- Defining the Variable
One of the first steps in preparing your dataset for analysis is to define your variable type. The variable type will communicate to the software how you categorize that data whether is nominal, ordinal, or scale. Unfortunately, it is not an option to highlight all the variables you want with the same variable type and complete this task in one step like you can in Microsoft Excel. Defining variables has to be completed by clicking on each variable one by one and selecting the variable type. If you have a dataset with hundreds of variables, this can become a time consuming task. A shortcut you can use is once you have defined a few variables in a row you can highlight the defined variable types, right click “copy”, highlight the next variables that need to be defined, and right click “paste”. This way you are naming multiple variables with the same variable type at once.
- Transforming Missing Data to 0
For different reasons there may be missing data or blanks in your data that make it inappropriate to use the series mean function to fill in these missing data cells. Blank cells can be problematic when computing composite scores in SPSS. The solution to this problem is to fill in each of these blank cells with a marker value that is compatible with other functions (e.g. compute function). Use syntax to code for filling in system missing data with a marker value by typing in “Recode Variable2 to Variable20 (SYSMIS=0).” This code communicates to the software that you would like to fill in all blank cells from Variable2 to Variable20 with 0. The first variable mentioned in the code is the starting point for your recode range and the second variable is the end point for your recode range. In syntax, highlight the code, click “run”, and click “save”. This shortcut will reduce the amount of time taken to recode purposefully blank data cells.
Photo Credit: Flickr – Eva the Weaver