High Stakes and Low Stakes in Assigning and Responding to Writing, NEW DIRECTIONS FOR TEACHING AND LEARNING, no. 69, Spring 1997 © Jossey-Bass Publishers
Author: Peter Elbow
0 General Document comments
0 Sentence and Paragraph comments
0 Image and Video comments
This essay argues that we will have an easier and more productive time with student writing if we make a distinction between high stakes and low stakes assignments and also between high stakes and low stakes ways of responding to student writing.
As I try to understand my own experience of writing and the experience of my students and as I try to plan my teaching, nothing has been more useful to me than the simple and crude distinction between high and low stakes writing— the question of how much a piece of writing matters or counts.
Assigning Writing
The goal of low stakes assignments is not so much to produce excellent pieces of writing as to get students to think, learn, and understand more of the course material. Low stakes writing is often informal and tends to be graded informally. In a sense, we get to throw away the low stakes writing itself but keep the neural changes it produced in students’ heads. High stakes assignments also produce learning, but they are more loaded because we judge the writing carefully for soundness of content and clarity of presentation.
It’s obvious why we need high stakes assignments in our courses. We can’t give trustworthy final grades that reflect whether students actually understand what we want them to understand unless we get them to articulate in writing what they have learned. If students take only short-answer tests or machine-graded exams, they will often appear to have learned what we are teaching when in fact they have not.
Am I saying that if students can’t explain something in writing, they don’t know it? Not quite. That is, I acknowledge that some students can understand something well and yet be hindered from explaining it in writing because of their fear of writing or lack of skill. In fact, it sometimes happens that we understand something well that we can’t even explain in speech—much less in writing. Nonverbal knowing is most obvious in realms like music, art, and dance (mathematics?), but it can occur in any realm. That is, we can know something at a felt, nonverbal level before we find words for what we know.
But even though students can sometimes know things they can’t explain in writing, I still argue for high stakes writing. I think good college grades should reflect more than nonverbal and nonwritten understanding. They should also reflect the ability to convey that understanding in writing. (Conceivably we should relax this demand in music, art, and dance classes.) I hasten to add that my tough position rests on two gentler premises. We should honor nonverbal knowing, inviting students to use low stakes writing to fumble and fish for words for what they sense and intuit but cannot yet clearly say. And if we assign lots of low stakes writing, students are much less liable to be held back by fear or inability to put what they know on paper when they come to high stakes writing.
Students may complain, “But how can you grade on the basis of writing when this isn’t a writing course?” We mustn’t forget here a basic pedagogical principle: we are not obliged to teach everything we require. We don’t teach typing, yet we often require it. Must we stop requiring skilled reading unless we explicitly teach it? Besides, if we require students to explain their learning on paper, we will be doing a big favor to our campus writing program and writing teachers. Writing courses only work well if students need writing to prosper in their other courses. (For more about assigning high stakes writing, see, in particular, Chapters Five and Six.)
Importance of Low Stakes Assignments
Writing feels like an inherently high stakes activity—especially because most people learn and use writing primarily in school, where it is virtually always evaluated, usually with a grade. Writing tends to be used for more serious occasions than speaking. (“Are you prepared to put that in writing?” ) Speech feels more like a low stakes activity because we learn it in the home and on the playground and use it casually everywhere. We don’t usually think of our speech as being graded.
But speech can be used in formal and evaluative settings—as when we are interviewed for a job or give a talk. In fact, if we pause and reflect for a moment, we will realize that our speech is almost always evaluated, even if not formally graded. How we talk and what we say are probably the main basis on which people we meet look down on us or are impressed with us.
And writing can be used informally, even casually, and in a nonevaluative setting. In truth, if we are looking for the best possible low stakes arena for lan-guage—for using language to learn, explore, take risks, or commune with our-selves and not have our language be evaluated—writing is much better than speaking. Writing permits us to keep our words private or to revise them before showing them to anyone else. Speech is riskier because it is almost always heard by someone in its first bloom; it can never be taken back.
