Procedures for Evaluating Writing: Assumptions and Needed Research

Abstract

NOw THAT ALMOST EVERYONE-parents, teachers, researchers, and politicians alike-is concerned about the quality of at all levels of our society, it seems appropriate that we examine critically the procedures by which we assess the quality of writing. Our ability to resolve the current writing crisis will depend upon our ability to make valid, useful judgments about the that has prompted us to claim that we do, in fact, have a crisis. In examining our current procedures for evaluating writing, we must, of course, ask whether those procedures can be used reliably to accomplish such purposes as rank ordering responses to a given assignment and predicting students' success in future English courses. But even more important, teachers, researchers, and evaluation specialists must be concerned with three additional questions: 1) What assumptions are implicit or explicit in our evaluation procedures? 2) Are those assumptions consistent with current discourse theory? 3) Will the results of using these procedures help us with the problem of improving students' writing? With these three questions foremost in mind, we shall examine four different approaches to evaluation: The General Impression scoring procedure used by Educational Testing Service (ETS) in scoring the English Composition Test, one of the achievement test series of the College Entrance Examination Board;1 the Analytic Scale developed by Paul B. Diederich;2 the assessment of relative readability pro-