Someone asked me by email about the advantages and disadvantages of various statistical models (multinomial logistic regression, ordinal logistic regression, and structural equations models). This is a somewhat difficult question to answer by email, but as a general rule, I think that people worry too much about the particular model that they choose.

So rather than deciding which among these models is best, perhaps you should just choose an approach that produces easily understood results. In other words, you don't have to have a perfect statistical model. You just have to have one that helps you get published. As long as you avoid using a very bad statistical model, you should be just fine.

George Box, a famous Statistician, has a well known quote about this. He says

All models are wrong, but some are useful-- en.wikiquote.org/wiki/George_E._P._Box

You can find an earlier version of this page on my original website.