I have to compute a variable using the median from separate items to create a subscale, l've used the
'compute variable' function and selected median to do it. It worked for my first subscale but gave incorrect values for the second. I'm assuming it's because of missing values but ideally if SPSS considers them as zero l'm still supposed to get the result.
How do I go about this? Would mean a lot if someone could help me out!
Hello Fiza,
In spss, the syntax:
compute newvar = median.3(x1 to x5) .
would compute and post median scores for all cases having at least 3 of the scores with non-missing values from the set x1-x5 in the data set.
Cases with two or fewer non-missing scores would have a missing value recorded for the created variable (here, "newvar").
SPSS does not consider missing cases to have a value of zero. (Generally, doing so would not be a good ideal.) However, if that is your intention, you should first recode missing values for the target set of variables to zero.
Good luck with your work.
I think you have to try some alternative methods. This method is much criticized in the literature
- Badges
- Science topic
- Similar topics
- Philosophy
- Esthetics
- Interpretation