This paper addresses the methodological issues encountered when attempting to evaluate a college course's effectiveness by determining whether students who enroll in it perform at higher levels than students who did not enroll. The paper identifies resource people who may be helpful in the evaluation, and offers guidelines for the research question, which defines the purpose of the project, and the null or research hypothesis. The necessity of operational definitions is also emphasized, with examples and definitions of key words. Suggested factors to consider in planning the research project include cost, ethics, time, choice of students for the study, availability of students, commitment, faculty, the experimenter effect, student retention during the study, outcome measures and when to examine them, treatment rigor, and necessary sample size. The four research models provided for evaluation of the college success course are: (1) the experiment with randomly assigned partially matched students; (2) the experiment with randomly assigned unmatched students; (3) the post-facto design; and (4) the time series design. Selection of a particular model depends on the collective purpose of the evaluation. The paper concludes with a suggested outline for the ensuing research report.(YKH)