Thinking Through Controversy: The Effect of Writing on the Argument Evaluation Processes of First-Year College Students

Author: Charles Hill
Degree: Ph.D. in Rhetoric, Carnegie Mellon University, 1992

This dissertation reports on a study designed to answer two questions: (1) How do college students and writing instructors evaluate written arguments? (2) What are the effects of writing on students' argument-evaluation processes?

Twenty first-year college students and 10 writing instructors were first questioned about their opinions on the issue of legalizing drugs. They then read two argumentative essays (pro and con) on the issue of legalization. The participants were asked to evaluate the arguments in the essays, and to verbalize the criteria they were using to evaluate each argument. Their comments were recorded and transcribed. The students then worked for two more sessions. In the second session, 10 of the students answered a set of critical thinking questions about the essays. The other 10 students wrote an essay describing what they felt were the weakest and strongest arguments in each essay, and defending their choices. In the third session, all of the students read and evaluated the essays again. A coding scheme was inductively derived from the transcripts in order to determine what evaluative criteria the participants were using. This scheme consisted of categories of argumentative criteria, such as Logic and Support; Comparison with the Reader's Prior Beliefs; and Probable Audience Reactions. There was a significant difference between the criteria- using patterns of the two groups. For instance, students made more comments about how other readers might respond to the arguments, while instructors commented more on matters of logic and support. Both groups made a large number of comments in which they compared the arguments in the essays against their own prior knowledge and beliefs.

The instructors rated the pro-legalization essay higher, while the students preferred the anti-legalization essay. A descriptive analysis of the transcripts shows that this was at least partly caused by the differences in the criteria that the two groups were using. The writing tasks had no significant effects on the students' evaluative ratings of the essays. However, after writing, students made even more comments about audience and even fewer comments about logic and support than in the first session.