I wanted to investigate if there was an association between 3 categories (good compliance, average compliance and bad compliance) and 4 populations (no studies, medicine 1ste and 2nd grade bachelor, medicine 1st, 2nd, 3th master, other university studies). So I used a chi square test to do that. But my problem is that the requirements are not ok. 40% of my expected counts are <5, what can I do to solve this problem other than taking two populations togheter? I really need a result because it is for my master thesis.
Thank you in advance!