Select Variables

The variable selection dialog box comprises one or more frames that contain a list of variables in the active spreadsheet. Formats applied to variable names (i.e., bold, italic, underline, and font color, but not font size) are also displayed in the variable list to make it easier to recognize and select key variables. When the list of variables is too large for the display windows, the lists are scrollable. Alternatively, you can resize the dialog box by dragging its corners or sides to increase the size of the display windows for better viewing. Note that ToolTips can be displayed (by pointing to the variable name) making it easier to see names that are longer than the column width.

In order to apply a set of variables to an analysis, you can select variables in the following ways:

  • To select a continuous list of variables, click on the first variable in the list, hold down the Shift key, and click on the last variable in the list. Or, click and drag the mouse over the contiguous list of variables to be selected.

  • To select a noncontinuous list of  variables, hold down the Ctrl key while selecting the desired variables.

  • See Variable Selection Filtering for details about using textual expressions (wildcard expressions and open-ended ranges) to select variables.

Select variables. Alternatively, you can enter variable numbers in the Select variables box. A continuous list of variables should be separated by a hyphen (e.g., 1-10), while a list of discontinuous variables should be separated by spaces (e.g., 5 7 12 20). To select a bundle, click the bundle name in the display window. Bundles are displayed in brackets and listed in alphabetical order at the top of the variable list.

Select All. Click the Select All button to select all the variables displayed in the list(s) of variables..

Spread. Click the Spread button to display more information (the variable type and the long description) of each variable.

Shrink. Click the Shrink button (which is the alternate function of the Spread button) to display concise information (the variable number and short name) for each variable.

Zoom. Click the Zoom button to display a dialog box containing detailed information about the selected variable.

Bundles. Click this button to display the Variable Bundles Manager. Using the options in this dialog box, you can create, edit, delete, and rename variable bundles as well as generate a spreadsheet containing bundle information. Bundles can be used to organize large sets of variables and to facilitate the repeated selection of the same set of variables. By creating bundles, you make it possible to quickly and easily locate a subset of data in a large data file.

Screening bundles. Note that you can pre-screen or filter bundles according to the type of variables (categorical or continuous) by selecting the Show appropriate variables only check box (see below). When this option is selected, Statistica will only show bundles that contain appropriate variables for the selected analysis (e.g., for Correspondence Analysis, only bundles containing categorical variables would be displayed, since Statistica expects categorical data to be used as inputs). If a bundle contains a mixture of continuous and categorical variables, the bundle would be considered only as continuous bundle, since categorical analyses cannot process continuous variables, but continuous variable analyses can evaluate categorical variables.

Show appropriate variables only. Select this check box if you want to pre-screen or filter variables according to their Measurement Type; if that type is Auto, the Automatic variable pre-screening and classification option (located in the Options dialog box) determines how Statistica will automatically determine the Measurement Type. You can also assign an explicit Measurement Type in the Variable specifications dialog box.

The state of this check box will be remembered and restored between invocations of variable selection dialog boxes. When the Show appropriate variables only option does not apply (for some statistics and all graph analyses), it will be disabled.

If this check box is selected, and one of the Automatic variable pre-screening and classification options [other than Off (variables with Measurement Type 'Auto' will not be classified)] is selected in the Options dialog box - Analyses/Graphs - Display tab, only variables are shown (from which you can select) that are appropriate for the respective analysis and list (for example, when computing correlation coefficients, you would not be able to select text variables). Clear this check box to review and be able to select from among all variables.

Note: (This note is applicable when the Treat a variable as categorical if it has labels or is of type Text, Integer, or One-byte; or contains only integer values in the first _ cases option button is selected in the Automatic variable pre-screening and classification group box in the Options dialog box - Analyses/Graphs - Display tab.) Statistica considers variables that contain integers to be categorical. Integer variables that contain numerous unique values may inadvertently be misclassified as categorical, when the intention is that they should be considered continuous. Although this is an exceptional situation, it is recommended that integer variables that are meant to be seen as continuous should have their Measurement Type set to Continuous.