"I feel that I learned more in this physics class than in any of my other classes because of the way classwork and grading was conducted. I was never worried about what my grade would be at the end of the quarter or semester, because I had faith in the way you graded and evaluated us. In all of my other classes I continually stressed out about my grade and always had to make sure that they were good enough." —Gabriella B., student
Keep your eyes on the prize.
We communicate our values with grades. In a grading system based on the curriculum transfer model of school, the only thing that matters is how much of the content the student has nominally mastered. Test scores and other assessments of learning will typically carry great weight in determining a grade.
If, however, we start from the bedrock belief that the purpose of school is to prepare students to live their lives well, we must shape our grading system accordingly. Every aspect of the grading system must been seen through this filter. Because we understand that tests rarely distinguish between genuine, long-lasting learning and “mastery” that will be forgotten in a few days or weeks, we must change the way tests are used. We must also look for other means of assessing learning that can supplement testing.
Furthermore, with our new framework — the Student Agency Model — the personal growth of students becomes much more vital. Training students to be internally motivated, responsible learners is more important than, say, whether they are punctual. When designing the “formula” for grades, all our priorities must be founded on self-directed learning. This must be the basis for the relative weights of what we are measuring.
Grades are of one of the most important structural factors that will affect whether your students are doing school or becoming self-directed learners. If we are preparing them for life, our grading structures must be designed to steer them away from meaningless activity and towards a relationship with learning that is intentional and rewarding.
An effective grading system fosters communication.
Your grading system should include regular communication between you and your students. Optimally, this feedback will come in a range of formats, not just points. Self-evaluated contracts, portfolios, quizzes and tests, test re-submittals, and grade conferences are all possible pieces of a final grade. Creating a form that summarizes all these aspects can serve as a basis for generating appropriate and authentic grades. An example of such a structure is described in detail in the section 9.13 below.
Grades should reflect a nonjudgmental posture.
If we want students to acquire the skill of learning from their mistakes, our grading system must avoid penalizing mistake-making as much as possible. This is even true for mistakes in judgment, like copying someone else’s homework. It is more important to teach students to act responsibly than to train them to not make mistakes out of fear of getting caught. In other words, be sure your grading scheme is not overly punitive in nature. Our goal is to cultivate integrity, not compliance.
If you want to be nonjudgmental in your working relationship with your students, your grading system must reflect that. Generosity of spirit can be built into an honest system of evaluation.
Recognizing the limits of precision in grading certainly runs counter to the current focus on data collection. But quantification is, in itself, often misguided and counterproductive. Using a 13-step scale from A through F, for instance, leads both student and teacher to focus on the difference between a B- and a C+, rather than on how much the student has learned. Besides, how precisely can learning be measured? Even more problematic, how precisely can traits such as grit or self-directedness or creativity be measured?
Whenever possible, use means of measuring learning with as few decimal places as possible. Having a student receive a 3 on a 1 - 5 scale, for instance, is less fraught than getting a 62.6% on homework mastery. Creating “soft” rubrics — replacing a complex and overly defined structure with less specific quantifiers of evaluation — helps avoid student preoccupation with collecting points. Whenever possible, try to avoid direct translations to letter grades.
A flexible interpretation of the importance of different aspects of the grading system also helps ground feedback in personal meaning for the student. For instance, rather than defining the three factors in a marking period grade as
Evidence of Work - 40.0%
Evidence of Learning - 40.0%
Personal Outcomes - 20.0%,
the same information can be communicated effectively without the unnecessary precision, as seen in the diagram to the right.
Use points only when necessary.
Distinguish between the aspects of grades that are overtly quantifiable, like the number of correct answers on a multiple choice test, from those which are not, like how much effort a student has put into the learning process, or how well organized her written journal is, or how well she collaborates with her peers. The lack of quantification itself leads to more meaningful conversations about these topics.
One legitimate function of points is to quantify the relative weight of different activities. Another is to keep track of how many questions a student got right out of the total number of questions asked. In your subject, with your students, you may find other situations where points are the only practical way to derive grades. Nevertheless, there are some serious disadvantages to point-based grades that are worth thinking about.
It may be that your school has a required grading system that insists on measuring all grades in points. As long as your students understand how their grades are composed and are comfortable with the process within your classroom, translating the results to fit into the computer whenever required will not pose an impediment to their authenticity.
