Skip to content

Z-Score Calculator

Last updated

Your values

Find a z-score and its percentile, or the value for a percentile.

Values

Results update as you type

—

Enter a value, the mean and the standard deviation, and this z-score calculator gives the z-score, its percentile and the one- and two-tailed p-values. Switch to reverse mode to turn a percentile back into a z-score and raw value.

How to calculate a z-score

  1. Subtract the mean from your value.
  2. Divide by the standard deviation.
  3. Read the sign: positive means above the mean, negative below.

Example: you scored 85 on a test where the mean is 70 and the standard deviation is 10. z = (85 − 70) ÷ 10 = 1.5. Your score is 1.5 standard deviations above average. In a normal distribution, Φ(1.5) = 0.9332, so you scored higher than about 93.3% of test-takers.

Z-table: common z-scores

Table of z-scores and percentiles: z of minus 2 is the 2.28th percentile, 0 is the 50th, 1 is the 84.13th, 1.645 the 95th, 1.96 the 97.5th and 2.576 the 99.5th percentile
Values for the standard normal distribution.

A z-table lists the area under the standard normal curve to the left of each z-score – the percentile. The calculator does the lookup for you, with more decimal places than a printed table.

Z-scores and p-values

In hypothesis testing, the p-value is the probability of a result at least as extreme as yours if the null hypothesis is true. For z = 1.5: the right-tail p-value is 1 − 0.9332 = 0.0668; the two-tailed p-value (extreme in either direction) is 0.1336. Use a one-tailed test only when you predicted the direction before seeing the data. Critical values: |z| > 1.96 is significant at the 5% level (two-tailed), and z > 1.645 at the 5% level (one-tailed).

Z-score for a sample mean

When you test an average rather than one value, divide by the standard error instead of the standard deviation: z = (x̄ − μ) ÷ (σ ÷ √n). If a population has mean 100 and SD 15, and a sample of 36 people averages 105, z = 5 ÷ (15 ÷ 6) = 2.0. The right-tail p-value is 0.0228 – a sample mean that high would be rare by chance. If σ is unknown and the sample is small, a t-test is usually more appropriate.

From percentile back to a score

Reverse the formula: x = μ + z × σ. To find the score needed for the 90th percentile with mean 70 and SD 10, look up the z for 0.90 (1.282): x = 70 + 1.282 × 10 = 82.8.

Comparing scores from different tests

Z-scores put different scales on the same footing. A score of 85 on a test with mean 70 and SD 10 (z = 1.5) is relatively better than 650 on a test with mean 600 and SD 50 (z = 1.0), even though 650 is the bigger number. This is why standardised tests and growth charts report z-scores or percentiles.

When z-scores can mislead

  • Percentiles from the normal table assume the data is roughly normal. For skewed data, the true percentile can be quite different.
  • With small samples and an estimated standard deviation, use the t-distribution.
  • Outliers inflate the standard deviation and shrink everyone else's z-scores.
  • A z-score of 0 means 'average', not 'zero'.

Z-scores and confidence intervals

Critical z-values build confidence intervals for a mean when σ is known: mean ± z × σ ÷ √n. With a sample mean of 105, σ = 15 and n = 36, the standard error is 2.5, so the 95% interval is 105 ± 1.96 × 2.5 = 100.1 to 109.9. For a 90% interval use z = 1.645; for 99%, z = 2.576.

Z-scores in Excel and Google Sheets

TaskFormula
z-score=STANDARDIZE(x, mean, sd)
Percentile from z=NORM.S.DIST(z, TRUE)
z from percentile=NORM.S.INV(p)
Raw value from percentile=NORM.INV(p, mean, sd)

Z-scores for spotting outliers

A common rule of thumb flags values with |z| above 3 as possible outliers, because in a normal distribution only about 0.3% of values fall that far from the mean. Check flagged values for data-entry errors before removing anything, and be careful with small samples, where a single extreme value strongly affects the mean and standard deviation.

How to interpret a z-score

z-scorePositionShare of values above (normal data)
0Exactly average50%
±0.5Slightly above / below average30.9% / 69.1%
±1Clearly above / below average15.9% / 84.1%
±2Unusual2.3% / 97.7%
±3Very rare0.13% / 99.87%

Z-score vs t-score

A z-score uses a known population standard deviation. When you only have a sample's standard deviation, and especially when the sample is small (under about 30), the t-distribution is used instead. It has heavier tails, so the critical values are larger – about 2.06 instead of 1.96 for a 95% two-sided test with 25 degrees of freedom. As samples grow, t-values approach z-values.

Sources

  1. NIST/SEMATECH e-Handbook of Statistical Methods – Measures of scale — itl.nist.gov/div898/handbook/eda/section3/eda356.htm
  2. OpenStax Introductory Statistics (Rice University) – free textbook — openstax.org/details/books/introductory-statistics-2e

Results are estimates for general information. Last reviewed: .

Zeeshan Nawaz

Written & built by

Zeeshan Nawaz

Calculator Developer · Tuff Search

Builds and tests Tuff Search calculators, turning published formulas and official sources into tools anyone can check.

FAQ

Frequently asked questions

Quick answers about the Z-Score Calculator.

The number of standard deviations a value is above or below the mean.

Yes. A negative z-score means the value is below the mean.

It depends on context. Above 0 is above average; above 2 is unusually high (about the top 2.3%).

Find the area to the left of z under the standard normal curve – the calculator does this for you.

About 1.645.

1.96 for a two-sided 95% confidence interval.

Still have a question? Email support@tuffsearch.com