Interactions in Binary Logistic Regression (SPSS)

[jhunter]

Hi all,

I am trying to figure out how best to calculate and interaction term for two dummy variables.

I have Shape (S=0/F=1) and Color (BW=0/G=1). I cannot simply take the product of these variables--if I do I end up with:

S*BW=0
S*G=0
F*BW=0
F*G=1

Has anyone had experience with this issue before? Any ideas?

Thanks

[epidem_delhi]

As far As i know, it would not make any difference as to what coding you use for categories for calculation of Beta coefficients. It will be at the time of interpretation that it becomes tricky. In SPSS, just select the two variables for which you need to check for interaction and Click on th [> a*b>] button to include them as interaction terms