A z-score (standard score) measures how many standard deviations a value is from the mean. z = (x − μ) / σ. A z-score of 0 means the value equals the mean. z = 1 means one SD above the mean. z = −2 means two SDs below the mean.

How do I calculate a z-score using this calculator?

Enter your full dataset in the Descriptive Stats tab and press Calculate. Note the mean (μ) and sample SD (s) from the results. Then apply z = (x − mean) / SD for any specific value x. Example: mean=50, SD=10, x=65: z = (65−50)/10 = 1.5.

What does a z-score of 1.5 mean?

z = 1.5 means the value is 1.5 standard deviations above the mean. For a normal distribution this corresponds to approximately the 93.3rd percentile — the value is higher than 93.3% of the data.

What is the 68-95-99.7 rule for z-scores?

For a normal distribution: 68% of values have |z| less than 1, 95% have |z| less than 1.96, and 99.7% have |z| less than 3. A z-score beyond ±3 is extremely rare (0.3% of data).

Should I use population SD or sample SD for z-scores?

Use population SD (σ) when you have the entire population. Use sample SD (s) when your data is a sample — the resulting score is sometimes called a t-score for small samples. The calculator shows both; for large samples (n > 30) the difference is negligible.

How do I convert a z-score to a percentile?

Use the standard normal table. z=0 → 50th percentile. z=1.0 → 84.1th. z=1.645 → 95th. z=1.96 → 97.5th. z=2.576 → 99.5th. z=−1.0 → 15.9th. For negative z, percentile = 1 minus the percentile of the positive z.

What is z-score used for in real life?

Standardized tests (SAT, GRE) report z-scores as scaled scores. Finance uses z-scores to measure how unusual a return is (Altman Z-score for bankruptcy risk). Quality control uses z-scores to detect out-of-spec products. Any field comparing values from different scales uses z-scores to standardize.

Z-Score Calculator

Enter your dataset to get mean and standard deviation, then apply z = (x − μ) / σ for any value. Includes percentile interpretation and the 68-95-99.7 rule.

Dataset (comma, space, or newline separated)

You might also need

Average Calculator Standard Deviation Calculator Confidence Interval Calculator P-Value Calculator Density Calculator Molarity Calculator

Guides & Reference

How It Works

Step 1 — Get mean and SD from your datasetAny dataset: test scores, measurements, financial returns.

Enter all your values in the Descriptive Stats tab separated by commas. Press Calculate. The results panel shows Mean (μ) and both Sample SD (s) and Population SD (σ). Use these to compute z-scores for any individual value in — or outside — your dataset.

z = (x − μ) / σ | use Sample SD for sample data

Scores: 60,70,72,80,88 → mean=74, SD=10.3 → z(88)=(88−74)/10.3=1.36

Z-score interpretationAssessing how unusual a value is relative to the group.

Positive z: value is above the mean. Negative z: below the mean. |z| less than 1: typical (within 68% of data). |z| between 1 and 2: moderately unusual. |z| between 2 and 3: rare (within top/bottom 5%). |z| above 3: very rare (beyond 99.7%). The skewness value in the results panel tells you if the 68-95-99.7 rule applies to your data.

z = 0: mean | |z| < 1: 68% range | |z| < 2: 95% | |z| < 3: 99.7%

z=−2.1 → below mean, rarer than 98% of values, bottom ~1.8%

Comparing across datasetsComparing performance on tests with different scales.

Z-scores remove units and scale. If Alice scores 85/100 on a math test (mean=75, SD=8) and Bob scores 620/800 on a different test (mean=500, SD=100), who performed better? Alice: z=(85−75)/8=1.25. Bob: z=(620−500)/100=1.20. Alice performed slightly better relative to her group despite the different scales.

z standardizes: z = (score − group mean) / group SD

Alice z=1.25 > Bob z=1.20 → Alice outperformed relative to her group

Using SEM (Standard Error of Mean)Assessing whether a sample mean differs from a known population mean.

The calculator also shows SEM = SD / √n. SEM measures uncertainty in the sample mean itself. A z-score based on SEM: z = (sample mean − population mean) / SEM. This is the foundation of z-tests. Example: sample mean=52, population mean=50, SEM=1.5: z=(52−50)/1.5=1.33 — not statistically significant at p=0.05 (needs |z| greater than 1.96).

SEM = SD / √n | z_test = (x̄ − μ₀) / SEM

n=25, SD=7.5: SEM=1.5 | z=(52−50)/1.5=1.33 (p>0.05, not significant)

Outlier detection using z-scoresData cleaning, quality control, anomaly detection.

Values with |z| greater than 3 are statistical outliers (beyond 99.7% of a normal distribution). The calculator uses the Tukey IQR method for outlier detection (shown in results), which is more robust for non-normal data. For normal data, compare: if |z| exceeds 3 for a value that Tukey does not flag, your distribution likely has heavy tails. Use both methods together.

SD outlier: |z| > 3 | Tukey: Q1−1.5×IQR or Q3+1.5×IQR

Data with mean=100, SD=10: value 135 has z=3.5 → statistical outlier

Quick Reference

Key z-score values and their percentile equivalents.

