In Defense of Subjective Grading
In Defense of Subjective Grading
What is wrong, after all, with "writing for the teacher"? As traditionally undervalued instructors of everyone's "worst subject," composition teachers become too quickly defensive when their scientific colleagues talk of the reliability of composition grades, retreat too easily into "We know the variation from teacher to teacher is bad, but we're working on it." Without denigrating the work of those who strive to find reliable composition grading methods, I think we can stop feeling guilty about what we do and admit that individuality, lack of uniformity, may be among our discipline's greatest assets.[i]
Our fear of our own subjectivity seems well-founded. Many composition experts would agree with E.D. Hirsch that "Until we have reliable means of rating the quality of a student's prose, we lack a sound basis for determining the teaching methods which will raise that quality most efficiently.... We cannot progress in other lines of pedagogical research until we solve the assessment problem."[ii] How can we teach if we can't seem to agree on how and what we desire as the products of our teaching? That we don't agree on how and what we grade has been painfully demonstrated, most convincingly by Paul Diederich.[iii] Any writing staff that subjects itself to grading sessions knows that this variation exists, but the traditional response to our apparent unreliability has been to hide the variation, bemoan its existence, or talk about everyone "getting tough."
Diederich has shown conclusively that the writing staff can be trained to grade reliably; his success with staff grading robs us of the traditional response to the grading problem: "There just isn't any better [read 'more reliable'] way." His methods yield results which are reproducible and more valid than those of objective tests, and they help students view their teachers as coaches rather than judges, because teachers no longer wield the dreaded grade.
Yet despite the success of staff grading, many English departments continue to encourage teachers to grade in the traditional way, each teacher evaluating his or her students' work. It may be that these departments have examined Diederich's studies and worry about the time training sessions would take or the problems heavy staff turnover would create, or sympathize with the inconsistent student who may not work well under time pressure, with assigned topics, or on a particular day, and would therefore suffer in most staff grading programs. Some departments may resist the change to staff grading for less thoughtful reasons: the transition would be unsettling, time consuming, or politically dangerous; inertia is too strong.
Or maybe we sense that grading reliability, like any kind of enforced uniformity, would take its toll.
Most writing in educational settings continues to occur in the context of a long-standing relationship between teacher and student. The tendency, especially in recent years, has been to downplay this relationship by encouraging writers to ignore the fact that the teacher will probably be a paper's only reader, and to fight the implications of the bond between student and teacher by trying to depersonalize the grading process, removing student names from their papers, grading according to an established set of priorities, or passing final papers on to other English teachers so that the paper itself can be judged without reference to the personality of student, teacher, or class.
These approaches may have value in certain situations, but they ignore many of the benefits that can be derived from the close student-teacher relationship. A student working with a teacher is likely to write for an audience, and that sense of audience is crucial to any student's becoming a successful writer. Very few of our students will go on to write for readers as anonymous as those at Educational Testing Service, and we may find that training writers to write for unknown anyones--something that staff grading may encourage--produces writers similar to those who work in the government, writing for that great faceless "general audience." Our students will more likely be writing for a particular professor or boss or committee or group of co-workers. Students who learn to identify the quirks and desires of a particular teacher are not just "polishing the apple"; they will be ready in the future to write for the boss who loves passives and obfuscation, the professor who hates the first person, or the school board impressed with pedantry. The writer aware of his or her audience, writing for that audience, will be likely to get the job, the raise, or the "A" before another writer technically superior but unable to adjust to suit a reader.
Of course, writing with the teacher in mind as audience or editor differs substantially from writing what the teacher is perceived to "want." Although some teachers do successfully assign subjects, only rarely can we justify judging a student's ideas as "right" or "wrong"; our subjectivity should not extend that far. We judge form and style, the presentation of ideas, the methods and logic used to reach a conclusion, but seldom the conclusion itself. We encourage students to think well and say something, but we should never assign that something a low grade just because we disagree with it.
When dealing with content, the teacher's most productive role may be as an elusive devil's advocate, taking the stance of a hypothetical audience for the piece of writing. Students who can discriminate between the demands of the intended audience and those of the editor can aim a paper at an imagined audience, keeping the teacher-as-editor in mind as a side concern. Less sophisticated students may not be able to write with an audience more abstract than "grade-giver" in mind, but imagining the teacher as audience is better than imagining no audience at all; it will help the student to, in Linda Flower's terms, move from writing writer-based prose to writing reader-based prose.
The problem we have traditionally faced in writing classes, then, is not that students write for the teacher, but that we have refused to admit that fact. In our desire to appear as believers in the absolute of "good writing," we have deceived students into thinking that everyone agrees with each of us. Students taught the "right" way to write by Ms. X will naturally be frustrated when Mr. Y's "right" way is different, and last year's A work gets Cs this year. No wonder students become suspicious about their English grades.
