How to Interpret a Z Score
A z score is a compressed sentence about a value: where it stands relative to its group, in units everybody shares. Interpreting one is a three-step routine — sign, size, rarity — that ends with the sentence written out in plain words.
The routine: (1) Sign — positive means above the mean, negative below. (2) Size — the absolute value counts standard deviations from the mean. (3) Rarity — convert to a percentile to say how unusual the position is. z = 1.4 → above the mean, by 1.4 standard deviations, standing at the 91.92nd percentile.
The size bands
- |z| under 0.5 — the ordinary middle. Between the 31st and 69th percentiles. "Perfectly typical" is the correct interpretation; most values of most datasets live here.
- |z| between 0.5 and 1 — noticeable, unremarkable. Up to the 84th (or down to the 16th) percentile. Worth a word in a report, not a conclusion.
- |z| between 1 and 2 — clearly above or below par. The 84th–97.7th percentile band (or its mirror). Selection, flagging and follow-up decisions usually start here.
- |z| between 2 and 3 — unusual. Beyond the 97.7th (or below the 2.3rd) percentile: fewer than 1 in 40 values. In test statistics this is the significance neighbourhood.
- |z| beyond 3 — exceptional. Rarer than 1 in 740 on its side of the curve. Interpret as an event demanding explanation — see the z score outliers page before concluding anything.
Three interpretations, written properly
- Exam, z = 1.40: "The score is 1.4 standard deviations above the class mean, at about the 92nd percentile — a strong result; roughly 8% of the class scored higher." (Φ(1.40) = 0.9192 from the positive table.)
- Height, z = −0.80: "The measurement is 0.8 standard deviations below the reference mean, at about the 21st percentile — shorter than average, well within the normal range." (Φ(−0.80) = 0.2119.)
- Process measurement, z = 2.90: "The reading is 2.9 standard deviations above the process mean; only about 0.19% of in-control readings reach this far. Investigate before the next batch." (Right tail at 2.90 = 0.0019.)
The misreadings that matter
"Higher z means a bigger value." A higher z score means a higher position within its own distribution — nothing cross-distribution follows. z = 2 on an easy quiz can be a lower raw attainment than z = 0.5 on a hard paper; this is also the honest answer to "what does a higher z score mean?": further above its mean and rarer, not automatically better or larger. "z is a percentage." z = 1.4 is not 140% of anything; the percentile (91.92nd) is the percentage-like quantity, and confusing the two is the commonest interpretation error in student work — the z score vs percentile page separates them permanently. "The sign is a grade." Below-mean is bad for marks, good for golf scores and blood pressure, and neutral for temperatures. Interpretation ends with the context sentence, not the number.
Also asked as: how to interpret the z score · how to interpret z-score · what does a higher z score mean · what does z score tell you · how to read a z score.
Related questions
Frequently asked questions
What is a good z score?
There is no universal "good" — it depends on the direction the field values. For exam performance, above +1 (84th percentile) is strong and above +2 exceptional. For defect counts or blood pressure, negative is good. Interpret against the purpose, then the bands above.
What does a z score of 0 mean?
The value equals the mean exactly — the 50th percentile, the centre of the distribution. It is the most ordinary position possible, neither above nor below average.