In my logistic regression model, I have one independent variable that has a B value of 0, p = 0.006 and exp(B) = 1. Goodness-of-fit measures indicate that the model is acceptably "good".
Intuitively, I would imagine a significant variable to have a non-zero coefficient and some effect. Statistically though, I am not sure.
So, I am not sure if the outcome on this variable is an acceptable outcome. If yes, what is the explanation? If this type of outcome [B=0, p=0.006, exp(B)=1] is not acceptable, what could it be symptomatic of -- what is the problem?