A quantile is a value that divides a dataset at a specific point, so a known fraction of the data falls below it. If your test score is at the 0.3 quantile (or 30th percentile), that means 30% of scores fall below yours and 70% fall above. Quantiles are one of the most practical tools in statistics because they tell you exactly where a single value sits relative to everything else.
How Quantiles Work
Think of quantiles as cut points. You take all the values in a dataset, line them up from smallest to largest, and then slice at specific positions. The value at each slice point is a quantile. The simplest quantile you already know: the median. It’s the 0.5 quantile, meaning 50% of values fall below it and 50% fall above.
What makes quantiles useful is that they describe position, not magnitude. Saying a household earns $75,000 a year tells you one thing. Saying that income falls at the 0.60 quantile tells you something different: that 60% of households earn less. The raw number gives you a fact; the quantile gives you context.
Quantile vs. Percentile
These two terms describe the same idea with different scales. A quantile is expressed as a fraction between 0 and 1. A percentile is expressed as a number between 0 and 100. The 0.25 quantile is the same as the 25th percentile. Statisticians tend to use “quantile” because fractions work more cleanly in formulas, while doctors, teachers, and most everyday contexts use “percentile” because whole numbers feel more intuitive.
One common mistake: confusing a percentile rank with a percentage score. Scoring at the 90th percentile on a test does not mean you answered 90% of questions correctly. It means you scored higher than 90% of people who took the test. On a difficult exam, a 90th percentile result might correspond to getting only 60% of questions right. On an easy one, it might require 99%. The percentile tells you where you rank, not how many answers you got.
Common Types of Quantiles
Different fields slice data into different numbers of groups, and each has its own name:
- Quartiles divide data into 4 equal parts. The first quartile (Q1) is the 25th percentile, the second (Q2) is the median, and the third (Q3) is the 75th percentile.
- Quintiles divide data into 5 equal parts, each containing 20% of values.
- Deciles divide data into 10 equal parts, each containing 10%.
- Percentiles divide data into 100 equal parts, giving the finest commonly used resolution.
Which one gets used depends on the situation. Economists studying income inequality often group households into quintiles. Standardized tests report scores as percentiles. Medical charts for children use percentile curves. Financial analysts sometimes use deciles to rank stock performance.
How to Calculate a Quantile
For a small dataset, the process is straightforward. First, sort all the values from lowest to highest. Then find the position in that ordered list that corresponds to the fraction you want.
If you have a dataset with n values and want to find the position i for a given percentile P, the formula is: i = (P × (n − 1) / 100) + 1. So in a dataset of 21 values, the 25th percentile falls at position (25 × 20 / 100) + 1 = 6. You’d look at the 6th value in your sorted list.
When the formula points between two positions (say, position 6.5), you average the values at positions 6 and 7. This interpolation step is why different software tools sometimes give slightly different quantile results for the same data. There are actually multiple accepted interpolation methods, which is why you might see tiny discrepancies between Excel, R, and Python. For large datasets, these differences are negligible.
For quartiles specifically, there’s a shortcut that skips the formula. Find the median of your dataset: that’s Q2. Then find the median of all values below Q2: that’s Q1. Find the median of all values above Q2: that’s Q3. The minimum value is Q0, and the maximum is Q4.
Quantiles in Box Plots
The most common way you’ll see quantiles displayed visually is in a box plot (sometimes called a box-and-whisker plot). A box plot is built from five numbers: the minimum, the 25th percentile, the median, the 75th percentile, and the maximum.
The bottom edge of the box sits at Q1 (the 25th percentile) and the top edge at Q3 (the 75th percentile). A line through the middle marks the median. The height of the box represents the interquartile range, or IQR, which is simply the difference between Q3 and Q1. This middle box contains the central 50% of your data.
The “whiskers” extend from the box outward, typically reaching 1.5 times the IQR above and below the box. Any data points beyond the whiskers appear as individual dots and are treated as potential outliers. If the median line sits closer to the bottom of the box than the top, the data is skewed toward higher values. If it’s centered, the data is roughly symmetrical. A single box plot can tell you the spread, center, and shape of a dataset at a glance.
Quantiles in Medicine
If you’ve ever seen a pediatric growth chart, you’ve seen quantiles in action. The curved lines on those charts represent specific percentiles from a reference population. When a pediatrician plots your child’s weight and it lands on the 90th percentile for their age and sex, that means only 10 out of 100 children in the reference group weigh more.
No single percentile is automatically “good” or “bad.” What matters is consistency. In most children, height and weight measurements track along the same percentile channel over time. A child who has been near the 25th percentile since infancy is likely just smaller than average, not unhealthy. The red flag is unexpected crossing of percentile lines. A child who drops across two or more percentile lines may be experiencing growth failure, chronic illness, or nutritional problems. Similarly, a rapid upward crossing could signal excess weight gain.
Specific thresholds do trigger closer evaluation. Height-for-age below the 3rd percentile may indicate stunting. BMI-for-age between the 85th and 95th percentiles is classified as overweight in children, while above the 95th percentile is classified as obesity. Head circumference below the 3rd or above the 97th percentile in infants prompts screening for developmental or neurological concerns.
Quantiles in Public Health Research
Researchers studying health inequalities frequently divide populations into quintiles based on income, education, or neighborhood deprivation. A large study of 17 common health conditions in England, for example, assigned geographic areas to deprivation quintiles and then compared disease rates between the most deprived group (Q1) and the least deprived (Q5). Dividing the prevalence in Q1 by the prevalence in Q5 produces a rate ratio that quantifies inequality in a single number.
This approach works because quantiles create groups of equal size, making comparisons fair. You’re always comparing the bottom 20% to the top 20%, regardless of how income or deprivation scores are distributed. That consistency is why quantile-based grouping has become a standard tool for informing where public health resources should be directed.
Why Quantiles Handle Outliers Well
One practical advantage of quantiles over averages is their resistance to extreme values. If a company’s median salary is $65,000, that number won’t budge much if the CEO’s pay doubles. The median (the 0.5 quantile) depends only on position in the sorted data, not on how far away the extremes are. The same is true for any quantile. This makes quantile-based statistics especially useful for datasets with outliers, such as income distributions, housing prices, or hospital billing data, where a handful of extreme values can drag an average far from what’s typical.