### Origin: Box Plots

The box plot (sometimes called a box and whisker plot or box chart) is a graphical representation of key values from summary statistics. Typically, the values represented are the minimum, 25th percentile, median, 75th percentile, and the maximum. Sometimes these values are called the five-number summary.

Figure 1: Values represented in Origin's default Box Plot

Additional values are represented in Origin's default box chart. These include the 1st and 5th percentiles, mean, and 95th and 99th percentiles.

Box and Whiskers Plot Examples

Example plots from our Graph Gallery
[click on the image to see original graph]

 Figure 2: Box chart using the 'diamond box' option with data and normal curve overlay Figure 3: Box chart employing an incrementing custom color list for the box fill option

Box plots are helpful in Quality Analysis for interpreting the distribution of data since it can easily show whether the data is skewed and if there are unusual observations (outliers) in the dataset. Box plots are also very useful when large numbers of observations are involved and when two or more datasets are being compared.

Box and Whisker Plot Options

Box Display Type : Display for the Box chart can include the box and/or the actual data in a column scatter plot. Display options are as follows:

• Box
• Data
• Box + Data (overlapped, data on right/left)

When the data is included, its bin values (bin size, begin, end) can be altered. Display of the data can be snapped into bins. Also, a distribution curve can be superimposed on the data.

 When the data is displayed as a column scatter, the data can be plotted using its specific Y values. Alternatively, the data can be organized into bins.

Box Options: Although the default box chart is generated using the upper and lower quartiles, it can easily be changed to one of the following and saved to the box chart template for future use:

• standard deviation
• standard error
• 25th/75th percentiles (default)
• 10th/25th/75th/90th percentiles
• a constant value
• notched box (notches on the box represent the confidence interval about the median value)

Whisker Range Options: Although the range represented by the whiskers in Origin's default box and whisker diagram is between the 5th and 95th percentiles, it can be changed to any of the following and saved as the default in the box chart template:

• 10-90
• 5-95 (default)
• 1-99
• standard deviation
• standard error
• outlier
• min/max (extremes)
• constant
• quantile

Several of the options enable the use of a Coefficient field. If you select Outlier and accept the default coefficient value (1.5), the whisker length defines what are commonly referred to as the upper inner and lower inner fence values (a coefficient = 3 would define what are commonly referred to as the upper outer and lower outer fences).

Upper Inner Fence = 75th Percentile + (1.5 * Interquartile Range)
Lower Inner Fence = 25th Percentile - (1.5 * Interquartile Range)

Data points that lie outside of the fence values are considered to be outliers. An Outliers check box is provided to allow you to display these outliers as individual data points.

Miscellaneous:

• A Diamond Box option is provided.
• Labels can be associated with Box and Whisker Range values.
• Formatting for the Minimum, Maximum, 1%, 99%, and Mean value symbols/data points can be updated.

Basic Data Plot Properties

The following basic data plot properties can be edited in a Box and Whisker diagram depending on the Box and/or Whisker range settings:

• Symbol Properties: shape, interior, size, and color (more...*)
• Line Properties: style, width, and color (more...*)
• Fill Properties: color, pattern (more...*)
• Pattern Properties: border style, width, and color; pattern width and color (details are the same as those for Line Properties)

All can be updated quickly and easily through Plot Details, an intuitive tab-based dialog.

* Javascript must be enabled and pop-up windows must be allowed in order to view the details of a specific graph type.