Hello fellow listers. Another newbie question: I have two corresponding variables (b101sat c101sat); per case, only one of them has a value. In order to condense them into one variable (b101all) I use the following syntax RECODE b101sat c101sat (SYSMIS=96) (0=0) (1=1) (2=2) (3=3) (4=4) INTO b101miss c101miss . VARIABLE LABELS b101miss 'Overall Satisfaction Employees w/ missing values' /c101miss 'Overall Satisfaction Mgmt w/ missing values'. EXECUTE . COMPUTE b101all = b101miss + c101miss - 96 . VARIABLE LABELS b101all 'Overall satisfaction of all respondents (employee &'+ ' mgmt)' . EXECUTE . I have about 200 of those corresponding pairs and cannot imagine that this makes the sheet very clear/ practicable ... though it worked. Is there a way to avoid the detour of displaying the two (useless) new variables (b101miss c101miss)? - I tried simply overwriting the original variables (b101sat c101sat) with the following syntax RECODE b101sat (SYSMIS=96) (0=0) (1=1) (2=2) (3=3) (4=4) . EXECUTE . RECODE c101sat (SYSMIS=96) (0=0) (1=1) (2=2) (3=3) (4=4) . EXECUTE . COMPUTE b101all = b101sat + c101sat - 96 . VARIABLE LABELS b101all 'Overall satisfaction of all respondents (employee &'+ ' mgmt)' . EXECUTE . But this left me with ALL cases as 'system missing' in the computed variable (b101all) - does anyone see my blatant (?) mistake? Any help greatly appreciated- and many thx for your patience. Sabina