Z-Score Calculator
Convert a raw value or sample mean into a z-score, look up the corresponding percentile and p-value, and translate between z and IQ, T, SAT, GRE, or any custom scale. Includes a built-in one-sample z-test and reproducible R, Python, and APA output.
The observed score or measurement.
Mean of the reference population.
Standard deviation of the reference population.
Result
Interpretation
Within one standard deviation either side of the mean (1 ≤ |z| < 2): typical variation.
How the Z-Score Works and Where It Shows Up in Research
A z-score, or standard score, is the answer to one question: how many standard deviations is this value from the mean? The arithmetic is simple, subtract the mean, divide by the standard deviation, but the consequences are large because that one transformation puts every measurement onto the same dimensionless scale. A height in centimeters, an SAT verbal score, and a salary in dollars all become comparable once expressed as z-scores. The technique traces to Karl Pearson and his work on standardization in the 1890s, formalized in the modern textbook form by R. A. Fisher (1925) in Statistical Methods for Research Workers.
In descriptive statistics the z-score is a percentile lookup tool. Because the standard normal distribution is fully tabulated, a z of 1.5 immediately tells you the value sits at the 93.3rd percentile of a normal distribution; a z of −2 sits at the 2.3rd percentile. The 68-95-99.7 empirical rule (Cleveland 1985) summarizes this: roughly 68% of normally distributed values lie within ±1 SD of the mean, 95% within ±2 SD, and 99.7% within ±3 SD. The Score Conversion tab leverages this to translate between z and the standardized scales that consumer-facing tests use: IQ (M = 100, SD = 15), SAT section (M = 500, SD = 100), GRE section (M = 150, SD = 10), T-scores (M = 50, SD = 10), and stanines (M = 5, SD = 2). The transformation is always linear: score = M + z × SD.
In inferential statistics the z-score becomes a test statistic. The one-sample z-test compares an observed sample mean x̄ to a hypothesized population mean μ when σ is known: z = (x̄ − μ) / (σ / √n), evaluated against the standard normal. Two-proportion z-tests, Wald large-sample tests for regression coefficients, and standardized residuals in chi-square tables all reduce to z-statistics with the same critical values: |z| = 1.96 for α = 0.05 two-tailed, 2.58 for α = 0.01, and 3.29 for α = 0.001. The calculator's Sample Mean tab runs the full computation and reports the standard error explicitly so you can see what is being divided by what.
Three caveats matter in practice. First, the z-test requires the population SD to be known, which is rare outside calibrated reference distributions. When σ must be estimated from the sample, use the t-test instead, which follows Student's t with n − 1 df and accounts for the extra uncertainty. Second, the z-score assumes a normal underlying distribution; for heavily skewed data with small n, the Central Limit Theorem may not yet have kicked in and the nominal probabilities are wrong. Third, outlier detection by the |z| ≥ 3 rule is non-robust because the sample mean and SD are themselves pulled by extreme values. Iglewicz and Hoaglin (1993) recommend the modified z-score based on the median and MAD, which is unaffected by the very outliers you are trying to flag.
For full inferential analysis with assumption checks and effect sizes, pair this calculator with the p-value calculator (which handles t, chi-square, F, and r distributions), the confidence interval calculator for means, proportions, and ratio measures, and the two-sample t-test calculator when σ is unknown. For full study design and APA-formatted reporting from a PhD statistician, the statistical analysis service covers everything from assumption diagnostics to robust alternatives.
How to Use This Calculator
Pick the right tab
Raw Value for a single score, Sample Mean for a one-sample z-test, z ↔ percentile for distribution lookups, Score Conversion for IQ/SAT/T/GRE/stanine translation.
Enter your inputs
Raw value plus μ and σ, or sample mean plus μ, σ, and n. The percentile tabs need only one number; the conversion tab needs the target scale or a custom M and SD.
Read z, percentile, and p
Every numeric tab returns the z-score, the percentile (area to the left), the two-tailed and one-tailed p-values, and a plain-English outlier classification.
Translate to scaled scores
Use the Score Conversion tab to map between z and IQ (M=100, SD=15), SAT (M=500, SD=100), GRE (M=150, SD=10), T (M=50, SD=10), stanines (M=5, SD=2), or a custom scale.
