What is a z-score?
A z-score, also called a standard score, tells you how many standard deviations a value is from the mean. A z-score of 0 is exactly average, +1 is one standard deviation above, and −2 is two standard deviations below. Because it removes the original units, a z-score lets you compare values from different scales, such as a math score and a reading score, or a child's height against peers of a different age.
Here x is the raw score, μ is the population mean and σ is the population standard deviation. When you only have a sample, use the sample mean and sample standard deviation as estimates.
Worked example
Exam scores have a mean of 80 and a standard deviation of 6.4. A student who scored 71.3 has z = (71.3 − 80) / 6.4 ≈ −1.359. About 8.7% of students scored lower (P(x<71.3) ≈ 0.087014), about 91.3% scored higher, and 41.3% fall between the student's score and the mean.
From z-score to probability
If the data is normally distributed, each z-score corresponds to an area under the bell curve. The calculator reports five related areas:
- P(x < Z): the left tail, the share of values below Z. This is the cumulative probability or percentile.
- P(x > Z): the right tail, equal to 1 − P(x < Z).
- P(0 < x < Z): the area between the mean and Z.
- P(−Z < x < Z): the central area within ±Z of the mean.
- P(x < −Z or x > Z): both tails combined, which is the two-tailed p-value.
The converter also works in reverse: enter any one of these probabilities and it returns the matching z-score. For example, a central area of 0.95 corresponds to z ≈ 1.96, the familiar cutoff for 95% confidence.
Z-table
Before software, these areas were read from printed z-tables. A few landmark values for the left tail are below:
| z | P(x < z) | P(0 < x < z) |
|---|---|---|
| 0 | 0.5 | 0 |
| 0.5 | 0.69146 | 0.19146 |
| 1 | 0.84134 | 0.34134 |
| 1.5 | 0.93319 | 0.43319 |
| 2 | 0.97725 | 0.47725 |
| 2.5 | 0.99379 | 0.49379 |
| 3 | 0.99865 | 0.49865 |
For negative z, use symmetry: P(x < −z) = 1 − P(x < z).
The empirical rule
For normal data, about 68.27% of values fall within one standard deviation of the mean (|z| < 1), about 95.45% within two, and about 99.73% within three. A value with |z| above 3 is rare enough that it is often treated as an outlier worth checking.
Common uses
- Standardized testing, where scores are reported as percentiles derived from z-scores.
- Growth charts that place a child's height or weight relative to peers.
- Quality control, where process measurements beyond ±3σ trigger investigation.
- Hypothesis testing, where a test statistic's z-score is turned into a p-value.
- Finance, where the Altman Z-score and similar ratios flag unusual results.
Limitations
The probabilities assume the data is normal. Skewed data such as incomes or waiting times can make the tail areas misleading even though the z-score itself is still a valid measure of distance from the mean.
Comparing scores on different scales
Suppose a student scores 82 on a chemistry exam where the class mean is 74 with a standard deviation of 5, and 88 on a history exam where the mean is 85 with a standard deviation of 6. The chemistry z-score is (82 − 74) / 5 = 1.6 and the history z-score is (88 − 85) / 6 = 0.5. Even though the raw history score is higher, the chemistry result is far more exceptional: it beats about 94.5% of the class, while the history score beats about 69%.
Using the three tools
Use the first tool when you have a raw value, a mean and a standard deviation. Use the converter when you already have a z-score, or when you have a probability and need the z-score, such as a cutoff for the top 10%. Use the last tool for the probability that a value lands between two z-scores.