Introduction – The Mystery of the Failing Grades

Author Joe Feldman begins with a real-life mystery. A middle school principal discovers that students in some teachers’ classes are far more likely to receive a failing grade than students in other teachers’ classes. The difference is stark. In the math department, Teacher 1 and Teacher 2 had 2% or 3% of their students fail. Teacher 3 had an 11% fail. In the English department, 11% of students in Teacher 1’s class were failing, compared to just 2% in Teacher 2’s class, and a whopping 20% in Teacher 3’s class. What was the cause?

Students in these classes were learning the same skills, reading the same books, getting the same homework, and taking the same tests. Their learning experiences and levels of support were very similar. The student population in each class was similar too: they had similar standardized test scores and similar absentee rates. The teachers even scored tests as a team to ensure fairness. Yet somehow, having one teacher versus another could dramatically increase the likelihood that a student would receive a failing grade. Why?

The principal found the answer in grading policy. Each teacher had created their own syllabus, listing consistent assignments, tests, and so forth, but wildly inconsistent policies when it came to grading. One teacher did not accept homework after the bell, another deducted points if homework was late, and a third accepted late work at any time without penalty. One teacher gave full credit for completed homework whether the answers were right or not. Another teacher allowed a week for students to fix incorrect answers to improve their scores. Some teachers deducted points if students turned in assignments with ripped edges or left off their last name. Some graded effort and participation. Others did not.

These widely varying policies made it possible for two students with identical academic performance to receive different grades.

So what? Here’s why that matters: grades impact placement in higher level classes like honors or AP, graduation, eligibility for athletics or other clubs, retention and promotion, college admissions and scholarships, and the list goes on.

The principal, predictably, faced resistance and doubt when she tried to discuss grading policy with the teachers at her school. They had crafted their grading policies thoughtfully, attempting to teach students how they believed was best. Naturally, they were resistant to throwing that out. But in reality, policies do not always have the desired results, and they do have an impact on equity.

The good news is that grading can be done differently, accurately, equitably, and without bias. Grading done well can motivate students to learn and persevere and take ownership of their achievements. It will take time to learn a new way and even to convince ourselves that the change is worth it, but it will be worth it.

In studies of classrooms where the new grading policies were adopted, grades became more accurate, students were less stressed, and teachers felt the emphasis shift from due dates to learning. Assignments were for the purpose of mastering skills and gaining knowledge, instead of earning points.

“Doing this work really changed my perspective,” says middle school history and English teacher Cathy. “This helped me realize that the main purpose of grading is to see how much the students know, to assess their learning instead of their efforts; do they really understand the work, as opposed to did they do all of the assignment.”

Part I – Why We Grade the Way We Do

The reaction faced by the principal in the previous section is a familiar one to many of us: Why would you question grading of all things? Grading is so embedded in our concept of school that it almost seems silly to question it. But if grading is so essential, how did it become that way? Where did grading come from?

A Brief History of Grading

Our current grading practices in the U.S. began about a century ago, and they were developed in the context of that era.

The rise of manufacturing meant that factories needed workers. Ideal factory workers would be efficient, specialized, quiet, attentive, and accustomed to timed routines and operating within a chain of command. Industry put pressure on public institutions like schools to cultivate these characteristics.

Intelligence testing and behaviorism were popular at this time. The behaviorist idea that humans could be taught to behave in certain ways through external conditioning was appealing to those running schools and factories. Intelligence testing seemed to lend support to racist and classist beliefs, and it was welcomed into schools and used for placement and tracking.

Migration to cities as well as immigration to the United States meant increased class sizes and diversity. Narrative feedback, delivered in person or in writing, was no longer as realistic as it might have been for a teacher who knew the families in her small community. Teachers needed a more efficient and standardized method of reporting progress.

Enter the system of letter grades A-F, adopted from colleges and universities, alongside norm-referenced grading. Norm-referenced grading reported how well a student performed in comparison to other students in the course. The belief was that grades would be more objective if they were organized in a bell-curve, reflecting the assumed distribution of intelligence across the student population.

Teachers’ Beliefs About Grading

And yet, ask any teacher today whether they believe their grading policy has been handed down to them from some captain of industry who lived a hundred years ago, and the answer will certainly be no. Without knowing the history of grading, most teachers have put a lot of thought, and even years of trial and error into developing grading policies that they believe work and reflect their philosophies.

