Z-Score Explained: What It Means and How to Calculate It
Z-score explained: the (x - mean) / standard deviation formula, how to read sign and magnitude, a worked test-score example, and a z-score to percentile table.
The formula
z = (x - mu) / sigma
Where x is your value, mu is the population mean, and sigma is the standard deviation. A z-score answers one question: how many standard deviations from the mean is this value?
That is what makes it useful. Raw numbers from different scales — a SAT score, a cholesterol reading, a stock return — become directly comparable once expressed in standard deviations.
For a sample rather than a full population, use the sample mean and sample standard deviation: z = (x - x-bar) / s.
Reading the result
- Sign — positive means above the mean, negative means below. Zero means exactly average.
- Magnitude — how unusual. Under 1 is unremarkable, 2 is uncommon, 3 or more is rare.
In a normal distribution, roughly 68% of values fall within z = +/-1, 95% within +/-2, and 99.7% within +/-3. This is the empirical rule.
Worked example: two test scores
Ana scored 82 on a biology exam where the class mean was 74 with a standard deviation of 6. Ben scored 88 on chemistry, mean 81, standard deviation 12. Who did better relative to their class?
- Ana: z = (82 - 74) / 6 = 8 / 6 = +1.33
- Ben: z = (88 - 81) / 12 = 7 / 12 = +0.58
Ben's raw score is higher, but Ana outperformed her class by more than twice as many standard deviations. Ana sits near the 91st percentile; Ben near the 72nd.
Z-score to percentile
| z | Percentile | Interpretation |
|---|---|---|
| -3.0 | 0.1% | Extremely low |
| -2.0 | 2.3% | Well below average |
| -1.0 | 15.9% | Below average |
| -0.5 | 30.9% | Slightly below |
| 0 | 50% | Exactly average |
| +0.5 | 69.1% | Slightly above |
| +1.0 | 84.1% | Above average |
| +1.65 | 95% | Top 5% |
| +1.96 | 97.5% | Two-tailed 95% cutoff |
| +2.0 | 97.7% | Well above average |
| +3.0 | 99.9% | Extremely high |
These percentiles only hold when the underlying data is approximately normal. For a heavily skewed distribution, a z-score still measures distance from the mean but no longer maps cleanly onto a percentile.
Where z-scores are used
- Grading curves — converting raw marks to a standardized distribution.
- Medical labs — bone density is reported as a Z-score (versus age-matched peers) and a T-score (versus young adults).
- Hypothesis testing — the 1.96 cutoff for a two-tailed test at 95% confidence is a z-score.
- Finance — the Altman Z-score for bankruptcy risk; standardizing returns across assets.
- Data preprocessing — standardizing features before regression or clustering.
FAQ
What is a good z-score? It depends on direction. For test performance, above +1 is strong. For a risk or error metric, near zero or negative is better. Interpretation always depends on what x measures.
Can a z-score be greater than 3? Yes, though in normal data it happens for about 1 in 370 observations. Very large z-scores often flag outliers or data-entry errors.
What is the difference between a z-score and a t-score? Use z when the population standard deviation is known or the sample is large (n above ~30). Use t for small samples with an estimated standard deviation; the t-distribution has fatter tails.
Compute one directly with the Z-Score Calculator and get sigma first with the Standard Deviation Calculator. For the intuition behind sigma itself, read standard deviation explained.
---
*Convert any value to a z-score and percentile with the Z-Score Calculator.*