Copy R, Python, or APA
Reproducible snippets paste straight into RStudio, Jupyter, or your manuscript. APA text uses italicized statistic letters and the conventional p < .001 truncation.
Next step
Got the z-score. Want the full normative or inferential analysis behind it?
A PhD statistician handles reference distribution selection, age and sex stratification, confidence intervals around the z, and a publication-ready APA results section.
Our promise: Free re-run and re-write if reviewers question the analysis or reporting.
Timeline
Most projects deliver in under 2 weeks. We confirm an exact date in your quote.
If reviewers push back
If reviewers question the analysis, assumptions, or reporting, we re-run and re-write free.
Confidentiality
NDA available on request before any project discussion. Your data, study design, and manuscript stay private either way.
Want a PhD methodologist to handle the whole project?
Get a complete normative analysis with reference distributions, age and sex stratification, confidence intervals, and APA-formatted reporting by a PhD statistician. Free re-run and re-write if reviewers question the analysis or reporting. Pay only after you approve your quote.
Reference Critical Values
1.645
α = 0.05 one-tailed; 95th percentile
1.960
α = 0.05 two-tailed; 97.5th percentile
2.326
α = 0.01 one-tailed; 99th percentile
2.576
α = 0.01 two-tailed; 99.5th percentile
3.090
α = 0.001 one-tailed; 99.9th percentile
3.291
α = 0.001 two-tailed; 99.95th percentile
±1.00
Central 68.27% (one SD interval)
±2.00
Central 95.45% (two SD interval)
Frequently Asked Questions
What is a z-score?
A z-score (also called a standard score) expresses how many standard deviations a value sits above or below the mean of its reference distribution. The formula is z = (x − μ) / σ where x is the raw value, μ is the population mean, and σ is the population standard deviation. A z of 0 is exactly at the mean; z = 1 is one SD above; z = −1.5 is 1.5 SDs below. Z-scores standardize values from different distributions onto a common scale (mean 0, SD 1) so they can be compared directly.
How do I calculate a z-score by hand?
Subtract the population mean from the raw value, then divide by the population standard deviation. For example, if a student scored 85 on a test with class mean 70 and SD 10, the z-score is (85 − 70) / 10 = 1.5, meaning the student scored 1.5 standard deviations above the class average. When you have a sample mean instead of an individual value, divide by the standard error σ / √n rather than σ alone, which is the formula the calculator uses in the Sample Mean tab.
What is the difference between a z-score and a t-score?
A z-score uses the population SD (σ), which is assumed known; the resulting statistic follows a standard normal distribution N(0, 1). A t-score uses the sample SD (s), which estimates σ from the data, and follows Student's t distribution with n − 1 degrees of freedom. For large samples (n > 30) the two are nearly identical because t converges to z. For small samples or when σ is genuinely unknown, the t-test is the correct choice. APA also uses 'T-score' to refer to a standardized score with mean 50 and SD 10, which is a linear transformation of z (T = 50 + 10z); the calculator's Score Conversion tab handles this.
How do I convert a z-score to a percentile?
A z-score maps to a percentile through the cumulative distribution function (CDF) of the standard normal. The percentile equals the area under the curve to the left of z, multiplied by 100. For example, z = 0 is the 50th percentile, z = 1 is the 84.13th percentile, z = 1.96 is the 97.5th percentile, and z = 2.58 is the 99.5th percentile. The calculator's z → percentile tab computes this directly, and the percentile → z tab reverses it using the inverse normal (quantile) function.
What is considered an unusual or outlier z-score?
By convention, |z| ≥ 2 marks a value as moderately unusual (it falls outside the central 95% of the distribution), and |z| ≥ 3 marks it as extreme (outside the central 99.7%). These follow from the 68-95-99.7 empirical rule for normal distributions. In quality control (Shewhart charts) the alarm threshold is |z| ≥ 3. For outlier detection in research, the modified z-score using median and MAD (Iglewicz and Hoaglin 1993) is more robust than the classical z-score because it is not pulled by the very outliers it is trying to detect.
What is a one-sample z-test?
