I conducted an ordered probit regression. As predictors, it included a factor with 4 levels (0,1,2, and 3) and a continuous covariate.
When I look at the resulting parameter estimates, it has an estimate of 0 for the factor level 3, with the note that "This parameter is set to zero because it is redundant".
I have two questions.
First, is it correct to interpret the parameter estimates for the other levels of the factor as relative to level 3? (As in, if the parameter for level 0 is significant, it is significantly different from level 3)?
Second, how can I change which factor level is left out, (thereby choosing what the other factor levels are compared against?)