Box Plot Comparison Builder — Free Canvas Box Plots

This tool draws box plots for comparing groups. A box plot summarizes a distribution in five numbers: minimum, first quartile, median, third quartile, and maximum. The box spans the middle half of the data (first to third quartile), the line inside is the median, and the whiskers reach the extremes. Placed side by side, box plots let you compare centres, spreads, and skew across groups at a glance — far more honestly than comparing means alone, because means hide skew and outliers that boxes show. You paste groups and the tool draws the plots plus the five-number table.

How to use it

Write one group per line as Name: value1, value2, value3.
Click Draw box plots.
Compare box positions (centres) and box widths (spreads).
Read the five-number table for exact values.
Note any group whose box barely overlaps another’s.

Worked example

Control scores run 11–16 with median around 13.5, while Treatment scores run 17–22 with median around 19.5. The boxes do not overlap at all: every treatment value exceeds every control value in this small sample. That visual separation is stronger evidence than the mean difference alone, because it shows the whole distributions are apart, not just their centres. A formal test would still be needed for inference, but the plot already tells the story a reader will remember — and it warns against the opposite case, where means differ but boxes overlap heavily, meaning individual variation swamps the group difference.

Honest limits

Box plots hide distribution shape: a bimodal group (two clusters) and a uniform group can share identical five numbers and look the same as boxes while meaning different things. With very small samples (under about 8 per group), the quartiles are unstable and the plot is more decoration than summary — show the individual points instead. The tool draws whiskers to min and max rather than the Tukey 1.5×IQR rule with outlier dots, stated plainly, so extreme values stretch whiskers instead of appearing as dots. For publication, re-plot with a statistics package that marks outliers explicitly.

Who it is for

This tool is for students learning distributions, researchers comparing groups before formal testing, and teachers showing why medians and quartiles beat means for skewed data. Use it for exploration and for the first figure in a report — then confirm with tests and, for small samples, dot plots.

Going further

Read boxes in pairs, not alone: ask three questions of every comparison — do the medians differ, do the spreads differ, and does either distribution skew? A difference in spread is itself a finding (a treatment that helps some and harms others widens the box), and skew toward high values in one group can signal a ceiling effect or a subgroup. When boxes overlap, resist declaring “no difference” from the picture alone: overlapping boxes are compatible with real differences that a test on paired or larger data could detect. The plot guides the question; the test, with its assumptions checked, answers it. When samples are small, pair every box plot with the individual data points drawn on top of it. A box built from six values is a fragile summary: one value moving changes the quartiles noticeably, and readers who see only the box cannot tell whether the distribution is smooth, clustered, or driven by a single extreme. Showing points restores that information without clutter, and it disciplines the writer too, because a box that looks convincing while its points scatter wildly is a warning to soften the claim. For larger samples, points still help at the tails, where outliers live, but the box earns its keep: with fifty or more values per group, the five numbers stabilize and the visual comparison becomes trustworthy. The rule of thumb is simple: the smaller the sample, the more the raw data must stay visible.

Embed this tool on your website — free

Copy the code below to add this working tool to your own page. A small credit link stays with it — that’s all we ask.