How to calculate a z-score
Subtract the mean from the value, then divide by the standard deviation. An exam with mean 70 and SD 10: a score of 85 has z = (85 − 70) ÷ 10 = 1.5. The area to the left of z = 1.5 is 0.9332, so 85 is about the 93rd percentile, assuming scores are roughly normal.
Positive z is above the mean, negative is below, 0 is exactly average. |z| > 2 happens about 5% of the time for normal data; |z| > 3 about 0.3%.
Comparing across different tests
z-scores put different scales on a common footing. An 85 on a test with mean 70, SD 10 (z = 1.5) is more impressive than a 90 on a test with mean 82, SD 8 (z = 1.0). To look up areas by hand, use the z-table.
Frequently asked questions
What is a good z-score?
It depends on context. For test scores, anything above 0 is above average; z = 1 is about the 84th percentile and z = 2 about the 98th.
Can I use z-scores if the data are not normal?
You can always compute z as a standardised distance, but turning z into a percentile or probability assumes a normal distribution.
How do I convert a z-score back to a value?
x = μ + z·σ. The normal distribution calculator’s inverse mode does this from a percentile.