This study was undertaken to determine whether significant differences exist in the performance of students based on three methods for teaching and evaluating writing: faculty review, group review, and Calibrated Peer Review (CPR), an Internet software program designed to teach writing. Classes were randomly assigned a method; 148 students were given a pretest, treatment, and posttest. Each is comprised of four cases requiring students to prepare two documents (letter or memo) adapted to four situations: direct request, goodwill, negative, and persuasive. Students who had the faculty review method of evaluation scored higher, but not significantly, on all four documents. No significant differences were found among the teaching and evaluating methods for the goodwill, negative, and persuasive messages; these findings seem to support the conclusion that any of the methods will enable faculty to help students become proficient writers. (Contains 4 tables.)