Some teachers choose to give points for punctuality, teamwork, participation, and meeting deadlines because they believe these are important life skills and that their job as teacher is not just to teach math or history but to prepare students for life. Teachers who heavily weigh homework when calculating grades often do so because they think it is important to teach responsibility. Teachers might grade on a curve in order to distinguish which students they ought to recommend for the “honors” class next year.

Well-intentioned teachers may even have precisely the same goal in mind – fairness – but disagree about which grading policy results in fair grades. One might believe that student effort should count for a lot and a fair grading system should reward a student who tries hard, whether or not they get everything right. Another teacher may feel that the only fair way to grade is purely based on academic achievement, regardless of student effort, otherwise hardworking students might be given the false impression they’ve done well, when they have not, in fact, learned the required content.

Expect Strong Emotions When You Discuss Changes to Grading

If you ask teachers to change their grading policies, they will likely feel you are questioning their professional judgement. They may even feel an emotional or psychological threat. Fairness, responsibility, effort, and so on are core values the teachers are attempting to teach through use of their carefully chosen grading policies.

While the new grading policies proposed in this book are founded in research, don’t expect teachers’ minds to be easily changed by facing a mountain of evidence. Evidence against their former practices is likely to bring on strong and unpleasant emotions.

“As I am learning the improved grading practices, I’m thinking about how many students I may have hurt in the past,” says Jillian, a middle school math and science teacher. “I don’t want to go there.”

“Going there” – accepting evidence that contradicts your current, well-intentioned beliefs and practices – is hard. If you are asking teachers to do this work, expect that it will be hard. Some of the most devoted and thoughtful teachers may find it the hardest because it asks them to cast doubt on dearly held beliefs, and to consider that they themselves may have done unintentional harm in the past to students they were trying to help. Teachers who define themselves by their commitment to equity may need to face their own contributions to inequity. It will not be easy.

As Feldman writes, “When the concepts in this book challenge you in uncomfortable ways, stay open to new evidence and possibilities, imagine what could be, and be less conservative in your web of belief. Consider equitable approaches to grading that you may have previously believed were impossible.”

Part II – How Current Grading Practices Undermine Learning

So, what exactly were these teachers in the previous section discovering was so bad – so inequitable or harmful – about their current grading practices? Standard grading practices undermine learning in three serious ways:

1. They damage trust in student-teacher relationships
2. They shift motivation from intrinsic to extrinsic
3. They don’t accurately represent student abilities

1. Student-Teacher Trust

It has been shown in numerous studies that students have higher achievement, are more motivated, and are more engaged, when they have a positive relationship with their teacher. Learning involves vulnerability, openness, and the willingness to make mistakes. Consciously or not, students evaluate teachers for their trustworthiness. Is this someone who believes in me, responds to my needs, respects me, and treats me fairly? Or is this someone who might ridicule or punish me if I make a mistake? Can I reveal my weaknesses to them safely?

Students of color or others who have experienced being treated unfairly, may be even less willing to be vulnerable.

But learning requires vulnerability. Making mistakes and revealing what we don’t already know are essential for learning and growth. When teachers penalize mistakes made during the learning process (as opposed to on the test) they are sending the message that mistakes aren’t acceptable. Instead of trusting their teachers to help them learn, students feel pressure to conceal or avoid error. This often leads to even less-desirable behaviors such as skipping class (because they feel embarrassed by their mistakes in class), or copying homework (because that is the only way they believe they can turn in mistake-free work).

Rather than a high-trust zone of comfort where learning happens, classrooms become low-trust zones where every action and every error “counts” and will be penalized. Even non-academic actions like arriving late or struggling to get along with your group might lead to grade penalties and the feeling that one must always be perfect in order to get a good grade.

2. Intrinsic vs. Extrinsic Motivation

By the latter half of elementary school, learning and behavior are often commodified into a system of points. These miniature economies help teachers manage the classroom environment and they are also meant to motivate learning. But do they?

Research has shown that extrinsic motivation (points, grades, or any external reward) is not an effective motivation strategy for authentic learning. It can stifle deep thinking and isn’t well-suited to the type of work we are asking of students today. B.F. Skinner and other early behaviorists used extrinsic rewards in their research to motivate rote tasks (think factory work), but for work that requires higher-order thinking, extrinsic motivation is not effective.

