A z-score tells you how far a value sits from the average, measured in standard deviations. A z-score of zero means the value is exactly at the mean. A positive z-score means it’s above average, and a negative z-score means it’s below. The number itself tells you the distance: a z-score of +2 means the value is two standard deviations above the mean, while a z-score of -1.5 means it’s one and a half standard deviations below.
How Z-Scores Work
The idea behind a z-score is simple: raw numbers are hard to compare across different scales, so you convert them into a common language. That language is standard deviations from the mean. If you scored 76 on an exam where the class average was 70 and the standard deviation was 3, your z-score would be +2.0. You didn’t just do “a little better than average.” You performed two full standard deviations above the class, which puts you well ahead of most test-takers.
The formula is straightforward: subtract the mean from your value, then divide by the standard deviation. That’s it. The result strips away the original units (points, pounds, dollars) and replaces them with a universal scale centered on zero.
Once converted, z-scores always follow the same distribution. The mean is 0, and the standard deviation is 1. This makes it possible to compare values from completely different datasets. You could compare your performance on a math test with your performance on a reading test, even if one was scored out of 50 and the other out of 200, because both z-scores speak the same language.
The 68-95-99.7 Rule
When data follows a bell curve (normal distribution), z-scores map neatly onto predictable percentages. About 68% of all values fall between z-scores of -1 and +1. Roughly 95% fall between -2 and +2. And 99.7% fall between -3 and +3. This is sometimes called the empirical rule, and it gives you an instant sense of how unusual any z-score is.
A z-score of +1 isn’t especially remarkable. About a third of the population scores higher. But a z-score of +3 is extremely rare, sitting beyond where 99.7% of values land. If you see a z-score beyond +3 or -3 in a dataset, you’re looking at something highly unusual.
Converting Z-Scores to Percentiles
A z-score of zero corresponds to the 50th percentile, meaning you’re right at the midpoint. From there, every shift in z-score moves you up or down the percentile ladder, though not in equal steps. A z-score of +1 puts you around the 84th percentile. A z-score of -2 drops you to roughly the 2.3rd percentile, meaning only about 2 out of 100 values fall that low.
The conversion isn’t perfectly intuitive because the bell curve bunches most values near the center. Moving from a z-score of 0 to +1 covers a large chunk of the population (about 34 percentage points), while moving from +2 to +3 covers only a tiny sliver. Statisticians use a standard normal table or calculator to look up exact percentiles, but the rough landmarks above are enough for most practical purposes.
Z-Scores in Health and Medicine
Doctors use z-scores in several areas where comparing an individual to a reference group matters. Two of the most common are pediatric growth charts and bone density scans.
The World Health Organization uses z-score cutoffs to flag abnormal growth in children. A child whose weight-for-length falls below a z-score of -2 (the 2nd percentile) is classified as having low weight-for-length. A child above +2 (the 98th percentile) is classified as having high weight-for-length. The same threshold applies to height: a length-for-age below -2 indicates short stature. These cutoffs help pediatricians identify children who may need nutritional or medical attention.
For bone density, a z-score compares your scan results to the average for someone of your same age and sex. A z-score below -2.0 signals that you have significantly less bone mass than expected, and your doctor will likely investigate further to understand why. (You may also hear about a T-score on bone density reports. The T-score compares you to a healthy young adult rather than your own age group, and it’s used to diagnose osteoporosis in older adults. The z-score is more useful for identifying unusual bone loss relative to your peers.)
Z-Scores for Spotting Outliers
In data analysis, z-scores are a common first step for identifying outliers. The general practice is to flag any data point with a z-score beyond +3 or -3 as a potential outlier, since fewer than 0.3% of values in a normal distribution land that far from the mean.
This method has limits. For small datasets, the maximum possible z-score shrinks, so a true outlier might not reach the +/-3 threshold. The National Institute of Standards and Technology recommends a modified version of the z-score for more reliable outlier detection, using the median and median absolute deviation instead of the mean and standard deviation. With this modified approach, values beyond +/-3.5 are flagged as potential outliers.
Z-Scores in Hypothesis Testing
When researchers want to know whether a result is statistically significant, z-scores play a central role. The core question is: “Is this result far enough from what we’d expect by chance to be meaningful?”
The most widely used threshold is a z-score of +/-1.96, which corresponds to a 5% significance level in a two-tailed test. If the z-score from an experiment exceeds 1.96 in either direction, the result is considered statistically significant at the 95% confidence level. For stricter testing at the 1% level, the threshold rises to +/-2.58. In one-tailed tests (where researchers are only looking for an effect in one direction), the cutoffs are lower: +/-1.64 for the 5% level and +/-2.33 for 1%.
These thresholds are arbitrary conventions, not magic numbers, but they’re deeply embedded in scientific practice. When you read that a study’s findings were “significant at p < 0.05," it means the test statistic landed beyond that 1.96 boundary.
Why Z-Scores Are Useful
The real power of a z-score is that it answers a question raw numbers can’t: “How unusual is this?” A blood pressure reading of 135 doesn’t tell you much on its own. But knowing it sits 1.5 standard deviations above the mean for your age group immediately communicates where you stand relative to everyone else. Z-scores turn context-dependent numbers into something comparable and interpretable, whether you’re looking at test results, medical measurements, or financial data.