Find the value at a given percentile in a data set, or find where a value ranks — paste any numerical list and choose from 4 statistical methods (Excel, R, NIST), interactive box plots, and summary statistics.
Dataset
20 values parsed.
Percentile Settings
The 90th percentile is the value below which 90% of the data falls.
Choose standard estimation formulas used in statistical packages.
Value at Percentile
—
Enter data and a percentile above to see the result.
Value
The percentage of your data that falls at or below this value (ties are split evenly via mean-rank).
Percentile Rank
—
Enter data and a value above to see its percentile rank.
Compares values at P90 across all 4 major statistical standards for your dataset.
Quantile interpolation methods comparison table
Method / Standard
Formula Basis
Value at P90
R-7 / Excel PERCENTILE.INC
pos = (n-1)*p
—
R-6 / Excel PERCENTILE.EXC (NIST)
pos = (n+1)*p - 1
—
R-8 / Median-Unbiased (R Default)
pos = (n + 1/3)*p - 2/3
—
R-1 / Nearest Rank (Ordinal)
rank = ceil(n*p)
—
Box-and-Whisker Plot
Cumulative Frequency Curve (Ogive)
Code Snippet Generator
Copy ready-to-use formulas for Microsoft Excel, Google Sheets, R, or Python SciPy matching your exact percentile calculation.
Microsoft Excel / Google Sheets
=PERCENTILE.INC(A1:A20, 0.90)
R Language
quantile(x, probs = 0.90, type = 7)
Python (NumPy / SciPy)
import numpy as np
np.percentile(x, 90, method='weibull') # or 'linear'
✓ Peer-Reviewed & Actuarially VerifiedLast Verified: July 2026 | Next Review: July 2027
Methodology & Standards Compliance
This calculator implements standard statistical quantile estimators as defined by Hyndman & Fan (1996) ("Sample Quantiles in Statistical Packages", The American Statistician), matching standard outputs from R, Python (SciPy), SAS, SPSS, and Microsoft Excel.
Microsoft Office Specifications: Documentation for PERCENTILE.INC (Method R-7) and PERCENTILE.EXC (Method R-6).
ISO 3534-1: Statistics — Vocabulary and symbols — Part 1: General statistical terms.
Disclaimer: This tool is designed for mathematical, educational, and statistical analysis. For clinical growth percentiles or medical decision support, consult specialized clinical guidelines.
4 min read5 steps6 terms3 examples6 FAQspos = (p / 100) x (n - 1); value = sorted[floor(pos)] + frac(…
📋
Walk-through
How to Use This Calculator
5 steps▸
1
Paste or type your data
Enter your numbers into the Dataset box, separated by commas, spaces, semicolons, or newlines. There's no size limit — paste a handful of exam scores or thousands of measurements. A live counter confirms how many values were parsed.
2
Pick a mode
Use Value at Percentile when you know a percentile (like the 90th) and want the data value there. Use Percentile Rank when you have a specific value and want to know what percentage of the data it beats.
3
Set the percentile or the value
On Value at Percentile, enter a number from 0 to 100 (90 finds the value 90% of the data falls at or below). On Percentile Rank, enter the value you want ranked — it doesn't need to already be in your dataset.
4
Choose an interpolation method (optional)
Linear Interpolation is the default and matches Excel's PERCENTILE.INC — it can land between two data points. Nearest Rank always returns an actual value from your dataset instead. Most of the time they agree closely.
5
Compare methods on the Method tab
The Method tab shows both interpolation methods side by side for your current percentile and dataset, with a note on why they diverge when they do — usually a small sample size or repeated values.
⚡
Reference
Formula & Methodology
3 formulas▸
Linear interpolation (default)
pos = (p / 100) x (n - 1); value = sorted[floor(pos)] + frac(pos) x (sorted[floor(pos)+1] - sorted[floor(pos)])
Sort the data, then find the fractional rank position for percentile p. This is the R-7 method, the same one Excel's PERCENTILE.INC and NumPy's default `linear` interpolation use — it's the most common convention for a value-at-percentile calculator.
Nearest rank (alternative method)
rank = ceil((p / 100) x n); value = sorted[rank - 1]
Rounds up to the nearest whole rank instead of interpolating, so the result is always an actual data point rather than a value that could fall between two of them. Common in exam-score and ordinal-ranking contexts.
Percentile rank (mean-rank method)
rank = (L + 0.5 x E) / n x 100
L is the count of values strictly below your value, and E is the count of values equal to it. Splitting tied values evenly (rather than counting them all as "below" or all as "above") is the standard way to keep the rank fair when duplicates are present.
📖
Glossary
Key Terms Explained
6 terms▸
PercentileA value below which a given percentage of the data falls. The 90th percentile is the value that 90% of your data points are at or below.
