Find a Z Score from a Mean and Standard Deviation

Identify the three quantities in the formula

A z score measures a value's signed distance from a reference mean in standard-deviation units. The formula is z=(x−μ)/σ, where x is the value, μ is the reference mean and σ is the positive reference standard deviation. All three inputs must describe the same variable on the same scale. If x is measured in minutes, the mean and standard deviation must also be in minutes. Dividing their common units makes the resulting z score dimensionless.

Calculate a value above the mean

Suppose x=82, the reference mean is 70 and the standard deviation is 6. First subtract:82−70=12. Then divide by 6 to obtain z=2. Enter those three values in the local form and check the output. The result means 82 is two standard deviations above 70. It does not mean 82 is twice the mean or that the value increased by 200 percent. The scale is the standard deviation, not the original value or a percentage denominator.

Calculate a value below the mean

Keep mean 70 and standard deviation 6, but change x to 58. The difference is −12, so z=−2. The negative sign indicates below the mean, and the magnitude indicates a two-standard-deviation separation. Values 82 and 58 are equally far from the mean in opposite directions. If the tool gives a positive score for 58 under this reference, the subtraction order or an entered sign is wrong. The formula subtracts the mean from the value, not the other way around.

Use the stated standard deviation rather than its square

If a problem supplies variance 36, take its positive square root 6 before using the z formula. Dividing the difference 12 by variance 36 would give 1/3 and use squared units in the denominator. Likewise, a standard error is not automatically interchangeable with an individual-observation standard deviation. A problem about a sample mean may use σ/√n as its reference spread, while a problem about one observation usesσ. Identify what random quantity is being standardized before choosing the denominator.

Choose a reference population or sample consistently

When a population mean and standard deviation are given, use those parameters as stated. If you estimate a reference from a sample, document the dataset and the sample-standard-deviation convention. Recomputing the reference after adding the new observation changes the question because both centre and spread can shift. A z score relative to a training dataset should use that same fitted reference for later values when comparability is intended, rather than standardizing every value against its own newly assembled group.

Separate standardization from a probability statement

The arithmetic z=(x−μ)/σ can describe distance for many kinds of numerical data; it does not make a non-normal distribution normal. A standard-normal table assigns probabilities only when the normal model is appropriate for the quantity under study. Under that model, z=2 has a cumulative probability near 0.97725, but a skewed or heavy-tailed distribution can have a different proportion below the same standardized distance. State the distributional assumption before translating a z score into a percentile.

Handle zero spread and inconsistent inputs

If every reference value is identical, its standard deviation is zero and ordinary z scores are undefined. There is no standard-deviation unit by which to divide. A negative standard deviation is also invalid. Do not replace zero with an arbitrary tiny number merely to obtain a result. Check whether the variable was recorded correctly, whether a constant group is expected, and whether another comparison is meaningful. A finite numeric answer is useful only when its inputs describe the intended reference.

Recover the original value as a check

Rearranging gives x=μ+zσ. With mean 70, standard deviation 6 and z=2, this returns 70+12=82; with z=−2 it returns 58. This reverse calculation is a quick check on sign and scale. Report the value, reference mean, standard deviation and resulting z together so another reader can reproduce the comparison. Round the final score as required while retaining enough original input precision to avoid changing a threshold comparison prematurely.

Related guides

Find a Z Score Step by Step with Signed Examples

Follow subtraction and division to calculate positive and negative z scores, enter the correct fields, and check answers without reversing the sign.

Find Standard Deviation with a Checked Worked Example

Calculate squared deviations, choose the sample or population denominator, take the square root, and interpret standard deviation in original units.

How To Determine Normal Distribution?

How To Determine Normal Distribution: calculation method, required inputs, worked check, limitations and a link to Normal Distribution Calculator.