Ready to calculate?
Use our free Z-Score Calculator — no sign-up required.
A z-score (also called a standard score) measures how many standard deviations a data point is from the mean of its distribution. Z-scores are essential in statistics for comparing values from different distributions, identifying outliers, computing probabilities, finding percentiles, and conducting hypothesis tests. Whether you're analyzing test scores, quality control data, medical measurements, or financial returns, understanding z-scores unlocks powerful statistical insights.
Key Takeaways
- Z-score = (value − mean) ÷ standard deviation — measures SDs from the mean
- 68% of normally distributed data falls within z = ±1; 95% within z = ±2
- z = 0: 50th percentile | z = 1: 84th percentile | z = −1: 16th percentile
- Use z-scores to compare across different distributions or scales
- Outliers: |z| > 2 = unusual; |z| > 3 = very unusual, investigate for errors
The Z-Score Formula
The z-score formula converts any observation to a standardized scale:
For a population: z = (x − μ) / σ
For a sample: z = (x − x̄) / s
Where: • x = the observed value • μ = population mean (or x̄ = sample mean) • σ = population standard deviation (or s = sample standard deviation) • z = the z-score
Interpretation: • z = 0: the value equals the mean • z = 1: the value is 1 standard deviation above the mean • z = −1: the value is 1 standard deviation below the mean • z = 2: the value is 2 standard deviations above the mean
Example: A test has mean = 70 and standard deviation = 10. A student scored 85. • z = (85 − 70) / 10 = 15 / 10 = 1.5 • Interpretation: the score is 1.5 standard deviations above the mean
Negative z-scores: any value below the mean has a negative z-score. z = −2 means 2 standard deviations below average.
- Formula: z = (x − mean) ÷ standard deviation
- z = 0: value equals the mean
- Positive z: above average | Negative z: below average
- z = 1.5 means 1.5 standard deviations above the mean
The Standard Normal Distribution
When data follows a normal (bell curve) distribution, z-scores follow the standard normal distribution: • Mean = 0 • Standard deviation = 1 • Symmetric around 0
The Empirical Rule (68-95-99.7 Rule): • 68.27% of data falls within ±1 standard deviation (z between −1 and +1) • 95.45% of data falls within ±2 standard deviations • 99.73% of data falls within ±3 standard deviations • 99.99% falls within ±4 standard deviations
Probability from z-scores: Using the standard normal table (or calculator): • P(Z < 1.5) = 0.9332 (93.32% of data falls below z = 1.5) • P(Z > 1.5) = 1 − 0.9332 = 0.0668 (6.68% falls above) • P(−1 < Z < 1) = P(Z < 1) − P(Z < −1) = 0.8413 − 0.1587 = 0.6827 (68.27%)
Key z-values to memorize: • z = ±1.645: 90% confidence interval boundaries • z = ±1.960: 95% confidence interval boundaries • z = ±2.326: 98% CI boundaries • z = ±2.576: 99% CI boundaries
- 68% of data within ±1σ | 95% within ±2σ | 99.7% within ±3σ
- P(Z < 1.96) = 0.975 — used for 95% confidence intervals
- z = ±1.645: 90% CI | z = ±1.96: 95% CI | z = ±2.576: 99% CI
- Standard normal: mean = 0, standard deviation = 1
Percentiles and Z-Scores
Z-scores directly correspond to percentile ranks in a normal distribution:
Z-score to percentile: • Use the cumulative standard normal distribution table (z-table) • Percentile = P(Z ≤ z) × 100
Common z-score to percentile conversions: | Z-Score | Percentile | |---------|------------| | −3 | 0.13% | | −2 | 2.28% | | −1.645 | 5% | | −1 | 15.87% | | 0 | 50% | | +1 | 84.13% | | +1.282 | 90% | | +1.645 | 95% | | +1.960 | 97.5% | | +2 | 97.72% | | +2.326 | 99% | | +3 | 99.87% |
Percentile to z-score: • 75th percentile → z = 0.674 • 90th percentile → z = 1.282 • 95th percentile → z = 1.645 • 99th percentile → z = 2.326
Example application: SAT score of 1350, mean = 1060, SD = 195 • z = (1350 − 1060) / 195 = 290/195 = 1.49 • Percentile ≈ 93rd percentile (top 7% of test-takers)
- z = 0: 50th percentile | z = 1: 84th percentile | z = −1: 16th percentile
- 90th percentile: z = 1.282 | 95th percentile: z = 1.645 | 99th: z = 2.326
- Percentile = area under normal curve to the left of z-score
- SAT/ACT scores are often reported as z-scores converted to scale scores
Using Z-Scores for Comparison
Z-scores enable comparison across different scales and distributions — one of their most powerful uses.
