Predicted and Predictor Variables

By default, all quantitative variables are used for EM and regression estimation. If needed, you can choose specific variables as predicted and predictor variables in the estimation(s). A given variable can be in both lists, but there are situations in which you might want to restrict the use of a variable. For example, some analysts are uncomfortable estimating values of outcome variables. You may also want to use different variables for different estimations and run the procedure multiple times. For example, if you have a set of items that are nurses' ratings and another set that are doctors' ratings, you may want to make one run using the nurses' item to estimate missing nurses' items and another run for estimates of the doctors' items.

Another consideration arises when using the regression method. In multiple regression, the use of a large subset of independent variables can produce poorer predicted values than a smaller subset. Therefore, a variable must achieve an F-to-enter limit of 4.0 to be used. This limit can be changed with syntax.

To Specify Predicted and Predictor Variables

This feature requires the Missing Values option.

  1. From the menus choose:

    Analyze > Missing Value Analysis...

  2. In the main Missing Value Analysis dialog box, select the variable(s) for which you want to estimate missing values using the regression method.
  3. Select EM or Regression in the Estimation group.
  4. Click Variables.
  5. If you want to use specific rather than all variables as predicted and predictor variables, select Select variables and move variables to the appropriate list(s).