Ordered logit or OLS?

#1
Hi there,

I am using survey data and my dependent variable has 5 levels ranging from 'Strongly disagree' to 'Strongly agree.' My understanding is that it is appropriate to use an ordered logit model in this situation, as equal distances can not be assumed to lie between each level of the dependent variable. However, when I compare the results of the ordered logit model to an OLS regression, the results are very similar.

Is there a test I can perform to determine if I am able to use OLS rather than ordered logit in this situation?
 
Last edited: