Training

When computing a new variable in SPSS Statistics (previously known as SPSS for Windows®), be sure to use statistical functions to account for missing data that might be present in one or all of the variables that are included in the calculation.

For example, perhaps you'd like to compute a composite variable for Education using the average of education data across four variables. As shown below, one approach is to drag each of the four variables into the Numeric Expression box, separate each variable by an addition sign, and then divide them by four. In this example, the Compute Variable window looks like this:

There's a problem with this method of computing a new variable, however: if any of the four variables have missing data, they will not be included. You can resolve this dilemma by using the functions available in the Compute Variable window:

• Go to Function group and select Statistical
• Go to Functions and Special Variables and choose Mean

Next, go to the Numeric Expression box and place each variable within the parentheses for Mean. When you calculate the new variable, it will include all of the variables—even if there is missing data in each of the variables. In this example, the Compute Variable window looks like this:

The statistical functions make computing new variables easier, and they also have an important advantage over manual entry. For more information on using syntax within SPSS Statistics, please refer to the SPSS Statistics Command Syntax Reference that accompanies your software.