Designed to present a rationale for evaluating the implementation of computer assisted instruction (CAI) in higher education, this paper describes a multi-paradigmatic approach to the evaluation of CAI pilot implementation efforts and its application in various math and English courses at the New York Institute of Technology (NYIT). Students in two math courses--Developmental Math and College Algebra--and four different English courses--Communication Skills, Composition, Business Writing, and Technical Writing--spent significant portions of class time working in a microcomputer laboratory utilizing Apple/Franklin and Commodore 64 computers. Evaluations of instructional effectiveness focused on a process analysis, but a quantitative evaluation was also conducted. Included are discussions of: (1) CAI and student role changes; (2) CAI and faculty role changes; (3) faculty professional development; (4) points unique to the math experience; (5) points unique to the English experience; (6) student attitudes toward computers and math CAI; (7) students' perceptions of word processing in the learning of writing; and (8) unanticipated negative consequences. Findings indicate that: CAI students demonstrated a good conceptual grasp of subject matter; their difficulties were identified quickly; they often worked in pairs, which provided mutual support; and faculty were able to interact individually with students, observe the learning process closely, and intervene directly and immediately. (JB)