A one-sample z-test compares a sample mean to a hypothesized population mean when the population standard deviation σ is known. The test statistic is z = (x̄ − μ) / (σ / √n) and is compared to the standard normal under H0. The two-tailed p-value is 2 × P(Z > |z|). Most real-world applications use the t-test because σ is rarely known; the z-test is the textbook benchmark and remains correct for very large samples or when σ comes from a calibrated reference (for example, IQ tests calibrated to σ = 15 in the population). The calculator's Sample Mean tab runs this test.
Can z-scores be negative?
Yes. A negative z-score means the raw value is below the mean. There is nothing special about negative versus positive z-scores in terms of unusualness; what matters is the absolute value |z| for two-tailed inference. For example, z = −2.33 and z = 2.33 both correspond to two-tailed p ≈ 0.02. Educational testing companies (SAT, ACT) and clinical assessments (IQ, T-scores) deliberately use scaled scores with non-zero means specifically to avoid negative numbers for laypeople, which is what the Score Conversion tab translates between.
How do z-scores work for IQ, SAT, and other test scores?
Most norm-referenced standardized tests rescale the underlying z-score onto a more interpretable scale: IQ uses M = 100, SD = 15 (so z = 1 corresponds to IQ 115); SAT sections use M = 500, SD = 100 (so z = 1 corresponds to SAT 600); GRE sections use M = 150, SD = 10; T-scores in clinical psychology use M = 50, SD = 10; and stanines use M = 5, SD = 2. The transformation is linear: score = M + z × SD. The calculator's Score Conversion tab handles all five plus custom scales bidirectionally.
What sample size do I need for a z-test to be valid?
The z-test assumes either (a) the underlying population is normally distributed, or (b) the sample size is large enough for the Central Limit Theorem to make x̄ approximately normal. The CLT rule of thumb is n ≥ 30 for moderately skewed populations; for heavily skewed populations (counts, durations, income) n ≥ 100 is safer. With small n and unknown population SD, use the t-test instead. If the underlying distribution is heavily non-normal and n is small, consider a non-parametric alternative (Wilcoxon signed-rank) or a bootstrap.
How does a z-score relate to a p-value?
A two-tailed p-value is 2 × P(Z > |z|) = 2 × (1 − Φ(|z|)) where Φ is the standard normal CDF. For example, z = 1.96 gives p = 0.05, z = 2.58 gives p = 0.01, and z = 3.29 gives p = 0.001. One-tailed p-values use the upper or lower tail alone. The calculator's z → percentile tab returns the full set (two-tailed p, both one-tailed p-values, and percentile) so you can pick the one your analysis requires. For more general test statistics (t, χ², F, r), the p-value calculator handles them directly.
Related Research Tools
For converting a test statistic into a p-value across distributions (z, t, χ², F, r), use the p-value calculator. When the population SD is unknown and σ is estimated from the sample, switch to the two-sample t-test calculator. For confidence intervals on means, proportions, odds ratios, and risk ratios, the confidence interval calculator covers Wald, Wilson, and log-Wald methods. For effect-size standardization (Cohen's d, Hedges' g, log odds, Fisher's z), see the effect size calculator. For a priori sample-size planning with z-based formulae, the sample size calculator handles means, proportions, and survival outcomes.
Reviewed by
Dr. Sarah Mitchell
Dr. Sarah Mitchell holds a PhD in Biostatistics from Johns Hopkins Bloomberg School of Public Health and has over 15 years of experience in systematic review methodology and meta-analysis. She has authored or co-authored 40+ peer-reviewed publications in journals including the Journal of Clinical Epidemiology, BMC Medical Research Methodology, and Research Synthesis Methods. A former Cochrane Review Group statistician and current editorial board member of Systematic Reviews, Dr. Mitchell has supervised 200+ evidence synthesis projects across clinical medicine, public health, and social sciences. She reviews all Research Gold tools to ensure statistical accuracy and compliance with Cochrane Handbook and PRISMA 2020 standards.
Need a Statistician? Our PhD Team Is Standing By.
From data cleaning and assumption checks to the full analysis and a publication-ready results section, we handle the numbers so you can focus on the science.
Our promise: Free re-run and re-write if reviewers question the analysis or reporting.
The methodologists behind your review
Your project is led by a named PhD methodologist with real credentials and published work.
4.9 / 5 across 1,194+ delivered projects
