I have 2 samples (they are social research studies), one large (2000-approx) the other small (56) they are responses to various questions.
I wish to test for significant differences between the two studies, for each question.
Traditionally for 2 large sample I would employ the familiar hypothesis test difference between 2 proportions (with pooled proportion etc.) based around z-distribution approx. to binomial.
However in my case (one large, one small) the small sample does not satisfy the np > 5, n(1-p) > 5 approximation requirement.
Any recommendations how best to proceed?
You may leave the list at any time by sending the command
to [log in to unmask], leaving the subject line blank.