Z-Score Calculator

Turn a raw value into standard deviations from the mean so you can compare scores across different scales.

Z-Score Calculator

Calculate z-scores to standardize values and determine how many standard deviations from the mean

Show Percentile
Show Probability
Show Interpretation

Example Scenarios

What a Z-score is

A Z-score (standard score) says how many standard deviations a value sits from the mean. Z = 0 is exactly average. Z = 1 is one SD above; Z = −1 is one SD below.

That makes cross-scale comparisons fair. An SAT score and an ACT score live on different number lines; both become comparable once you standardize them.

Formula

Z = (X − μ) / σ

X = raw value, μ = mean, σ = standard deviation

Reading the number

Positive

Above the mean. +1, +2, +3 = one, two, three SDs up. Beyond ±3 is rare under a normal curve.

Negative

Below the mean. Same scale in the other direction.

68-95-99.7 rule

In a normal distribution, about 68% of values fall within ±1 SD, about 95% within ±2, and about 99.7% within ±3.

Z-scores and percentiles

Under normality, Z maps to a percentile (share of the population below that value):

Z = −2
~2.3%
Z = −1
~15.9%
Z = 0
50%
Z = +1
~84.1%

Z = +2 ≈ 97.7th percentile.

Worked examples

SAT: X=1250, μ=1060, σ=195

Z = (1250−1060)/195 ≈ 0.97 → roughly 83rd percentile

Height: X=74 in, μ=69, σ=3

Z = 5/3 ≈ 1.67

Significance cutoffs

In two-sided tests, |Z| > 1.96 ≈ 5% level; |Z| > 2.58 ≈ 1%; |Z| > 3.29 ≈ 0.1%. Statistical significance is not the same as practical importance—especially with huge n.

Assumptions matter: roughly normal data, known or well-estimated μ and σ, independent observations. Skewed data still give a computable Z; the percentile story does not hold.

Frequently Asked Questions

Z-score vs T-score?

Use Z when population σ is known (or n is large). Use t when σ is estimated from the sample, especially with smaller n—t has heavier tails.

Can |Z| be greater than 3?

Yes. Beyond ±3 is uncommon under normality; beyond ±4 often means an outlier or a data error.

What does Z = 0 mean?

The value equals the mean—50th percentile under a symmetric normal curve.

Data not normal?

You can still compute Z; do not trust normal-table percentiles. Prefer ranks, or transform first if that fits the analysis.

Sample instead of population?

Using sample mean and SD gives a standardized score. Fine for large n; for small samples, lean on t-based methods.

Related statistical tools

Standard Deviation

Get σ for the Z formula

Open tool →

Average Calculator

Find the mean

Open tool →

Confidence Interval

Intervals that use Z or t

Open tool →

Related tools

Related tools