Al Rull posted on Thursday, November 04, 2010 - 10:48 am
I would like to estimate simultaneously with a path analysis the influence of six success drivers (x1-x6) as independent variables (all are metric scaled) on four dependent variables (2 of them are metric scaled: y1 y2 and two binary scaled: u1 u2). Besides, I would also like to estimate within the model the influence of the dependent variables y1 --> y2; y2 --> u1 and u1 --> u2. My question is now, how it is possible to estimate the influence of one binary coded variable (i.e., u1 (trial purchase no = 0; yes = 1) on the other binary variable u2 (repeat purchase with no = 0; yes = 1) within this model?
u2 ON u1; is not working when these variables are coded as NOMINAL.
*** ERROR in MODEL command A nominal variable may not appear on the right-hand side of an ON statement: U1
Does there exist any alternative which may work? Thank you very much in advance. Al