A z score is a standard score: it tells you how many standard deviations a value lies above or below the mean. If you want to convert a z score to a standard score on a different scale, use the scale’s mean and standard deviation: standard score = (z score × new standard deviation) + new mean.
What “standard score” means
“Standard score” can mean either a score expressed in standard-deviation units or a score transformed to a particular reporting scale. A z score is the first kind. It has a mean of 0 and a standard deviation of 1 in the reference distribution. A transformed score may instead use a familiar scale such as a mean of 100 and a standard deviation of 15.
That terminology causes confusion: there is no single conversion from a z score to a standard score unless you know which target scale is intended. If someone simply asks for a standard score, clarify the scale or check the test manual. Different tests may use different means and standard deviations.
How to convert a z score
To express a z score on a new scale, multiply it by the target scale’s standard deviation, then add the target scale’s mean:
Target score = (z × target standard deviation) + target mean
For example, suppose a person has a z score of 1.2 and a test reports standard scores with a mean of 100 and standard deviation of 15. The calculation is (1.2 × 15) + 100 = 118. The score is 1.2 standard deviations above the mean on that reporting scale.
For a scale with a mean of 50 and standard deviation of 10, the same z score becomes (1.2 × 10) + 50 = 62. The underlying relative position is unchanged; only the units and labels differ.
- Identify the z score.
- Find the target scale’s mean and standard deviation from the relevant test or instructions.
- Multiply the z score by the target standard deviation.
- Add the target mean, keeping the sign of the z score.
How to calculate a z score from raw data
If you have a raw value rather than a z score, first compare it with the appropriate mean and standard deviation. For a population, the formula is z = (x − μ) / σ, where x is the value, μ is the population mean, and σ is the population standard deviation. For a sample, calculations often use the sample mean and sample standard deviation, written x̄ and s.
For instance, if a value is 70, the mean is 60, and the standard deviation is 5, then z = (70 − 60) / 5 = 2. On a scale with mean 100 and standard deviation 15, that becomes 2 × 15 + 100 = 130.
Use statistics that match the data and reference group. A score standardized against one population may not be comparable to a score standardized against another. In formal testing, use the test’s published conversion rules rather than assuming a generic mean and standard deviation.
Examples of common standard-score scales
Many familiar scales are linear transformations of z scores. Their conventions are common, not universal; always verify the specific measure’s documentation.
- T scores: often have a mean of 50 and standard deviation of 10. A z score of −1.5 converts to 50 + (−1.5 × 10) = 35.
- IQ-style scores: many use a mean of 100 and standard deviation of 15. A z score of −1 converts to 85 on that scale.
- Some educational or test scales: may use other means and standard deviations, such as 500 and 100. The appropriate conversion depends on the particular test.
These examples show why the phrase “standard score” needs context. A value of 85 could be one standard deviation below the mean on a 100/15 scale, while the same number has a different interpretation on another scale.
Interpreting converted scores and percentiles
A linear conversion preserves relative position: a positive z score remains above the target mean, a negative one remains below, and a zero z score maps to the target mean. It does not change a person’s rank within the same reference distribution. If the original distribution is normal, a z score can also be related to a percentile; for example, z = 0 is the 50th percentile, while z = 1 is approximately the 84th percentile.
However, a z score alone does not prove that the underlying data are normally distributed. Percentile interpretations based on the normal curve are approximate unless that model is appropriate. In a skewed or otherwise unusual distribution, the same z score need not correspond to the expected normal percentile.
Also, “one standard deviation above average” describes distance from the mean, not a percentage increase. A z score of 1 is not “1% higher.” The size of a raw-score difference represented by one standard deviation depends on the original measurement and its variability.
Common conversion mistakes
- Using the wrong target scale: a conversion requires both the intended mean and standard deviation.
- Reversing the formula: convert to a target scale by multiplying z by the target standard deviation, then adding the target mean.
- Dropping a negative sign: z = −1 on a 100/15 scale gives 85, not 115.
- Confusing raw scores with standard scores: a raw score cannot be converted correctly without the relevant reference mean and standard deviation.
- Assuming all tests use the same conventions: reporting scales and norm groups vary by assessment.
Rounding can also create small differences. Keep full precision through the calculation and round only the final result, following the test’s reporting guidance.
Frequently Asked Questions
Is a z score the same as a standard score?
Yes, in the broad sense: a z score is a type of standard score, expressed with a mean of 0 and standard deviation of 1. Some fields use “standard score” to mean a transformed score on another scale, so context matters.
What is the formula to convert a z score to a standard score?
Use target score = (z × target standard deviation) + target mean. You need the mean and standard deviation of the scale you are converting to.
What is a z score of 1 on a scale with mean 100 and standard deviation 15?
It is 115: (1 × 15) + 100. That is one standard deviation above the scale’s mean.
Can a z score be negative?
Yes. A negative z score means the value is below the reference mean. For example, z = −2 is two standard deviations below that mean.
Does converting a z score change its percentile?
No, a linear conversion to another scale preserves the score’s relative position and percentile within the same reference distribution. The percentile estimate itself depends on the distribution and may rely on a normality assumption.
