What Is a Z Score Calculator?

A Z Score Calculator is a statistics tool that converts a raw value into a standardized score. A z-score tells how many standard deviations a value is above or below the mean. If a score has \(z=1\), it is one standard deviation above the mean. If a score has \(z=-2\), it is two standard deviations below the mean. If \(z=0\), the value is exactly equal to the mean.

Z-scores are used in statistics, probability, psychology, education, standardized testing, business analytics, finance, quality control, data science, machine learning, SAT/ACT preparation, AP Statistics, IB Math, GCSE/IGCSE math, college statistics, research methods, and hypothesis testing. They are useful because they put values from different distributions onto a common scale. A raw score of 85 might be excellent on one test and average on another. A z-score tells where the value stands relative to its own distribution.

This calculator includes several z-score tools in one section. The basic z-score mode calculates \(z=(x-\mu)/\sigma\). The reverse z-score mode solves for raw value, mean, or standard deviation. The probability mode calculates areas under the standard normal curve. The inverse percentile mode finds the z-score that matches a percentile or tail probability. The dataset mode calculates z-scores for every value in a list. The compare mode lets you compare two scores from different distributions. The p-value mode calculates one-tailed or two-tailed p-values from a z statistic.

The calculator also gives left-tail probability, right-tail probability, percentile, two-tailed area, and interpretation. For probability results, it uses the standard normal distribution. This is appropriate when the underlying variable is approximately normal or when the test statistic follows a standard normal model. If a distribution is highly skewed, discrete, or has strong outliers, a z-score can still describe standardized distance, but normal probability interpretation may not be appropriate.

How to Use This Z Score Calculator

Use the Z-Score tab when you know a raw value, a mean, and a standard deviation. Enter \(x\), \(\mu\), and \(\sigma\), then calculate. The result tells how far the value is from the mean in standard deviation units. The result panel also shows the percentile, left-tail area, right-tail area, and two-tailed area.

Use the Reverse Z tab when you know a z-score and want to find the raw value. For example, if a test has mean 100 and standard deviation 15, a z-score of 1.5 corresponds to \(x=100+1.5(15)=122.5\). This mode can also rearrange the formula to solve for the mean or standard deviation.

Use the Probability tab for standard normal areas. You can calculate \(P(Z\le z)\), \(P(Z\ge z)\), \(P(z_1\le Z\le z_2)\), or the outside probability beyond two z-values. Use the Inverse Percentile tab when you know the probability or percentile and need the corresponding z-score. For example, the 95th percentile of the standard normal distribution is about \(z=1.645\).

Use the Dataset Z-Scores tab when you have a list of values and want to compute a z-score for each item. Choose sample or population standard deviation depending on your data. Use the Compare Scores tab when two scores come from different distributions. The higher z-score is relatively stronger compared with its own distribution. Use P-Value for hypothesis-test style z statistics.

Z-Score Formulas

The standard z-score formula is:

For sample data, the same idea is often written as:

Solving for the raw value gives:

Solving for the mean gives:

Solving for standard deviation gives:

The standard normal cumulative distribution is:

Right-tail probability is:

The probability between two z-scores is:

How to Interpret a Z-Score

A z-score is a standardized distance from the mean. The sign tells direction. A positive z-score means the value is above the mean. A negative z-score means the value is below the mean. The magnitude tells how far away the value is in standard deviation units. A z-score of 0.25 is close to the mean, while a z-score of 2.5 is far from the mean.

For approximately normal distributions, the empirical rule is useful. About 68% of values fall within one standard deviation of the mean, about 95% fall within two standard deviations, and about 99.7% fall within three standard deviations. Therefore, a value with \(|z|\ge2\) is often considered unusual, and a value with \(|z|\ge3\) is often considered very unusual. These are practical rules, not automatic proof of error.

A z-score does not say whether a value is good or bad. It only says where the value is relative to the distribution. In a test score context, a positive z-score may be good. In a medical measurement context, a high z-score could be concerning. Interpretation always depends on context.

