Wielding the Red Pen

Mark Shaver for The Chronicle

February 07, 2011

I've never heard any academics say that they love grading. Yet, as graduate teaching assistants, we talk about it often—which is not surprising, since it is our primary job.

We discuss our grading systems, such as how much value we assign each portion or element of a test or paper. We share the outlandish answers that students offer, like warping General George Armstrong Custer a century back in time and placing him at the head of a charge up Bunker Hill. We compare grade distributions. We joke about simplifying our task by awarding everyone a B, or throwing the papers in the air and giving grades based on where they fall. And when we face a stack of exams over a weekend, naturally we grouse a bit. Yet seldom do we broach the subject of our personal stake in grading.

In my six years of TA-ing—first as a master's student in anthropology, and then as a Ph.D. candidate in history—I spent far more time assessing students' work than I did on any other teaching activity. And I probably learned more from the experience of grading than my students learned from the feedback I gave them. Mostly I learned to use grading to help students learn rather than intimidate them from questioning my assessments.

The first time I encountered Scantron exams, I loved them. Amazingly, that first encounter wasn't until about six years ago, when, as a master's student, I worked as a TA in an introductory anthropology course that drew nearly 500 undergraduates. I loved the ease of grading Scantron exams, which was a matter of alphabetizing the narrow, rectangular forms and then feeding them, with all of their bubbled-in answers, into a machine. As long as the forms didn't crumple and jam the machine, I could sneak in a little reading as I "graded."

But Scantrons had other benefits. Students rarely questioned their grades; rather, they came to see me asking how they could do better next time (though that, too, was rare). The point is, they accepted a Scantron exam as objective and focused on their performance. And yet the exams suffered from the obvious criticisms: They tested rote memorization rather than writing and analysis. They favored a good guess over a command of the material sufficient to encourage creative analysis and use.

In spite of the nearly burden-free grading offered by Scantron, I felt that I was missing out on an important part of the teacher-student relationship: personalized assessment.

So, when I started my stint as a doctoral student, which included a four-year teaching assistantship, and got my first pile of bluebooks full of handwritten short answers and essays, it was as if I had moved up a notch in the academic world. Jeff, the professor I was TA-ing for, described his grading rubric and turned me loose, trusting my knowledge of the subject and my objectivity.

When it came to grading those first exams, as well as history essays and research papers for the next few semesters, I let the red ink flow. I circled every passive-verb construction, drew squiggles through each split infinitive, underlined any awkward phrases. I scrawled comments all over the margins, to call attention to weak or strong thesis statements, abundant or flimsy evidence, and both clear and poorly organized paragraphs. Not to mention all the marks I made signaling right and wrong facts. Great papers bled almost as profusely as poor ones.

When Jeff and I compared our graded work, in an attempt to standardize our efforts, our agreement relieved me. But I noticed that he hadn't written nearly as many compliments and corrections as I had. His grading seemed sparse compared with mine. Why was I writing so much?

There are lots of ways I could have justified my many markings: I was being thorough. The more I crossed out as wrong and the more I complimented as right, the more I was teaching. I was showing that I cared about all the details of a paper or essay exam. All of which were true, to an extent.

By my third year as TA in a few of the same classes, I had caught on to Jeff's more concise method of grading. I began identifying common and repetitive errors within the first few paragraphs of an assignment or exam, then directed the student to look for the same errors thereafter. I noted problems and successes with simple X's and check marks. I wrote "vague" in the margin rather than an elaborate explanation of why the meaning of a passage was hard to understand. And I started withholding my comments until I had finished reading the work, and then provided overarching compliments and criticisms at the end. I did that to save myself time, but I was able to do so because of a new sense of confidence I had in my authority.

As my knowledge of the course material increased, so did my self-confidence. That may be an obvious thing, but it revealed less obvious things about my early approach to grading.

On reflection, I saw all of the slashing and hacking I'd done with my red pen as more in the service of my need for authority than as an effective teaching tool. More specifically, the marks I made on students' work served as a barrier or a fence, perhaps, to prevent students from questioning my assessment of their work and to keep them from challenging my authority, or, more truthfully, the lack thereof. The more I marked, the higher I built the fence. And early on I barbed that fence as much as possible. I feared students' questions about my grading, so I discouraged those questions. I buried them in red ink.

Once I became aware of doing that, I began seeing it happen in other ways.

Listening to my fellow TA's talk about their discussion sections, I heard about what I call "the wall of words." Similar to building red-ink fences in grading, we, as unsure TA's, often talk too much as a way of filling our allotted time in front of students, instead of fostering good discussions with them. By doing most of the talking, by lecturing rather than spurring conversation, we ensure our authority in two ways. First, we prevent students from posing questions that might reveal the limits of our knowledge or challenge our understanding of the material. Second, we are building our authority by filling the room with our own voices, our own knowledge of the subject at hand.

Last semester I taught an American-history survey course on my own and had five teaching assistants working for me. When it came time to exchange graded bluebooks, so that we could standardize our grading, I saw the same struggle for authority played out. My exams looked virtually untouched compared with most of theirs.

As we got to know one another better, I mentioned my thoughts on this to one of the TA's whom I knew well as a fellow graduate student and friend. She quickly agreed that she wrote a lot on exams as a way of justifying the grades, or, put another way, to deter questioning and shore up her authority.

Similarly, when I asked my TA's how their discussion sections were going, I heard a lot about what they had said to students and far less about what the students had to say. Each week we devoted time to figuring out ways to stimulate discussion, and yet each week I got the sense that the TA's did most of the talking. Some of that had to do with the difficulty of running a seminar-type environment with undergraduates who are not specializing in the field or subject. But a large part of that difficulty, I believe, rests in the uneasiness we feel as graduate students guiding undergraduates, which I first recognized through grading—or overgrading.

After serving six years as a teaching assistant and having taught a few of my own classes, I am not sure I have an easy solution to this problem.

I still catch myself filling margins that perhaps are better left unfilled, or talking too much as a way of drowning out a student's question.

But I've learned to do otherwise. I've learned to point out recurrent problems in students' writing once or twice and ask the writer to identify other examples of the problem. In the margins of a test or paper, I pose questions that encourage students to decipher their errors, rather than bludgeon them with my recognition of those errors. I favor a few comments at the end of a paper that assess the work as a whole.

To seasoned graders and teachers, those may seem like obvious techniques. To those of us relatively new to the practice, they are more than techniques. They are at the core of transforming the grading process from one of self-defense to one of promoting students' willingness to question, to be curious.

More students come to talk during my office hours these days. And more frequently than ever, they begin with the introduction, "I don't have any problem with my grade, I just wanted to talk about one of your comments." We tend to talk more about the substance of their work than fuss about point totals.

When I have dared to ask students how much attention they pay to my grading and comments, they almost always tell me that the more I write, the less they read. When faced with a paper drenched in grader's ink, they search for the final grade and rarely consider questioning it. The stifling of questions cannot possibly be good teaching.

David Brooks is an A.B.D. doctoral student in history at the University of Montana. If you would like to contribute a First Person essay to our series on graduate-school work and life, please e-mail your ideas and essays to