I have two groups of students who answered a question on the pre-survey about their savings behavior, i.e. choose one answer a/b/c/e/d/f: (a) Save the same amount each week, (b) Save varied amount each week, (c) Save only when have money, (d) Save only when want to buy something, (e) Do not save.

One group of students (intervention) then went through in intervention to instill positive savings mindset, and then they answered the same question again as a post-survey. The other group was the control group.

I am wondering how to analyse this. I used the Chi-squared or Fisher's exact test to compare between intervention and control groups at each time section, (1) pre, (2) post.

But how do I compare within/between the groups for the paired pre-post longitudinal results? I read up on McNemar's but this seems to be only for 2x2 table. I have a 2x5 table here for each of the intervention and control pre-post?

Also, the intervention group sample size is small, 24 students, so I might need an exact paired test. I used the Fisher's exact for the pre and the post cross-sectional analysis, because some categories have less than 5. The control group size is larger, about 70 students, but some categories might still be less than 5 and so needing an exact test.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.