A z score of -1 means a value is one standard deviation below the mean of its reference distribution. If the data are normally distributed, it corresponds to about the 16th percentile: roughly 16% of values fall below it and 84% fall above it.
What a z score tells you
A z score, also called a standard score, expresses how far a value is from the mean in units of standard deviation. Its sign shows direction: a positive score is above the mean, a negative score is below it, and zero is at the mean. Its size shows the distance.
The usual formula is z = (x − μ) / σ, where x is the value, μ is the population mean, and σ is the population standard deviation. A z score of -1 therefore means the difference between the value and the mean is negative one standard deviation. It does not, by itself, tell you the original measurement or whether the result is good, bad, or unusual.
How to interpret -1
Interpretation depends on the distribution and the comparison group. For a normally distributed variable, the standard normal curve puts approximately 15.9% of observations below z = -1. This is often rounded to the 16th percentile. The score is below average, but it is not especially far into the lower tail: about 84.1% of normally distributed observations are higher.
Percentile language can be confusing. “16th percentile” means a value is at or above about 16% of the distribution, not that it is 16% below the mean. A z score measures distance in standard deviations; a percentile measures relative rank. The conversion between the two relies on a distribution, commonly the normal distribution.
- z = 0: at the mean.
- z = -1: one standard deviation below the mean.
- z = +1: one standard deviation above the mean.
- z = -2: two standard deviations below the mean, which is near the 2.3rd percentile under a normal model.
A calculation example
Suppose test scores have a mean of 70 and a standard deviation of 10. A score of 60 has z = (60 − 70) / 10 = -1. The score is 10 points below the mean, and those 10 points equal one standard deviation in this group.
The same z score can describe a very different raw difference in another context. If a measure has a mean of 500 and a standard deviation of 50, a value of 450 also has z = -1. Standardization makes relative positions comparable, but it does not make the measurements themselves interchangeable.
To convert a z score back to a raw value, rearrange the formula: x = μ + zσ. For z = -1, this becomes the mean minus one standard deviation. For example, if μ = 70 and σ = 10, then x = 70 + (-1 × 10) = 60.
When the percentile interpretation applies
The arithmetic meaning of z = -1 is one standard deviation below the chosen mean whenever the standard deviation is positive. The familiar 16th-percentile interpretation, however, assumes a normal distribution or a suitable normal model. In a skewed distribution, the share of observations below a value one standard deviation under the mean may be quite different.
Even in a normal distribution, a calculated z score is only as meaningful as its reference data. The mean and standard deviation should come from the relevant population or comparison group, measured in a comparable way. A score standardized against one age group, class, laboratory reference, or time period may not have the same interpretation against another.
There is also a distinction between a population z score and a sample-based score. In practice, analysts often estimate the mean and standard deviation from a sample. The resulting standardized value is useful, but estimates can be imprecise when the sample is small or unrepresentative. Some statistical procedures use a t score rather than a z score to account for uncertainty in estimated variability.
Common mistakes and limits
- Calling it a raw score: A z score of -1 is not a score of -1 units. It is a standardized position; the original units depend on the mean and standard deviation.
- Assuming it means “bad”: Below average is not automatically undesirable. A low time to complete a task might be favorable, while a low measurement could be concerning in another setting.
- Treating the percentile as exact for every dataset: About the 16th percentile is a normal-distribution result, not a universal rule.
- Ignoring outliers or unusual shapes: The mean and standard deviation can be strongly affected by extreme values, which may make z scores less representative.
- Comparing incompatible groups: A standardized score does not correct for differences in how groups were selected, measured, or defined.
A z score also does not establish statistical significance or diagnose a health condition on its own. Those conclusions require context, an appropriate reference standard, and—in formal testing—an understanding of the method and uncertainty involved.
Frequently Asked Questions
Is a z score of -1 good or bad?
It is one standard deviation below the reference mean. Whether that is good, bad, or neutral depends on what is measured and what the comparison group represents.
What percentile is a z score of -1?
For a standard normal distribution, it is approximately the 15.9th percentile, usually rounded to the 16th. For a non-normal distribution, the percentile may differ.
Is a z score of -1 unusual?
Usually not under a normal model. About 16% of observations lie below it, so it is lower than average but not rare in the way a score several standard deviations below the mean would be.
How do I find the raw value for z = -1?
Use x = μ + zσ. With z = -1, subtract one standard deviation from the mean. You need the relevant mean and standard deviation to calculate a specific value.
Can a z score be negative?
Yes. A negative z score simply means the value is below the reference mean. It does not mean the original value itself is negative.
What to remember
A z score of -1 places a value one standard deviation below its reference mean. Under a normal distribution, that position is around the 16th percentile, but the percentile and practical meaning depend on the distribution, the comparison group, and what the measurement represents. Check those details before drawing conclusions from the number.
