Using SPSS and PASW/Creating New Variables

The previous page showed how to create a new variable by Transform of an existing one. In this exegesis we will show you how to create entirely new variables.

Creating a new variable is a foundational skill for users of SPSS, as it is necessary to perform many statistical tests and procedures (see Part II).

Step one in creating a variable is switching the Data Editor window to “Variable View”, by clicking on its tab:

Then click in the first empty row and provide your nascent variable with a name, e.g., “Age”:

You can specify the type of the variable by clicking in the cell under the column labeled “Type.” An ellipsis will appear:

Click on the ellipsis to open the Variable Type Dialog box, which looks like:

Most variables will be “Numeric,” meaning they will be number based, either because the number itself is meaningful (interval/ratio variables) or because it represents distinct categories. Other common variable types are String, for text, Date, and Currency. Select the desired type and click OK.

Next, input a label (description) for your variable as a whole:

Moving right in the Variable View, the next field is Values. Click there and then on its ellipsis to expose

If the variable is ordinal you must assign a unique number to each label. E.g., for a variable holding participant race you might specify "1" as Value with "White" as the Label, and continue with such mappings as 2=Black, 3=Native American, etc.

Next up is Missing. Click here and as usual on the ellipsis to specify how missing (did not respond, not applicable, lost, etc) values will be denoted so they are not used in any calculations by default:

While you have free hand to choose your designation, it is common to uses iterations of 9 to indicate missing values in numeric fields, where that value would otherwise be impossible. For instance, if a variable holds an integer in range 1 - 4 plus a missing value, you can assign a “9” to indicate that a value is missing. If there are 20 categories, "20" is two decimal places and <99, so you could use two “9”s to indicate a missing value in that variable, and so on. If the variable is of type String, you would usually use more description missing values, such as "NA".

Finally, you must specify your level of measurement. This may be done by clicking on the cell under the column labeled “Measure” and choosing from Nominal, Ordinal, and Scale:

Nominal and ordinal variables should be labeled under their corresponding names, while interval and ratio variables should be labeled as “Scale.”

Repeat this process for as many variables as you need to create.

Chapter contributed by Damian Patrinostro.