Some teachers may argue, But in my experience, the points are effective. More likely, the point system creates an illusion of effectiveness. Students may be quiet (a rote behavior) when the teacher is speaking because they fear she will deduct points. This creates the illusion of engagement in the lesson, but it does not lead to real engagement. Students may be eager for points, making them dependent on the teacher’s approval of their behavior and work tasks, but they are less independent and less focused on learning.

Disturbingly, point economies are often touted as most necessary for low-income and historically under-served students, when there is no research supporting this assumption. There is no reason to assume students who are low-income or have fewer advantages would have any less capacity for intrinsic interest in learning than their peers with more advantages.

Not only is an extrinsic motivation system not the most effective, but it can also be detrimental to students’ engagement. When students experience school as a jumble of subjective judgements with teachers awarding or deducting points based on any action of the students, some students – especially those from marginalized backgrounds – may become suspicious of school’s fairness and choose to stop putting forth effort rather than subject themselves to arbitrary judgements with no clear path to success.

Motivating students is a key component of a teacher’s work, but for too long we have depended on carrot-and-stick practices that simply don’t motivate students, engage them, or lead to learning. In fact, all research studies have shown that grades as extrinsic rewards have the opposite effect – demotivating students, robbing their intrinsic motivation to learn, and even causing unethical behaviors like cheating and skipping class. We need to stop thinking that points = motivation and refocus students on learning.

3. All-in-One Grading

Frequently, the single grade that appears on the report card (A, B, C, D, or F) encompasses an extraordinary amount of information about students, including performance on homework, tests, quizzes, projects, classwork, participation, effort, punctuality, attitude, rule-following, and so on. What is the impact of attempting to roll all these categories up into one grade?

One impact is that some of the categories of the all-in-one grade introduce a lot of subjectivity because they are evaluated purely through the lens of teacher observation. When has a student displayed sufficient respect? How do you know if a student is listening? The window is now open to teachers’ implicit biases. Even teachers who do not intend it have been shown to have more negative interpretations of the behaviors of students of color – either due to the teacher’s implicit bias or variations in cultural expectations about what is considered respectful or appropriate. For example, research shows that white teachers rate Black students lower on “citizenship” than white students. When grades include subjective categories, biases are woven in, even when we don’t intend to.

A second impact of all-in-one grades is that grades do not clearly communicate accurate information about students’ academic abilities and achievements. Let’s look at two imaginary students with opposite profiles.

Depending upon the weights the teacher (or grading software) assigns to various categories, these two students might end up with the same grade. Perhaps both receive a C (Student 1’s participation, effort, and homework completion pulling up his failing academic grades, and Student 2’s lack of participation, poor effort, and missing homework pulling down her excellent academic grades).

If these two students both receive a C, Student 1 may not be flagged for the academic support he clearly needs. He may not realize himself how poorly he is doing, and next year’s teacher may be annoyed at how unprepared this supposedly average student is. On the other hand, Student 2 may not receive the help she needs with social skills or executive function. With a C, Student 2 might be passed over for placement in an honors or gifted class, which might be a better fit considering her excellent academic performance with very little effort. Student 2 is likely to become even less motivated to work if her abilities are not recognized and she is kept in “easy” classes year after year. Clearly, the grade of “C” conceals the actual strengths and weaknesses of these two students.

While no teacher would dispute that effort, teamwork, responsibility and so on are important, it is unhelpful to try to lump them in together with academic achievement resulting in a single grade that is so complex as to be nearly meaningless. If we want grades to communicate accurate information about a student’s academic performance (to the students, to their parents, to future teachers, to colleges, etc.) then we need to limit what is included in the student’s grade. It is time to leave out subjective (and bias-ridden) categories, simplifying and clarifying the grade.

Part III – 3 Pillars of Better Grading Practices

If so many of our familiar grading practices undermine student learning, what should we do instead? Embrace new practices that are:

1. Accurate
2. Bias-Resistant
3. Motivating

1. Accurate Grading Practices

Our first pillar of equitable grading practices is Accuracy. That is, we must ensure that our grades accurately represent what a student knows and can do academically. As Feldman describes this first pillar:

Problems with the 100-Point Scale

One of the biggest barriers to accuracy is the 100-point grading scale. It’s easy to use because you can convert any grade into a percentage. If a student gets 47 points on a 60-point quiz, you just divide 47 by 60, and the student gets a 78%. But there are some problems with this scale. First, can anyone really tell the difference in academic achievement between a student who gets a 77 or a 78?

