Hello! I am working with Qualtrics survey data on SPSS. My survey has a pre-recommended 0-4 scale with values ranging from 0=not at all - 4=very much. I have been unsuccessful in trying to recode some reversed items. Could somebody kindly explain whether the zero is the problem here and what would be the best way to do this.
Thank you so much for your help 🌼
Hello Kareen,
No, the zero value shouldn't be a problem, _unless_ you declared for that variable that zero is to be considered a missing value.
The method outlines by Natha should work just fine, assuming you correctly reverse the scores for the recoded version of your variable.
An even faster way, is: compute newscore = X - oldscore .
where: newscore is the recoded/reversed variable name;
X is highest possible score on your response scale plus the lowest possible score on your response scale (e.g., for you: "X" = 4 + 0 = 4);
oldscore is the unreversed/original variable name.
Good luck with your work.
- Badges
- Science topic
- Similar topics
- Statistics
- Survey