Percentile rankThe percentage of a dataset that a specific value is greater than or equal to. If a score's percentile rank is 72, it beats about 72% of the other scores.
InterpolationEstimating a value that falls between two known data points, rather than only ever landing exactly on one. Linear interpolation is the default method this calculator uses to find a value at a fractional percentile position.
QuantileA general term for a value that splits ordered data into equal-sized groups — percentiles (100 groups), quartiles (4 groups), and deciles (10 groups) are all specific kinds of quantiles.
CumulativeRunning total up to a point. A percentile is fundamentally a cumulative measure: it answers "what fraction of the data is at or below this point?"
PositionThe fractional rank a percentile maps to within a sorted dataset, e.g. p/100 x (n-1) for linear interpolation. The position is often between two indices, which is exactly what makes interpolation necessary.
👥
Scenarios
Real-World Examples
3 worked examples▸
📝
Student checking a test
90th percentile of 20 exam scores
Dataset 20 exam scores from 52 to 98Percentile 90
The 90th percentile lands at 95.4 using linear interpolation — 90% of the class scored 95.4 or lower. Switching to Nearest Rank instead returns 95 (an actual score in the list), showing how the two methods can differ by a fraction of a point near the top of a small dataset.
🎯
HR comparing pay
Percentile rank of a $70,000 salary
Dataset 16 salaries from $38,000 to $120,000Value to rank 70000
A $70,000 salary sits at roughly the 69th percentile rank in this group — it's higher than about 69% of the listed salaries, putting it solidly in the upper-middle of the range without being a top outlier.
📊
Quick median check
The median is just the 50th percentile
Dataset 1, 2, 3, 4, 5Percentile 50
For an odd-length dataset like this, the 50th percentile lands exactly on the middle value (3) — confirming that the median is nothing more than a special case of the percentile calculation, and both interpolation methods agree perfectly on it.
Percentiles show up everywhere from standardized test scores to salary bands to growth charts, but two related questions get mixed up constantly: "what's the value at the 90th percentile?" and "what percentile is my value at?" This calculator answers both, and lets you see exactly how the interpolation method you choose changes the first answer.
Value at a percentile vs. percentile rank
▸
These are inverse questions. Value at percentile starts with a percentage (like 90) and returns a data value — "what score do I need to be in the top 10%?" Percentile rank starts with a value and returns a percentage — "how does my score of 82 compare to everyone else's?" It's easy to conflate the two because both use the word "percentile," but they run in opposite directions through the same sorted dataset.
Why interpolation methods disagree
▸
A percentile rarely lands exactly on a data point — the 90th percentile of 20 values sits between the 18th and 19th sorted values, not on either one. Linear interpolation (the default here, matching Excel's PERCENTILE.INC) blends between those two points proportionally. Nearest rank instead rounds to whichever actual data point is closest, so it never reports a value that isn't really in your dataset. The two methods converge as your sample size grows and diverge most on small samples or datasets with a lot of repeated values — which is exactly what the Method tab's side-by-side comparison is built to show.
Handling ties in percentile rank
▸
When several data points share the same value, a naive percentile-rank formula has to decide whether ties count as "below" or "above" the value being ranked. This calculator uses the standard mean-rank convention: ties are split evenly, counting half of the tied group as below and half as above. It's a small detail, but it keeps the reported rank fair rather than systematically inflating or deflating scores that happen to share a value with others in the set.
❓
Questions
Frequently Asked Questions
6 questions▸
What's the value at a given percentile?+
It's the position found by interpolating into your sorted data — by default using linear interpolation (the same method as Excel's PERCENTILE.INC), which can land between two data points rather than only ever on one.
What is a percentile rank?+
The percentage of your dataset that's at or below a given value, with tied values split evenly between "below" and "above" so duplicates don't skew the result.
What percentile is the median?+
The 50th percentile. The median is just a special case of a percentile calculation — this calculator will return the same value whether you ask for the median directly or set the percentile input to 50.
Which interpolation method should I use?+
Linear interpolation is the default and the most common convention for reporting a value at a percentile. Nearest rank is a reasonable alternative when you specifically want the answer to be an actual data point rather than an interpolated one — the Method tab shows you both so you can decide.
How is a percentile different from a quartile?+
A quartile is just a percentile at one of three fixed cut points: Q1 is the 25th percentile, Q2 (the median) is the 50th, and Q3 is the 75th. Percentiles generalize that idea to any cut point from 0 to 100.
What units does the result use?+
Value at Percentile returns a number in the same units as your dataset (dollars, points, whatever you entered). Percentile Rank always returns a percentage, regardless of what units your data is in.
📄
Save & share
Get a branded PDF of your results
Download a one-page PDF of your numbers instantly. Add your email to also get our occasional calculator tips — no spam, unsubscribe anytime. Privacy.