How do I read a box plot?

The box covers the middle 50% of the data: its lower edge is Q1, its upper edge Q3, and its length is the interquartile range IQR = Q3 − Q1. The line inside is the median. Whiskers reach the furthest points still within Q1 − 1.5×IQR and Q3 + 1.5×IQR, and anything beyond is drawn as an outlier. A longer box means more spread, an off-centre median suggests skew, and comparing boxes shows differences between groups at a glance.

Why do different programs give different quartiles?

There is no single definition of a quartile. For the 11 values 6, 7, 15, 36, 39, 40, 41, 42, 43, 47, 49, a TI-84 (which leaves the median out of both halves) gives Q1 = 15 and Q3 = 43; Tukey’s hinges and Excel’s QUARTILE.INC give 25.5 and 42.5; Excel’s QUARTILE.EXC and Minitab, which interpolate at position (n+1)p, give 15 and 43. Pick the method that matches your textbook or software.

How are outliers identified?

The usual rule is Tukey’s: compute the IQR, and any point below Q1 − 1.5×IQR or above Q3 + 1.5×IQR is an outlier; if it’s also beyond the outer fences at 3×IQR it’s an extreme outlier. The plot uses different symbols for the two and the table lists the fences. You can change the 1.5 factor, or switch the whiskers to min–max, which flags no outliers.

What are notches for?

A notch shows an approximate 95% confidence interval for the median, median ± 1.57×IQR/√n. If the notches of two boxes don’t overlap, that’s rough evidence the medians differ. With small samples the notch can be wider than the box; the plot then keeps it within the box.

🗳️

🗳️Box Plot Maker - Box and Whisker Plot with Quartiles

Paste a few columns from Excel or any spreadsheet — one group per column, with an optional header row — and get side-by-side box plots, or put one group per line instead. The box runs from the first quartile Q1 to the third quartile Q3 with a bold line at the median; by default the whiskers reach the furthest data points within 1.5 interquartile ranges (IQR), and points beyond are flagged as mild outliers (1.5–3 × IQR, open circles) or extreme outliers (more than 3 × IQR, stars). You can extend the whiskers to the minimum and maximum instead, or change the factor k. Quartiles are computed differently by different software, which matters for small samples, so five methods are offered: Tukey’s hinges, Moore & McCabe (used by the TI-84 and many textbooks), Excel’s QUARTILE.INC and QUARTILE.EXC, and Minitab. The chart can show the mean (diamond), notches for a 95% interval around the median and all data points with jitter, horizontally or vertically, with a title and axis label, and exports as PNG or SVG. The table below lists each group’s count, five-number summary, IQR, mean, standard deviation, fences, whisker ends and outliers, ready to copy.

Free · No sign-up · Runs in your browserLast updated: 2026-10-26

Loading tool...

How to use Box Plot Maker

1Paste your data: one group per column with optional names in the first row (or switch to one group per line)
2Choose the quartile method and whisker rule, and turn on the mean, notches or data points if you like
3Read the box plot and the statistics table, then export PNG/SVG or copy the table

FAQ

How do I read a box plot?: The box covers the middle 50% of the data: its lower edge is Q1, its upper edge Q3, and its length is the interquartile range IQR = Q3 − Q1. The line inside is the median. Whiskers reach the furthest points still within Q1 − 1.5×IQR and Q3 + 1.5×IQR, and anything beyond is drawn as an outlier. A longer box means more spread, an off-centre median suggests skew, and comparing boxes shows differences between groups at a glance.
Why do different programs give different quartiles?: There is no single definition of a quartile. For the 11 values 6, 7, 15, 36, 39, 40, 41, 42, 43, 47, 49, a TI-84 (which leaves the median out of both halves) gives Q1 = 15 and Q3 = 43; Tukey’s hinges and Excel’s QUARTILE.INC give 25.5 and 42.5; Excel’s QUARTILE.EXC and Minitab, which interpolate at position (n+1)p, give 15 and 43. Pick the method that matches your textbook or software.
How are outliers identified?: The usual rule is Tukey’s: compute the IQR, and any point below Q1 − 1.5×IQR or above Q3 + 1.5×IQR is an outlier; if it’s also beyond the outer fences at 3×IQR it’s an extreme outlier. The plot uses different symbols for the two and the table lists the fences. You can change the 1.5 factor, or switch the whiskers to min–max, which flags no outliers.
What are notches for?: A notch shows an approximate 95% confidence interval for the median, median ± 1.57×IQR/√n. If the notches of two boxes don’t overlap, that’s rough evidence the medians differ. With small samples the notch can be wider than the box; the plot then keeps it within the box.