Example: Comparing performance across two different tests • Student scored 78 on Test A (mean 65, SD 12) and 82 on Test B (mean 75, SD 10) • Test A z-score: (78 − 65)/12 = 1.08 • Test B z-score: (82 − 75)/10 = 0.70 • Conclusion: Test A performance was relatively better despite the lower raw score
Comparing distributions: • SAT vs. ACT scores (different scales, both approximately normal) • Heights in different countries (different means, different SDs) • Investment returns from different assets
Detecting outliers: • Values with |z| > 2: somewhat unusual (outside 95% of data) • Values with |z| > 3: very unusual (outside 99.7% of data) — likely an outlier • Values with |z| > 3.5: extreme outlier — investigate for data entry errors
In quality control (Six Sigma): • Six Sigma means 6 standard deviations between the mean and the nearest spec limit • This corresponds to only 3.4 defects per million opportunities
- Z-scores standardize different scales — enabling apples-to-apples comparison
- Higher z-score = further above the mean relative to peers
- Outlier detection: |z| > 2 is unusual; |z| > 3 is very unusual
- Six Sigma quality: 6 SD between mean and spec limit = 3.4 defects per million
Z-Score Applications in Hypothesis Testing
Z-scores are the foundation of z-tests in statistics:
One-sample z-test: • Tests if a sample mean differs significantly from a population mean • z = (x̄ − μ) / (σ / √n) • Where n = sample size and σ/√n = standard error
Example: A factory claims mean weight = 500g (σ = 20g). Sample of 36 items: x̄ = 508g. • z = (508 − 500) / (20/√36) = 8 / (20/6) = 8 / 3.33 = 2.40 • P-value for z = 2.40 (two-tailed): 0.016 (1.6%) • At α = 0.05: reject H₀ — evidence the population mean ≠ 500g
Critical values for common significance levels: • α = 0.10: |z| > 1.645 → reject null (one-tailed) or |z| > 1.282 (two-tailed 0.10) • α = 0.05: |z| > 1.960 (two-tailed) • α = 0.01: |z| > 2.576 (two-tailed)
When to use z-test vs. t-test: • Z-test: population SD (σ) is known, or sample n > 30 • T-test: population SD unknown and n < 30
Z-scores in confidence intervals: • 95% CI: x̄ ± 1.96 × (σ/√n) • 99% CI: x̄ ± 2.576 × (σ/√n)
- Z-test formula: z = (x̄ − μ) ÷ (σ/√n) — accounts for sample size
- Critical values: z = ±1.96 for 95% CI; z = ±2.576 for 99% CI
- Use z-test when σ is known or n ≥ 30; use t-test when σ unknown and n < 30
- P-value < α (e.g., 0.05): reject the null hypothesis
Real-World Applications of Z-Scores
Z-scores appear in many practical contexts:
Education: • Standardized test scores (SAT, GRE, IQ tests) often expressed as z-scores • Class rank percentile calculations • Identifying students who need extra support (z < −1.5) or gifted programs (z > 2)
Medicine: • BMI and growth charts: pediatric height and weight expressed as z-scores vs. age norms • Clinical test reference ranges: values beyond ±2 SD flagged as outside normal range • Bone density (DEXA) T-score and Z-score for osteoporosis diagnosis
Finance: • Altman Z-Score: bankruptcy prediction model using financial ratios • Stock returns: z-score of daily return vs. historical mean — signals unusual activity • Portfolio risk: z-scores of position sizes relative to mean allocation
Quality control: • Process capability: z-scores measure how far specification limits are from process mean • Control charts (3-sigma rule): points beyond ±3 sigma trigger investigation
Social sciences: • Survey response normalization across different scales • Comparing diverse demographic groups on standardized metrics
- Standardized tests: SAT, GRE, IQ all use z-score principles for norming
- Medical: growth charts and lab values referenced as z-scores vs. age norms
- Finance: Altman Z-Score predicts bankruptcy risk from accounting ratios
- Quality control: 3-sigma (z > 3) rule triggers process investigation
Frequently Asked Questions
What does a z-score of 1.5 mean?
A z-score of 1.5 means the value is 1.5 standard deviations above the mean of the distribution. In a normal distribution, approximately 93.3% of values fall below this point (it's at the 93rd percentile). Only about 6.7% of data points score higher. For example, a test score with z = 1.5 when mean = 70 and SD = 10 corresponds to a raw score of 70 + 1.5(10) = 85.
How do I convert a z-score to a percentile?
Use a standard normal distribution table (z-table) or a statistical calculator. Look up the z-score to find P(Z ≤ z), then multiply by 100 for the percentile. Example: z = 1.28 → P(Z ≤ 1.28) = 0.8997 ≈ 90th percentile. For negative z-scores: z = −0.5 → P(Z ≤ −0.5) = 0.3085 = 30.85th percentile.
Can a z-score be negative?
Yes — any value below the mean has a negative z-score. A z-score of −2 means the value is 2 standard deviations below the mean, placing it at approximately the 2.3rd percentile. Z-scores range from negative infinity to positive infinity, though in practice most data falls between −3 and +3.
What is the difference between a z-score and a t-score?
A z-score uses the population standard deviation (σ) and applies when the population SD is known or the sample is large (n ≥ 30). A t-score (from Student's t-distribution) uses the sample standard deviation (s) and is appropriate for small samples (n < 30) when the population SD is unknown. As sample size increases, the t-distribution approaches the normal distribution, so z and t give nearly identical results for large samples.
What is a good or bad z-score?
There's no universally good or bad z-score — it depends on context. For a test score, higher (more positive) is better. For a toxicity measurement, lower is better. For quality control, values near 0 are ideal. As an outlier indicator: |z| < 2 is typical; |z| between 2–3 is unusual; |z| > 3 suggests an outlier. In hypothesis testing, |z| > 1.96 indicates statistical significance at the 5% level.
Try the Z-Score Calculator
Free, instant, and accurate — calculate now.