Z-Score Calculator with Percentile and Probability
Z-score calculator: percentile, tail probabilities and p-value for any raw score or z-score, or the z-score behind any percentile.
Z-score
- Area to the left, P(Z < z)
- 0.933193
- Area to the right, P(Z > z)
- 0.066807
- Two-tailed, P(|Z| > |z|)
- 0.133614
- Between −|z| and |z|
- 0.866386
- Percentile
- 93.3193%
Formula: z = (85 − 70) ÷ 10 = 1.5
How to use the Z-Score Calculator
- Choose what you know: a raw score with its mean and standard deviation, a z-score, or a percentile.
- Enter the values. The standard deviation must be greater than 0, and a percentile must lie between 0 and 100.
- Read the z-score (or the left-tail area) in the result panel, with a one-line explanation under it.
- Use the rows below for the right-tail, two-tailed and central areas and the percentile, ready for a z-table check or a hypothesis test.
Formula
z = (x − μ) ÷ σ P(Z < z) = Φ(z) P(Z > z) = 1 − Φ(z)
Where:
- z
- The z-score: how many standard deviations x lies above (positive) or below (negative) the mean
- x
- The raw score, the individual value you are standardizing, in the data's own units
- μ
- The mean (average) of the distribution, in the same units as x
- σ
- The standard deviation of the distribution, in the same units as x; must be greater than 0
- Φ(z)
- The standard normal cumulative distribution function: the area under the bell curve to the left of z
The probabilities assume the data follow a normal (bell-shaped) distribution. The z-score itself is valid for any distribution, but its percentile is only exact for normal data. Two-tailed area = 2 × Φ(−|z|).
Worked example
IQ scores are designed to have a mean of 100 and a standard deviation of 15. What is the z-score of an IQ of 130, and what share of people score lower?
- Subtract the mean from the score: 130 − 100 = 30.
- Divide by the standard deviation: z = 30 ÷ 15 = 2.
- Look up Φ(2), the area to the left of z = 2: 0.977250.
- Area to the right: 1 − 0.977250 = 0.022750, so about 2.3% of people score higher.
- Two-tailed area: 2 × 0.022750 = 0.045500, so 95.45% of scores fall between 70 and 130.
An IQ of 130 has a z-score of 2. It sits at about the 97.72nd percentile: roughly 97.7% of people score lower and 2.3% score higher.
How to convert a z-score to a percentile
A z-score re-expresses a value as its distance from the mean in standard deviations, so 0 is exactly average, +1.5 is one and a half standard deviations above and −2 is two below.
When the data follow a normal distribution, that distance maps to a percentile. The area under the bell curve to the left of z, written Φ(z), is the share of values below it; multiply by 100 for the percentile. Two modes do this without a printed table:
- Raw score → z takes a value, the mean and the standard deviation. This is what people look for as a percentile calculator with mean and standard deviation.
- z → probability starts from a z you already have, for example from a hypothesis test, and works as a z score to percentile calculator.
Both show the left-tail area, the right-tail area, the two-tailed area, the area between −|z| and |z|, and the percentile.
Worked example: is 450 hours unusually early?
A manufacturer finds that the life of a rechargeable battery is normally distributed with a mean of 500 hours and a standard deviation of 40 hours. A customer's battery died after 450 hours.
- Subtract the mean: 450 − 500 = −50 hours.
- Divide by the standard deviation: z = −50 ÷ 40 = −1.25.
- Area to the left: Φ(−1.25) = 0.105650, so about 10.6% of batteries fail at or before 450 hours.
- Area to the right: 1 − 0.105650 = 0.894350, so about 89.4% last longer.
A battery at the 10.6th percentile is below average but not rare: roughly one in nine fail that soon.
How to find a p-value from a z-score
Pick the area that matches your question.
| Question | Area to use |
|---|---|
| What share of values is below x? | Left tail, P(Z < z) |
| What share is above x? | Right tail, P(Z > z) |
| How unusual is x in either direction? | Two-tailed, 2 × Φ(−|z|) |
| What share lies within this distance of the mean? | Between −|z| and |z| |
The two-tailed area is the p-value of a two-sided z-test; a one-sided test uses a single tail. A two-sided result is statistically significant at the 5% level when |z| is above 1.96 and at the 1% level when it is above about 2.576.
The central area gives the 68–95–99.7 rule: about 68.27% of a normal distribution lies within one standard deviation of the mean, 95.45% within two and 99.73% within three.