Do we really need 101 different levels (including 0) to distinguish the academic levels of our students? The second and larger flaw is that so many of those 101 levels are devoted to failing grades. Take a look at our traditional grading scale:

Why are we using a scale that is tipped in favor of failing kids by having such a large range of numbers (60 points of out 101!) that equal failing? Plus, if a student is performing in the “B” range and wants a better grade, they only need to get 10 more points, while a student who is failing with a 30, let’s say, needs 30 more points just to go up one level to earn a “D.” At a minimum, this makes no sense, but at its worst, it’s demotivating as students see there is almost no hope of recovering from a failing grade.

On top of the inequity of using a 100-point scale, a few of our traditional grading practices make things even worse for students, namely using zeros and averaging grades. See below for ideas for what we might do instead.

Eliminate the Zero

It is common practice for teachers to enter a grade of zero for missing work. We are well aware that when averaged with other grades, the zero acts as a nuclear bomb – disproportionately dragging down the average of scores earned on other assignments. The teacher’s intention is to “send a message” that students must turn in required work or their grade will suffer the consequences. In our imaginations, we hope that students will be motivated to work harder and keep up with their assignments.

In reality, however, no single research study shows that zeros motivate students. Instead, zeros usually act as demotivators, leading students to withdraw from learning because they see no hope of earning a good grade now.

Use “Incomplete” Instead of the Zero

To further understand the inaccuracy of using zeros, consider this example. Imagine you have four fish in a pond, and you want to know the average weight. You weigh the first three but just can’t seem to catch and weigh the fourth. These are your results:

Fish 1: 4 pounds              Fish 2: 3 pounds              Fish 3: 3 pounds              Fish 4: Unknown weight

Using our traditional grading practices, we assign that fourth fish a zero and average 4, 3, 3, and 0, to get 2.5 pounds. Does this accurately represent the typical weight of the fish? We have no idea because we literally know nothing about the weight of the fourth fish. To be more accurate, we could average the weight of just the first three (it would be 3.33 pounds which seems more accurate) or we could wait until we actually catch that fish if we truly want to know the average weight of all.

The same applies to grading. If several assignments are missing, a teacher could assign a grade of “Incomplete” for the term and insist that work be completed in order to change the “Incomplete” to a standard grade. Completing the work is probably, in fact, the most appropriate consequence for missing work, and better for learning than simply receiving a low grade.

Stop Averaging Grades

We all just assume that an average is the best way to represent a series of grades. But why not the median or the mode? Imagine a student who gets these five scores on five assessments: 91, 92, 40, 94, and 94. How’s this student doing?

Using our traditional approach – the average – we would assign this student an 82 or a B-. However, is this more accurate than the median or mode? Is it accurate given that this student never scored in the B-minus range even once? What if the student just had a bad day (felt sick, bad news at home) on the day they received the 40? Does the average accurately represent what this student knows and can do? What about a different student who gets completely different scores (say, 82, 82, 82, 82, and 82) and has the same average? Does the grade of 82 tell you accurate information about these two different students? Or does it violate that first accuracy pillar?

Put More Weight into Recent Performance

Finally – and this is something we rarely think about – does it matter when in the semester the student earned the scores? For example, if a student came from a middle school that didn’t prepare her to write essays well, then she might get a D on her first essay (61), but as she learns to write better, she might get a few Cs (72,76, 72), then some B-minuses (82, 81), and eventually a B+ (89). Her average would be a C, but you can see she was slowly improving all semester, so does the average represent where she ended up? Or should we weigh her more recent scores more heavily – perhaps giving a B – to show what her skill level is now?

What’s more important: Where a student starts or ends up with her learning?

Think about these examples: If someone passes the bar on her third try, we don’t average in the first two failures. If a marathon runner enters a race, his final time is what describes his performance – we don’t average in his training sessions. Even as teachers – if we stumble at first but then get better at the craft, we are judged more on our more recent observations, not those initial missteps. Our earlier “grades” don’t reflect our achievement, they reflect our learning trajectory. What matters is ultimate mastery, and we need to allow for a diversity of ways to get there.

It’s a matter of equity because early grades can reflect a student’s circumstances. Compare the student above who came from a middle school that didn’t prepare her with another student whose background allows her to earn Bs right away. If they both end up earning a B+ on final assessments, should we punish the student who had the poorer preparation?

