Z-Score Calculator: Formula & How to Use It

This Z-score calculator works out how many standard deviations a given value sits from the mean of a dataset. You enter the value (x), the mean (μ), and the standard deviation (σ), and you get back the z-score plus the probability of a value being below it. The result tells you whether your data point is typical, unusually high, or unusually low in a normal distribution.

Formula

The z-score formula is:

z = (x − μ) / σ

• x – the raw value you are standardizing (same unit as the data).
• μ – the mean of the population or sample (same unit as x).
• σ – the standard deviation of the population (same unit as x).

The result z is dimensionless. A positive z means x is above the mean; negative means below. The probability below is computed from the standard normal cumulative distribution function, which assumes the data are normally distributed.

Worked Example

Using the defaults: x = 85, μ = 70, σ = 10.

1. Subtract the mean from the value: 85 − 70 = 15.
2. Divide by the standard deviation: 15 ÷ 10 = 1.5.
3. The z-score is 1.5.
4. For z = 1.5, the probability of a value below 85 is about 0.9332 (93.32%).

That matches the calculator output: z = 1.50, probability below = 0.9332.

When to Use a Z-Score

Z-scores are used whenever you need to compare a single measurement to a group. Common uses include:

• Standardized test scores (e.g., comparing an exam result to the class average).
• Quality control – checking if a product measurement is too far from the target.
• Anthropometric data – comparing a child’s height or weight to growth charts.
• Outlier detection – flagging values beyond z = ±2 or ±3.

Because z-scores are unitless, they allow comparisons across different scales. For example, a height of 180 cm in a group with mean 170 cm and σ 10 cm gives z = 1.0, while a test score of 85 in a class with mean 70 and σ 10 also gives z = 1.5 – both are equally “far� from their respective means.

How to Read the Result

The z-score itself tells you direction and distance:

• z = 0 – exactly at the mean.
• 0 < z < 1 – above the mean but within one standard deviation (common).
• z > 2 – unusually high (about 2.5% of data if normal).
• z < −2 – unusually low.

The probability below is the area under the normal curve to the left of your value. For z = 1.5, that is 0.9332 – meaning 93.32% of values are below 85. The probability above is simply 1 − that number (6.68%).

Common Mistakes and Misconceptions

• Using sample standard deviation instead of population. If you have a sample, use the sample SD (n−1 denominator) unless you know the population SD. The formula is the same, but the interpretation changes slightly.
• Assuming normality. The probability below is only valid if the data are normally distributed. If not, the z-score still measures distance, but the probability is not accurate.
• Unit mix-ups. x, μ, and σ must all be in the same unit. If your data is in centimeters, don’t mix in meters.
• Rounding too early. Round only at the end. For example, if σ = 7.3 and you round to 7, you get a different z.
• Confusing z-score with percentile. A z-score is not a percentile. The percentile is the probability below multiplied by 100. For z = 1.5, the percentile is 93.32, not 1.5.

Limitations of the Z-Score

The z-score is a simple transformation, but it has limits:

• It does not tell you if the distribution is normal – you must check that separately.
• It is sensitive to outliers in the data used to compute μ and σ. A single extreme value can skew the mean and inflate the SD, making z-scores less meaningful.
• For small samples, the normal distribution may not hold; the t-distribution might be more appropriate for probabilities.
• It only works for numeric data with a meaningful mean and standard deviation. It is not for categorical data.

Despite these, the z-score remains one of the most widely used standardization methods in statistics.

Frequently Asked Questions

What is a good z-score?

There is no universal “good� – it depends on context. In many fields, z-scores between −2 and 2 are considered normal. In quality control, a z-score beyond ±3 might trigger inspection. For standardized tests, a z-score of 0 is average, positive is above average, negative is below.

Can a z-score be negative?

Yes. A negative z-score simply means the value is below the mean. For example, if x = 60, μ = 70, σ = 10, then z = (60−70)/10 = −1.0. The magnitude still tells you how far, and the sign tells you the direction.

What is the difference between z-score and t-score?

A z-score uses the population standard deviation (σ) and is used when σ is known or the sample is large. A t-score uses the sample standard deviation (s) and is used when σ is unknown, typically with small samples. The t-distribution has heavier tails, so t-scores are slightly more conservative.

How do I find the probability from a z-score without a calculator?

You can use a standard normal table (z-table) found in most statistics textbooks. Look up the z-score in the table to get the cumulative probability. For example, z = 1.50 gives 0.9332. Many online tables also provide the area in the tail.

Does the z-score formula work for any distribution?

The formula (x−μ)/σ always gives a number, but the interpretation as a probability depends on normality. For non-normal distributions, the z-score still tells you how many standard deviations away, but you cannot use the standard normal table for probabilities. You would need the actual distribution.

What is the empirical rule?

The empirical rule (also called the 68-95-99.7 rule) states that for a normal distribution, about 68% of values lie within 1 standard deviation of the mean, 95% within 2, and 99.7% within 3. This is a quick way to interpret z-scores without a table.

Related

• Edexcel Grade Calculator: How to Predict Your Result
• Ed’s Heat Load Calculator: How It Works
• Edward Jones Life Calculator: How It Works

in x | Value (x) | - | 85 | 0
in m | Mean (μ) | - | 70 | 0
in s | Standard deviation (σ) | - | 10 | 0
out z | Z-score | - | (x - m) / s | 2
out p | Probability below (P(X ≤ x)) | - | 0.5 * (1 + erf(z / sqrt(2))) | 4