Ordinal variable difference between pre and post- age IV

#1
I have a self-efficacy scale ordinal 1 to 5. I ask this question in pre and post tests. I would like to know if age affect the change in self-efficacy between pre and post.
My attempt was to gen efficacy _change= efficacy_post - efficacy_pre. and the just regress efficacy _change age

Is this valid?

Thank you,
Marvin
 

hlsmith

Less is more. Stay pure. Stay poor.
#2
This is an approach given the error terms in the linear regression model are normally distributed.
 

hlsmith

Less is more. Stay pure. Stay poor.
#4
Run the linear regression model, test the normality of the error (residuals) terms. Do this with either a statistical test or by graphing them.