Two students each score 85 on a test. One took an easy exam where the class averaged 90; the other took a brutal one where the class averaged 70. The same raw score means very different things. A z-score fixes this by measuring how many standard deviations a value sits above or below the mean, putting scores from different distributions onto one shared scale. This tool computes it from your value, the mean, and the standard deviation.
The formula
The z-score is the distance from the mean, expressed in standard-deviation units instead of raw units.
Formula: z = (x - mu) / sigma, where x is the value, mu is the mean, and sigma is the standard deviation
Subtracting the mean centers the value at zero, and dividing by the standard deviation rescales it. A z-score of 1.0 means one standard deviation above average; -2.0 means two below. Because the units cancel, you can compare a test score, a manufacturing measurement, and a stock return on equal footing.
How to use it
- Enter x, the individual value you are evaluating.
- Enter mu, the mean of the distribution it belongs to.
- Enter sigma, the standard deviation of that distribution.
- Read the z-score and its plain-language meaning.
If you are working with raw data rather than ready-made statistics, the permutation and combination calculator is not what you need, but pairing z-scores with probability tables is the classic next step, and the factor calculator can help simplify the fractions in hand calculations.
Worked example
A student scores 75 on an exam. The class mean is 60 and the standard deviation is 15. Apply the formula: z = (75 - 60) / 15. The numerator is 15, so z = 15 / 15 = 1.0. The student sits exactly one standard deviation above the mean. For a roughly normal distribution, that beats about 84 percent of the class, a strong result the raw score of 75 alone did not make obvious. Now flip it: a score of 45 gives z = (45 - 60) / 15 = -1.0, one standard deviation below the mean.
Where z-scores earn their keep
Teachers use them to compare sections that took different exams. Quality engineers flag any part beyond 3 standard deviations as a likely defect, since such values are rare in stable processes. In finance, z-scores standardize returns so assets with different volatilities can be compared. Researchers use them to spot outliers before those points distort an analysis, and the combination calculator often appears in the same toolkit when the question turns to sampling odds.
Common mistakes
- Using the sample standard deviation as sigma. The classic formula uses the population standard deviation. With sample data, the result is an estimate, and small samples make it wobbly.
- Reading a negative z-score as a failure. Negative just means below the mean. A z of -0.1 is an utterly ordinary value.
- Assuming normality to get percentiles. The z-score itself needs no distribution assumption, but converting it to a percentile with the 68-95-99.7 rule only works for roughly bell-shaped data.
An honest limitation
A z-score standardizes, it does not normalize. On heavily skewed data, like incomes or house prices, a z of 2.0 does not carry the same "rare event" meaning it does on bell-shaped data. Always glance at the shape of your distribution before drawing strong conclusions from the score.