Hi everyone,
I want to combine two variables in spss, one is Twitter Use measured by yes/no with yes being 1 and no being 2. The other variable is Facebook Use measured by yes/no with yes being 1 and No being 2 as well. I want an overall variable that identifies whether the respondent uses social media or not.
My thinking is that I go to transform, sum the two variables, and then change the values of the new variable so that 2=Yes, 3=Yes and 4=no. Obviously this leaves me with 3 categories though, and I want only 2. How do I then recode this so that 2 and 3 = Yes, and 4=No. Is it recode into different variables? Would anyone know how to do this on spss?
Sorry for the long question, please tell me if this is incorrect or whether I have don't something very silly.
Many thanks,
Marnie (Novice spss user)
Hello Marnie,
This syntax should do the trick (though, I am recoding the combined variable such that 1 = Yes, uses either or both; 0 = No, uses neither). Be sure to check that the variable names below match with what your data set contains! The variable, xsum, is temporary, and may be deleted after running the syntax.
compute xsum = Twitter + Facebook.
recode xsum (sysmis = sysmis) (4 = 0) (2 thru 3 = 1) into use_any_social .
value labels use_any_social 1 "Yes" 0 "No" .
execute .
Good luck with your work.
David Morse Hi David, thank you that is extremely helpful. Im just wondering whether I need to remove all system missing values for all my variables? I googled and some said this may lead to bias. Would you be able to advise whether I should be removing all system missing cases?
Many thanks,
Marnie
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS