Skip to content

Z-Score Formula

How many standard deviations a value is from the mean.

Variables

x
The value
μ
Mean
σ
Standard deviation

Explanation

A z-score puts values from different scales on the same scale. Positive means above the mean, negative means below. For normally distributed data, a z-table converts to the share of values below it.

Example

A test score of 85 when and : , about the 93rd percentile.

How to use it

  1. Subtract the mean from the value.
  2. Divide by the standard deviation.
  3. Read the result as "how many standard deviations above (positive) or below (negative) the mean."

Another worked example

Adult heights in a group have mean 170 cm and standard deviation 8 cm. A person 158 cm tall has

They're 1.5 standard deviations below average, shorter than about 93% of the group.

Comparing different scales

z-scores let you compare apples and oranges. A 680 on a test with mean 500 and SD 100 has . A 27 on a test with mean 21 and SD 5 has . The first result is relatively stronger.

Working backwards

Rearrange to . To find the score at the 90th percentile () when and : .

Key z-values for normal data

  • : the mean, 50th percentile.
  • : about 68% of values lie between these.
  • : 95% lie between, used for 95% confidence intervals.
  • : 99.7% lie between. Values beyond are often treated as outliers.

For sample means

The z-score for a sample mean uses the standard error: . This is the basis of the z-test.

Common mistakes

  • Subtracting in the wrong order, which flips the sign.
  • Dividing by the variance instead of the standard deviation.
  • Reading percentiles from a z-table when the data aren't roughly normal.

Where it's used

  • Education: standardised tests report scores and percentiles using z-scores.
  • Health: children's growth charts compare height and weight to age norms in z-scores.
  • Finance: detecting unusual price moves or fraud by how many standard deviations a value sits from normal.
  • Manufacturing: Six Sigma quality aims for defects more than 6 standard deviations from the mean.

Key takeaways

  • counts standard deviations from the mean.
  • Positive is above average, negative below.
  • With normal data, a z-table turns into a percentile.

Try it: Z-Score Calculator

Calculate a z-score from a value, mean and standard deviation, and find the matching normal-distribution probability and percentile.

Open calculator

Practice problems

Try each one first, then open the answer to check your work.

  1. 1., , . Find .

    Show answer

    3.

  2. 2., , . Find .

    Show answer

    .

  3. 3., , . Find .

    Show answer

    122.5.

  4. 4.About what percentile is ?

    Show answer

    About the 84th.

  5. 5.A sample of 25 has mean 52 from a population with , . Find .

    Show answer

    .

Key terms

z-score
The number of standard deviations from the mean.
Standard normal distribution
The normal distribution with mean 0 and SD 1.
Percentile
The percentage of values below a given value.
Standard error
The standard deviation of a sample mean, .
Standardise
To convert a value to its z-score.

Frequently asked questions

What z-score is unusual?

Values with |z| > 2 are often called unusual, because only about 5% of normal data lies that far from the mean.

What does a z-score of 0 mean?

The value equals the mean exactly. Positive z-scores are above average; negative ones are below.

How do I turn a z-score into a percentile?

Look it up in a standard normal table or use a calculator's normal CDF. For example, is the 95th percentile.

What counts as an unusual z-score?

Values beyond are unusual (about 5% of normal data) and beyond are very rare (about 0.3%).

Do z-scores require normal data?

You can compute a z-score for any data. But converting it to a percentile with the normal table only works if the data are roughly normal.

What percentage of values lie between $z = -1$ and $z = 2$?

About 34% + 47.5% = 81.5% for normal data, from the 68-95-99.7 rule.

Is a z-score the same as a t-score?

They're computed the same way, but a t-score uses the sample standard deviation and the t-distribution, which matters for small samples.

Can a z-score be greater than 3?

Yes. It just means the value is very far from the mean. In normal data that's rare, but in skewed data it happens more often.

How do I compare z-scores from two groups?

Directly. Because both are measured in standard deviations, a z of 1.5 is equally far above average in either group, whatever the units.

What's the z-score for the top 5%?

About 1.645. For the top 1% it's about 2.326, and for the top 2.5% it's 1.96.

Related formulas

More statistics formulas