How it works
z = (x − μ) ÷ σ
A z-score of 0 is exactly average, 1 is one standard deviation above, −2 is two below. For data that is roughly normally distributed, about 68% of values have z between −1 and 1, about 95% between −2 and 2 and about 99.7% between −3 and 3.
Worked example
A test has mean 70 and standard deviation 10. A student scores 85. z = (85 − 70) ÷ 10 = 1.5, so the score is 1.5 standard deviations above the mean. A score of 62 on a test with mean 70 and SD 5 gives z = −8/5 = −1.6.
From a z-score to a percentile
For data that follows a normal distribution, the z-score fixes the share of values below it:
| z | Share below |
|---|---|
| −2 | 2.28% |
| −1 | 15.87% |
| 0 | 50% |
| 1 | 84.13% |
| 1.5 | 93.32% |
| 1.645 | 95.00% |
| 1.96 | 97.50% |
| 2 | 97.72% |
| 3 | 99.87% |
So a z-score of 1.5 is at about the 93rd percentile. These shares hold only for normally distributed data.
Comparing scores from different tests
85 on a test with mean 70 and standard deviation 10 gives z = 1.5. 80 on a test with mean 65 and standard deviation 12 gives z = (80 − 65) ÷ 12 = 1.25. The first score is further above its average, even though the raw scores are close.
Going back from z to a score
Rearrange the formula: x = μ + zσ. With mean 70 and standard deviation 10, a z-score of 2 is a score of 70 + 2 × 10 = 90.
Tips
- The standard deviation must be positive.
- Use the sample standard deviation s in place of σ when you only have a sample.
- z-scores let you compare results from tests with different scales.
Questions
What does a negative z-score mean?
The value is below the mean.
What is a good z-score?
It depends on context. Above 2 is unusually high for roughly normal data; below −2 is unusually low.
How do I get a percentile from a z-score?
Use a standard normal table or the normal distribution function: z = 1.5 is about the 93rd percentile.
How do I find the raw score from a z-score?
Multiply the z-score by the standard deviation and add the mean: x = μ + zσ.
What z-score marks the top 5%?
About 1.645 for normally distributed data: 95% of values lie below it.
Can a z-score be bigger than 3?
Yes. It just means the value is more than three standard deviations above the mean, which is rare for normal data (about 0.13% of values).
Sources
- NIST/SEMATECH e-Handbook of Statistical Methods, §1.3.6.7.1 Cumulative Distribution Function of the Standard Normal Distribution. Standardize by subtracting the mean and dividing by the standard deviation; table of areas (rows 1.0, 1.5, 1.9, 2.0, 3.0); critical values 0.950 → 1.645, 0.975 → 1.960. Checked 22 November 2026.
- NIST/SEMATECH e-Handbook of Statistical Methods, §1.3.6.6.1 Normal Distribution. μ is the location and σ the scale; μ = 0, σ = 1 is the standard normal; parameters are estimated with the sample mean and sample standard deviation. Checked 22 November 2026.
- NIST/SEMATECH e-Handbook of Statistical Methods, §1.3.6.4 Location and Scale Parameters. Standard form has location 0 and scale 1; non-positive scale parameters are not allowed (σ must be positive). Checked 22 November 2026.
- Casio fx-115ES PLUS / fx-991ES PLUS C (2nd edition) user’s guide, online manual. Performing Normal Distribution Calculations: ▶t gives the normalized variate t = (X − x̄) ÷ σx; P(, Q(, R( give standard normal probabilities. Checked 22 November 2026.