As the supervisor of English and Communication at a large suburban high school, I was required to initial the electronic grade books of the 40 teachers in our program on the last day of school. In my cursory review, I was being the quintessential bureaucrat, simply ensuring that the school had a paper copy of all grades. I often wondered if I should do more. If I really looked at the grade books, could I learn something important about teachers and student performance?
I also wondered how teachers might feel if I examined their grade books? I realized that I had never discussed my grade book practices with a supervisor. For more than twenty years, I had made and sequenced assignments, established point values, and implemented grading policies, and no one had said a word to me. Indeed, I worried that teachers might be suspicious of me, fearing that I was questioning their judgment and treading on their territory.
I started with beginning teachers. My student-teaching supervisor had once showed me how to use a scatter gram to assign grades on a 60-point test, a really helpful skill, and I figured rookies might appreciate an opportunity to talk with a veteran. I had no protocol for these early meetings, but quickly found a pattern of questions. How many different types of assignments do you have? How do you assign points for these assignments? How many total points are linked to each type of assignment?
It seems a set of simple questions, but it’s really hard for new teachers, who rarely know the number of assignments in a quarter. After all, they are making them up on the fly, sometimes spontaneously as lessons unfold. One teacher had twenty-three grades in the first quarter, another twelve. Not knowing the total number of points in advance, teachers must guess at the point value for each assignment. Sometimes they logically assign point values based on the principle of size: major projects earn 50 points while minor projects earn 25. At other times, they count the number of questions: ten-question quizzes earn 10 points while fifty question-tests earn 50 points. I discovered that teachers were pretty honest in these conversations, explaining their thinking or lack of it, often admitting that they had just grabbed a number from mid-air.
One teacher assigned 75 points to an early project and then reduced point values on subsequent assessments. In effect, one assignment had taken on massive significance in his students’ overall grades, a problem compounded by the fact that the assignment focused on less important skills in the curriculum. Another teacher had given ten reading quizzes (100 points) and a major three-draft project (100 points). In her percentage-based system, the default on all electronic grade books, her quizzes of literal comprehension weighed as much as a multi-genre project measuring a wide range of literacy skills.
As I uncovered problems, I worked with teachers to reveal certain principles of grading. The previous examples demonstrate that teachers must establish a clear relationship between point totals assigned and critical learning skills. A simple system of weighting applies the major-minor project approach, and it is easy for students to understand. A more effective weighting system is available through electronic grade books. Here teachers categorize assignments and determine the weight of each curricular component. Reading quizzes, for example, might comprise five percent of the grade, and it does not matter whether the teacher assigns one quiz or twenty.
Another principle of grading is that teachers should establish some means to acknowledge learning and improvement over time. Do teachers see a difference between a student who earns scores of 90, 80, and 70 during a term and one who earns scores of 70, 80, and 90? In a percentage-based system, these students earn identical grades, yet one can imagine many scenarios where such a grade is unfair. Perhaps the student has a learning disability and needs extra time to acquire a particular skill. Perhaps the student is a recent transfer and needs a period of adjustment. Perhaps the student has benefitted from private tutoring or parent intervention? I showed teachers how to solve the problem by increasing the value of similar projects in succession and how to place additional weight on late-term work.
With developed confidence, I turned my attention to veterans. I met with freshman teachers to discover their policies for late work. One wrote, “When I started teaching, I gave an automatic F (59%) for late work. Now I use a one-letter-grade-deduction-per-day system, with a cap of five school days late.” Another wrote, “I think my practices are moderate. I have a three-day-late policy for all assignments. Additionally, I provide students with a homework coupon that acts as a get-out-of-jail-free card.” A third said, “I don’t ever want students to feel freed from the responsibility of submitting missing work, so I accept late work almost always, reducing its grade somewhat or not at all if the submitted work is of excellent quality.” I was shocked by the variability in teachers’ policies, and not surprisingly, teachers were also. Initially, those with harsh penalties defended their practices until someone asked if they had ever missed deadlines in their own lives. One admitted that she had submitted her grades a day late and another hadn’t completed a homework request for a homebound student. A third talked about the crazy pressure her daughter felt with three major projects due on the same day.
Ironically, our group settled on having no late-work policy. We agreed that teachers must sometimes apply a late penalty, especially if lateness is chronic, but we acknowledged that students have complex lives and schedules. We also rejected any policy that might discourage students from applying themselves to their studies. We affirmed that the quality of late work should matter.
While some of the problems in teachers’ grade book practices are caused by mathematical misunderstandings (How do we convert rubric scores to percentage scores?) and confusion about grade book software (Should we turn rounding on?), most issues come down to a teacher’s philosophy of grading. Why do we grade students? I always ask teachers to rank order the following purposes for grading in terms of their importance to them.
|Rank||Purposes of Grading|
|Provide an accurate measure of student performance|
|Characterize a student’s grasp of course content and skills|
|Hold students accountable for their learning|
|Offer expected feedback to students and parents|
|Motivate students to submit their best work|
|Reward and punish students for their efforts|
|Demonstrate student progress in learning|
|Sort students by achievement|
Every teacher has a set of beliefs about grading, and these beliefs influence the mathematics of grading. Nowhere is this more evident than in the assignment of zero grades. If teachers believe that grades should reward and punish students, they will find no stronger punishment than a zero grade. It is the death sentence in a percentage-based system, having the power to announce failure on one assignment and turn other projects into failures as well. For example, a student earning 90-point A-grades on three projects will see his average drop to a 67 percent D grade if he does not complete the final project. Note how such an outcome would trouble teachers who believe that grades should provide an accurate measure of student performance. In this case, a D-grade vastly mischaracterizes the skills of the student.
Nor does a zero grade make much sense mathematically. If the interval between an A-grade and B-grade is 10 points and between a C-grade and a D-grade is 10 points, how can the interval between a D-grade and an F-grade be 60 points? If we adjust the zero grade to 50, a score we might give to a student who completely lacks the skills being measured, the student above now receives an 82.5%, a B-grade for the course, and surely a much more accurate depiction of his learning. Ironically, in the absence of a grade in an electronic grade book, the score always defaults to zero unless teachers instruct the program to do otherwise, and teachers who may not seek to punish students end up doing so anyway.
Indeed, one of the most startling discoveries of my work with teachers is that many do not see the conflict between their grading practices and their values. One teacher professed a deep commitment to minority education, having once taught in the Teach-for-America program, yet she applied grading practices that harmed the very students she wanted to help. Later, she added extra credit assignments, dropped the lowest grade in a term, and reduced the weight of homework assignments, all practices that helped her non-traditional learners.
I used to think that the demon in this story was electronic grade books. They forced teachers into mathematical processes that complicate communication with students and parents and undermine the teaching-learning process. Later, I decided that it was the lack of training in the use of these software programs, for every problem identified above can be avoided if teachers program the software carefully. Today, however, I think the demon in the story is the value of accountability itself. It’s a loaded word in education today, but those who believe it want to reward students (and teachers) for their successes and punish them for their failures. But truly, what does a failing student deserve? Surely he deserves a dedicated and skillful teacher who works hard on his behalf, and definitely one that understands the mechanics of grading. Surely he deserves a guidance counselor, who investigates his lack of achievement and seeks to coordinate the efforts of the professional staff on his behalf. And surely he deserves opportunities to pursue his interests through the curriculum and experience the curricular opportunities open to other students. Grading is one of things we can’t escape in schools, yet if we actually talk about it, examining small practices for their underlying beliefs, we can learn to do it fairly and with humanity.
CEL Past Chair