gizmobench

Box and Whisker Plot Maker

Paste one column of numbers, or a label and its values per line for several series, and get a box and whisker plot that states its method on the chart. Quartiles are placed by the median of halves and outliers by the 1.5 IQR rule. The box runs from the first to the third quartile with the median drawn across it, whiskers reach the last values inside the fences, and outliers are drawn as dots. Download the plot as a PNG at three sizes or as an editable SVG, with nothing uploaded and no account.

Box plot1200×800 · document or slide

Box plot. No data yet.

Type
Size
Palette

Paste data on the left to draw a chart.

One row per line, the label first, then a comma, a tab or two spaces, then the number. A first line whose second field is not a number is read as a header and used to title the axes. Scatter plots take x, y on each line; box plots take one column of numbers, or a label followed by that series' values. The chart, the title, the size and the palette are remembered in this browser only.

Quartiles by median of halves; outliers beyond 1.5 IQR. The median splits the sorted values into two halves, leaving itself out when the count is odd; the first quartile is the median of the lower half, the third of the upper, so 1 to 9 gives 2.5 and 7.5. A value more than 1.5 times the interquartile range beyond either quartile is drawn as a dot, and the whiskers stop at the last value inside that fence. Other software may use a different quartile method, which is why this one is printed on the chart.
Accuracy. Every mark is drawn from the numbers you paste; nothing is smoothed or estimated. Box plots use the median-of-halves quartile method and the 1.5 IQR outlier rule, both stated on the chart.

Common questions

How are the quartiles calculated?
By the median-of-halves method. The values are sorted and the median found; when the count is odd the median itself is set aside, and the remaining values split into a lower and an upper half. The first quartile is the median of the lower half and the third quartile the median of the upper half, so 1 to 9 gives a median of 5, a first quartile of 2.5 and a third of 7.5, and 1 to 8 gives 2.5, 4.5 and 6.5. The method is printed at the foot of every plot.
Why do my quartiles differ from a spreadsheet's?
Because there are several accepted ways to place a quartile and they disagree on small samples. Excel's QUARTILE.INC and R's default interpolate between values and give 3 and 7 for 1 to 9, where the median-of-halves method gives 2.5 and 7.5; many textbooks and calculators use the halves method. Neither is wrong, which is why this plot says which one it used.
How are outliers decided?
By the 1.5 IQR rule. The interquartile range is the third quartile minus the first, and any value more than 1.5 times that range below the first quartile or above the third is an outlier, drawn as a hollow dot. The whiskers do not reach the outliers: each stops at the smallest or largest value that lies inside the fence. The text description under the chart lists the outliers by value.
What format does the data take?
For one series, one number per line; a first line of text, such as Scores, becomes its name. For several series, one per line: the label, a comma, then the values separated by spaces or commas, like Class A, 71 78 82 90 95. A value that is not a number stops the chart and the error names the line and the value. A series with a single value draws a flat box at that value.
How do I read the plot?
Each series is one box. The bottom of the box is the first quartile and the top is the third, so the box holds the middle half of the values; the line across it is the median. The whiskers show how far the rest of the data reaches, and any dot beyond a whisker is a value the 1.5 IQR rule set apart. A taller box means more spread in the middle half, and a median close to one end means the values bunch toward the other.

Every mark is drawn from the numbers you paste; nothing is smoothed or estimated. Box plots use the median-of-halves quartile method and the 1.5 IQR outlier rule, both stated on the chart.