Five Number Summary Calculator

Introduction

Long numeric lists rarely tell their story at a glance. A five-number summary gives the list a small set of ordered landmarks, so the reader can see the lowest value, the lower quartile, the median, the upper quartile, and the highest value without reading every observation.

The main advantage is resistance to extremes. A mean can move sharply when one salary, response time, sale price, or test score sits far away from the rest of the data. The median and quartiles still show where the middle of the sorted list sits, which makes a five-number summary useful for school statistics, operations checks, quality reviews, survey results, and quick comparisons where distribution shape matters.

Minimum and maximum

The two endpoints of the sorted dataset. They define the full observed range, but they may be unusual values rather than typical cases.

Median

The middle of the sorted dataset. Half the observations sit at or below it and half sit at or above it, allowing for ties.

Q1 and Q3

The first and third quartiles. They frame the middle half of the data and support the interquartile range, usually shortened to IQR.

Box plots are the visual form of the same idea. The box runs from Q1 to Q3, the median line sits inside the box, and whiskers show how far the remaining non-outlier values extend. If outliers are plotted separately, the whiskers may stop before the minimum or maximum, so the outlier rule matters when comparing charts from different sources.

Quartiles are not defined by one universal calculation rule. Some courses use Tukey hinges, many spreadsheets use inclusive percentiles, and some reports ask for exclusive percentiles. The same sorted list can therefore have slightly different Q1 and Q3 values while the median, minimum, and maximum stay the same.

A five-number summary does not prove why a dataset is spread out or skewed. It can point to a long upper tail, a tight middle, repeated values, or a possible data entry problem, but context still decides whether a value is an error, a rare but valid case, or part of the pattern being studied.

How to Use This Tool:

Use the calculator as an auditable statistics workspace. Paste the values, choose the quartile convention that matches your source, then review the ledger and box plot before exporting anything.

1. Paste the dataset into Number list. Commas, spaces, tabs, line breaks, negative signs, decimals, scientific notation, and thousands separators are accepted. Check the n= badge or the Count row to confirm that the intended values were found.
2. Choose Quartile method. Tukey hinges matches many hand-calculated box-plot lessons, Inclusive percentile matches spreadsheet-style inclusive quartiles, and Exclusive percentile is useful when a worksheet or report explicitly calls for an exclusive rule.
3. Set Outlier fence. The standard option uses 1.5 x IQR, the outer option uses 3 x IQR, and No outlier fence leaves outlier flags off while still showing the five-number summary.
4. Open Five-Number Ledger for the summary values, Fence Review for fences and whiskers, Sorted Values for rank-by-rank checking, and Box Plot Spread for the chart view.
5. Adjust Decimal places only after the method and fence rule are correct. Rounding changes the displayed tables, chart labels, CSV files, DOCX exports, and JSON output, not the parsed values used for the calculations.

Interpreting Results:

Read the median and IQR together. The median gives the resistant center, while the IQR shows the width of the middle half of the dataset. A small IQR with a large full range usually means the endpoints are doing most of the spreading; a large IQR means the typical middle values are spread out too.

The result can also show a shape cue. High-side stretch means the upper side of the middle spread is longer, Low-side stretch means the lower side is longer, Flat middle means Q1, median, and Q3 collapse together, and Small sample warns that the quartile picture has little data behind it.

For graded work, reconciliation, or repeated reporting, save the quartile method and fence rule with the result. A method mismatch can make a correct calculation look wrong even when every value in the sorted list matches.

Technical Details:

The calculation starts by extracting numeric tokens from the pasted text, converting thousands separators, sorting the values from low to high, and applying the selected quartile convention. Non-numeric labels are ignored, so the count is the first check against accidentally missing or including data.

Quartile convention controls only the way Q1 and Q3 are chosen. Minimum and maximum come directly from the sorted endpoints. The median is the middle sorted value for an odd count, or the average of the two middle sorted values for an even count.

Formula Core

The spread, fence, skew, and mean-based comparison values use these core relationships.

In the fence formulas, m is 1.5 for the standard fence and 3 for the outer fence. Fences are active only when a fence rule is selected, the dataset has at least four values, and the IQR is greater than zero. Values equal to a fence stay inside the fence; only values below the lower fence or above the upper fence are flagged.

Quartile Rules

Boundary Behavior

Limits, Privacy, and Accuracy Notes:

Five-number summaries are descriptive. They do not test normality, prove causation, estimate confidence intervals, or decide whether an outlier should be removed. A flagged value can be a data-entry mistake, a rare event, or the most important observation in the dataset.

Small samples need extra care because one value can change a quartile or fence sharply. Repeated values can also collapse the IQR to zero, which disables fence-based outlier checks because the lower and upper fences would not provide useful separation.

Pasted numbers are parsed and calculated in the browser after the page loads. The export buttons create CSV, DOCX, image, and JSON files from the current result, so private datasets still need normal care on shared computers and managed browsers.

Worked Examples:

Balanced class dataset

For 12, 18, 19, 21, 24, 26, 31, 34, 38, 42, 47, 55 with Tukey hinges, the five-number summary is minimum 12, Q1 20, median 28.5, Q3 40, and maximum 55. The IQR is 20, so the standard fence runs from -10 to 70 and no value is flagged.

High response-time value

For 8, 9, 10, 10, 11, 12, 13, 14, 15, 65, Tukey hinges return Q1 10, median 11.5, and Q3 14. The IQR is 4, so the standard upper fence is 20. The value 65 is flagged as a potential high outlier while still remaining the dataset maximum.

Quartile method mismatch

For 2, 4, 7, 9, 11, 30, Tukey hinges return Q1 4 and Q3 11. Inclusive percentile returns Q1 4.75 and Q3 10.5. Exclusive percentile returns Q1 3.5 and Q3 15.75. The dataset did not change; the quartile rule changed.

FAQ:

Glossary:

Five-number summary

A compact dataset summary made from minimum, Q1, median, Q3, and maximum.

Median

The center value of the sorted dataset, or the average of the two center values when the count is even.

Q1

The first quartile, used as the lower edge of the middle half of the data.

Q3

The third quartile, used as the upper edge of the middle half of the data.

IQR

Interquartile range, calculated as Q3 minus Q1.

Fence

An IQR-based cutoff for flagging values that sit far from the quartile box.

Whisker

The box-plot line from Q1 or Q3 to the nearest endpoint or nearest non-outlier endpoint.

References:

• What are outliers in the data?, NIST/SEMATECH e-Handbook of Statistical Methods.
• 2.4 Box Plots, OpenStax Statistics.
• QUARTILE.INC function, Microsoft Support.
• QUARTILE.EXC function, Microsoft Support.