A study investigated the effects of instruction on students' persuasive writing at two grade levels (third and fifth), and determined the categories and types of written persuasion used by students at four grade levels (3rd, 5th, 10th, and 12th). The first objective, determining instructional effects, was accomplished by specifically instructing 139 students in 8 classes in a southern California school in oral and written argument/persuasion. There were no significant main effects for instructional strategy or for the presence of the oral interaction component. The second purpose of the study (to categorize persuasive responses from 308 students and to determine age and gender differences, if any) was addressed using D. M. Weiss and J. Sachs' (1991) classification system. There was no significant main effect for gender, but there was a significant main effect for age. Students in grade 3 did not use compromise at all, while 11.1% of the 12th grade students' responses were compromise. Simple statements were used more by students in grade 3 than by students in grades 5, 10, or 12. There were also differences in the types of responses by grade level. (Two tables of data are included; 21 references and 2 appendixes, presenting the scoring guide and a description of persuasive categories, are attached.) (Author/SR)