UPDATE
so i thought i had found the answer, however the syntax below only works when there is no missing data in the variables. if there is missing data it removes the case from the file. i.e. if 6 cases and one with missing response, it creates a new variable with only 5 cases. does anyone know what can be done to make the syntax work and preserve record of which cases are missing data?
ORIGINAL QUESTION
I have an spss datafile which separated responses from two groups of participants on the same survey question into two variables in SPSS (i.e. Variable 1 = Males responses to question 1, and Variable 2 = Females responses to question 1). Does anyone know of syntax I can use in SPSS to merge these two variables into one variable in SPSS so that the new variable, e.g. Question 1, would contain the responses from both males and females?
I could sort by gender (i have this variable) and then cut and paste the data but am reluctant to do this as it introduces possibility of human error and is not so easily replicable (prefer to have the syntax as evidence of what was done)
Thanks,
Nicola
for example, data looks like:
So I found the answer to my own question. Posted here for interested parties.
syntax is varstocases
varstocases
/make v from v1 v2
/index q (v).
where v is the new variable that will be created
v1 and v2 are the two original variables in the dataset that will be merged to create the new variable (v)
q is a newly created variable which indexes which variable each value in the newly created variable came from (i.e. either v1 or v2).
for example:
varstocases
/make Q1 from Q1_Male Q1_Female
/index q(Q1).
running this script changes the datafile from the question to the one included in this answer. NB: it removes the original variables from the datafile and replaces them with var q. not sure how I can keep the originals.
Hi,
try this - it preserves all cases! It creates the new variable for all cases that are valid either in Q1_Male or Q1_Female.
Q1_all is not calculated if Q1_Male AND Q1_Female are missing (pairwise exclusion), but the case is not deleted!
DATASET ACTIVATE DataSet1.
RECODE Q1_Male Q1_Female (1 thru 2=Copy) (ELSE=0) INTO Q1_Male_temp Q1_Female_temp.
VARIABLE LABELS Q1_Male_temp 'Q1_Male_temp' /Q1_Female_temp 'Q1_Female_temp'.
COMPUTE Q1_all=Q1_Male_temp + Q1_Female_temp.
RECODE Q1_all (0=SYSMIS).
EXECUTE.
Best regards,
Stephan
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS