The two objectives of this study were to compare the results of the teaching of written composition to high school students, using a problem-solving approach, with the results obtained using a traditional approach, and to determine if a problem-solving approach produces a significant difference in reflective thinking skills. Three null hypotheses and one alternative hypothesis were tested in a Posttest Only Control Group Design, using a sample of 306 eleventh-grade students (six experimental sections and six control sections, each composed of 153 students). Five teachers of junior English instructed both experimental and control classes. Analysis of variance, two-way classification, was used to examine the data; evaluation questionnaires completed by students and teachers were treated descriptively. The study findings indicate that the problem-solving method of teaching composition produced learning of reflective thinking skills to a significantly greater degree than did the control method; it also produced skills needed for writing to a significantly greater degree than did the control method. (Author/DB)