I have three categories of data (no change, uplisted, down listed). For each category I have a value for the frequency of occurrence. How do I compare these frequencies?

It is essentially a one-way ANOVA type analysis that I want to do, but since the data are frequencies not means this test cannot be performed.

I have tried a chi-square analysis (observed compared to expected frequency) but this does not say where the significant difference is, only that there is a difference somewhere. Is there a post-hoc test that could be used to find this?