Finally, the use of averages with zeros together with the 100-point scale is particularly deadly. Imagine a student gets a B, then a B, and then doesn’t turn in an assignment. Using the 100-point scale, zeros, and averaging, that would be 85 + 85 + 0 (which is 170) divided by 3 (which is 57) so the student would receive a grade of F (57). A student who can achieve Bs (85) sees her grade drop almost 30 points by using the zero and an average. And even if the student had earned two more Bs after not turning in an assignment: B, B, didn’t turn in one assignment, B, and B, the average of 85, 85, 0, 85, and 85 is still a D (68). The zero makes it virtually impossible to bring up your grade. As Feldman puts it, this “atomic F” simply doesn’t “fit the crime of a missing assignment.”

Assign a Minimum Grade

So, if you absolutely must continue with a 100-point grading scale and averaging, what can you do? You could leave a blank in the gradebook or enter something non-numerical such as “M” for missing that will leave this grade out of the average, allowing the teacher to calculate a grade that more accurately represents what the student has learned (like the “incomplete” described above).

Or you could assign a minimum grade, such as 50. Many educators have significant reservations about this because they say the student didn’t master 50% of the material. But on the other side, the student also didn’t show lack of mastery either. We simply do not know. So, we would use the 50 to correct the disproportionate focus on failure of our current system. Instead, it would look like this:

Entering a 50 in the gradebook instead of a 0 would both increase the accuracy of the average as a representation of what the student has actually learned, and it would prevent the use of “0” as a nuclear bomb by allowing students to still have the possibility to improve their grade. Research has also shown that minimum grades do not contribute significantly to grade inflation or social promotion, but rather they encourage students who have failed to complete an assignment, stay committed to learning, and not give up. 

Replace the 100-Point Scale with a 4-Point Scale

If it seems overcomplicated or would be difficult to get buy-in for eliminating zeros, assigning a minimum grade, and using “incompletes,” then another path is to replace your 100-point grading scale entirely with the more accurate 4-point scale.

There are numerous ways we measure mastery in life that do not use the 100-point scale: drivers’ exams, the ACT/SAT, the bar exam, medical boards, etc. And we already use a 4-point scale for Grade Point Average (GPA). Why not use this easier grading scale?

Teachers are more likely to agree on what a “2” level of work means rather than distinguishing between a 77 and a 78. Plus, you can keep your zeros because now it’s proportionate. The average of B (3), B (3), and F (0) is now a C – much more accurate than the F (57) above as the average using the 100-point scale. Overall, both agreement and accuracy improve when you use a 4-point scale.

Nix the Group Grade

Lastly, we come to a well-known source of inaccuracy: the group grade. Often, teachers grade groups as a way to encourage teamwork and participation in a group project. When some students don’t contribute their fair share, we may still stand by the group grade because we see it as a representation of “the real world” in which coworkers do have to rely on one another and be accountable for their share. Unfortunately, we may be underestimating how complex it is to learn life skills like managing others, supporting peers, and handling conflicts, and using a grade as a carrot isn’t enough for students to learn these challenging skills.

Isn’t the purpose of group work for all students to learn the content?

Further, higher-achieving students may prefer to do more or all of the work to get a good grade while weaker students prefer to allow those students to do that work rather than reveal gaps in their own skills and knowledge. In any case, the purpose of group work is for all students to learn the content through the group’s work together and you cannot determine what each student has learned if the group grade is a “B.” This violates the ACCURACY PILLAR’s basic principles.

If students receive individual grades for their contributions in group work, then how do we motivate students to participate in and contribute to the group’s work? This is a much larger issue of motivation which will be addressed later, but for now:

• Ensure that students understand that the purpose of the group work is to increase the learning of each team member.
• Explain what the group is to do and how they will each be individually evaluated on what they have learned.
• Restructure projects to focus on the learning that takes place using rubrics like those found here: www.gradingforequity.org

2. Bias-Resistant Practices

In addition to Accuracy, our second pillar of equitable grading practices is Bias-Resistant. We need to ensure that our grades truly represent a student’s academic performance – not their behavior, circumstances, study habits, or our biased views of them.

Whether we realize it or not, we are including student behaviors and teacher bias when we do the following in our final grades: include extra credit, subtract points for late work, include behaviors like effort and participation, and include grades for formative assessment (like homework). Below are some ways to address all of this that may shock you, but that are actually much more equitable.

