Percentile Rank Calculator
Enter your dataset and a target value to find its percentile rank: the percentage of values in the dataset that fall at or below it. Choose from three standard formulas, see the step-by-step working, and use the reverse-lookup mode to find which value sits at a given percentile. The distribution curve updates live as you type.
Formula
Worked example
Dataset: 45 52 58 61 63 67 70 72 75 78 80 83 85 88 92 (N=15). Target value: 75. Sorted values at or below 75: 45 52 58 61 63 67 70 72 75, so L=9. Cumulative PR = (9/15) x 100 = 60.0%. Midpoint PR: F=1 (one value equals 75), so PR = ((9 - 0.5) / 15) x 100 = 56.7%. Exclusive PR: CFE=8 (below 75), CFG=6 (above 75), PR = (8/(8+6)) x 100 = 57.1%.
What is a percentile rank?
A percentile rank tells you what percentage of values in a dataset fall at or below a specific number. If your test score has a percentile rank of 75, it means 75% of all scores in the comparison group are at or below yours. Unlike a raw score, a percentile rank always depends on the reference group: the same score can have a very different percentile rank in one class versus another. Percentile ranks are widely used in education, standardised testing, clinical assessment, and anywhere you need to compare an individual measurement against a broader distribution.
The three formula methods explained
There is no single universally agreed formula for percentile rank; three methods are in common use. The cumulative method (L/N x 100) is the most taught in textbooks and counts every value at or below the target. It gives a rank of 100 for the maximum value, which some find counterintuitive. The midpoint method ((L - 0.5F)/N x 100) halves the contribution of tied values and is popular in educational testing because it avoids assigning 100 to a non-unique maximum. The exclusive method (CFE/(CFE+CFG) x 100) matches Excels PERCENTRANK.INC function and is preferred in data analysis contexts; it can return 0 for the minimum and 100 for the maximum when no ties are present, just like the PERCENTILE functions in most spreadsheet software. Choose the method that matches your field or reporting standard.
Reverse lookup: finding the value at a percentile
Sometimes you know the percentile rank and need the data value - for example, to find the cutoff score for the top 10% of a cohort. This calculator uses linear interpolation to find the exact value. It computes a fractional index position in the sorted dataset and interpolates between the two nearest values. This is the same algorithm used by Excels PERCENTILE.INC function and by most statistical software. The result is a precise estimate even when the target percentile does not correspond to an exact data point.
Percentile rank vs percentile vs z-score
These three statistics are related but distinct. A percentile rank is a percentage (0-100) describing position in a specific dataset. A percentile is the value itself at a given percentage point, for example the 75th percentile value in this dataset is 83. A z-score is a standardised distance from the mean measured in standard deviations; it assumes an underlying distribution (usually normal) and is not tied to any particular dataset. Percentile ranks make no distributional assumption, which is why they are preferred when you cannot verify normality or when the dataset is small.
Percentile rank interpretation guide
| Percentile rank | Interpretation | Performance band |
|---|---|---|
| 99 and above | Top 1% - exceptional | Advanced |
| 90 to 98 | Top 10% - well above average | Advanced |
| 75 to 89 | Top 25% - above average | Proficient |
| 50 to 74 | Above median - average to above | Proficient |
| 25 to 49 | Below median - below average | Basic |
| 10 to 24 | Bottom 25% - well below average | Below basic |
| 1 to 9 | Bottom 10% - significantly below | Below basic |
Standard interpretation of percentile ranks used in education and assessment contexts.
Frequently asked questions
What does a percentile rank of 80 mean?
A percentile rank of 80 means that 80% of the values in the comparison dataset are at or below your value (using the cumulative formula). Equivalently, your value is higher than 80% of the group and lower than the remaining 20%.
What is the difference between percentile rank and percentage score?
A percentage score measures how many questions you answered correctly as a fraction of the total questions, for example 85%. A percentile rank measures where that score falls relative to other people who took the same test, for example 92nd percentile. You can have a high percentage score but a low percentile rank if everyone else in the group also scored highly.
Why do the three formula methods give different results?
The three methods differ in how they handle the target value itself and any ties. The cumulative method counts all values at or below the target, so the maximum value in a dataset always gets a rank of 100%. The midpoint method counts ties at half-weight to avoid this. The exclusive method ignores the target value entirely, counting only values strictly below or above it, which matches spreadsheet software but can give 0% for the minimum value. For most practical purposes the differences are small, especially in large datasets.
Can I use this calculator for standardised test scores?
Yes, with one important caveat: the percentile rank you get will only be relative to the dataset you enter. If you enter your class scores, you get your rank within that class. For nationally normed tests like the SAT or ACT, the reference dataset is hundreds of thousands of students, and the percentile tables published by the test makers already account for that. For small classroom or internal datasets, this calculator is ideal.
How does the reverse lookup work?
Switch to reverse-lookup mode, enter a percentile rank between 0 and 100, and the calculator returns the interpolated value at that percentile. It sorts the dataset, computes a fractional position index as (percentile/100) x (N-1), then linearly interpolates between the two surrounding values. A rank of 0 returns the minimum and a rank of 100 returns the maximum.
What is the z-score shown in the results?
The z-score shows how many standard deviations the target value is from the dataset mean. A z-score of 1.5 means the value is 1.5 standard deviations above the mean. This is useful for comparing position across datasets with different scales, but it assumes the data are roughly normally distributed to be fully meaningful.
What happens if the target value is not in the dataset?
All three formulas work with values outside the dataset. The cumulative and midpoint methods count how many values are at or below the target, whether or not it appears in the data. The exclusive method counts values strictly below and strictly above. A value below the dataset minimum gets a rank of 0% (exclusive) or very low (cumulative), and a value above the maximum gets 100% (cumulative) or 100% (exclusive when CFG is 0).