With skewed or heavy-tailed data, the percentiles drift from the z-based estimate. The z-score itself — the distance in standard deviations — stays valid regardless; it is only the precise percentile claim that leans on normality. The 93.3rd percentile for z = 1.5, for example, is exactly right only for normal data.

Common mistakes

• Swapping mean and value — the sign flips and a genuinely strong score hides as a weak one.
• Dividing the difference by the wrong spread measure — use the standard deviation (σ), not the variance or the range.
• Reading a negative z as "bad" — below-average is not automatically undesirable; it depends entirely on what is being measured.
• Assuming percentiles are exact for non-normal data, then quoting them to a decimal point.

Getting the inputs

If the mean and standard deviation of your dataset are not at hand, compute the mean first — the average calculator reports it directly from a comma-separated list — then feed the value, mean and σ here.