Selecting Data Rows Without Missing Values

Several of the plotting and statistical tools require that the input data set have complete cases–that is, no missing values for any row or variable. These tools will eliminate missing values either automatically by row or according to the user’s specification of whether rows or columns with missing data should be removed. This data selection tool allows the complete cases for any set of variables to be easily viewed on the map and in the data table.

This data selection option is avalable from the Selections / Non-missing values menu item, which displays the following dialog.

The dialog for selecting data rows without missing values for one or more variables.

This dialog displays a list of all variable names in the data table. One or more of these can be selected by left-clicking with the mouse. Control-click and shift-click can be used to select multiple rows.

Below the prompt for the column(s) of interest is a drop-down list that specifies the action to be taken with the results of the data selection. The available actions are:

  • Replace: All currently selected data are replaced with the newly selected data.

  • Union: The results of the new data selection will be added to any previously-selected set of values.

  • Intersection: Only the table rows that are in both the previously-selected set and the newly-selected set will be retained.

  • Difference O-N: The results of the new data selection (New) will be subtracted from any previously-selected set of values (Old).

  • Difference N-O: The set of previously-selected values (Old) will be subtracted from the newly-selected set of values (Nnew).

The Replace action is the default.

When the ‘Select’ button is clicked, all of the rows with complete cases for the selected variable(s) will be highlighted on the map and in the data table.