The process for deriving a grade should be transparent and clear. Whatever structure you use, a student should never be surprised by how the pieces are combined to make the whole grade. The relative weights of the different factors that make up grades should be built in to the structure.
On the other hand, a certain flexibility is important in the process. This would seem to contradict the need for transparency, but intentionally building in flexibility in deriving a grade acknowledges the inherent subjectivity of grades. The amount of flexibility is, of course, something that every teacher must decide for herself.
Find a balance that works for you.
The approach you take to grades will be a mixture of student self-evaluation and teacher evaluation. In general, more self-evaluation leads to more independence and metacognitive thinking on the part of the students. On the other hand, teacher-generated evaluations allow you to stay in control of the process. You should be comfortable with the balance between the two. That balance can and probably will also evolve over time as you become accustomed to this new way of dealing with grades.
Valuing the learning process.
One of our goals is to cultivate attributes like responsibility, willingness to learn from mistakes, and honest self-appraisal. Therefore, how a student learns can be as important as what she learns. Meeting deadlines, being prepared to participate in conversational learning, choosing to work at the appropriate level of challenge, and collaborating well with other students — these are all skills to be valued and our grading systems should reflect it.
Valuing the evidence of learning.
Our intent is to measure what the student has genuinely learned, but to do so we must distinguish between doing school, which is a simulation of learning, and the real thing. In particular, test scores by themselves are not reliable tools, since they do not distinguish between material that the student has integrated and that which she will forget in days or weeks.
When a student has genuinely learned new material, she will be able to apply the new knowledge appropriately and understand its relevance and connection to other knowledge, and she will still know it months or years later. Determining whether such learning has taken place is not easy. Even retesting the same material at a later date (as often happens with semester exams) can be “gamed” by a student if she crams for the same material a second time. Therefore, assessing self-directed learning requires multiple measures of proficiency, including various forms of feedback and evaluation.
Most importantly, to make testing a better measure of learning, a full assault on the habits of doing school must be built into the classroom culture. When the community is grounded in the act of self-directed learning, a student is much less likely to indulge in cramming and regurgitating on tests because doing so feels pointless.
Valuing personal attributes.
This is important because the person is important. Her development as a responsible, self-directed learner may, in her life, be more important than any other factor that grades in your classroom are based on. These attributes are legitimate aspects of a student’s grade, even if they are harder to quantify than, say, the percentage of homework completed. Including these factors in our grading systems lets students know that we value them.
Student self-evaluation is particularly well-suited to assessing personal outcomes. This can be done periodically throughout the year, especially at the end of each marking period. It’s important not to assess these factors too frequently, however, because students may stop taking them seriously and they will start to lose their meaning.
Develop simple alternatives to points in evaluating these non-curricular aspects. This can be a survey, a set of questions (as described below), or a reflective essay. Using such techniques may at first seem to complicate your grading system, but it will prove to be a more authentic and reasonable approach than reducing everything to points. In a short while, both you and your students will become used to it.
To begin, you will need to choose those attributes that you believe are the most important to focus on with your students. Factors in making that choice include how mature students are, how motivated to learn they are, and how self-reliant they are. How many of them are suffering from a fixed mindset (“I’m no good at math”)? How well do they collaborate with each other? How ready are they to form a community?
The next step is to determine how each trait can be recognized.Where is it visible? How can it be recognized by you and by the student herself? Develop a mechanism for the student to evaluate herself on each trait. I found the most direct approach was to have students answer a set of simple questions. If you use this approach, think about how complex the questionnaire should be. Too few questions can lead to miscommunication, while too many can be cumbersome and repetitive.I used three questions for each major trait, which worked well with my students, but that is fairly arbitrary.
Living within constraints.
Any discussion of grading policy must acknowledge the hard truth that teachers rarely have the autonomy they once had in shaping how they use grades. When I retired from the classroom in 2010, I was still able to keep a manual grade book and enter the grades on an as-needed basis, including, of course, at the ends of marking periods. In many schools, that is no longer an option — grades are entered immediately and very frequently, sometimes daily.
The question of how to reconcile the looser, less point-driven approach that I am recommending with the constraints by your your school’s grading system is likely to be an ongoing issue. One way to resolve it is to determine the minimum frequency that grades are required by the school’s system, and develop a means to translate your system into something that can be entered into a computer only at those times.
The next section describes the grading system I used in my classroom.