multiple regression analysis on likertscale

#1
Dear All,

I am doing the analysis of my data obtained through a questionaire. i want to find the relationship of different forms justice percieved by the phd researchers (predictors) and the commitment of students with the university (independent variable) where as mediator is LMX(leader member exchange, ie relationship and qualities of supervisors with the students), I also want to note down if there is any significance relationship present in case of demographic variables which are Age, Gender and Years of working at university which are not in lickert scale. The problem is my age variable comprises of Age= 24,25,27,28,29, 30, 31,33,35,38, 39, 44 and 45. similarly years of experience are 1,2,3,4,5 and 6 and the Gender is male and female.
i am confused here how to add these three demographic variables (predictors) in stepwise regression.....i am using SPSS.

Best Regards

and thanks

K
 
#2
well, if you are asking only for demographic variables,
i prefer enter method, rather than step wise.
the reason is, in step wise, we are entering one independent variable each time in the equation and see the effect. but that is done to compare the strength of independent variables.
for demographics, since they are not important, enter method is useful.

by the way, even for multiple independent variables, i prefer enter than step wise method.

if your question is how to proceed in spss, : analyse>>regression>>linear
and you will have a selector box of different methods.

what i did not get through your question is the dependent variable. what is your DV? fyi, IV and predictor are the same