If, on the other hand, Ms. X said, "I value active sentences and smooth transitions," and Mr. Y, "You must have a clear thesis statement and a personal writing style," students would realize that they were being taught not two different "right" ways, but two different subsets of the general goal "good writing." They would understand why the same paper could be graded differently, and although they might agree more with Ms. X's or Mr. Y's emphasis, they would be able to learn from both without becoming cynical, and they would be ready for Professor Z to pay attention to something completely different. Compared to students kept in the dark about our differing standards, students able--and, indeed, encouraged--to write differently for particular teachers will not only be more flexible and aware of their audiences in the future, but will most likely learn more from their various teachers because they won't be confused trying to reconcile what appear to be contradictory dicta.
This acceptance of diverse approaches and evaluations in the writing classroom cannot be absolute, for certain emphases can harm students' writing or destroy any pleasure students may derive from it. Focusing on margin size, typographical perfection, or rigid adherence to a prescribed form would seem out of place in most courses, as would an emphasis on nothing but content, although students taught under either regimen might learn the crucial lesson about audience. But the issue of such idiosyncratic emphases is more theoretical than real. As S.W. Freedman has shown, most teachers value the same elements in a paper: content and organization.[iv] Admonishing teachers to make explicit their writing values might actually reduce the differences between teachers, because teachers would become more introspective about what they do in class, and those who tend to grade spelling, punctuation, and other easily assessed features might realize, and try to eliminate, the disparity between the values they truly believe in and those that seem to inform their grading practices.
Those who worry about teacher bias for or against particular students should analyze that bias before condemning it. It would be absurd to claim that any writing grade reflects solely a writer's ability--class participation, attendance, attitude, improvement, and changes in the writer's composing process may influence a teacher, and in many cases probably should. The brilliant writer who is the class pest may suffer from the personalized grading relationship, but on occasion such "unfairness" may help both the student and class, as long as the reasons for it are made explicit. The benefits of a familiar teacher's superior ability to judge purpose and intent outweigh whatever loss of objectivity such personalization of grading may entail. And only a teacher who knows a student well can look closely at a student's writing process and help that process evolve, an evolution which may be central to a writer's improvement.
Acceptance of the subjectivity of composition grading leads to several implications for the classroom. First and most important: we must make clear to students our own sets of priorities, letting them know what we will be looking for when we grade. We can elucidate our desires both in class, as we spend time on those elements we expect to see in their writing, and when we comment to students about their papers, letting them know what we feel they should work on first and with most effort.
As a logical extension of this attitude, we should put off grading student papers for as long into the term as possible, so that students will have time to learn the system, figure out what the teacher's writing priorities are, and respond to them. A grade on the first paper might be the most effective method to let students know that they have to mend their ways quickly, but the resentment and anxiety such a quick grade may foster make such grading a pedagogical mistake.
Unsettling as it may sound, we must impress upon our students that "good writing" is at least partially relative, and to drill into them the maxim "know thy audience." We can all agree that clear, concise writing with active verbs outshines murky, passive jargon, but politicians, boards of directors, and perhaps some of our colleagues in other fields may disagree with us. Rather than send our students out as lively idealists in a world of passives and casuistry, we should prepare them to be flexible. Flexibility does not mean lowering or changing our standards; it simply means letting students know that these standards are ours, and that others exist, and will continue to exist, despite all efforts to eradicate them.
It should be clear from the above that I feel teachers should grade the papers of their own students. But if a department uses group grading techniques to regularize its biases and establish a uniform set of priorities, it must make every aspect of the standardization explicit, so that all composition teachers can pass the priorities on to their students. Departments which set priorities may need to train teachers specifically to teach and evaluate certain writing skills; otherwise, students' successes or failures at reaching certain uniform goals may reflect their teachers' understanding of those goals rather than their own.
Despite all attempts to objectify the teaching of writing, I don't think writing teachers need to worry about computers and teaching machines stealing their jobs in the near future. As long as there are writers, unique individuals will continue to generate unique sentences which require unique responses. None of us would want to destroy the individuality writers put into their papers, and we should be equally wary of destroying the individuality with which we read them.
[i].Throughout this paper, I will use "reliable" and "valid" in the narrow technical sense--"reliability" referring to the agreement from assessor to assessor about how a particular writer or a particular work should be evaluated; "validity" indicating that the assessing method accurately tests the skills it was designed to test.
[ii].E.D. Hirsch, Jr., The Philosophy of Composition. Chicago: University of Chicago, 1977, 11.
[iii].See Paul B. Diederich, Measuring Growth in English. Urbana: NCTE, 1974.
[iv].S.W. Freedman, "Why Do Teachers Give the Grades They Do?" CCC XXX May 1979, 161-64.
Originally published in North Carolina English Teacher XXX: 4, Summer 1983.