Whether to categorically code likert predictors in binary logistic regression in SPSS and whether to use the Enter method? I am working on a project trying to predict the antecedents of adoption for a new technology. I am using IBM SPSS software but I am not confident with statistics analysis.  
I have 10 independent variables (each of them is formed of 3 likert scale items. I just summed the three items to compute each latent variable) and one dependent variable (a dummy variable with 0 not adopting and 1 adopting).
I plan to perform a binary logistic regression. 
My questions regard the SPSS software functions and they are: 


*

*When I put my independent variables in the covariates box should I also select the option Categorical Covariates because they are not continuous variables?

*In the option Method should I leave the option Enter selected or is preferable one other method?

 A: Your second question is relatively straightforward; enter is better than any automated method, if you  have some sensible order to enter them in or if you think they all should be entered. See many posts on here on variable selection. Note that it is only straightforward because of how you worded it: Variable selection is not easy.
Your first question is trickier, but if each independent variable is the sum of 3 Likert scales, and each Likert scale ranges from 1 to 5 or 1 to 7, and the range of the sums is adequate, then a linear option is probably better than a categorical one, even though it is not, strictly speaking, correct; by summing the Likert items you are already assuming they are interval scaled (otherwise adding makes no sense), so, in for a penny, in for a pound and go for it.
But if the sums are very oddly distributed, then a linear choice might not be so good. However, categorical might not be so good either. You might need something more complex. I don't know what's available in SPSS. 
