Writing and Scoring Guides

Kimberly Swygert has a post on the SAT essay. The core problem in the post is the difficulty of scoring writing. Indeed, scoring writing is not nearly as transparent as scoring the more objective kinds of test questions. There is a movement to develope better scoring guides, but I think that can only go so far. I think scoring guides only conceal just how subjective writing is. In Bloom's Taxonomy, Evaluation is the highest level, and so must involve cumulative disagreements about choices made at lower levels. Yet, good writers can generally distnguish between low, middle, and high quality writing. Getting more accurate is probabaly an illusion. But is it even neccesary? Further, different instructors will emphasize different writing goals, which is said to undermine validity of the scoring, but is it? In the real world, writing requirements are often specific and sometimes arbitrary. Is learning to please an instructor any different than learning to please and employer? Indeed, it may be a valuable skill. It would be nice to imagine an ideal rhetoric to aspire to, but its also not how the world actually works.

When teaching writing, the scoring guide has some uses, but its not by itself a teaching tool. Much more useful are examples of high quality writing and a process of revision and critique. Students should have access to one to three pieces of quality writing for every piece of writing they make (this ratio can decrease as you move into more and more advanced students) of their own. Revision should go beyond a single re-write to two or three revisions. The skills reqired at the third revision are different from those at then first. All require practice. Students should review each other's writing in a randomly or systematically selected process, so that students recieve a wide critique. With such criteria met, a scoring guide can be useful to guide students to the next step in quality. Without models, revision, and critique, a scoring guide creates a pseudo-objectivity.