Ever scored 85 on one exam and 78 on another, and felt oddly prouder of the 78? That’s the z-score at work — the statistic that tells you how impressive a number really is relative to everyone else. A z-score measures how many standard deviations a value sits above or below the mean, which makes it the fairest way to compare scores from completely different distributions.
Once you can calculate z-scores, you can compare test results across different classes, spot outliers in a dataset, and read standard normal tables to find percentiles and probabilities. Here’s the full method, start to finish.
What a Z-Score Actually Tells You
The sign and size of a z-score carry all the meaning:
- Positive z-score: the value is above the mean. Bigger positive = further above average.
- Negative z-score: the value is below the mean. Keep the minus sign — it matters.
- Zero z-score: the value is exactly equal to the mean.
- A z-score near ±3: the value is extremely unusual — roughly 99.7% of data in a normal distribution falls within 3 standard deviations of the mean.
Think of a z-score as a universal ruler. It converts any normal distribution into the standard normal distribution — mean 0, standard deviation 1 — so scores from totally different tests become directly comparable.
The Z-Score Formula
The formula looks slightly different depending on whether you’re working with an entire population or just a sample, though the arithmetic is identical.
Population z-score
z = (x – μ)/(σ)
- x = the raw observation
- μ = the population mean
- σ = the population standard deviation
Sample z-score
z = (x – x̄)/(s)
- x̄ = the sample mean
- s = the sample standard deviation
You can also compute statistical spreads, sum of squared differences, and verify radical values directly on CalcSolver.
How to Calculate a Z-Score in 4 Steps
- Identify the three values: the raw score (x), the mean (μ or x̄), and the standard deviation (σ or s).
- Find the deviation: subtract the mean from the raw score, (x − μ). If the score is below the mean, this will be negative — keep it that way.
- Divide by the standard deviation: this rescales the deviation into “standard deviation units.”
- Round to two decimal places: z-tables are indexed to the hundredths place (e.g., 1.45 or −0.82), so two decimals is the standard.
Worked Example 1: Which Test Score Is Really Better?
A student scored 85 on a Biology exam and 78 on a History exam. Which performance was relatively stronger?
- Biology: mean (μ) = 75, standard deviation (σ) = 8
- History: mean (μ) = 66, standard deviation (σ) = 5
Step 1: Biology z-score
z = (85 – 75)/(8) = 10/8 = +1.25
The student scored 1.25 standard deviations above the Biology class mean.
Step 2: History z-score
z = (78 – 66)/(5) = 12/5 = +2.40
The student scored 2.40 standard deviations above the History class mean.
Verdict: the 78 in History (z = +2.40) is the far better relative performance, even though 85 > 78 as raw numbers. The z-score exposes what raw scores hide.
Worked Example 2: Spotting an Unusual Value
On a national exam, scores have a mean of 70 and a standard deviation of 10. A student scores 92. Is that unusual?
z = (92 – 70)/(10) = 22/10 = +2.20
A z-score of +2.20 means the student beat roughly 98.6% of test-takers (we’ll see how to read that from a table below). Anything beyond ±2 is generally considered unusual; beyond ±3, a genuine outlier.
Reading a Z-Table and Finding Percentiles
A standard normal table (z-table) converts a z-score into a cumulative probability — the area under the bell curve to the left of that score:
- Find the row and column: the row holds the first two digits (e.g., 1.2), the column holds the hundredths digit (e.g., 0.05).
- Read the intersection: for z = 1.25, the table value is approximately 0.8944 — the score sits at the 89.44th percentile.
- Area to the right (upper-tail p-value): subtract from 1:
P(Z > 1.25) = 1 – 0.8944 = 0.1056 (or 10.56\%)
3 Mistakes That Wreck Z-Score Problems
- Reversing the subtraction: always compute (x − μ), not (μ − x). Swapping the order flips the sign and puts your value on the wrong side of the distribution.
- Dropping a negative sign: a below-average value must keep its negative z-score. Losing the sign misplaces the point entirely.
- Confusing the z-score with the probability: the z-score (1.25) is a distance along the axis; the probability (0.8944) is the area under the curve. They answer different questions.
The most common exam trap is asking “which student did better?” with raw scores designed to mislead. Standardize first, compare second — the z-score is the only fair judge.