The five number summary calculator above reduces a data set to the five order statistics that describe its position and spread: minimum, first quartile, median, third quartile and maximum. It draws them as a box plot, so you see the shape of the distribution rather than reading five numbers off a line. Paste values in any format, pick a quartile method, and both the numbers and the picture update as you type.
Arb Digital publishes this alongside the rest of the statistics set because the five number summary is the fastest honest description of a data set there is. It survives outliers, it assumes nothing about the shape of the distribution, and it takes seconds to read. Before running any test on a set of numbers, it is worth knowing what those numbers look like.
What This Five Number Summary Calculator Does
It sorts your values, then reports the smallest, the largest, the middle, and the two quartiles that cut the rest into quarters. Those five numbers divide the sorted set into four groups of roughly equal size: the bottom quarter below Q1, the second up to the median, the third up to Q3, and the top quarter above it.
The box plot renders that structure directly. The box spans Q1 to Q3, so its width is the interquartile range and it holds the middle half of the data. The line inside it is the median, and the whiskers reach out to the minimum and the maximum. The drawing is scaled so the minimum sits at the left edge and the maximum at the right, which means the position of the box tells you at a glance whether the data leans low or high. The same summary is printed underneath in plain text, so the numbers stay readable and copyable with styling switched off.
The tool also shows the count of values it parsed. If you paste 40 values and it reports 38, something in the paste was not a number, and you have found the problem before it reaches your results.
How to Use It
- Paste or type your values. Commas, spaces, tabs, semicolons and line breaks all work in any combination, so a spreadsheet column needs no cleaning first.
- Check the parsed count under the headline result against how many values you meant to enter.
- Choose a quartile method. Exclusive Tukey is the default because it is the one taught with box plots. Type 7 is the Excel, R and NumPy alternative.
- Read the five numbers from the headline and the grid, and the shape from the box plot below them.
- Compare the methods with the dropdown. If the box changes noticeably, your set is small and the method is worth stating alongside your result.
The Formula: How the Five Numbers Are Found
Sort the data first. Each of the five is an order statistic, defined by position in the sorted list rather than by arithmetic on the values, and the minimum and maximum are simply the first and last entries.
The median is the middle value when the count is odd, and the average of the two middle values when it is even. The quartiles are where the methods diverge. Under the exclusive or Tukey rule, split the sorted data at the median into a lower and an upper half — leaving the median itself out when the count is odd — then take the median of each half. Those are the hinges, Q1 and Q3.
Under Type 7, the quartile comes from linear interpolation on a position index. For the p-th quantile of n values the index is h = (n − 1) × p, counting from zero, and the answer is the value at position floor(h) plus the fraction of h times the gap to the next value. This is what Excel PERCENTILE.INC returns, and what R and NumPy return by default.
Neither is wrong. They answer slightly different questions about what a quartile of a finite sample means, and the gap shrinks as the sample grows. The quartile calculator is the deep dive on how the methods differ and which number each one gives; this page takes one method at a time, adds the minimum and maximum, and draws the result. If you need a percentile other than the 25th, 50th or 75th, the percentile calculator handles any percentile you name under one consistent convention.
A Worked Example You Can Check Yourself
The calculator loads with eleven values, already in order: 1, 2, 5, 6, 7, 9, 12, 15, 18, 19, 27. Eleven is odd, so the median is the sixth value counting from either end. Count them: 1, 2, 5, 6, 7, then 9. The median is 9.
For the Tukey quartiles, remove that median and split what remains. The lower half is 1, 2, 5, 6, 7 — five values, so its median is the third, which is 5. That is Q1. The upper half is 12, 15, 18, 19, 27, and its median is likewise the third, 18. That is Q3. The minimum is 1 and the maximum is 27, so the full summary reads 1, 5, 9, 18, 27. The range is 27 − 1 = 26 and the interquartile range is 18 − 5 = 13, exactly half the range.
Now switch the dropdown to Type 7 on the same eleven numbers. The median stays at 9. For Q1 the index is h = (11 − 1) × 0.25 = 2.5, so the answer sits halfway between the third and fourth sorted values, 5 and 6, giving 5.5. For Q3, h = (11 − 1) × 0.75 = 7.5, halfway between the eighth and ninth values, 15 and 18, giving 16.5. The box narrows from a width of 13 to a width of 11 on identical data, and you can watch it happen in the plot. That is the whole argument for stating your method.
The NIST/SEMATECH e-Handbook chapter on exploratory data analysis sets out the box plot conventions formally, including the hinge definition used here.
What the Shape of the Box Tells You
A box plot is not decoration. Its geometry carries three separate signals about skew, and when they agree you can trust the reading.
Where the median sits inside the box. A median nearer Q1 means the middle half is squeezed below it and stretched above: right skew. Nearer Q3 means left skew. In the default data the median 9 sits 4 above Q1 and 9 below Q3, so the upper part of the box is more than twice the lower — right skew.
Whisker lengths. The whiskers say the same about the outer quarters, usually more loudly, because extremes move further than middles. Here the lower whisker runs 1 to 5, a span of 4; the upper runs 18 to 27, a span of 9. The long whisker is on the high side, again right skew.
