Z Score of -0.8: Percentile and What It Means

Home » Blog » Z Score of -0.8: Percentile and What It Means

A z score of -0.8 is approximately at the 21.2nd percentile in a standard normal distribution. In other words, about 21% of values fall below it, while about 79% fall above it.

What the Percentile Means

A percentile describes a position in a distribution: the 21.2nd percentile is a value equal to or greater than roughly 21.2% of observations. It does not mean that the score is 21.2% of the average, or that it is 21.2 points below the mean. The percentile is a rank, not a measurement in the original units.

For a standard normal distribution, the mean is 0 and the standard deviation is 1. A z score of -0.8 therefore places the value eight-tenths of a standard deviation below the mean. The corresponding cumulative probability is approximately 0.2119, or 21.19%, usually rounded to the 21st percentile or 21.2nd percentile.

How the Calculation Works

The percentile for a z score is found using the cumulative distribution function of the standard normal distribution, often written as Φ(z). It gives the area under the normal curve to the left of the selected z score:

Percentile = Φ(z) × 100

For z = -0.8, Φ(-0.8) ≈ 0.2119. Multiplying by 100 gives approximately 21.19%. The area is a probability under the curve; interpreting it as a percentile assumes the reference distribution is normal, or that a normal model is appropriate for the data.

You can obtain the result in several ways:

  • Standard normal table: Find the row for -0.8 and the column for 0.00. Depending on the table, it may show 0.2119 or a related area that needs conversion.
  • Calculator or statistics software: Use the standard normal cumulative distribution function, not the probability density function.
  • Spreadsheet: In many spreadsheet programs, a standard normal cumulative function returns about 0.2119 for -0.8; multiply by 100 to express it as a percentage.

What a Negative Z Score Tells You

The sign indicates direction relative to the mean. A negative z score is below the mean, zero is at the mean, and a positive z score is above it. The size of the z score tells how many standard deviations away the value is. Thus, -0.8 is below average, but less than one standard deviation below average.

For example, suppose test scores have a mean of 70 and a standard deviation of 10, and scores are reasonably modeled by a normal distribution. A score of 62 has z = (62 − 70) ÷ 10 = -0.8. Under that model, it is around the 21st percentile. The same z score could describe a height, measurement, or other variable; the percentile remains the same as long as the standardization and normal model apply.

Percentile, Percentage, and Position

Percentile rank is easy to confuse with percentage correct or a percentage of the mean. These are different ideas. If a student’s score is at the 21st percentile, that does not say the student answered 21% of questions correctly. It says the score is higher than about 21% of the reference distribution and lower than about 79% of it, subject to the conventions used to calculate ranks.

Likewise, “21st percentile” does not mean the score is exactly the 21st observation in a group. It is a relative position. In a finite dataset, percentile ranks may be computed with different conventions, and ties can affect the result. A z score converted through the normal curve is a model-based percentile, while a percentile calculated directly from observed data is an empirical percentile. They may not match exactly.

How It Compares With Nearby Z Scores

These approximate values help show how a z score maps to cumulative percentile under the standard normal model:

  • z = -1.0: about the 15.9th percentile.
  • z = -0.8: about the 21.2nd percentile.
  • z = 0: the 50th percentile.
  • z = +0.8: about the 78.8th percentile.
  • z = +1.0: about the 84.1st percentile.

The normal curve is symmetric around zero. That is why the percentile for +0.8 is approximately 100% − 21.2% = 78.8%. The percentages are cumulative areas, so equal changes in z do not produce equal changes in percentile everywhere on the curve.

When the Answer May Not Apply

The value 21.2% is correct for the standard normal distribution. A z score can also be calculated from data that are not normally distributed. Standardizing a value—subtracting the mean and dividing by the standard deviation—does not by itself make the underlying distribution normal. If the data are strongly skewed, have unusual tails, or contain multiple clusters, the normal-curve percentile may be a poor estimate of the value’s actual rank.

For a percentile based on a particular sample or population, use the reference data when available. Also check what “z score” means in context: some reports use a transformed score or a specialized reference distribution. If a result matters for a clinical, educational, or employment decision, consult the report’s documentation rather than treating a normal approximation as a definitive ranking.

Frequently Asked Questions

Is a z score of -0.8 below average?
Yes. It is 0.8 standard deviations below the mean. The score’s original-unit difference from the mean depends on the standard deviation of the variable.

Is -0.8 the 21st or 22nd percentile?
Its normal-curve percentile is about 21.19%. It is commonly reported as the 21st percentile, or as 21.2% when a decimal is useful. Rounding conventions account for small differences in wording.

What percentage of values are above z = -0.8?
About 78.8% are above it in a standard normal distribution, because the area to the left is about 21.2%.

How do I find the z score for the 21st percentile?
Use the inverse standard normal function. The z score for the 21st percentile is approximately -0.81, close to -0.8; small differences depend on rounding.

Does a z score of -0.8 mean a poor result?
Not by itself. It only describes position relative to a mean and standard deviation. Whether that position is concerning or meaningful depends on the measure, the reference group, and the purpose of the assessment.

Related

Scroll to Top