linearity of logit - extremely high Exp(B)

[Chui Ling Liew]

I am trying to test the linearity of logit assumption. Previously there is 0 cell issue, but I have re-coded all 0 cells to a small value (0.0001) to enable the Ln transformation. Still, the warning sign appear but the test seems ok. 

Question:

1) The only solution for violation is to transform the variable from continuous to categorical?

2) One Exp(B) (ACSIZE) is extremely large (15203.835). What does this means? What caused this?

3) There is one interaction has Exp(B) = 0.00. Something wrong here?

Thanks. 

p/s: attached also the data and output file. 

[<Joanne Tse>]

Hi,

 

Our office will be closed for Martin Luther King Holiday, 1/17/11.  I will reply to your email on Tuesday.

 

Thank you.

 

Joanne Han

Marketing Research Manager

UCLA Extension

www.uclaextension.edu