How do I get more meaningful variable names in SPSS

Many researchers are used to having to add this meaning within SPSS (or other packages) after data collection using variable  and value labels, but Qualtrics can actually do a lot to help with this. As default behaviour Qualtrics actually codes your responses for you (that is, the VALUE LABELS part of the traditional SPSS process), and it also uses the question text as the default VARIABLE LABELS.

When it comes to the variable names themselves, Qualtrics just numbers the questions in your survey according to the creation process.  If you move stuff around, create things in different orders, or bring in questions from a library, that can mean that the numbers are not in order.  In any case, they are rarely meaningful.  The VARIABLE LABELS are a big help, but if you want your variable names to be meaningful there are a few things you can do.  The simplest is just to name the variables yourself in the creation process.

Example1 export survey narrow1


Consider a simple example survey excerpt with a multiple-choice question and a matrix block of 3 questions on a 5-point scale.  By default Qualtrics just assigns numbers to the questions as you can see here.

When we download a dataset to SPSS based on this definition, Qualtrics provides the VARIABLE and VALUE labels, but the variable names are just Q-numbers.  Note that For the matrix question, Qualtrics just adds a “.1”, “.2” etc. to the items in the scale in sequence.


Example1 SPSS Variables Default highlight

Example1 Edit Question Names

For the researcher though, Qn.k is not a terribly descriptive name and in variables lists it is useful to have something more meaningful to work with. The simplest way to do this in Qualtrics is just to edit the name of the questions in the main survey edit page. After editing the names here, when we download the data again the new names appearas variable names in SPSS.
Example1 SPSS Variables with names focus





Qualtrics limits the names to 10 characters.  This simple tip will work with most question-types.

Qualtrics will add additional numbering whenever a question types requires it such as matrix questions as we see here, or if multiple responses are allowed.

Taking the time to name each of your variables in a sensible way will save a lot of time and confusion when it comes to analysing your data, so is well worth it. When this simple trick is not enough, Qualtrics has some advanced techniques for specifying variable names, response values and response labels.