Z-scores

A z-score tells you how many standard deviations a value is above or below the mean, calculated as z = (x − μ)/σ. On the AP Stats exam, z-scores standardize values so you can find normal probabilities, compare scores from different distributions, and build test statistics.

Verified for the 2027 AP Statistics examLast updated June 2026

What are Z-scores?

A z-score (also called a standardized score) answers one question. How unusual is this value? You compute it with z = (x − μ)/σ, which means you subtract the mean and divide by the standard deviation. A z-score of +2 means the value sits two standard deviations above the mean. A z-score of −1.5 means one and a half standard deviations below. The sign gives direction, the size gives distance.

The real power is that z-scores are unit-free. A fish length in centimeters and a shampoo fill in liters can both be converted to z-scores and compared on the same scale. That's the whole trick behind the standard normal distribution. Once you standardize, every normal distribution becomes the same N(0, 1) curve, so one table (or one calculator function) handles every normal probability problem you'll ever see. Per AP Stats 5.2.A and 5.2.B, you use z-scores both directions, turning a value into an area (probability) and turning a given area back into a boundary value.

Why Z-scores matter in AP Statistics

Z-scores show up in almost every unit of AP Stats, which makes them one of the highest-leverage ideas in the course. In Unit 5, Topic 5.2 (AP Stats 5.2.A and 5.2.B), z-scores are how you calculate the probability that a normal random variable lands in an interval, and how you reverse-engineer a cutoff value from a percentage (like finding the heights that mark the most extreme 12% of a population). In Unit 2, the correlation coefficient is secretly built from z-scores. The formula in AP Stats 2.5.A averages the products of x and y z-scores, which is why r is unit-free and trapped between −1 and 1. In Units 6 and 7, every test statistic and critical value is a z-score in spirit. The formula (statistic − parameter)/standard error is just z = (x − μ)/σ wearing inference clothes. If you genuinely understand z-scores, half the formula sheet stops looking like memorization and starts looking like one idea repeated.

How Z-scores connect across the course

Normal Distribution (Unit 5)

Z-scores and the normal curve are partners. Standardizing converts any N(μ, σ) problem into a standard normal N(0, 1) problem, so the area under the curve between two z-scores gives you the probability for that interval (AP Stats 5.2.A).

Correlation Coefficient (Unit 2)

The formula for r in AP Stats 2.5.A multiplies each point's x z-score by its y z-score and averages the products. That's why correlation has no units and stays between −1 and 1. It is built entirely from standardized values.

Critical Value (Units 6-7)

Critical values like z* = 1.96 are just z-scores chosen to capture a specific central area, such as the middle 95%. Confidence intervals take a point estimate and add or subtract a critical value times a standard error, which is the z-score logic run in reverse.

Empirical Rule (Unit 1)

The 68-95-99.7 rule is just three pre-computed z-score facts. About 68% of values fall within z = ±1, 95% within z = ±2, and 99.7% within z = ±3. It's the fast mental-math version of a z-table.

Are Z-scores on the AP Statistics exam?

Z-scores get tested in two directions, and you need both. Direction one gives you a value and asks for a probability. The 2022 FRQ Q3 did exactly this with shampoo bottles filled at a mean of 0.60 liter and standard deviation of 0.04 liter; you standardize the boundary value, then find the normal area. Direction two gives you an area and asks for the value, like finding the cutoff heights for the most extreme 12% of women's heights when μ = 65 inches and σ = 3.5 inches. That requires invNorm (or a table read backwards) to get the z-score, then un-standardizing with x = μ + zσ. Multiple-choice questions love intervals with asymmetric z boundaries, like asking what percentage of data falls between μ − 2.33σ and μ + 1.28σ, which forces you to actually use z-scores instead of leaning on the Empirical Rule. On FRQs, show the standardization step and shade or describe the area. A bare calculator answer with no setup loses credit. In Unit 7, the same skill reappears with t in place of z when you use s instead of σ.

Z-scores vs t-statistic

Both measure how many standard-deviation-sized steps a statistic is from a parameter, and the formulas look nearly identical. The difference is what's in the denominator. A z-score uses the known population standard deviation σ, while a t-statistic uses the sample standard deviation s as a stand-in. Because s adds extra uncertainty, the t-distribution has fatter tails than the normal curve (AP Stats 7.2.A). Practical rule for inference on means: σ is almost never known in real life, so means problems use t, while proportions problems use z.

Key things to remember about Z-scores

  • A z-score is computed as z = (x − μ)/σ and tells you how many standard deviations a value sits above (positive) or below (negative) the mean.

  • Z-scores are unit-free, which lets you compare values from completely different distributions, like an SAT score versus an ACT score.

  • To find a normal probability, standardize the boundary value into a z-score, then find the corresponding area under the standard normal curve.

  • To find a value from a given area (like 'the top 10%'), work backwards with invNorm to get the z-score, then un-standardize using x = μ + zσ.

  • The correlation coefficient r is the average product of x and y z-scores, which is exactly why r has no units and is always between −1 and 1.

  • When the population standard deviation σ is unknown and you use s instead, the z-score becomes a t-statistic and you switch to the heavier-tailed t-distribution.

Frequently asked questions about Z-scores

What is a z-score in AP Stats?

A z-score measures how many standard deviations a value is from the mean, using z = (x − μ)/σ. A z-score of 1.5 means the value is 1.5 standard deviations above the mean; −1.5 means 1.5 below.

Can a z-score be negative or bigger than 3?

Yes to both. A negative z-score just means the value is below the mean, and z-scores have no maximum. Values beyond z = ±3 are rare in a normal distribution (only about 0.3% of data) but completely valid, and they often signal an unusual or outlier-worthy value.

What's the difference between a z-score and a percentile?

A z-score measures distance from the mean in standard deviations, while a percentile tells you the percent of data at or below a value. In a normal distribution they're linked. A z-score of 0 is the 50th percentile, and z = 1.28 is roughly the 90th. The percentile is the area to the left of the z-score.

When do I use a z-score versus a t-statistic?

Use z when the population standard deviation σ is known or when working with proportions. Use t when you estimate σ with the sample standard deviation s, which is the standard situation for inference about means (AP Stats 7.2). The t-distribution has fatter tails to account for that extra uncertainty.

Do z-scores only work for normal distributions?

No. You can calculate a z-score for any distribution since it only needs a mean and standard deviation. But translating a z-score into a probability using the standard normal table or normalcdf only works when the distribution is approximately normal (AP Stats 5.2.C).