Normal Distribution Calculator
Convert a value or z-score to a cumulative probability, or find the z-score for a given probability.
z = (x − μ) / σ
Example: a test score of 85 with class mean 75 and standard deviation 10 gives z = 1.0.
Z-Score
Turning Any Value Into a Z-Score
The normal (bell curve) distribution shows up constantly — test scores, heights, measurement errors — and the z-score is what lets you compare a specific value against the whole distribution on one universal scale:
Where x is your value, μ (mu) is the distribution's mean, and σ (sigma) is its standard deviation. A test score of 85 in a class with a mean of 75 and a standard deviation of 10 gives z = (85−75)/10 = 1.0 — exactly one standard deviation above average.
From Z-Score to Probability (and Back)
Once you have a z-score, the standard normal distribution tells you what fraction of values fall below it — this is the cumulative distribution function (CDF), traditionally looked up in a printed z-table. This calculator computes it directly instead:
- z = 0 → 50% (exactly at the mean, by definition).
- z = 1.96 → 97.5% (the upper bound of a 95% two-tailed confidence interval).
- z = 1.645 → 95% (the boundary for a 90% two-tailed or 95% one-tailed interval).
- z = 2.576 → 99.5% (the boundary for a 99% two-tailed interval).
The "Probability → Z-Score" tab runs this backwards — given a target probability, it returns the z-score cutoff, which is exactly what you need when setting up a confidence interval or hypothesis test threshold.
How to Use This Calculator
- Value → Probability: enter your raw value, the distribution's mean, and its standard deviation — get the z-score and the probability above/below it.
- Z-Score → Probability: already have a z-score? Enter it directly.
- Probability → Z-Score: enter a target cumulative probability (e.g. 0.95) to find the corresponding z-score cutoff.
The Empirical Rule: A Quick Sanity Check Without a Calculator
Before reaching for exact z-score math, the "68-95-99.7 rule" gives a fast approximate check: roughly 68% of values in a normal distribution fall within 1 standard deviation of the mean, about 95% fall within 2 standard deviations, and about 99.7% fall within 3. So a value more than 2 standard deviations from the mean is already in the outer ~5% of the distribution — a useful mental shortcut for spotting an unusual result before running the precise calculation. This rule is an approximation, not exact (the real 2-standard-deviation boundary is closer to 95.4%, and the precise 95% cutoff is actually at 1.96 standard deviations, not exactly 2) — which is exactly why 1.96 rather than a round 2.0 shows up as the standard z-score for a 95% confidence interval.
Where This Math Shows Up Beyond the Classroom
Z-scores and normal-distribution probability aren't just a statistics-class exercise — quality control processes use them to flag manufactured parts that fall outside an acceptable range of measurements, standardized testing uses them to convert raw scores into percentile rankings comparable across different test administrations, and A/B testing in product and marketing analysis uses the same underlying math to determine whether an observed difference between two groups is large enough to be statistically meaningful rather than random noise. In each case, the same core question repeats: given a value's distance from the average in standard-deviation units, how likely or unlikely is a result like this to occur by chance.
Frequently Asked Questions
What is a z-score?
How many standard deviations a value is from the mean: z = (x − μ) / σ. A z-score of 1.0 is one standard deviation above the mean; -2.0 is two below.
How do I convert a z-score into a probability?
Look up the cumulative probability under the standard normal curve up to that z-score. z=0 gives 50%, z=1.96 gives 97.5%, z=-1.96 gives 2.5%.
Why are 1.96 and 1.645 such common z-scores?
1.96 marks the 97.5th percentile (95% two-tailed confidence interval boundary); 1.645 marks the 95th percentile (90% two-tailed or 95% one-tailed). These come from the conventional 90/95/99% confidence levels used across statistics.
What does 'inverse' mean in Probability → Z-Score mode?
It runs the lookup backwards — given a target probability, it finds the z-score below which that proportion of the distribution falls. This is what a z-table's critical-value lookup does.
Does this only work for test scores or IQ?
Any approximately normally-distributed data — test scores, height, measurement error. It doesn't apply to skewed or multi-peaked data (like income, which is typically right-skewed).