Excel Box and Whisker Plot: What Column Names Mean

Excel box and whisker plots do not require special or reserved column names. Put each group of raw numeric observations in its own column, and use the first cell of each column for the group name; Excel uses that header as the label for the corresponding box.

What the column names represent

A column name in this chart is a label, not an instruction to Excel. If the header says Control, Week 1, or North, Excel displays that text to identify the group represented by the numbers underneath it. There is no required set of names such as Minimum, Q1, Median, Q3, and Maximum. Excel calculates the quartiles and whiskers from the observations you supply.

For example, a worksheet might have a header row like this:

A1: Control; A2:A8: measurements from the control group.
B1: Treatment; B2:B8: measurements from the treatment group.
C1: Follow-up; C2:C8: measurements from the follow-up group.

With that layout, Control, Treatment, and Follow-up are the names associated with the three boxes. Choose concise labels that readers can understand without a separate key. Include the unit in a header only when it helps distinguish the measurements, such as Temperature (°C); the values in a given chart should still use a consistent unit and measurement scale.

Headers may contain spaces, numbers, or punctuation. They do not have to match a specific naming pattern. If you omit a header or Excel does not recognize one, the chart may use generic labels such as Series 1. Clear headers make the chart easier to interpret, particularly when it has several groups.

How Excel interprets the data

A box and whisker chart summarizes the distribution of the values in each group. The box shows the lower and upper quartiles, with a line for the median. The whiskers extend to the most extreme observations that are not treated as outliers under the chart’s rule; points beyond the whiskers can appear separately as outliers. The chart’s purpose is to compare distributions, not just to show one typical value.

Excel needs the observations themselves so it can calculate the distribution. A simple, reliable layout is one group per column, with the group name in the top cell and one numeric observation in each cell below. Each row can represent a different record, but the row position does not have to mean the same thing in every group. For instance, the fifth measurement in one column need not be paired with the fifth measurement in another unless the study design says those observations are matched.

Groups do not have to have the same number of observations. If one column has fewer values, leave the unused cells blank rather than filling them with zero. A zero is a real measurement and can change the distribution; a blank is not an observation. Keep notes, totals, and other text outside the data range so Excel does not mistake them for part of the chart input.

Create a box and whisker chart from columns

In Excel for Microsoft 365 on Windows, select the header cells and the numeric data beneath them, then choose Insert > Insert Statistic Chart > Box and Whisker. Excel creates a chart using the selected data. If the labels or groups appear in the wrong direction, select the chart and choose Chart Design > Switch Row/Column. The precise button placement can vary slightly with the ribbon window size.

Place each group in a separate column and put its descriptive name in the first row.
Select the complete rectangular range, including the headers and the observations. If your data columns are separated by unrelated material, move or copy them together first.
Choose the Box and Whisker chart from the Insert tab’s statistical chart options.
Check the horizontal labels and the number of boxes. Use Switch Row/Column if Excel has treated rows as groups instead of columns.
Add or edit the chart title and axis titles if a reader needs more context about the measurement or units.

For example, selecting A1:C8 in the layout above should produce three boxes labelled Control, Treatment, and Follow-up, provided the cells contain the stated headers and numeric observations. If you select only A2:C8, the values can still be plotted, but the labels may be generic because the header row was left out. If the chart already exists, you can inspect its source through Chart Design > Select Data and adjust the data range.

Excel for Mac has a corresponding box-and-whisker chart option under the Insert tab’s statistical chart choices, although ribbon labels and layout can differ by release. Excel for the web has different feature availability and interface details; if the chart type is unavailable in your browser version, open the workbook in the desktop application. The guidance here is for current Excel for Microsoft 365 on Windows.

Choosing useful names and arranging groups

A good header identifies the group, condition, time point, location, or other variable being compared. Prefer labels such as Before, After, and Month 3 over vague labels such as Set A, Set B, and Set C when the meaning is known. If the name alone is not enough, use a chart title or axis title to explain what was measured. For instance, a plot titled Reaction time by training condition is clearer than a title that merely says Results.

Use one consistent naming style. If the columns represent locations, name them all as locations; if they represent treatments, make the treatment distinction explicit. Avoid long sentences in headers because they can crowd the chart. You can keep full descriptions in nearby worksheet notes and use shorter chart labels, so long as the labels remain unambiguous.

Think carefully about what each box represents before organizing the columns. A column for each age group can compare distributions across age groups; a column for each measurement date can compare dates. Do not combine values from different units or fundamentally different measures in a single box plot just because they fit in the same worksheet. A chart label cannot correct a mismatch in the underlying data.

Duplicate names can make boxes hard to distinguish even when the data are correctly plotted. Add a distinguishing word, such as Site A—Morning and Site A—Evening, or use a consistent short code and explain it in the chart or worksheet. If headers are too long, adjust the chart width or shorten the labels rather than deleting useful distinctions.

Rows, columns, and chart orientation

Excel can sometimes interpret a selected range differently from what you intended. In the recommended layout, each column is a group and its header is the group name. If your source data instead has groups in rows, the chart may initially show a box for each column, or it may use the first row or column as labels in an unexpected way. The preview of the finished chart is the practical check: count the boxes and read their labels against the worksheet.

