Your data
Paste a table straight from Excel or Google Sheets, or drop a CSV file. Nothing is uploaded anywhere — the whole calculation runs inside this browser tab.
CSV, TSV or plain text. Files saved in windows-1251 are detected and re-read automatically, so Cyrillic headers do not turn into garbage.
Parsing options
Columns
Result
Save and share
The drawing library is downloaded only when you press the button, and only on this page.
This page has no server side at all. The table you paste is parsed by JavaScript inside your own browser, the chart is drawn on a canvas element on your machine, and the export file is assembled locally. Nothing is uploaded, stored or written to any log. You can disconnect from the network after the page has loaded and everything will still work — which is the simplest way to verify the claim yourself.
Facts and limits of this method
| Box | From Q1 to Q3 — the middle 50% of observations |
|---|---|
| Line inside | The median, P50 |
| Whiskers | Tukey convention: to the furthest value within 1.5 · IQR of the box |
| Points beyond whiskers | Individual outliers, drawn separately |
| Quantile definition | R type 7, matching the quartile and five-number tools on this site |
| Minimum rows | 4 before the quartiles mean anything |
When it misleads you
- A box plot cannot show two peaks. A distribution with a gap in the middle produces exactly the same box as a smooth one, which is why a histogram belongs next to it whenever the shape is in question.
- The whiskers are a convention, not a discovery. The 1.5 multiplier is Tukey’s rule of thumb; on perfectly normal data it flags about 0.7% of observations as outliers purely by construction, so a few points beyond the whiskers is normal rather than alarming.
- The box hides the sample size completely. A box drawn from 10 values looks identical in authority to one drawn from 10,000, so the count belongs in the caption.
- On heavily skewed data the median line sits far from the centre of the box, and readers routinely misread that asymmetry as an error in the drawing rather than as the finding it is.
How it is calculated
The five numbers come from the same quantile routine as the rest of the section — R type 7 interpolation — so a box plot and the five-number summary of the same column always agree exactly.
The whiskers do not extend a fixed distance. They reach to the most extreme observation that still lies within 1.5 interquartile ranges of the box, which means the whisker always ends on a real data point rather than at an abstract boundary.
Values beyond that limit are drawn individually as points. This is deliberate: an outlier is a specific row you can go and inspect, not a statistical category, and showing them one by one invites exactly that inspection.
The sample data contains one invoice of 8900 among amounts near 1200. It is there to show what the format does with a genuine outlier: the box stays readable, the whisker stops at the last ordinary value, and the extreme sits alone far above.
Questions and answers
Why do the whiskers stop where they do?
They stop at the last observation within 1.5 IQR of the box. That is the Tukey convention and it is the most widely used one, but it is a convention: other tools sometimes draw whiskers to the minimum and maximum, which produces a very different picture.
Are the points beyond the whiskers errors?
Not necessarily. They are values far from the middle half of the data, which is a reason to look at them, not a verdict. The outlier detection tool lists them with their z-scores so you can judge.
Box plot or histogram?
Box plot for position, spread and outliers, especially across several groups. Histogram for shape. A box plot will never reveal that your data has two peaks.
Why is the median not in the middle of my box?
Because your data is skewed, and that is the box plot doing its job. A median close to Q1 means the lower half of the middle is compressed and the upper half is stretched.
Is anything sent to a server?
No. Quartiles and drawing both happen locally in your browser.