In this volume, Toby Fulwiler, Art Young, and M. Elizabeth Sargent in particular (Chapters Two, Three, and Four, respectively) describe low stakes writ-ing assignments: frequent, informal assignments that make students spend time regularly reflecting in written language on what they are learning from discussions, readings, lectures, and their own thinking. These informal pieces of writing are sometimes done in class and sometimes for homework. These pieces are low stakes because individually they don’t have much effect on the final grade. Teachers tend to distinguish these assignments by calling them not essays but quickwrites, letters, freewrites, thinkpieces, or inkshedding. (When we require students to turn in a draft of a high stakes essay a week or more before the final version is due, the draft tends to function as a low stakes piece.) Stephen Fishman and Anne Herrington (Chapters Five and Six) describe a mixture of high and low stakes writing assignments.
Because it is so ingrained to treat writing as a high stakes activity, especially in schools and colleges, I want to summarize here some of the special benefits of low stakes writing.
• Low stakes writing helps students involve themselves more in the ideas or subject matter of a course. It helps them find their own language for the issues of the course; they stumble into their own analogies and metaphors for academic concepts. Theorists are fond of saying that learning a discipline means learning its discourse, but learning a discipline also means learning not to use that discourse. That is, students don’t know a field until they can write and talk about what is in the textbook and the lectures in their own lingo, in their informal home or personal language—language that, as Vygotsky famously observed, is saturated with sense or experience.
• When students do high stakes writing they often struggle in nonproductive ways and produce terrible and tangled prose. When they do low stakes writing, their prose is usually livelier, clearer, and more natural—often more interesting—in spite of any carelessness and mistakes. They don’t tie their syn-tax in so many knots or defensively restrict themselves to simple “Dick and Jane” sentences, because they aren’t worrying so much about the grade or whether they are writing exactly what the teacher was looking for. I’ve almost never seen a piece of low stakes writing I couldn’t easily understand. But I’ve seen lots of high stakes writing that students worked very hard on that was impenetrable.
• Low stakes writing improves the quality of students’ high stakes writ-ing. By assigning frequent low stakes pieces, we ensure that students have already done lots of writing before we have to grade a high stakes piece—so that they are already warmed up and more fluent. Their high stakes pieces are more likely to have a clear, alive voice. And it’s no small help to their high stakes writing that we have seen a number of their low stakes pieces. For then, when they turn in a high stakes essay that is awkwardly tangled or even impenetrable, we don’t have to panic or despair; we can just say, “Come on. You can say all this in the clear, lively voice I’ve already seen you using.”
• Low stakes writing gives us a better view of how students are understanding the course material and reacting to our teaching. We get a better sense of how their minds work. We can see better the interactions between their thinking about course material and their thinking about other realms of their life, between their thinking and their feeling. We get better glimpses of them as people.
• Probably the main practical benefit of frequent low stakes assignments is to force students to keep up with the assigned reading every week. When students put off the reading till an exam or major paper is due, they learn much less from discussions and lectures. And when only the teacher and a few diligent students have done the reading, the whole course tends to lose steam.
Responding to Writing
When we assign writing, we can trust that we are helping students learn more and probably even write better. But when we respond or comment, we can’t be so confident. The news from researchers is not encouraging. They have discovered how often teacher’s comments are not clear, how often comments are misunderstood by students even when they are clear, and how often comments cannot be trusted (for example, the teacher writes, “You should omit this section,” or, “You need a comma here,” or, “This hypothesis has been discredited,” when in fact many or even most authorities would disagree). Researchers have trouble finding good evidence that our comments on student writing actually help students learn more or write better. (Elizabeth Hodges is one of these researchers, and in Chapter Seven, she gives us interesting and practical glimpses into the essential sequence of events: the teacher’s reactions to a student paper, that teacher’s actual comment on the paper, and the student’s reading and understanding of that comment.)
