Z Score Distribution Graph: Meaning, Uses, and Examples

Charts & pivot tables · Updated November 22, 2026

A z score distribution graph shows values on a standard normal curve, where the horizontal axis is measured in standard deviations from the mean. A z score of 0 marks the mean, positive scores lie above it, and negative scores lie below it; the area under the curve represents probability when the data follow a normal distribution.

What a z score distribution graph shows

The graph most people mean is the standard normal distribution: a symmetric, bell-shaped curve centered at z = 0. Its horizontal axis uses z scores rather than the original units of a measurement, such as centimeters, test points, or dollars. Its vertical axis shows probability density, not the probability of one exact score. The total area beneath the curve is 1, or 100% of the probability.

Because the curve is symmetric, equal distances to the left and right of zero have matching heights and areas. For example, the area below z = −1 matches the area above z = 1. The curve gets closer and closer to the horizontal axis in both directions, but in the mathematical model it does not suddenly stop at a minimum or maximum z score.

A graph may show the full curve, a histogram of observed z scores, or a curve with a region shaded to illustrate a probability. These are related but not identical. The standard normal curve is a theoretical model; a histogram displays actual data and can look uneven, skewed, or unlike a bell.

How z scores are calculated

A z score tells how far a value is from a mean in standard-deviation units. For a population with mean μ and standard deviation σ, the calculation is z = (x − μ) / σ, where x is the value being standardized. If a measurement equals the mean, its z score is 0; if it is one standard deviation above the mean, its z score is 1.

For example, suppose a test has a mean score of 70 and a standard deviation of 10. A score of 85 has z = (85 − 70) / 10 = 1.5. On the graph, place it 1.5 standard deviations to the right of the center. A score of 55 has z = (55 − 70) / 10 = −1.5 and sits the same distance to the left.

When working with a sample, a common version uses the sample mean and sample standard deviation: z = (x − x̄) / s. The precise convention depends on the question and the data. In inferential tests, a z statistic can be calculated from a statistic’s estimate and its standard error; that is not necessarily the same calculation as standardizing one raw observation.

How to read the axes and position

The center of a standard normal graph is z = 0. Tick marks are often labeled −3, −2, −1, 0, 1, 2, and 3, though a graph may use finer intervals. Each whole-number step represents one standard deviation. A score of 2 is not two raw measurement units above the mean; it is two standard deviations above it.

Useful landmarks help make a graph easier to interpret:

  • About 68% of values fall between z = −1 and z = 1 in a normal distribution.
  • About 95% fall between z = −2 and z = 2.
  • About 99.7% fall between z = −3 and z = 3.

These are the approximate 68–95–99.7 rule, also called the empirical rule. They describe intervals around the center, not the area to the left of each individual tick. For instance, the central 95% runs from roughly −1.96 to 1.96, while the interval from −2 to 2 is a convenient approximation.

Shaded areas and probability

On a continuous distribution graph, probability corresponds to area under the curve over an interval. A single exact value, such as z = 1, has zero area and therefore zero probability in the continuous model; meaningful questions ask for a range, such as the probability of a value being below 1 or between 0 and 1.

The cumulative standard normal probability to the left of z = 1 is about 0.8413. In other words, approximately 84.13% of a normal distribution lies below that point. The area between z = 0 and z = 1 is about 0.3413, because the half-curve to the left of zero contains 0.5 and 0.8413 − 0.5 = 0.3413.

For a right-tail question, subtract the left-tail cumulative probability from 1. Thus, the area above z = 1 is about 1 − 0.8413 = 0.1587, or 15.87%. For an interval, subtract the cumulative probability at the lower endpoint from the cumulative probability at the upper endpoint. The graph’s shading should match the wording: below means shade left, above means shade right, and between means shade the region bounded by both scores.

A worked example: interpreting a test score

Suppose exam scores are approximately normal with a mean of 500 and a standard deviation of 100. A score of 650 has z = (650 − 500) / 100 = 1.5. On a standard normal graph, it is 1.5 standard deviations above the mean. The cumulative probability for z = 1.5 is about 0.9332, so under the model about 93.32% of scores are below 650 and about 6.68% are above it.

