How to make a prediction from baseline categorical variable – Logistic Regression - SPSS

Question

For example:

Let‘s say I have a categorical variable 'Hand' with two values 0=”Left” and 1=”Right”. DV is 'ReactionTestResult': 0=”Fail”, 1=”Pass”. One of the categories from variable 'Hand' will automatically be chosen as baseline and won’t be included in the model, thus I will end up with only one B coefficient after regression is complete. Let’s assume that Left was chosen as baseline. Then equation for predicting would be log(𝑝̂/( 1 − 𝑝̂ )) = Bconstant+HandBcoefficient*Hand.

However, because left hand is baseline (and thus excluded from the model) and I only have coefficient for right hand, how can I then proceed to make predictions about how likely left handed participants would be to pass the supposed reaction test? Do I somehow need to derive coefficient for baseline, perhaps using Exp(B) values or something? Or is there nothing I can do and I simply can’t make predictions from left handed participants?

Any help is appreciated. Started learning about logistic regression fairly recently and this is confusing.

Answer 1

When you have binary predictors, the intercept is the value of the DV when the IV is 0. In your case, the intercept will be the log odds of passing the test for left-handers.