These sobering results are not really so surprising once we stop and reflect on the conditions in which we write our comments and the conditions in which students read them. After all, we write comments in great quantity— working slowly down through thick stacks of papers on our desk. It is often late at night and we are usually in a hurry. And truth be told, we are often writ-ing in a discouraged or downright grumpy mood. Writing comments on papers and exams is a major portion of the “academic writing” of most academics, yet it’s not the writing we really care about. It seldom has much effect on our careers, and we seldom do any revising of it. No wonder it is seldom our best writing. And let’s face it: it’s not feasible to write our comments really slowly and to revise them carefully. We are surely going to continue to write comments fast, late at night, and not always in the best mood. Still, we can learn to do it better—thus the efforts in this book.1
Even when we write clear, accurate, valid, and helpful comments, our students often read them through a distorting lens of resistance or discouragement—or downright denial. (Don’t we sometimes read responses to our own articles by professional reviewers through similar lenses?) When students read what we write, they are usually reacting at the same time to all the past teacher comments they have received on their writing. The most obvious example of this is that students tend to take almost anything we write as criticism—even if we are just asking them a question or making an observation, or even making a low-key statement of mild praise. (“I’m curious how you managed to be so dispassionate on such a controversial issue,” or, “I was interested that you were able to quote from a book that I didn’t assign.” “Uh oh, I’ll never do those things again.” ) And when we include a grade with our comment, we increase the likelihood of a distorted reading—sometimes no reading at all!
What discouraging news. But I think we need to hear it. It helps us ask some very practical questions as we respond to student writing: “Am I wasting my time with this comment? What are the chances that it will be understood as I intend it? That it will help?” Perhaps we could adopt the principle of our better-paid fellow professionals: “At least do no harm.” When we assign writing, at least we do no harm.
Continuum Between High and Low Stakes Responding
In the face of this bleak situation, I call again on the distinction between high and low stakes. But here I am emphasizing a continuum with many intermediate points. Just as important, it is also a continuum from the least respond-ing to the most responding.
Zero response (lowest stakes). When I am clear and honest with students about the fact that I need to require more writing from them than I can comment on, I help them fairly quickly get over any feelings of deprivation or resentment. Most students come to appreciate the chance to write with the knowledge that they will be heard but will not have to deal with my response. In fact, many teachers require some low stakes writing that they don’t even read. Students can appreciate and benefit from the freedom of this private writ-ing. (See Sargent and Elbow, Chapters Four and Eleven, respectively, on ways to deal with private writing.)
Minimal, nonverbal, noncritical response. We can note effective or strong or correct passages by simply putting a straight line underneath particular words or phrases or alongside longer sections. (Teachers often use check marks in the margin for this purpose, but I find straight lines are more specific markers.) I can respond in this way virtually as quickly as I can read. Almost every student needs some encouragement, and some students on some occasions need lots. Even in very poor pieces of writing, certain parts are always better than others; students benefit from having them pointed out. To find strong points, even in weak writing, is a skill that will help us improve student learning and writing.
Supportive response—no criticism. There are usually things that students do well that are hard to point to with simple straight lines (for example, “You chose a good approach to your topic,” or, “You write with a clear and lively voice.” ) Whether we call it praise or positive reinforcement, the fact remains that this kind of response does the most good with the least effort. That is, we are most likely to cause learning and least likely to do harm if the message of our response is, in effect, “Please do more of this thing you are already doing here.” We are least likely to cause learning and most likely to do harm if we give the message that is all too often implied in critical feedback: “Start doing something you’ve never done before.”
Descriptive or observational response. An example of this response: “You begin with an anecdote from your own experience; then show us how it throws light on your academic topic. Then you make your case—which really rests on a process of definition—showing what fits and what is excluded.” One of the hardest things for student writers is simply to see their own text, to understand the logical and rhetorical strategies they have used. Neutral and noncritical observations can be very effective because students don’t need to resist them.
Minimal, nonverbal critical response. Just as quickly as we can read and put in straight lines, we can also put wavy or wiggly lines underneath words or alongside passages that are unclear or problematic or wrong. It’s remarkable what a strong sense of our readerly presence and response we can give to stu-dents when we note five or six phrases or passages per page with straight and wiggly lines: they get a felt sense of what is working and not working for us.
