Z-Score Calculator
Enter a value, mean, and standard deviation to find the z-score and percentile rank instantly — or enter a z-score to find the corresponding raw value. Includes probability interpretation and full step-by-step calculation.
Enter your values above to see the results.
Tips & Notes
- ✓Standard deviation must be positive and non-zero. A σ of 0 makes the z-score formula undefined — you cannot standardize a dataset where all values are identical.
- ✓Z-scores are dimensionless — they have no units. A z-score of 1.5 means the same thing regardless of whether the original data was in kilograms, dollars, or test points.
- ✓Most z-scores in real-world data fall between −3 and +3. A z-score beyond ±3 should prompt you to verify the data point — it may be an outlier, an entry error, or genuinely exceptional.
- ✓Use sample standard deviation (s) when your data is a sample, and population SD (σ) when you have complete population data. The z-score formula is the same, but the SD value and interpretation differ.
- ✓Z-scores to percentiles only work correctly for normally distributed data. For skewed or non-normal distributions, use the rank-based percentile method instead.
Common Mistakes
- ✗Subtracting in the wrong order. Z-score = (x − μ)/σ, not (μ − x)/σ. A value below the mean gives a negative z-score; above the mean gives positive. Swapping the order flips the sign and misidentifies the direction.
- ✗Using the wrong standard deviation. Sample SD (s) and population SD (σ) produce different z-scores. For a dataset that is itself a sample from a larger population, always use sample SD. Using population SD underestimates spread.
- ✗Applying z-score percentiles to non-normal data. Φ(z) gives accurate percentiles only for normally distributed data. For skewed distributions, a z-score of 2.0 may not correspond to the 97.72nd percentile at all.
- ✗Confusing z-score with percentile. z=1.0 is at the 84th percentile, not the 100th percentile. Z-score and percentile are related but not equal — always convert using Φ(z) rather than equating them directly.
- ✗Forgetting that z-scores can be negative. Values below the mean have negative z-scores. A z-score of −1.5 means 1.5 standard deviations below the mean — it is not an error, it is simply a below-average value.
Z-Score Calculator Overview
A z-score measures how many standard deviations a value sits above or below the mean. It converts any value from any distribution into a universal scale where 0 means exactly average, +1 means one standard deviation above average, and −2 means two standard deviations below average. This standardization is what allows comparing a student's math score with their science score, or a patient's blood pressure with their cholesterol level — even though these measurements use completely different scales.
Z-score — number of standard deviations from the mean:
z = (x − μ) / σ
EX: Exam with μ=70, σ=10. Score x=85 → z = (85−70)/10 = 15/10 = 1.50 → 1.5 standard deviations above the meanFinding the raw value from a z-score:
x = μ + z × σ
EX: μ=70, σ=10, z=−0.5 → x = 70 + (−0.5)(10) = 70−5 = 65 → A z-score of −0.5 corresponds to a raw score of 65Z-score to percentile — probability a random value falls below x:
P(X ≤ x) = Φ(z) — standard normal cumulative distribution
EX: z = 1.50 → Φ(1.50) = 0.9332 → Score of 85 is at the 93.32nd percentile — higher than 93.32% of studentsZ-score interpretation reference:
| Z-Score | Percentile | Interpretation | Probability Above |
|---|---|---|---|
| −3.0 | 0.13% | Extremely low — rare | 99.87% |
| −2.0 | 2.28% | Well below average | 97.72% |
| −1.0 | 15.87% | Below average | 84.13% |
| 0.0 | 50.00% | Exactly average | 50.00% |
| +1.0 | 84.13% | Above average | 15.87% |
| +2.0 | 97.72% | Well above average | 2.28% |
| +3.0 | 99.87% | Extremely high — rare | 0.13% |
Frequently Asked Questions
A z-score measures how many standard deviations a value is from the mean: z=(x−μ)/σ. Example: exam scores with μ=75, σ=10. A score of 90 gives z=(90−75)/10=1.5 — 1.5 standard deviations above average, better than about 93.3% of test-takers. A score of 60 gives z=(60−75)/10=−1.5 — 1.5 SDs below average, better than only about 6.7% of test-takers. The sign shows direction; the magnitude shows how unusual the value is.
Look up Φ(z) in a standard normal table — it gives the proportion of the distribution below z. Multiply by 100 for the percentile. z=1.28 → Φ(1.28)=0.8997 → 89.97th percentile. For negative z: Φ(−1.28)=1−Φ(1.28)=0.1003 → 10.03rd percentile. Using the symmetry of the normal distribution, Φ(−z)=1−Φ(z) always holds. This conversion only works correctly for normally distributed data.
Context determines whether a z-score is desirable. In academic grading, z=+2.0 (top 2.3%) is excellent; z=−2.0 (bottom 2.3%) indicates poor performance. In quality control, any product with |z|>3 falls outside specifications and signals a defect. In clinical medicine, a growth chart z-score beyond ±2 warrants monitoring. The z-score is a position on the scale — the field determines whether that position represents success or failure.
Yes — that is the primary purpose of z-scores. A student scoring z=+1.5 in math and z=+0.8 in English performed relatively better in math, even if the raw math score was numerically lower. Standardization removes the influence of different scales and units. This comparison only works reliably if both datasets are approximately normally distributed with meaningful means and standard deviations representing similar populations.
A z-score of exactly 0 means the value equals the mean: z=(x−μ)/σ=0 only when x=μ. If an exam mean is 75 and you score 75, your z-score is 0.00 — you are at the 50th percentile, exactly average. For a normal distribution, z=0 means half the population scored below you and half above. A z-score of 0 is not bad — it simply means the value is typical for the group.
Z-scores convert any normal distribution to the standard normal (μ=0, σ=1). The empirical rule applies directly to z-scores: 68% of data has |z|<1, 95% has |z|<2, 99.7% has |z|<3. The critical z-score of ±1.96 corresponds to the central 95% of the distribution — which is why 1.96 appears in every 95% confidence interval formula. Z-scores are the bridge between raw data values and standard normal probabilities.