Stop Offering Extra Credit

Teachers sometimes offer extra credit to encourage non-academic or academic-related (but non-essential) activities. These extra-credit worthy actions by students could range from donating supplies to attempting challenging extra work to attending some outside function like a museum exhibit that might contribute to their learning. This type of extra credit, however, is more accessible to students with more resources and more family support (money for the supplies, a ride to the museum, etc.). Challenging extra work is most accessible to the students who have already learned the most and who therefore should be least in need of extra credit. In fact, research shows that students who are lower-achieving simply don’t do extra credit as frequently as their higher-achieving peers.

Further, it’s not honest. If a student has not yet learned the quadratic equation and then brings extra tissues to the classroom for make-up work to achieve a “B,” the grade is inaccurate, and the lack of achievement will hurt the student in the future. In the best case, extra credit is frivolous and in the worst case it works against equity and hurts students who are the most vulnerable.

So, what are the alternatives to extra credit? If the learning was important in the first place, then require that the student do the work. Students shouldn’t be able to skip work that is essential and just complete an unrelated task. This will mean accepting late work (see more on this below) because some students learn more slowly and may need more time or more support to learn. “Extras” like going to a museum or attempting challenging work would be better as suggestions than activities than can inequitably impact the grade.

Allow Late Work Without Reducing Grades

When we reduce a grade (or assign a zero) for late work we are introducing both inaccuracy and bias into the grade. The day on which an assignment is turned in does not tell us anything about how much the student has learned (inaccuracy), and it may well be a reflection of a student’s environment (bias). Students might turn in late work for a whole range of reasons from domestic violence incidents at home to overwhelming extracurricular involvement to simple forgetfulness.

Rather than including any of these nonacademic behaviors or student circumstances in a grade, it is more equitable to simply allow the late work, at least up to a point such as the date of the test or a week before the term ends. Punctuality is a “soft skills” behavior, not an academic one. It’s not to say that it isn’t important to be on time, but “B” quality work that becomes a “C” because it is late does not accurately represent what the student knows and can do. There’s a whole chapter on teaching soft skills at the end of the book.

Instead of penalizing lateness, help students to see the inherent value of timeliness – work does not accumulate and become overwhelming. You might track which assignments were late and report this to them and their parents outside of the grade. Then if questions arise about their achievement, you can point out that completing all assignments on time will help them learn the material.

An incredible thing happens when teachers stop taking off points for lateness – they not only receive more work from students, but they receive higher quality work because students don’t cut off their learning (due to events, stress, etc.) just to turn it in.

Keep Nonacademic Student Behaviors out of the Grade

In addition to avoiding grading a student’s punctuality with work, another way to make your grades more equitable is to avoid grading soft skills that rely on your subjective observations to evaluate. Who’s to say whether the student put “effort” into an assignment? Your personal likes and dislikes and your cultural perspective are statistically very likely to seep into the grade when you rely on your own evaluations of things like participation, effort, and citizenship, no matter how hard you try to keep them out.

As teachers we have biased ideas about what makes a student successful – being quiet, keeping your notebook out, etc. – usually based on what made us successful. We need to allow for a diversity of learning styles and not give points for obedience and control. While we want to encourage behaviors that help students succeed, and there should be consequences for rule breaking, these encouragements and consequences should not be part of academic grades. Address these skills and violations separately from grading.

Don’t Grade Homework

Similar to allowing late work, leaving homework out of the grade entirely is better for avoiding the bias of inadvertently grading students on their home environments. Help with homework, supplies for homework, and encouragement to do homework varies widely from one household to the next, and it is information that does not belong in a student’s grade.

Further, homework is not supposed to be an accurate representation of academic mastery; it is an opportunity to practice a concept that has just been taught. When we give homework for a topic we have just taught, we want students to grapple with it, we assume they will make mistakes in order to learn. The two purposes of homework are for the students to practice new learning and for teachers to get a sense of where their students are still struggling. Like in the previous section about averages, it doesn’t make sense to include practice (homework is really formative assessment) in a grade.

And when we do grade homework (either for a grade or for completion), students often skip it because they haven’t mastered the material yet, or they cheat (research shows about 80% of students admit to having copied homework). Both of these responses hurt the two purposes of homework – practice and teacher insight intro struggles. Further, when we grade homework that has been copied or that others have helped with at home it is not an accurate representation of what the student knows and can do. And grading homework serves to exacerbate inequities and widen the achievement gap. So we should stop giving a grade for homework.

Instead, we need to show students that homework is for the student, not the teacher, and helps with learning. Students understand the importance of practice – they willingly complete hours of free throws or dance moves for the performance. We’ve just taken away their intrinsic motivation and replaced it with the non-functioning extrinsic reward of points. But we can fix this.

