Techniques for Exploring Data |
Details |
This section describes the Selection Mode dialog box, shown in Figure 11.13. To open the Selection Mode dialog box, right-click on a plot or data table, and select Selection Mode from the pop-up menu. Alternatively, click a data view’s title bar to activate it, and select Edit Selection Mode from the main menu.
The Selection Mode dialog box has the following elements:
sets the selection mode to be local selection mode. The active window becomes either a selector view or an observer view. All other data views linked to the active window become selector views.
sets the active window to be an observer view.
sets the active window to be an observer of the union of selector views. An observation is displayed as selected if it is selected in any selector view.
sets the active window to be an observer of the intersection of selector views. An observation is displayed as selected if it is selected in all selector views.
The following list presents a few additional details about using local selection mode:
There is a limit of 31 selector views that can be linked to an observer view. There is no limit to the number of observer views.
It is often useful to have multiple selector views but only one observer view. In this case it is quickest to activate the plot that is to become the observer view, and then to select Edit Selection Mode from the main menu. Configure that plot as a local observer view, and click OK. All of the other data views are automatically changed to selector views. This technique was used in the example.
If the observer view is a plot that displays individual observation markers (for example, a scatter plot), it is often useful to configure the plot to show only the selected observations. See the section Displaying Only Selected Observations for details. This technique is sometimes called graphical filtering, because selected observations do not "reach" the observer view until they have passed through all of the "filters" (criteria) imposed by the selector views.
Copyright © SAS Institute, Inc. All Rights Reserved.