**Select menu: Stats | Multivariate Analysis | Principal Components Analysis or Canonical Variates Analysis | Biplot**

Use this to produce a biplot of the results from a principal components or canonical variates analysis.

- After you have
**imported your data, from the menu select**.

**Stats**| Multivariate Analysis | Principal Components

*OR*

**Stats | Multivariate Analysis |****Canonical Variates**. - After you run the analysis the
**Biplot**button will become enabled. Click to open the dialog below.

## Available data

This lists data structures appropriate to the current input field. It lists texts for use in specifying the labels for the individuals and/or the variables. Double-click a name to copy it to the current input field; alternatively, you can type the name directly into the input field.

## Display in plot

Specifies additional items to display in the graph.

Convex hull |
draws a convex hull around the scores. |

Means |
plots the means for each group (canonical variates analysis only). |

## Type of axes

Lets you select either predictive or interpolative axes.

## Multiplier for vector loadings

Arrows are plotted on the axes to represent their loadings (or inverse loadings); the loadings show the approximate contribution of each variable in the first two dimensions. If the loadings are all close to the origin, they are multiplied by a scalar to make them easier to read. By default, the multiplier is calculated automatically, but you can supply a specific value using this option.

## Labels for individuals

Can be used to specify what type of labels to use to identify the individuals on the plot. Select **Numbers** to display the unit numbers, **None** for no labels, **Unit labels** to display the unit labels, or **Labels** to supply alternative labels using the name of a text structure in the space provided.

## Labels for variates

Can be used to specify what type of labels to use to identify the variates on the plot. Select **Numbers** to display the sequential numbering of the variates, **None** for no labels, **Identifiers** to display the identifier names, or **Labels** to supply alternative labels using the name of a text structure in the space provided.

## Grouping factor

For scores from a principal components analysis this specifies a grouping factor that can be used to identify groupings of individuals by plotting the groups using different colours.

## Action Icons

Clear |
Clear all fields and list boxes. | |

Help |
Open the Help topic for this dialog. |

## See also

- Principal components analysis menu for principal components analysis
- Canonical variates analysis menu for canonical variates analysis
- DBIPLOT and BIPLOT procedures in command mode