Not Sure Which Stats Test to use in SPSS I have two groups: Students (1) who did and (2) who did not successfully complete an ENG-101 course within two years of taking an ENG-102 course. 
I would like to see if there is a difference in their success in ENG-102 (Success = A,B,C; Unsuccessful = D,F,W).
Further, I would like to look at three groups: 
(A) Students who successfully completed an ENG-101 course within two years of taking an ENG-102 course
(B) Students who did not take an ENG-101 course prior to ENG-102 but had an ACT waiver (cut-off 26) 
(C) students who did not successfully complete ENG-101 within two years of taking an ENG-102 course (No ACT or one < 26).
I would like to see if there is a difference in their success in ENG-102 (Success = A,B,C; Unsuccessful = D,F,W).
Which test would be best to use?
 A: One way to go about solving your first problem would be to define the following two variables for each student: 


*

*An outcome variable Y such that Y = 1 if the student was successful at completing ENG-102 or Y = 0 if the student was unsuccessful at completing ENG-102;

*An explanatory variable X such that X = 1 if the student was successful at completing ENG-101 or X = 0 if the student was unsuccessful at completing ENG-101. 
Then you could perform a binary logistic regression in SPSS to determine whether the odds of successful completion of ENG-102 are different among students who successfully completed ENG-101 and those who did not. 
For your second problem, the outcome variable Y would be defined as above, but the explanatory variable would be defined so that it would have 3 categories - one for each category of students you are interested in. (You could assign numerical codes to each category, such as 1, 2 and 3.) But you could still use binary logistic regression to relate Y to this new explanatory variable. The binary logistic regression model would then allow you to compare the odds of successful completion of ENG-102 among every pairwise combination of groups corresponding to the explanatory variable. 
See here for an example of how to perform binary logistic regression in SPSS: http://core.ecu.edu/psyc/wuenschk/MV/Multreg/Logistic-SPSS.PDF. 
