Hello friends,
I'm trying to compare a set of proportions, each variable has more than two possibilities so, sometimes, I have a 5 x 5 sets.
I need to verify if a proportion in the column is different from a proportion in the row. When we run a test, for example, a z-test, the comparison is among the different columns.
Thanks in advance.
I don't think such a thing is possible.
What is your actual research question? More information needed.
Hi Mehmet,
Thanks for asking. Actually, I have a dataset with MRI scans for both shoulders, symptomatic and asymptomatic. For each side, I have many categorical variables that describe the MRI findings. So, I have information about a tear, for instance, and I have many different types of the tear. I have the same categories for type of lesion for both shoulders, so if I build a crosstab and run a z-test I have a comparison of the proportion of one category (row) inside the categories that are in the column. So, the SPSS run a comparison among the columns, but not between the variable from the first column and the first row. Is it makes sense?
I've just found the answer. I'm using another software now. Minitab seems to be quite simple for some analysis. At least, more than SPSS.
Thanks anyway.
Hi Mehmet,
In Minitab, you can put the number of subjects and in another field the number of occurrence for an outcome. Now I'm trying to figure out how don't manually do it. Because I have to manually put all pairs of comparison, but it's a start, right? I've never worked with Minitab before.
Thanks!
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS