Process and form in the teaching of college-level writing: From polarization to synthesis

The traditional paradigm in rhetoric and composition has been under assault since the 1970s for overemphasizing form and failing to deal adequately with process. Only recently has the field considered the potential value of a form/process synthesis, and only a few studies have investigated methodologies that combine form with process. Even these studies were found to be limited, both in their conceptualization of this synthesis and in their failure to account for form-based methods and process approaches within the same study.

While much of the literature on process-oriented instruction downplays the potential use of rhetorical models, assessment practices continue to deemphasize the importance of process. In most studies, performance on an impromptu writing test is the sole measure of a student's writing ability. No studies were found that systematically investigated the correlation between first drafting ability and revising ability to assess whether impromptu test scores are an acceptable measure of writing quality.

This dissertation has attempted to articulate more fully the relationship between form and process. It used an experimental design to address the following questions: (1) How do three different types of instruction (traditional form-based, process-centered, and a form/process synthesis) affect writing performance? (2) Will the measures of first-drafting ability and revising ability vary significantly in identifying the differences among the three treatment effects?

The findings of this study indicate that the formative process approach was the most effective of the three treatments. This difference was not evident on the impromptu pre- and post-tests which measured first-drafting ability, but showed clearly when the revised pre- and post-tests were compared. These results not only confirmed the greater effectiveness of the formative process method, but also demonstrated the ability of revision tests to reveal important information which may not be evident to an evaluator relying solely on impromptu tests.