Z-Score Calculator

A test score of 115 sounds good - but is it? It depends entirely on how everyone else did. If the average was 100 and the typical spread (standard deviation) was 15, then 115 is exactly one standard deviation above average, which works out to a z-score of 1.0 and roughly the 84th percentile. That single number - the z-score, or standard score - is what lets you compare a value against the rest of its group, or even against values measured on a completely different scale. This calculator computes it instantly and reads it back to you in plain language.

The z-score formula

A z-score measures how many standard deviations a value sits above or below the mean. The formula is:

where x is your raw value, μ (mu) is the mean of the distribution, and σ (sigma) is the standard deviation. To go the other way - from a z-score back to a raw value - rearrange the same equation:

That is the whole idea. Subtracting the mean centers the value on zero; dividing by the standard deviation rescales it so that one unit equals one standard deviation. The result is a pure number with no units, which is exactly why z-scores are so useful for comparison.

How to use this z-score calculator

You only need three pieces of information. Work through the fields in order:

1. Pick a direction: use the Find z from x tab to standardize a raw value, or Find x from z to convert a z-score back into a raw value.
2. Enter your value: type the raw value (x) or the z-score, depending on the tab.
3. Enter the mean (μ): the average of the group your value belongs to.
4. Enter the standard deviation (σ): the typical spread of that group. It must be greater than zero.

The result updates as you type. The large number at the top is your answer; below it you get the approximate percentile, the area above and below the value, and a step-by-step view of the arithmetic so you can check the math by hand.

Worked example 1: an IQ score

IQ tests are scaled to a mean of 100 and a standard deviation of 15. Suppose someone scores 130. Then z = (130 − 100) ÷ 15 = 30 ÷ 15 = 2.0. A z-score of 2 means the score is two standard deviations above the mean, which places it at about the 97.7th percentile - higher than roughly 98 out of 100 people. The same calculation tells you that a score of 85 has z = (85 − 100) ÷ 15 = −1.0, sitting one standard deviation below the mean at about the 16th percentile.

Worked example 2: comparing two different tests

This is where z-scores really earn their keep. Imagine you scored 78 on a biology exam (class mean 70, SD 8) and 82 on a chemistry exam (class mean 75, SD 12). The raw scores suggest chemistry was your stronger subject - but standardize them and the picture flips. Biology: z = (78 − 70) ÷ 8 = 1.00. Chemistry: z = (82 − 75) ÷ 12 = 0.58. Relative to your classmates, you actually did better in biology. Z-scores let you compare apples to oranges by converting both to the same standardized ruler.

Worked example 3: going backward (find x from z)

Sometimes you start with the z-score. Say a university only admits applicants whose admission-test result is in the top 10%, which corresponds to a z-score of about 1.28. If the test has a mean of 500 and a standard deviation of 100, the minimum raw score needed is x = μ + z · σ = 500 + 1.28 × 100 = 628. Switch this tool to the Find x from z tab, enter z = 1.28 with that mean and SD, and you get the same answer instantly.

Z-score, percentile and the empirical rule

If the data is roughly bell-shaped (normally distributed), the empirical rule (also called the 68-95-99.7 rule) gives you a quick mental map of what a z-score means:

• About 68% of values fall between z = −1 and z = +1.
• About 95% fall between z = −2 and z = +2.
• About 99.7% fall between z = −3 and z = +3.

So a value with |z| greater than 2 is in the outer 5% of the distribution, and one beyond |z| = 3 is genuinely rare. The calculator converts your exact z into an exact percentile using the standard normal curve, but the empirical rule is a handy sanity check.

Reference table: common z-scores and percentiles

The table below shows where common z-scores land on a standard normal distribution. "Percentile" is the percentage of values below that z, and "Area in tail" is the percentage beyond it on one side.

Notice the symmetry: a z of +1 and a z of −1 are mirror images, so the area below +1 (84.1%) plus the area below −1 (15.9%) sums to 100%. The values 1.645 and 1.96 appear constantly in statistics because they mark the 5% and 2.5% one-tailed cutoffs used in hypothesis testing and confidence intervals.

Who uses z-scores

• Students and teachers standardizing exam scores or curving grades.
• Researchers identifying outliers and standardizing variables before analysis.
• Healthcare - growth charts report a child's height and weight as z-scores against age- and sex-based norms.
• Finance and quality control measuring how far a return or a measurement deviates from its expected value.
• Anyone comparing results measured on different scales, where raw numbers cannot be compared directly.

Key terms explained

• Mean (μ): the arithmetic average of all the values in the group.
• Standard deviation (σ): the typical distance of values from the mean - the "spread" of the data.
• Standard score: another name for a z-score - a value expressed in standard-deviation units.
• Percentile: the percentage of a distribution that falls below a given value.
• Standard normal distribution: a bell curve with mean 0 and standard deviation 1, the reference curve z-scores map onto.
• Outlier: a value far from the rest, often defined as one with |z| above 2 or 3.

Tips for getting it right

• Match the value to its group. The mean and standard deviation must describe the same population the value comes from - do not mix a math-test value with the spread of an English test.
• Mind the sign. A negative z is below average and a positive z is above; the size tells you how far.
• Use population values when you can. If you only have a sample, your z-scores are estimates - reasonable for large samples, shakier for small ones.
• Round at the end. Keep extra decimals during the calculation and round only the final z-score to avoid drift.

Related concepts

Z-scores connect to several neighboring ideas. A t-score rescales a z-score to a mean of 50 and standard deviation of 10 (T = 50 + 10z), common in psychology. Standardization in machine learning applies the same z formula to every feature so variables on different scales contribute equally. Confidence intervals and hypothesis tests use critical z-values like 1.96 to mark cutoffs. And the percentile you read here is the cumulative probability of the standard normal distribution - the foundation of much of inferential statistics.

Limitations and assumptions

This calculator does the arithmetic exactly, but a few assumptions are worth keeping in mind:

• The z-score itself is valid for any data, but the percentile assumes a normal distribution. For skewed data, treat the percentile as a rough guide only.
• You must supply the mean and standard deviation; the tool does not compute them from a raw data set.
• If you use a sample standard deviation rather than the true population value, the z-scores are approximate.
• A standard deviation of zero makes the z-score undefined (every value would be identical), so the calculator requires σ > 0.