Don’t Punish Cheating with a Grade

Cheating is one of the most common rule violations to be addressed via the grade. Teachers feel that cheating is such an enormous violation of trust between student and teacher that the common consequence is to give the student an atomic zero. But while this policy is an emotionally understandable response, it is not accurate in terms of representing what a student has learned, “You cheated on a test, so as your consequence, your grade isn’t going to reflect what you know!” Further, this policy invites bias because teachers may be more suspicious of some students than others and therefore monitor them more vigilantly for cheating.

Then what should be the consequences for cheating?

The most appropriate consequence for cheating is to require the student to redo the assignment with stricter monitoring so that cheating is not possible and so that their true level of content knowledge can be revealed. A non-grade consequence, such as an apology letter to the teacher or a call home, might also be applied. The most common form of cheating mentioned above – copying homework – will also be disincentivized if you accept late work and do not grade homework.

Grade Summative Assessments Only

If we should no longer be including grades on homework, effort, participation, citizenship, or extra credit, what should we be grading? 

The answer is that you should be grading summative assessments. Only. It’s radically different from our traditional practice and will take some getting used to, but it is the way forward if you intend your grade to be a valid reflection of what a student has learned. It may require you to rethink your summative assessments to be sure they cover the full range of content you expect students to learn and allow students to express their learning in various ways (written, visually, PowerPoint, etc.).

What about students who suffer from test anxiety? Will this put even more pressure on the end-of-unit test?

It does not need to. For one thing, you can and should offer retakes to ensure you are grading learning and not timing. When a student fails to demonstrate a skill on Tuesday, but can successfully demonstrate it on Friday, she has demonstrated her learning just as well as the student who could demonstrate it a few days earlier. Some students need more time, and some may need the experience of not knowing the answers to questions on the test as part of their learning process.

Additionally, it is up to you what you count as a summative assessment. A student who aces a pre-test has demonstrated mastery of the content – that could be her summative assessment. A student who doesn’t perform well on the test due to anxiety, but who is able to verbally explain the concepts to you the next day with great accuracy has also demonstrated mastery of the content – that could be his summative assessment (and if you are grading on a 4-point scale, it would be much easier to give a grade to his verbal explanation – a good explanation might get a 3, while an excellent one might get a 4 – than trying to assign it a score out of 100).

Will students stop trying on homework and classwork if nothing counts but the test?

At first, some might. They may need to learn through trial and error, but ultimately you will help them make the connection between effort and learning. Even if you are no longer grading homework and classwork, you might keep track of completion so they can begin to see the connection between working along the way and the final grade. This is by far a better life lesson than the lesson that comes from believing an authority figure is watching your every move and mistake and making it harder for you to succeed.

3. Practices that Motivate Learning

Beyond Accuracy and Bias-Resistance, our final pillar of equitable grading practices is Motivational. According to Feldman:

Overall, we want grades to motivate rather than discourage students. We want to focus students on the learning, not on collecting points, which is what we do now.

Unfortunately, we have an inaccurate view of the relationship between grading and motivation: we believe grades will motivate students to try harder and perform better. In reality, no research shows that grades motivate students. In fact, it is our own grading practices – with 60 points dedicated to failing out of 100 – that have taken away hope and encouragement from our students. The equitable practices introduced so far help to motivate students, and below are some additional practices that can help.

Require Retakes

Allowing students to retake assessments is not only humane by giving kids a second chance, but it is motivating. However, rather than simply allowing retakes, we ought to require them (in a class period or an “intervention block”). This emphasizes the fact that making mistakes is part of the learning process and therefore students should be given the chance to demonstrate what they’ve learned from their mistakes. Requiring retakes (rather than simply permitting them) is the only equitable way because it is often only the higher-performing students’ parents who pressure them to ask for an optional retake, but all students can benefit from the opportunity. And one more thing – allowing retakes is also an excellent antidote to test anxiety because the student always has another chance.

How can I possibly manage all these retakes?

If you are only grading summative assessments, then you need only offer retakes on summative assessments. If many students take them, you can allot class time. You might also consider structuring your summative assessments so that they are organized by learning standards and can be retaken in chunks. A student who performs well on the parts of the assessment that evaluate standards X and Y, but not Z, need only retake the part that evaluates Z.

What about preparing students for college or the real world?

