Box plots, also called Box and Whiskers (see wikipedia), are a graphical way to compare distributions of data. These are non-parametric approaches (i.e., do not make any assumption about the underlying shape of a distribution of data).

Use the **Box** **command** to create box plots, point distributions, sideways histograms, and violin plots.

## Values

Specify one column of data in the **Values** field.

## Type

In addition to the standard Box and Whisker, there are several other types of graphs you can create using a **Box** command.

### Whisker

By default, the **Box** command draws a standard box and whisker diagram, or Tukey plot. Further options in the detail view:

**IQR**: The box is drawn around the Inner Quartile Range (IQR), where the IQR is the difference between the first and third quartile. Outliers (1.5 times beyond the IQR ) are drawn as filled in circles, and extreme outliers (3 times beyond the IQR) are open circles. The whiskers are drawn to the smallest/largest non-outlier. Choose whether or not to draw outliers.

In this graph, 1.5 times the IQR is shaded, indicating the range, beyond which, a point is an outlier.

**Min/Max**: Draw the whiskers out to the minimum and maximum values.

The same graph as above drawn with the ‘Min/Max’ option.

**Percentages**: Specify one or more percentages to place the whiskers.

### Points

The Box command can draw a point cloud. You can change the **Maker** type and change the **Point color** to use a color scheme.

Here is the same example drawn with **Type** = ‘Points’.

### Sideways Histograms

**Probability/Histogram** options are used to draw sideways histograms. You can control the width of the **Bin** **width**.

**Probability** scales the height individually for each dataset. **Histogram** scales the height relative to the entire dataset.

### Violin Plots

There are two options for violin plots, **Violin PDF** (scales each individually) and **Violin Count** (scales based on the number of points across the dataset).

When you have the same number of data points in each group, **Violin PDF** and **Violin Count** produce similar results.

## Position

The **Position** indicates the location of the box on the X-axis.

When the **Position** is set to the default of ‘Single value’, the number to the right indicates the numerical location on the X-axis, and each command draws one box.

When the **Position** is set to a numerical column, the value of the column can be used to specify multiple locations, effectively binning (i.e., grouping) the data. Thus, multiple box plots are created with one command.

## Bins

When the **Position** is set to a numerical column, you have the option of further binning the data, using the **Bins** menu. For example, below the data are binned using a stride of 0.1.

## Labels

When the **Position** is set to a text column, you have the option of specifying the label on the x-axis, using the **Labels** menu. With the first menu, select a column to specify what groups to include and in what order. With the second menu, select a column that specifies a new label (optional).

## Direction

The **Direction** drop-down box allows you to position the boxes on the Y-axis.

## Width

You can vary the **Width** of the box plot.

## Fill

The **Fill** with drop-down box has options for solid colors, patterns, gradients, and color schemes.

## Summary Statistics

The **Box** command has a scrollable table is shown that provides a list of summary statistics.

If you have a single box (one bin of data), the statistics will be in a single scrollable column.

If you have multiple boxes drawn by a command, each box is listed as a row in the table.

Directly below this table, you can specify additional percentages that you would like to compute. There is also a check box that will add a label with the numerical value of the median to the box plot.