The mean against the median. The confirmation from outside the plot. The eleven values sum to 121: 1 + 2 + 5 + 6 + 7 = 21, then 9 + 12 + 15 = 36 for 57, then 18 + 19 + 27 = 64 for 121. Divide by 11 and the mean is exactly 11, against a median of 9. The high values have pulled the mean above the median, which is what right skew does. Three signals, one conclusion. The mean median mode calculator gives that comparison in one step.
When the three signals disagree, that itself is information — usually a bimodal set, or a single extreme value doing all the work.
Why Five Numbers Instead of a Mean and a Standard Deviation
The mean and standard deviation describe a data set with two numbers, which is efficient when the distribution is roughly symmetric and has no extreme values. When either condition fails, both move in ways that misrepresent the data. A single value ten times larger than the rest shifts the mean and inflates the standard deviation, while the median and the quartiles barely notice: they depend on position in the sorted order, not on magnitude.
This is resistance. Changing the largest value in the default data from 27 to 270 leaves the summary at 1, 5, 9, 18 and moves only the maximum. The mean jumps from 11 to about 33 — above every value in the set except the one that caused it. For revenue figures, response times and other right-skewed measurements, the five number summary describes what happened and the mean describes something no observation resembles.
The trade is that quartiles discard information: they know the order of your values but not the distances between them. Look at both. The standard deviation calculator measures spread around the mean and is itself distorted by the extremes the box plot exposes, so the two views are complementary rather than competing.
The Whisker Question, and Where Outliers Go
This tool draws whiskers to the true minimum and maximum, which is the plain five number summary. Many published box plots instead cap the whiskers at 1.5 times the interquartile range beyond each quartile and plot anything past that as an individual point. Both are called box plots, which causes real confusion when comparing charts from different sources.
The distinction matters when you read someone else's plot. If the whiskers are capped, a long whisker means the data really does extend that far; if not, it may be one lone value. Under the 1.5 rule the fences for the default data sit at 5 − 1.5 × 13 = −14.5 and 18 + 1.5 × 13 = 37.5, and every value falls inside that range, so there are no outliers here and both styles would look identical. The interquartile range calculator turns the box width into that outlier test; this page presents the same quartiles as a picture rather than as a rule.
The Summary Is Not Unique, and That Is Fine
Because the quartiles are not uniquely defined, the five number summary is not either — only the minimum, median and maximum are identical under every convention. Nine quantile definitions are in common use across statistical software, and on small samples they differ visibly, as the default data shows. On a few hundred values the gaps usually fall below the precision anyone reports.
The honest practice is to name the method whenever the sample is small. Reporting "Q1 = 5.5 (Type 7)" costs three words and removes the ambiguity. The Penn State Eberly College of Science statistics courses work through the competing definitions in detail.
Arb Digital builds measurement setups where spread and outliers are visible, so decisions rest on the shape of the data rather than on a single headline figure.
Browse All Free Tools Talk To Arb DigitalCommon Mistakes to Avoid
- Forgetting to sort first — all five numbers are defined by position in the sorted list. Hand calculations go wrong here more than anywhere else.
- Assuming the box is centred on the mean — it is bounded by quartiles and split by the median. On skewed data the mean need not lie inside it.
- Comparing quartiles from different methods — a Tukey Q1 and an Excel Q1 are different statistics on small samples, and the gap looks like a real group difference.
- Reading a long whisker as a wide spread — on an uncapped plot one extreme value can stretch a whisker while the rest of that quarter sits near the box.
- Pasting data with stray characters — currency symbols and thousands separators are dropped rather than counted, so always check the parsed count.
Related Free Tools From Arb Digital
Dig into the definitions with the quartile calculator, turn the box width into an outlier test with the interquartile range calculator, and find any other percentile with the percentile calculator. Get the centre three ways with the mean median mode calculator, spread around the mean with the standard deviation calculator, and one value's position with the z-score calculator. The free online tools hub lists the full set.
Frequently Asked Questions
It is the minimum, first quartile, median, third quartile and maximum of a data set. Those five values split the sorted data into four groups of roughly equal size and describe both its centre and its spread without assuming any particular distribution shape.
The box runs from Q1 to Q3 and holds the middle half of the data. The line inside it is the median. The whiskers reach out to the smallest and largest values, and each whisker covers a quarter of the data.
The exclusive Tukey method is the one usually taught with box plots and is the default here. Type 7 matches Excel PERCENTILE.INC, R and NumPy. Pick whichever matches the software you are comparing against and state your choice.
Because there is no single agreed definition of a quartile for a finite sample. Around nine conventions are in common use. The differences are largest on small data sets and usually negligible once you have a few hundred values.
This plot draws whiskers to the true minimum and maximum, so no points are separated out. Many published box plots cap the whiskers at 1.5 times the interquartile range beyond each quartile and mark anything further as an outlier.
If the median line sits nearer Q1 the data is right-skewed, and if it sits nearer Q3 the data is left-skewed. A noticeably longer whisker on one side points the same way, and the mean sitting above or below the median confirms it.
Any mix of commas, spaces, tabs, semicolons and line breaks. A column pasted straight from a spreadsheet works. Non-numeric entries are ignored, so check the parsed count shown with the result before trusting the output.
This calculator sorts the values you enter and reports order statistics from them. Whether those values are the right sample to summarise is a question the tool cannot answer for you.