Z-Scores, Percentiles, and Normal Probability

When a variable is normally distributed, z-scores connect directly to percentiles. The left-tail probability \(P(Z\le z)\) is the percentile as a decimal. For example, \(z=1\) corresponds to about the 84th percentile, meaning about 84% of values are below it. \(z=-1\) corresponds to about the 16th percentile.

The right-tail probability \(P(Z\ge z)\) tells how much of the distribution lies above the value. This is useful for exceedance questions, such as the probability that a score is above a cutoff. The area between two z-scores tells the probability that a value falls inside a range. The outside area tells the probability of falling below a low cutoff or above a high cutoff.

Percentile interpretation assumes the normal model. If the data are not normal, the z-score still measures standardized distance, but the normal percentile may not match the actual dataset percentile.

Reverse Z-Score and Inverse Percentile

Reverse z-score calculations are useful when you know the desired standardized position and want the raw value. The formula \(x=\mu+z\sigma\) converts from z-score back to the original scale. This is common in test scores, grading cutoffs, quality control limits, and normal distribution problems.

Inverse percentile calculations move from probability to z-score. If a problem asks for the cutoff that captures the top 5%, you need the z-score with 95% to the left, which is about 1.645. If the distribution has mean 100 and standard deviation 15, the raw cutoff is \(100+1.645(15)\approx124.7\).

Dataset Z-Scores

Dataset z-scores standardize every value in a list. First, calculate the mean. Next, calculate the standard deviation. Then apply \(z=(x-\bar{x})/s\) or \(z=(x-\mu)/\sigma\) to each value. Dataset z-scores are useful for detecting unusual values, comparing values within the same dataset, and preparing features for data analysis.

Choose sample standard deviation when your dataset is a sample from a larger population. Choose population standard deviation when the dataset includes the entire population of interest. The difference matters most for small datasets because sample standard deviation divides by \(n-1\), while population standard deviation divides by \(N\).

Comparing Scores from Different Distributions

Z-scores are powerful because they let you compare scores from different scales. Suppose one student scores 85 on a test with mean 75 and standard deviation 10, while another scores 110 on a test with mean 100 and standard deviation 15. The raw scores cannot be compared directly because the tests have different scales. The z-scores show relative position within each distribution.

If Score 1 has \(z=1.0\) and Score 2 has \(z=0.67\), Score 1 is relatively stronger even if its raw score is lower. This is why standardized scores are common in educational testing, psychology, and assessment.

Z Statistics and P-Values

In hypothesis testing, a z statistic measures how far a sample result is from the null hypothesis expectation in standard error units. A p-value measures how extreme the observed statistic is under the null model. For a right-tailed test, the p-value is \(P(Z\ge z)\). For a left-tailed test, it is \(P(Z\le z)\). For a two-tailed test, it is usually \(2P(Z\ge |z|)\).

A small p-value means the observed result would be unlikely under the null model. It does not prove a hypothesis is true. It also does not measure practical importance. Always interpret p-values with assumptions, study design, effect size, and context.

Z-Score Worked Examples

Example 1: Calculate a z-score. If \(x=85\), \(\mu=75\), and \(\sigma=10\), then:

This means 85 is one standard deviation above the mean.

Example 2: Convert z-score to raw value. If \(z=1.5\), \(\mu=75\), and \(\sigma=10\), then:

Example 3: Probability below a z-score. If \(z=1.96\), the left-tail area is about 0.975. This is why 1.96 is commonly used in 95% confidence interval work.

Example 4: Two-tailed p-value. If \(z=1.96\), the two-tailed p-value is approximately:

Common Z-Score Mistakes

The first common mistake is using the wrong standard deviation. Use the standard deviation that belongs to the same distribution as the raw value. The second mistake is reversing the subtraction. The formula is \(x-\mu\), not \(\mu-x\). Reversing the order changes the sign.

The third mistake is interpreting normal probabilities for a dataset that is not normal. A z-score can standardize any value, but normal probability tables assume a normal distribution. The fourth mistake is treating z-score magnitude as practical importance. A statistically unusual value may or may not be practically important.