Calculate Percentile Rank: Formula and Examples

A percentile rank tells you the percentage of scores in a group that are at or below a given score. Enter the score and the group’s scores to calculate its percentile rank, with ties included in the calculation.

Formula

This calculator uses the inclusive empirical percentile-rank formula: percentile rank = 100 × (number of scores at or below the target score ÷ total number of scores). In symbols, PR = 100B/N, where PR is percentile rank in percent, B is the count of observations less than or equal to the target score, and N is the total number of observations. The score list and target score must use the same scale and units.

The calculator counts scores at or below the target; it does not estimate a percentile value from a continuous distribution. If the target is absent from the list, it still calculates the proportion of listed scores that are no greater than that target.

Worked Example

Using the defaults, the scores are 60, 70, 80, 90, and 100, and the target score is 80.

1. Count scores at or below 80: 60, 70, and 80, so B = 3.
2. Count all scores: N = 5.
3. Calculate 100 × (3 ÷ 5) = 60%.

Thus, under this formula, a score of 80 has a percentile rank of 60. The result means 60% of the listed scores are 80 or lower; it does not mean the score itself is 60% of a maximum.

What Percentile Rank Means

Percentile rank describes relative standing within a particular group. A rank of 60 indicates that the target score is at least as high as 60% of the observations, according to the inclusive counting rule used here. The remaining 40% are above it.

It is not the same as a percentile score or cutoff. The 60th percentile is a score value intended to mark a position in an ordered dataset. A percentile rank takes a score value as its input and reports the proportion of observations at or below it. Terminology and calculation conventions can differ across tests and software, so use the convention specified by the relevant source when comparing published results.

How to Calculate It by Hand

For a small dataset, sort the scores if that makes them easier to inspect, then count the values at or below the score being ranked. Divide that count by the full number of observations and multiply by 100. Sorting is helpful but not required; the order of the list does not affect the result.

• Include every observation in the group in the total, including repeated scores.
• Count a score equal to the target as at or below it.
• Use a consistent definition of the group: changing who is included changes the rank.
• Keep the result as a percentage. The calculator rounds to a whole percent; the underlying fraction may be more precise.

For example, if 18 of 24 scores are at or below a target, the rank is 100 × 18/24 = 75%. If the target is lower than every score, the count is zero and the rank is 0%. If it is at least as high as every score, the rank is 100%.

Ties, Samples, and Interpretation

Ties matter because this calculator includes all scores equal to the target. Suppose the list is 4, 6, 6, 6, and 9, and the target is 6. Four scores are at or below 6, so the inclusive rank is 80%. A different convention may assign tied observations a midrank, count only scores strictly below the target, or apply a test-specific adjustment. Those methods can produce different answers, particularly when many scores are tied.

The result also depends on the dataset, not just the target. The same score can have different percentile ranks in two groups because their score distributions differ. A percentile rank therefore does not, by itself, describe how far apart scores are, how difficult a test was, or whether a difference is meaningful. It is a relative position, not a measure of the size of a score gap.

For a sample, the calculation describes the observations supplied. It does not automatically establish the rank the score would have in a larger population. To make population claims, the sampling and statistical method need to be considered separately.

Common Errors and Limits

• Using the wrong group: A rank is only interpretable alongside its reference group and the conditions under which scores were collected.
• Dropping duplicate scores: Each listed observation counts. Removing ties changes both the count and the total.
• Confusing rank with score: A 60th-percentile rank is not a score of 60 points, nor is it necessarily a 60% test grade.
• Ignoring the tie rule: Inclusive counting is clear and reproducible, but it may not match a published test’s convention.
• Rounding too early: Calculate the fraction first and round only the final percentage. Small groups produce larger jumps between possible ranks.

This simple empirical formula is suitable when you have the actual comparison scores and want the share at or below a target. It does not calculate a score cutoff for a requested percentile, interpolate between values, or convert a score to a normal-distribution percentile. Those tasks require a different method and, in some cases, additional assumptions.

Frequently Asked Questions

How do you calculate percentile rank?
Count the observations at or below the target score, divide by the total number of observations, and multiply by 100. State the group and tie convention so the result can be interpreted.

Is percentile rank the same as percentile?
No. A percentile rank is a percentage associated with a score’s position in a group. A percentile is commonly a score value at a specified position in an ordered distribution. The exact convention can vary by context.

How are ties handled in this calculator?
All scores equal to the target are included in the count. This is the inclusive method. Other sources may use a midrank or a different rule, so their results may not match.

Can a percentile rank be 0 or 100?
Yes. It is 0% if no scores are at or below the target, and 100% if every score is at or below it. These are possible outcomes under the formula used here.

What if my target score is not in the list?
The calculation still works: count the listed scores that are lower than the target, since none are equal, then divide by the list size. The result describes that target’s position relative to the supplied observations.

Related

• Grain Bin Size Calculator: Capacity and Dimensions
• Calculate Effect Size: Cohen’s d and Other Measures
• Confusion Matrix Accuracy: Formula and Example

list scores | Scores | - | 60, 70, 80, 90, 100
in x | Score to rank | - | 80 | 0
out rank | Percentile rank | % | 100 * count(scores) | 0