Understanding ordinal regression

#1
Hi!

This is a really simple question, but one for which the answer continues to elude me!
I understand that in a binary logit regression the resulting logit is the probability of an event occurring, defined as

Ln(P/1-P)

However, in an ordinal logit, in which there are multiple categories for the dependent, what does this represent?

Specially I am struggling to understand how an ordinal IV would be interpreted as affecting the ordinal DV.

Any help or advice is welcome, I am familiar with linear regression and attempting to teach myself some more complex models!

Thanks,