The reality is that retakes are frequently allowed in the real world (think driver’s test and bar exam, for example). They are often available in college, too, if students ask for them and have a good reason. The best way we can train students for the real world is to teach them how to learn from mistakes, how to study and improve, and how not to give up. Further, in the real world, retakes are often available for those who speak up. We can teach students to speak up and ask for that second chance.

Even if students fail early on, or take a long time to learn something, with retakes they can trust that their learning will not be cut off and that they can always have the chance to learn more and show that they have improved. What could be more motivating than that?

Rename Grades

A common demotivator among students is the feeling that they can’t understand how to succeed and therefore will never have a fair chance. When grades are derived by complicated methods, using weighted equations only computer software can solve, and roll many subjective judgements into the grade along with academic achievements, grades really can feel like a mystery – a mystery some students will give up on solving. We have the following driving principle to address this:

We can motivate students to stay engaged in learning when we “lift the veil” and are explicit about how grades are derived and what they mean. We can make it clear to every student how they can work their way to success. One practice that was already addressed is to ensure that grades only represent academic mastery, not a hodgepodge of nonacademic behaviors and personal attributes.

Another way to lift the veil is to rename grades. Instead of circular definitions where an A is defined as 95% and 95% is defined as an A, we can put the grades in plain language. The student is “Exceeding the Standards,” “Meeting the Standards,” “Approaching the Standards,” “Has Not Yet Met the Standards,” or there is “Insufficient Evidence.”

Use Rubrics to “Lift the Veil” on Grading

We can also use rubrics to clearly explain what the standards are and what it looks like to meet or exceed them. Rubrics democratize the power to evaluate. Keep them clear and objective (and translated into other languages when needed) and all parents and students can read and understand how a grade is earned and what it takes to improve. Students don’t have to guess. They can even self-assess.

Don’t make the mistake, however, of totaling the points on a rubric and applying a ratio to determine a grade out of 100 for the assignment. In several ways, that defeats the purpose of the rubric. The rubric gives the opportunity to evaluate and communicate whether learning standards have been met. That is the information you need for your grade (not points out of 100).

Additionally, the rubric could easily include categories such as “teamwork,” which shouldn’t figure into the grade, but can be noted nonetheless so students (and parents) know how students perform in these areas. See this or other sample rubrics at gradingforequity.org.

Standards-Based Scales and Grade Books

Imagine how much clearer it would be if instead of getting an 88% on a test, a student got a sense of how well she had mastered each standard. Take a look at this sample checklist of standards mastered from the book that a teacher uses to show the student which standards she has mastered.

Rather than getting the result of the correct number of points divided by the total points, say 77% – which is essentially meaningless – a student might see this on her test: “You have earned a C+ on the immigration standard because you did not yet demonstrate knowledge of both a government and a public response to immigration.”

Creating standards-based scales is complex and takes time: for each standard, the teacher would need to describe what it looks like to “meet,” “exceed,” and “approach” the standard. There are resources online to help. And with standards-based grading, rather than a student getting a “B” in math, she would get a grade or perhaps a descriptor (like “meets,” “exceeds,” or “approaching”) for each standard. Neither of these practices is easy to set up and there is a lot more detail in the book if you choose to pursue them. There is an example of a standards-based grade book on page 198, and many more examples at gradingforequity.org.

Building “Soft Skills” Without Grading Them

The whole idea of this book is that students don’t learn because we give them grades and we need to change the way we use grades if we truly want students to learn. Because educators – rightfully so – see their jobs as teaching “soft skills” as well as “hard” academic ones, the same thing goes for these types of skills. If we want to teach skills like timeliness, responsibility, participation, and teamwork, we need to stop awarding points and instead help students connect the dots between all these “soft skills” and their own learning and achievement. Doing homework and bringing your book to class helps you learn. Participating during a group project helps you learn and maintain positive relationships. In the real world, no one will follow your students around with points and penalties for every action. Instead, they will need to learn that their actions have meaning and an impact on the outcomes they achieve.

The final chapter of the book is an incredibly nuanced and beautiful argument for why and how to teach the most essential of 21st-century soft skills – self-regulation – which involves both goal setting and feedback. When we move the point-system out of the way and instead support students to develop the independence to know how well they’re doing, set goals, and make adjustments on their own, then they will know how to learn and can continue learning for life, without any grades necessary.

THE MAIN IDEA’s PD suggestions that go along with this book are not included in this html link. To see those ideas, go to www.TheMainIdea.net and download the full book summary.