Critical response, diagnosis, advice (highest stakes). This is our meat and potatoes—what we tend to assume is our main job. Obviously, we often need to give critical response to help with learning and to explain the basis of poor grades. But my premise here is that the higher we go on the continuum, the more we need to ask the crucial pragmatic questions: Is this comment worth it? How much response do I need? How much criticism will be useful? What is the likelihood of my effort doing good or harm?
I don’t mean to suggest that we can just mechanically match low stakes responses with low stakes assignments and high with high. Obviously, we will often mix levels of response—in particular, mixing praise and criticism. Even the highest stakes assignment merits some praise.
Nevertheless, it pays to notice the natural links between levels of assign-ment and response. That is, the lowest stakes response (zero response) goes most naturally with low stakes assignments: when the writing doesn’t much matter to the final grade, we can afford to withhold our response or criticism. Similarly the highest stakes response (critical response) goes most naturally with high stakes assignments: if our judgment of a student essay will have any significant impact on the final grade, we are obliged to explain any criticism we have. This critical response carries the highest stakes for many reasons: with critical response, we have to worry more about whether we are wrong or unsound; critical response is more likely to misfire or do harm because of how it is received—even if it is sound; and critical response is likely to cost us more work and more uncertainty. In contrast, low stakes minimal responding requires the least time and effort from us, requires the least expertise from us, takes the least time away from our teaching of the subject matter, and is least likely to turn teachers and students into adversaries
I am not trying to stamp out critical response; I’m just arguing that we should use less of it—and use more minimal and low stakes response instead. Note, for example (and this is another case of mixing), that we can use plenty of low stakes praise without giving up criticism—without pretending that a piece of writing is better than it is. For example, we can write something like this: “Your paper doesn’t work very well and the worst problem is confusing sentences. I often couldn’t understand you. Nevertheless you do have plenty of clear sentences and I’ve marked particularly strong ones with a straight line. To work on your serious problem, try to figure out what you were doing when you wrote those strong sentences—and do more of that.” It might seem hard to find examples of good organization in a disorganized paper, but not if we set our mind to it. For example, we can write: “I got lost a lot as I read your paper. It has big problems with organization. But I’ve put straight lines along several paragraphs that hang together just fine, and also lines between several paragraphs where they follow well and your transition works fine. Give us more of that! You’ve shown you can do it.”
It is important for us to realize that we don’t need to feel guilty if we use lots of low stakes and minimal response—especially if we are not teaching a writing course. Assigning more writing, using less response, and using more praise doesn’t mean leaving out all criticism or lowering standards. Students need the experience of writing a great deal and getting minimal and low stakes response because they tend to associate writing with criticism and high stakes. If we are not so much teaching writing as using writing to teach something else, it makes particularly good sense to use lots of minimal and low stakes response. When we assign a piece of writing and don’t comment on it, we are not not-teaching: we are actively setting up powerful conditions for learning by getting students to do something they wouldn’t do without the force of our teaching.
Conclusion: Concrete Suggestions
• For high stakes assignments, it can be very helpful to require a draft a week or more before the final version. Teachers handle drafts in a wide variety of ways depending on their circumstances and styles. At the very least, we can just collect drafts and not comment—simply checking that they are done— thus forcing students to carry their thinking through two steps. Of course, if our circumstances make it feasible, it is good to give comments on a draft. When we comment on a draft, our response becomes almost automatically low stakes, even if critical: we can write suggestions for revising rather than just an autopsy. (Notice in Chapter Six how Herrington describes the production of an essay that has very high stakes but one that students work up to along a path of lower-stakes drafts and comments on those drafts.) It is probably worth cutting back on the amount of responding on some assignments for the sake of giving students at least one experience of feedback on a draft aimed at a revision. If we can only do this once, it’s better to do it in the first half of the semester—with the goal that students can internalize some of our responses when they work on later high stakes assignments. But commenting on drafts may be more feasible than some teachers think: if we give good responses on a draft, we can make do with just a quick verdict on the revision (perhaps using the kind of grid that I suggest in Chapter Eleven).
