How to use the Box Plot Maker
- Paste one column per group, with group names in the first row.
- Choose "Box plot".
- Add a Y axis title, then download the figure.
How the box plot is drawn
The box runs from the first quartile (Q1) to the third quartile (Q3), so it contains the middle 50% of values. The thick line is the median, and the + marks the mean. Whiskers extend to the furthest value within 1.5 × IQR (interquartile range) of the box (Tukey’s rule). Points beyond the whiskers are drawn as open circles: these are potential outliers.
When to use a box plot
Box plots are ideal for comparing the spread of several groups side by side: treatments in an experiment, classes in a test, batches in production. They show skew and outliers that a bar chart of means hides. For one group with many values, a histogram shows the shape in more detail.
Worked example: test scores in three groups
The sample data has 40 scores for a control group and two treatments. Reading the plot from left to right: Treatment A has the highest median and the narrowest box, so its scores are both higher and more consistent. Treatment B has the lowest median and the tallest box and whiskers, so its results vary the most. When one group’s box sits entirely above another’s, the difference is usually large enough to be worth testing statistically.
| Group | Median | IQR (Q3 − Q1) | What it tells you |
|---|---|---|---|
| Control | ≈ 72 | ≈ 11 | Baseline spread |
| Treatment A | ≈ 78 | ≈ 8 | Higher and more consistent |
| Treatment B | ≈ 65 | ≈ 15 | Lower and much more variable |
Approximate values from the sample dataset. Your exact numbers appear in the statistics table under the chart.
Common mistakes with box plots
Too few values. With fewer than about 8 values per group the quartiles jump around, so show the individual points (a scatter or dot plot) instead. Mixing up the line and the plus. The line in the box is the median; the + is the mean. Treating every circle as an error. Points beyond the whiskers are only potential outliers; check whether they are real measurements before removing them, and say in your methods if you did.
Frequently asked questions
- How many values do I need for a box plot?
- At least 5, but box plots are most reliable with 10 or more values per group. With very small samples, show the individual data points as well or instead.
- How do I describe a box plot in a report?
- Give the median and interquartile range for each group (for example "median 78, IQR 74–82"), mention any outliers, and compare the positions of the boxes. Say that whiskers extend to 1.5 × IQR.
- How are quartiles calculated?
- With linear interpolation between ranked values (the same method as Excel QUARTILE.INC and NumPy’s default).
- Can groups have different numbers of values?
- Yes. Empty cells are ignored, so columns can have different lengths.
- What does the + inside the box mean?
- It marks the mean. When it is far from the median line, the data is skewed.
- Is my data private?
- No. Your data never leaves your browser. Plotiq parses and plots files locally with JavaScript, so nothing is uploaded to a server.
Last updated: · by the Plotiq team