This does not mean that the score is 93.32% correct, nor does it say that the student answered 93.32% of questions correctly. It describes the score’s position relative to the assumed distribution. The interpretation is only as good as the model: if scores are strongly skewed or the stated mean and standard deviation do not represent the relevant group, the probability estimate may mislead.

Using a z table or calculator

A standard normal table usually reports cumulative area to the left of a z score, though some tables instead show the area between zero and z. Check the table’s heading or key before using it. In a cumulative table, find the row for the first digits and tenths place and the column for the hundredths place. For z = 1.23, for example, use row 1.2 and column 0.03.

For negative scores, use the negative rows if the table includes them. Otherwise, symmetry can help: the area to the left of −z equals the area to the right of z. Always identify whether the question asks for a left tail, right tail, or interval before looking up a value; otherwise, a correct table entry can still answer the wrong probability question.

Calculators and statistical software can return cumulative probabilities or draw a curve with a shaded region. When checking a result, confirm the direction and units, and whether the output is a probability, percentage, or z score. A probability of 0.05 is 5%, not 0.05%.

When the graph is useful—and its limits

Z scores are useful for comparing positions measured on different scales. A height, an exam result, and a laboratory measurement cannot be compared directly in their original units, but their z scores indicate how far each observation is from its own group mean in standard deviations. Z scores also appear in probability calculations and many statistical procedures.

However, converting observations to z scores does not make their distribution normal. Standardizing shifts the center to zero and changes the scale to standard-deviation units; it does not remove skew, multiple peaks, outliers, or other features of the original data. A standard normal curve is appropriate for probability interpretation only when the underlying variable or relevant statistic has a normal distribution, or when a justified approximation applies.

For an individual observation from a small, skewed sample, a z score can still describe relative distance from the sample mean, but a normal-curve percentile may not be trustworthy. Check the data’s shape, the sample size, and the assumptions of the analysis. Also consider whether an outlier has inflated the standard deviation, since that can make other observations appear less unusual in z-score units.

Z score graph compared with a t distribution

A standard normal curve is often used when the population standard deviation is known or when a statistic has a normal distribution under the relevant assumptions. When estimating a population mean from a sample with an unknown population standard deviation, many procedures instead use the t distribution. The t curve is also centered at zero, but it has heavier tails, especially with fewer degrees of freedom.

As degrees of freedom increase, the t distribution approaches the standard normal distribution. This does not mean every graph centered at zero is a z graph: check whether the horizontal axis is labeled z or t and identify the method that produced the statistic. Using normal critical values where a t procedure is required can affect conclusions, particularly with small samples.

Common interpretation mistakes

  • Confusing z with a percentile: z = 1 is one standard deviation above the mean; it is not the 1st percentile. Under the standard normal model, its percentile is about the 84th.
  • Assuming every z score implies normal data: a z score can be calculated for non-normal data, but normal-curve probabilities may not fit.
  • Reading density as probability: the curve’s height at a point is not the chance of observing that exact point. Probability is an area across an interval.
  • Mixing up tails: the probability below a positive z score is usually much larger than the probability above it. Sketch the shaded region before calculating.
  • Ignoring which standard deviation was used: a population standard deviation, sample standard deviation, or standard error serves a different role. Follow the definition given in the problem.

Frequently Asked Questions

What does a z score of 0 mean?
A z score of 0 means the value equals the mean used in the calculation. It sits at the center of the standard normal graph.

What does a negative z score mean?
It means the value is below the reference mean. For example, z = −2 indicates a value two standard deviations below the mean, using the standard deviation in that calculation.

Is a z score of 2 unusual?
Under a standard normal model, z = 2 is above about 97.7% of values, and about 2.3% lie above it. Whether that is considered unusual depends on the context and the decision threshold.

How do I find the area under a z score graph?
Use a standard normal table, calculator, or statistical software, first deciding whether you need the area to the left, to the right, or between two z scores. A table’s convention matters, so check its heading.

Can a z score be greater than 3 or less than −3?
Yes. There is no fixed minimum or maximum z score. Such values are uncommon under a normal model, but they can occur, and a z score alone does not prove that an observation is an error.

Related

Try it: the Percentage Formula Generator: percent of, percent change, increase and decrease, with the formula.

More in Charts & pivot tables

Sources: Function behaviour and menu paths: Microsoft Support documentation for Excel. Method & sources

Spotted an error in this guide? Let us know.