Tips – Use subset of wide data files

Some users I met recently complained over the hard job to work with very wide data files (= a lot of variables or columns). I showed the user the “subset”-command and he was very impressed and wondered if this was a new command? No, this has been there for a while… if you are interested to learn these kind of tips and tricks – let your contact person know and we can put up some Consulting hours.

Here is an example of a wide data file with employee data mixed with attitudes:

Here is the command for creating the subset as a first step:

Choose a name of the first subset and mark the variables you want to be included there:

Then click the arrow so the variables moves in to the right box (Variables in Set):

After you have clicked “Add Set”-button you will see the following and can then continue to create the next subset, here “Attitudes” and select the attitude variables :

Then close the dialog box.

Now it’s time to choose the subset or subsets you want to work with in SPSS Statistics:

Mark the subset/subsets you want to work with. Start by unbox the “All Variables” ticket box.

As you see I keep “NEWVARIABLES”, this is recommended if you want to create new variables in the compute/recode command. Click OK.

As you see, only the variables from the chosen subsets are visible (the other is just temporary hidden).

Data View:

Variable view:

You can clearly see that variable number 4 is hidden as well as variable number 6 to 20.

Dialog box if you choose a command, in my example frequency command:

To choose all back again, just go back to the “Use Variable Sets”-command again and choose “All Variables” or following command:

Greetings Gunilla Rudander