Missing Data Replacement

You can access the Missing Data Replacement dialog box in the following ways:
  • Ribbon bar. Select the Data tab. In the Transformations group, click Filter/Recode and on the menu, select Replace Missing Data to display the Missing Data Replacement dialog box.
  • Classic menus. On the Data menu, select Replace Missing Data to display the Missing Data Replacement dialog box.
Use these options to permanently replace missing data values for selected variables with the means of those variables. When applicable, this option is also available in many types of Statistica analyses as one of the missing data handling options. If you perform a missing data replacement from within an analysis, the replacement is temporary and applies only to the current analysis.
Option Description
Variables Click the Variables button to display the Select Variable dialog box, where you select the variables in which missing data is replaced by means.
Cases By default, all cases are used in computing means for each variable. You can use case selection conditions to compute the mean for only the selected cases. Only those cases will have their missing data replaced. When you click the Cases button, the Case Selection Conditions dialog box is displayed. Here, you specify new case selection conditions or open existing conditions.
Weight By default, each (selected) case contributes equally to the computation of the variable means. Click the Weight button to display the Define Weight dialog box, where you select a weighting variable. By specifying one variable as containing weights, the influence of each case can be varied by an amount proportional to the value of the weight variable for the case. (Although Statistica actually uses a faster algorithm, this system can be interpreted as reading each case a number of times that is equivalent to the current value of its weight.)
OK Click the OK button to accept the settings and close the dialog box.
Cancel Click the Cancel button to close the dialog box. Any changes made will be disregarded.