Z-Score, Explained

A z-score (standard score) tells you how many standard deviations a value sits from the mean. It converts any normal measurement — test scores, heights, response times — onto one universal scale, which is what makes the z table possible.

z = (X − μ) / σ  where X is the value, μ the mean, σ the standard deviation.

Interpreting z-scores

z-scoreMeaning
0Exactly at the mean
+1One standard deviation above — beats ~84% of values
−1One below — beats only ~16%
+2Well above — top ~2.3%
±1.96The 95% boundary used in confidence intervals

Worked example

Your exam score is 82; the class mean is 74 with σ = 5. Then z = (82 − 74) / 5 = 1.6. The positive z table gives 0.94520 — you outscored about 94.5% of the class. That single conversion — raw number → z → percentile — is most of what introductory statistics asks you to do.

Why z-scores matter

They make different scales comparable (SAT vs ACT, cm vs inches), they power z-tests for hypothesis checking, they flag outliers (|z| > 3 is a common screen), and in finance a related idea drives the Altman Z-Score. The prerequisite is a roughly normal distribution — z-scores can be computed for any data, but their percentile meaning comes from normality.