3 Arm Study: Categorical Data needs a P-Value!


I conducted a 3 arm study and now I am trying to find signficance/lack of signififance. Some data I gathered, such as age was easy to calculate via ANOVA tests. But now I have smoe categorical data that neds to be analyzed. For example, I examined the quality of my procedure and it was ranked as 0 = good, 1 = ok, 2 = bad (on MS Excel). So now I have 3 groups of dummies I used in this case, but there all inputted with 0, 1, 2s and I need to calculate a p-value. Please let me know what test to do. Thank you very much. A regular chi-sq test yields this message:

Analysis of Contingency Table: Overview
Chi-square calculations are only valid when all expected
values are greater than 1.0 and at least 20% of the expected
values are greater than 5. These conditions have not been
met, and thus the chi-square calculations are not valid. To
avoid this problem, combine two or more rows or columns.

Thanks in advance. For any more information please ask and I'll post it accordingly.