• Even when we are commenting on a final version, we can frame our comments in a forward-looking way: instead of saying, “Here’s what didn’t work,” we can say, “Here’s what to work on in future papers.”
• I find it easier to comment on important assignments if I get students to turn in a short reflective cover letter or piece of process writing with the assignment itself. I invite something informal, even handwritten. I ask them to tell me what they see as their main points, how they went about writing and what happened, which parts they are most and least satisfied with, and what questions they have for me as a reader. Reading the cover letter usually helps me decide what to say in my comment. Often I can agree with much of what the student has said, and sometimes I can be more positive about the essay than the student was. Students may have difficulty at first with this self-reflective writing, but it promotes a skill worth working on. It gives them practice in trying to see their own thinking more clearly. (Herrington gives good examples in Chapter Six of cover letters for a mid-process draft and a final draft.)
• I find commenting much easier if I read the whole piece before making any comments except for straight and wiggly lines. I save lots of time by reminding myself that students can seldom benefit from criticism of more than two or three problems. Therefore, the most crucial decision in commenting is which problems to focus on, and I can’t make that decision till I read the whole paper through. Most of my bad commenting comes from jumping in with marginal comments as I am reading: I am more likely to waste my time on something that turns out to be a minor issue, or make some passing remark that the student misunderstands, or say something that’s actually wrong (“You obviously don’t understand x,” when later on it’s clear that she does understand x), or get caught up in a little spasm of unhelpful irritation. If I settle for just mak-ing straight and wiggly lines, these serve me as a map when I glance back over the paper after I have read the whole thing and I am trying to decide what are the few main things I need to say. (In Chapter Nine, Chris Anson points out an exception: when we put our comments on a tape cassette, we may want to tell the story of our reactions as we are actually in the process of reading. Yet Anson also points out that even for this kind of responding he sometimes does better by waiting till he has read the whole piece.)
• As Hodges points out in Chapter Seven, when we return papers to students with our comments attached, it’s a great help sometimes to ask students to take five minutes right then and write us a short note telling what they heard us saying and how they are reacting to it. This helps us learn when we are unclear or when students misinterpret our words or react in ways we don’t expect.
• If we are writing comments where the stakes aren’t too high, we can save time by waiting till we have two pieces in hand, read them together, and write only one comment on both. The comparison is often pedagogically useful. (“Notice how much clearer your point was on this paper compared to that one [or how much more carefully you argued]. What helped you?”)
• Though it sometimes costs me a few more words, I try to avoid an impersonal “God/truth voice” in my comments. Almost anything that we might say in response to a piece of writing is going to be affected by our own point of view. Even the main ideas in our discipline are arguable. If we are willing to say, “Unconvincing for me,” instead of “Unconvincing,” students are more likely to pause, listen, and think—instead of just resisting, or else unthinkingly giving in to authority. Besides, magisterial shorthand words like “Awk” are often extremely unclear. I have been trying to learn to write more accurate translations like, “I stumbled here,” or, “I’m lost,” or, “Wording feels unprofessional,” or, “Too slangy for my ear,” or, “Can you be less roundabout?”
I sum up this chapter with that useful dictum “At least do no harm.” Think how much good we do in assigning lots of writing, especially lots of low stakes writing. But this approach is only feasible when we realize that we can get by with far less response and criticism than we usually assume.
Note
1. It interests me as a writing teacher to note that though our commenting on student papers is undeniably “academic” and indeed “professional” writing, it is often very casual: we often write in incomplete sentences and use lots of “I” and “you.” I’m not saying that these features make our writing bad or unprofessional or unacademic. I’m just pointing out that many academics unthinkingly assume that casual informal writing is not academic and should not be used by students.
PETER ELBOW was a professor of English and director of the writing program at the University of Massachusetts, Amherst when he wrote this.
General Document Comments 0
icons on the left to see existing comments.
0 archived comments