Use Chart Design > Switch Row/Column to exchange which direction Excel treats as the series. This is a chart interpretation change; it does not rearrange or alter the worksheet cells. If switching produces confusing labels, return to the sheet and make a simple table with one group per column. A clean source layout is easier to audit and less likely to be misread when the data range changes later.

For a chart with one group, a single labelled column can be sufficient. Excel may show only one category label, so a descriptive chart title can help explain the context. For several groups, keep the names in a single header row rather than adding explanatory rows inside the observations. If you need a second descriptive detail, such as site and condition, use a clear combined label or reorganize the data so the intended comparison is obvious.

Quartiles, whiskers, and chart options

The column names tell readers which group they are looking at, but they do not control how the box is calculated. Excel provides options for how quartiles are calculated, and the choice can affect the plotted box, especially for small samples. Select or right-click a box and open Format Data Series to view the available series options. Depending on the chart and Excel build, options include inclusive or exclusive median calculations, displaying inner points or outlier points, and showing mean markers or a mean line.

Inclusive and exclusive median calculations are alternative conventions for calculating quartiles. They can produce different quartile positions for some sample sizes; neither setting is a new column-header requirement. If you are comparing plots, keep the same quartile setting across them and state the method when the analysis needs to be reproducible. Do not assume that two box plots use identical conventions just because their labels match.

Likewise, the display of individual observations and outliers is controlled by chart options, not by naming a column Outliers. An outlier point is identified from the data and the chart’s calculation, not from text in a header. Turning on inner points can make the underlying sample more visible, while a chart without them emphasizes the summary. Consider sample size and the purpose of the comparison when deciding which display is easiest to read.

Common label and data problems

Generic labels appear: Include the header row in the selected chart range. If the chart already exists, inspect its data range and confirm that the header cells are included.
There is one box per measurement instead of per group: Excel may have interpreted the data orientation incorrectly. Try Switch Row/Column or rearrange the data with one group per column.
A group is missing or looks wrong: Check that its cells contain numeric values rather than numbers stored as text, and make sure the selected range includes the entire group.
Blank rows become zeros in the result: Remove any zero placeholders that mean no observation. Leave missing observations blank instead.
Labels are crowded or cut off: Shorten headers without losing their meaning, widen the chart, or use a clearer chart title and supporting notes.

Also check for accidental totals, subtotals, or summary-statistic rows in the selected range. A total included among individual observations becomes part of the distribution and can distort the plot. Similarly, pasting a precomputed five-number summary under headers such as Minimum, Q1, Median, Q3, and Maximum does not tell Excel to draw a box from those five values. The built-in chart expects observations from which it can calculate the distribution.

When a box plot is the right choice

A box and whisker chart is useful when you want to compare the center, spread, and possible outliers of several groups. It can reveal that two groups have similar medians but very different variability, or that one group contains unusually distant observations. A column name should describe the category being compared, not the statistic displayed by the chart.

It is not a substitute for every other chart. If the order and shape of individual observations matter, a dot plot or histogram may show more detail. If the main goal is to compare averages over time, a line chart may be more suitable. Box plots compress many observations into a small number of summaries, so a reader may not see clusters, gaps, or the exact distribution shape. In a report, pair the chart with an explanation of the sample and measurement when those details affect interpretation.

For a custom chart built from already calculated quartiles, Excel’s standard Box and Whisker chart is generally not the right input method: it calculates its own summary from raw values. A custom construction may be possible with other chart types, but it requires deliberate setup and should be labelled clearly. Do not label raw-data columns Q1 or Median unless those cells actually contain the corresponding summary values and you are using a chart designed for that purpose.

Frequently Asked Questions

What should I name the columns in an Excel box plot?
Use the name of each group, such as Control, Treatment, or January. Excel does not require special terms; the header is used as the group’s label.

Does Excel need a header row for a box and whisker chart?
No, but headers are helpful. Without them, Excel may use generic series names, so include a clear header row and select it with the data.

Should I enter Q1, median, and Q3 as separate columns?
Not for the built-in Box and Whisker chart. Give Excel the individual numeric observations; it calculates the quartiles and median. Summary values alone are not a substitute for the raw data in this chart type.

Can each box have a different number of values?
Yes. Put each group in its own column and leave unused cells blank. Do not use zero to represent a missing observation, because zero is treated as a value.

Why does my chart show Series 1 instead of my column name?
The header may not have been included in the selected range, or Excel may be interpreting the range differently. Edit the source range to include the header, then check whether Switch Row/Column gives the intended labels.

Can I change what the whiskers mean by renaming a column?
No. Column names identify groups; they do not set the quartile or whisker calculation. Review the chart’s Format Data Series options for the available calculation and display settings.

Try it: the Date Difference Calculator: days, months, years and weekdays between two dates, with the DATEDIF formulas.

More in How-to basics

Sources: Function behaviour and menu paths: Microsoft Support documentation for Excel. Method & sources

Spotted an error in this guide? Let us know.