To convert a sample z score to a percentile, find the cumulative probability to the left of that z score in the standard normal distribution. Multiply that probability by 100: for example, a z score of 1.00 corresponds to about the 84th percentile, while a z score of −1.00 is about the 16th percentile.
What a Sample Z Score Means
A z score expresses how far an observation is from a mean in standard deviation units. A z score of 0 is exactly at the mean; +1 is one standard deviation above it; and −2 is two standard deviations below it. “Sample” may refer to a score calculated from sample data, but the conversion to a percentile still requires a reference distribution—usually the standard normal distribution.
A percentile describes relative position, not the percentage of points earned. Being at the 84th percentile means approximately 84% of values in the reference distribution are at or below that score. It does not mean the person answered 84% of questions correctly.
How to Convert a Z Score
Use the cumulative standard normal distribution function, often written as Φ(z). The percentile is:
Percentile = Φ(z) × 100
For example, Φ(1.00) is approximately 0.8413, so z = 1.00 is at about the 84.13th percentile. A z table, calculator, spreadsheet, or statistics program can provide the cumulative probability. Check that the tool reports the area to the left; some tables instead show the area between the mean and z, or the upper-tail area.
- Identify the z score, including its sign.
- Find the cumulative probability to its left under the standard normal curve.
- Multiply by 100 and report the result as a percentile, with sensible rounding.
Common Z Score Percentiles
These are approximate standard normal values. Small differences in rounding are normal.
- z = −2.00: 2.28th percentile
- z = −1.00: 15.87th percentile
- z = −0.50: 30.85th percentile
- z = 0: 50th percentile
- z = 0.50: 69.15th percentile
- z = 1.00: 84.13th percentile
- z = 1.96: 97.5th percentile
- z = 2.00: 97.72nd percentile
- z = 3.00: 99.87th percentile
The curve is symmetric around zero: the percentile for z = −1 is approximately 100% minus the percentile for z = +1. This symmetry can help check a calculation.
Using a Calculator or Spreadsheet
In Excel or Google Sheets, use NORM.S.DIST(z, TRUE) to get the cumulative probability, then multiply by 100. For z = 1, the formula is =NORM.S.DIST(1,TRUE)*100, which returns about 84.13. The TRUE argument matters: it requests cumulative probability. Setting it to FALSE returns the density at that point, not a percentile.
Many online calculators provide the same result. Confirm whether the input expects a z score and whether the output is a left-tail percentile, right-tail probability, or central area. These quantities are related but are not interchangeable.
What “Sample” Changes—and What It Doesn’t
If a z score was computed from sample statistics, its numerical value can still be converted using Φ(z) when the standard normal model is appropriate. For instance, a common sample z score is calculated as (observation − sample mean) ÷ sample standard deviation. That calculation places the observation in standard-deviation units; it does not, by itself, prove that the sample data follow a normal distribution.
For a small sample, or data that are skewed, heavy-tailed, or contain unusual values, the normal-curve percentile may not match the observation’s actual rank among the data. If you want the percentile within the observed sample, sort the data and calculate the observation’s empirical rank using a stated convention. That result can differ from the normal-model percentile.
There is another distinction in inference: when estimating a population mean from a small sample with unknown population standard deviation, a t distribution is often used instead of the standard normal distribution. A t statistic should generally be converted using the relevant t distribution and degrees of freedom, not a standard normal z table.
Interpreting the Result Carefully
A percentile is always relative to a reference. The 84th percentile under a normal model is not automatically the 84th percentile in a particular school, workplace, or clinical group. The reference population, measurement method, and assumptions determine what the comparison means.
- Do not confuse tails: z = 1 has about 84.13% below it and 15.87% above it.
- Do not confuse percentile with percent: a percentile is a rank-like position, not a score out of 100.
- Do not overstate precision: “about the 84th percentile” is usually more appropriate than reporting many decimal places.
- Check the model: normal-based percentiles can be misleading when the data or reference distribution is not approximately normal.
Frequently Asked Questions
What percentile is a z score of 1.5?
A z score of 1.5 has a cumulative standard normal probability of about 0.9332, so it is approximately the 93.32nd percentile.
Is a z score of 0 the 50th percentile?
Yes, for the standard normal distribution, z = 0 is the median and the 50th percentile. It is also the mean because the normal curve is symmetric.
How do I convert a negative z score to a percentile?
Use the same left-tail cumulative method. For example, z = −1 is about the 15.87th percentile. A negative z score indicates a position below the mean, not a negative percentile.
Does the 90th percentile equal z = 1.28?
Approximately. The 90th percentile of the standard normal distribution corresponds to z ≈ 1.2816. The exact value depends on the precision used.
Can every sample z score be converted with a z table?
It can be looked up mathematically, but whether the resulting normal percentile accurately describes the data depends on the model. For a t statistic or a non-normal sample, use the appropriate distribution or empirical method instead.