At mean

z when x=mean

Percentile

z=1.0 → percentile

84.1st

Percentile

z=1.96 → percentile

97.5th

Percentile

z=−1.0 → percentile

15.9th

Normal rule

|z| < 1 covers

68% of data

Normal rule

|z| < 1.96 covers

95% of data

Example

mean=72, SD=8, x=88

z = 2.0

Outlier rule

Outlier threshold

|z| > 3

Tips & Shortcuts

✓

Enter your full dataset first — the calculator gives you mean and SD in one step, then apply z = (x − mean) / SD manually for any specific value you want to assess.

✓

The SEM (Standard Error of Mean) in the results panel is used for z-tests — divide the difference between your sample mean and the population mean by SEM to get the test z-score.

✓

Check skewness before using the 68-95-99.7 rule. If skewness is above 1 or below −1, your data is not normal and z-score percentile interpretations are approximate.

✓

Use Sample SD (s) for most real-world data. The calculator shows it labeled as Sample SD — it is the value to use in z = (x − x̄) / s for sample data.

✓

Z-scores can exceed ±3 — they are not capped. A z-score of 4.5 simply means the value is 4.5 standard deviations from the mean, which occurs in less than 0.001% of normally distributed data.

Common Mistakes

Using population SD when data is a sample

For sample data, use Sample SD (s, divides by n−1). Population SD underestimates the true spread for samples, making z-scores slightly too large. For n greater than 30 the difference is small; for n=5 the difference is about 12%.

Applying the 68-95-99.7 rule to non-normal data

The 68-95-99.7 rule is specifically for normal distributions. Check the skewness value in the results — if |skewness| exceeds 1, your data is meaningfully skewed and z-score percentile conversions are only approximate.

Confusing z-score with percentile rank

z-score and percentile are related but different. z=1.0 corresponds to the 84.1st percentile — not the 100th. z=0 is the 50th percentile (median for symmetric data). Use a z-table or the critical values listed in the Quick Reference to convert accurately.

Using z-scores to compare within the same dataset without standardizing

Z-scores are most useful when comparing across different datasets or scales. Within a single dataset, simply comparing raw values works fine — z-scores add no information when all values share the same mean and SD.

Forgetting that z-scores can be negative

Any value below the mean has a negative z-score. z=−1.5 is perfectly valid — it means 1.5 SDs below the mean, at roughly the 6.7th percentile. Negative z-scores are not errors.

Frequently Asked Questions

A z-score (standard score) measures how many standard deviations a value lies from the mean: z = (x − μ) / σ. z = 0: value equals the mean. z = 1: one SD above the mean. z = −2: two SDs below the mean. Z-scores allow comparison across datasets with different scales and units.

Enter your full dataset in the Descriptive Stats tab and press Calculate. The results panel shows Mean and Sample SD. Apply z = (x − mean) / SD for any specific value. Example: test scores with mean=72, SD=8. A score of 88: z = (88−72)/8 = 2.0. That score is 2 standard deviations above the mean — approximately the 97.7th percentile.

z = 1.5 means the value is 1.5 standard deviations above the mean. For a normally distributed dataset this corresponds to roughly the 93.3rd percentile — higher than 93.3% of values. A z-score of −1.5 is the mirror: 1.5 SDs below the mean, at the 6.7th percentile.

For a normal distribution: |z| less than 1 covers 68.3% of values, |z| less than 1.96 covers 95%, |z| less than 2.576 covers 99%, |z| less than 3 covers 99.7%. A z-score above 3 or below −3 is extremely rare (0.3% probability). This rule lets you judge how unusual any value is without a lookup table.

Population SD (σ, divides by N): use when you have every member of the population. Sample SD (s, divides by n−1): use when your data is a sample. The resulting score with sample SD is technically a t-score for small samples. For n greater than 30, the difference between σ and s is negligible. The calculator shows both — use Sample SD (s) by default.

Key values: z=0 → 50th percentile. z=1.0 → 84.1th. z=1.645 → 95th. z=1.96 → 97.5th. z=2.326 → 99th. z=2.576 → 99.5th. For negative z: percentile = 100 minus percentile of the corresponding positive z. z=−1.0 → 15.9th percentile. For exact conversion use a z-table or the confidence interval calculator.

Standardized tests: SAT/GRE scores are scaled z-scores (mean 500, SD 100 for old SAT). Finance: Altman Z-score predicts corporate bankruptcy using five financial ratios. Quality control: Six Sigma targets processes where defects fall beyond 6 SDs (z=6). Epidemiology: z-scores compare child growth measurements across age groups. Any cross-dataset comparison benefits from z-score standardization.

Related Calculators

Variance Calculator

Calculate population and sample variance.

Statistics Calculator

Full descriptive statistics: mean, median, mode, range, std dev.

Molecular Weight Calculator

Calculate the molecular weight of any chemical formula.

Student Loan Calculator

Monthly payment and payoff timeline for student debt.

Scientific Calculator

Full scientific calculator with trig, log, and advanced functions.

Basic Calculator

Fast arithmetic: add, subtract, multiply, divide.