A study explored the effects of word processing on the revision strategies of 20 advanced college freshmen enrolled in a required writing course. Subjects were divided into two groups of 10, each of which alternately wrote a series of four essays on and off the computer. For each essay students submitted three drafts, which were analyzed for revisions. Each student produced two essays with word processing and two with pen and typewriter. In addition to undergoing text analysis, the essays were judged by trained raters using an analytical scale, so that the quality of the essays could be related to the number and kinds of revisions. Results of the analysis of 4,048 between-draft revisions of 80 essays indicated that writing on a computer did not lead to increased revision--at least not for these able students. The finding that there was no positive relationship between extensive revision and the quality ratings, however, suggests that manipulating text for the sake of revision has little value for students or their writing. Nor did the students make different kinds of revisions with a computer than they did with pen and typewriter. The essays produced with pen and typewriter, moreover, received ratings comparable in quality to those produced on the computer. (Author/HTH)