From a percentile back to a z-score
Percentile → z runs the lookup in reverse. Enter 90 and it returns z = 1.281552, the point with 90% of the distribution below it. The result is a z-score only, so turn it into a raw cutoff yourself with x = mean + z × standard deviation. For the batteries, the longest-lasting 10% begin at 500 + 1.281552 × 40 ≈ 551.26 hours.
The same mode gives critical values. A 95% two-sided confidence level leaves 2.5% in each tail, so enter 97.5 to get 1.959964, the familiar 1.96.
A short z-table of common values
A printed z-table lists Φ(z), the area to the left. These are the values the tool reproduces, to six decimals:
| z | Φ(z), area to the left |
|---|---|
| 0 | 0.500000 |
| 0.5 | 0.691462 |
| 1 | 0.841345 |
| 1.5 | 0.933193 |
| 1.96 | 0.975002 |
| 2 | 0.977250 |
| 2.5 | 0.993790 |
| 3 | 0.998650 |
For a negative z, use symmetry: Φ(−z) = 1 − Φ(z), so Φ(−1.5) = 0.066807. Tables in the "mean to z" style list the area from 0 to z, which is 0.5 smaller than these left-tail values.
Assumptions and limits
- Normal data. The z-score formula works for any data, but the percentiles are exact only for a normal distribution. For skewed data such as incomes or waiting times, use a percentile worked out from the data itself.
- Numbers only, no curve. There is no shaded bell-curve graph, and the only central area is the one between −|z| and |z|. For the area between two different values, subtract one left-tail area from the other.
- One value, not a sample mean. There is no sample size field. To standardize a sample mean, divide the standard deviation by √n yourself and enter that standard error as the standard deviation.
- Standard deviation, not variance. If the spread is given as a variance of 100, enter 10.
- Not a clinical score. Growth-chart, bone-density and exam z-scores are built on their own reference tables by age, sex or cohort, which this general tool does not contain.
Frequently asked questions
What is a z-score in statistics?
A z-score, or standard score, is a value expressed as a distance from the mean in standard deviations. Subtract the mean from your value and divide by the standard deviation: z = (x − μ) ÷ σ. A test score of 85 in a class with a mean of 70 and a standard deviation of 10 has z = (85 − 70) ÷ 10 = 1.5, one and a half standard deviations above average. A negative z-score means the value is below the mean.
How do I convert a z-score to a percentile?
Find the area under the standard normal curve to the left of z, written Φ(z), and multiply by 100. For z = 1.5, Φ(1.5) = 0.933193, so the value sits at about the 93.3rd percentile. Choose “z → probability” in the calculator to do this without a z-table.
What z-score is the 95th percentile?
z ≈ 1.644854. Other common ones: the 90th percentile is z ≈ 1.281552, the 97.5th is z ≈ 1.959964 (the familiar 1.96 used for 95% confidence intervals), and the 99th is z ≈ 2.326348. Choose “Percentile → z” to find any other.
How do I find a p-value from a z-score?
For a one-tailed test, the p-value is the area beyond z in the direction of your alternative hypothesis: P(Z > z) for an upper-tail test or P(Z < z) for a lower-tail test. For a two-tailed test, double the smaller tail: p = 2 × Φ(−|z|). A z of 1.96 gives a two-tailed p-value of about 0.05.
What z-score is an outlier or considered unusual?
There is no fixed rule, only conventions. A value more than 2 standard deviations from the mean (|z| > 2) is often called unusual: about 4.55% of a normal distribution lies that far out. A value beyond 3 standard deviations (|z| > 3) is commonly flagged as an outlier, since only about 0.27% of normal data falls there. Check such values before removing them; they may be real.
Does the percentile work for any data?
No. You can always compute the z-score, since it is just a distance in standard deviations, but the percentiles and probabilities shown here assume a normal distribution. For skewed data, such as incomes, a z of 2 may sit well below or above the 97.7th percentile. In that case, work out the percentile from the data itself.
Related calculators
- Standard Deviation CalculatorThis standard deviation calculator (sample vs population) turns a pasted list of numbers into s or σ, with the variance and mean alongside.
- Percentile CalculatorA percentile calculator from data set values: paste your numbers to find any percentile, the quartiles and IQR, or the percentile rank of one value.
- Confidence Interval CalculatorThis confidence interval calculator for mean values (t or z) and proportions turns summary statistics into lower and upper limits with the margin of error.
- Probability CalculatorThis probability calculator with multiple attempts finds the chance of at least one success in n tries, or combines two independent events.
- Sample Size CalculatorThis sample size calculator (finite population correction included) shows how many survey responses you need for a chosen confidence level and margin of error.
Last updated: