Percentile Calculator
Enter a dataset and a value to find its exact percentile rank — or enter a percentile to find the corresponding value. Covers both rank-based and z-score methods with step-by-step results.
Enter your values above to see the results.
Tips & Notes
- ✓Percentile rank and percentage score are different. Scoring 80% on a test is a percentage score. Being in the 80th percentile means you scored higher than 80% of the group — these are unrelated numbers.
- ✓The 50th percentile always equals the median. The 25th and 75th percentiles are Q1 and Q3. These relationships connect percentile analysis directly to box plot construction.
- ✓When two methods give different results (rank-based vs. z-score), use rank-based for small samples (under 30) and z-score only when you have confirmed the data follows a normal distribution.
- ✓For clinical data (growth charts, blood pressure), published percentile tables are based on large reference populations — don't compute percentiles from your small sample and call them clinical norms.
- ✓Percentile rank is between 0 and 100 inclusive. A value equal to the minimum of the dataset is at the 0th percentile; equal to the maximum is at the 100th. Some methods define these edge cases differently.
Common Mistakes
- ✗Confusing percentile with percentage. An 80th percentile score means you scored higher than 80% of people — not that you answered 80% of questions correctly. These are completely unrelated values.
- ✗Using z-score percentile for non-normal data. The z-score method only works when data is normally distributed. For skewed data, use the rank-based method — sort the data and count positions directly.
- ✗Reporting "90th percentile" without specifying the reference population. A child at the 90th percentile for height in one country may be at the 60th percentile internationally. Percentiles are meaningless without the comparison group.
- ✗Treating percentile differences as equal intervals. The difference between the 10th and 20th percentile is not the same size as between the 80th and 90th percentile in terms of raw values.
- ✗Assuming the highest score is at the 100th percentile. Some calculation methods place the maximum at a percentile below 100 — the exact value depends on the interpolation method used.
Percentile Calculator Overview
A percentile tells you what percentage of values in a dataset fall at or below a given point. Scoring in the 85th percentile on a standardized test means you outperformed 85% of test-takers. A baby at the 10th percentile for weight weighs more than only 10% of babies the same age. Percentiles convert raw values into relative standing, making them one of the most practically useful statistics in education, medicine, and data analysis.
Percentile rank — what percentage of values fall at or below x:
Percentile Rank = (Number of values ≤ x / Total values) × 100
EX: Scores [60,70,75,80,85,90,95,100]. Score = 85 → 5 values ≤ 85 out of 8 → Percentile = (5/8)×100 = 62.5th percentileFinding the value at a given percentile (percentile point):
L = (P/100) × n | If L is integer: average values at positions L and L+1 | If not: round up to next position
EX: Find 75th percentile of [60,70,75,80,85,90,95,100] → L = (75/100)×8 = 6 → Average positions 6 and 7: (90+95)/2 = 92.5Z-score method — for normally distributed data:
z = (x − μ) / σ → Percentile = Φ(z) × 100
EX: IQ test with μ=100, σ=15. IQ score = 130 → z=(130−100)/15=2.0 → Φ(2.0)=0.9772 → 97.72nd percentileKey percentile reference points — standard benchmarks:
| Percentile | Z-score | Meaning | Common Use |
|---|---|---|---|
| 10th | −1.28 | Bottom 10% | Minimum passing threshold |
| 25th (Q1) | −0.67 | Lower quartile | Box plot lower bound |
| 50th (Q2) | 0.00 | Median | Typical value |
| 75th (Q3) | +0.67 | Upper quartile | Box plot upper bound |
| 90th | +1.28 | Top 10% | High performers |
| 95th | +1.64 | Top 5% | Clinical reference ranges |
| 99th | +2.33 | Top 1% | Exceptional performance threshold |
Frequently Asked Questions
Percentage measures how much of something you have (score/total × 100). Percentile measures your relative rank in a group. A student scoring 90% answered 90% of questions correctly. If they are at the 99th percentile, they scored higher than 99% of students. A student scoring 60% might be at the 50th percentile if the test was hard. The two numbers are completely independent.
Sort the data in ascending order, count how many values are less than or equal to your target value, then divide by total count and multiply by 100. Example: scores [55,62,70,78,85,90,95,98], finding rank of 85. Values ≤ 85: five values (55,62,70,78,85). Rank = (5/8)×100 = 62.5th percentile — 85 scores higher than 62.5% of the group.
The 25th percentile is Q1 (lower quartile) — 25% of values fall below it. The 50th percentile is the median — exactly half of values fall below it. The 75th percentile is Q3 (upper quartile) — 75% of values fall below it. These three points divide the dataset into four equal quarters and form the basis of box-and-whisker plots. The gap Q3−Q1 is the interquartile range (IQR).
Use the z-score method when your data is approximately normally distributed and you have enough data points to verify this. z = (x−μ)/σ, then look up Φ(z) for the percentile. Example: IQ = 115 with μ=100, σ=15: z=(115−100)/15=1.0, Φ(1.0)=0.8413, so IQ=115 is at the 84th percentile. For small samples or non-normal data, use the direct rank-based method instead.
You scored higher than 99% of the reference group — only 1% scored the same or higher. The 99th percentile does NOT mean you scored 99% on the test. Example: on the SAT, a score of 1580/1600 (98.75%) is at roughly the 99th percentile. A score of 1200/1600 (75%) could be at the 74th percentile. Percentile rank depends entirely on how the entire group performed.
Only if the reference populations are the same. A child at the 90th percentile for height on a national growth chart is being compared to the national population. That same child's percentile on a different country's growth chart would differ. Similarly, being in the 95th percentile on one standardized test does not mean the same as 95th percentile on another test — the comparison group differs.