Over the last 10 years, classroom assessment specialists have been encouraging schools to make major changes in their grading and reporting systems. The problems with traditional grading are many, among them being the focus on mechanical processes and mathematical precision, often calculated to several decimal places. This focus unfortunately leads students — and parents — to fixate on the numbers rather than on the learning. Students become obsessed with how many points they need to earn on the next test to keep a B instead of what they need to learn to really master the subject.
Some may favor the traditional system of letter grades, percentages, and the 4.0 scale because it is familiar and aligned with the way that most colleges operate (Peterson’s Staff, 2018). But a growing number of college admissions officers find grade-point averages to be of little use (Marklein, 2013). The cynicism of these college admissions offices about grades is well warranted, as grades do not typically represent student achievement but rather an amalgam of achievement, behavior, compliance, and test-taking skill. In the competition to have the highest GPA, an artificial points game trumps learning and genuine accomplishment. In addition, this approach is responsible for many otherwise capable students failing courses and not completing high school — sometimes because of issues as simple as not completing homework. Even if high school GPA is the most accurate predictor we have of college students’ first-year performance, the current system still results in a high percentage of college students having to take remedial courses and a scandalous first-year failure rate, which is a good sign that the system isn’t working as well as it should.
This system of using points, often to reinforce and punish behavior, is not the best way to prepare students to be the self-directed, independent learners they need to be for lifelong success. Grades that include behavior and achievement not only fail to give an accurate picture of student achievement but also produce students who are more concerned about the accumulation of points through compliance and extra credit than they are about learning. By using grades to compel hard work and good behavior, teachers are leaning on extrinsic motivators, instead of promoting an intrinsic motivation to learn, which is more effective in the long term.
A better system would produce grades that are fair, accurate, specific, and timely (FAST). Reporting student achievement in this way has a much greater likelihood of producing knowledgeable and reflective learners who understand that learning is about more than crunching numbers and jumping through hoops. Let’s consider the FAST qualities that leaders can use to guide conversations on improving grading practices in their schools.
For grades to be assigned fairly, schools must agree on their purpose. We see the primary purpose of grades as communicating current student achievement to whomever has the need and right to know that information, including students. This means that all students must have equal opportunity to learn and to show what they know, understand, and can do.
One way to establish this agreement begins with setting forth a mission and vision statement and guiding principles that emphasize learning and success for all. See, for example, the vision statement from the Arts and Technology High School in the West Linn-Wilsonville School District in Oregon:
Arts & Technology High School is a dynamic learning community that supports the success of our diverse student population. All learners collaborate with staff to develop and commit to a personalized education and social/emotional support plan to ensure academic success.
To put this vision into action, the school states that:
At ATHS we approach assessment and reporting as an important part of a student’s understanding of his/her progress as a learner, as well as a tool for communicating progress to parents, colleges, and the community. In order to ensure that grades accurately reflect students’ learning, we are implementing a proficiency-based grading system in all our classes. This system separates academic grades from reports on student behavior.
Patterson (2003) stated that “fair does not mean equal; yet, when it comes to grading, we insist that it does” (p. 572). Traditionally, teachers have expected all students to do the same assessments in the same amount of time, but to provide fairness in assessment and grading, we must recognize individual differences by using practices that acknowledge that fairness is equity of opportunity, not uniformity. Treating all students the same when they are obviously different in many ways is unfair. This means, for example, that the time available on tests and exams must be flexible not fixed, and that students should almost always have a variety of ways to demonstrate their knowledge, understanding, and skills.
Leaders must ensure that the grades students receive in their school are an accurate representation of students’ achievement and that behaviors are reported separately.
The idea of fairness as equity of opportunity was brilliantly stated by the Ministry of Education, Citizenship and Youth in Manitoba, Canada. It stated, “All students are given an equal opportunity to demonstrate what they know and can do as part of the assessment process. Adaptations to assessment materials and procedures are available for students including but not restricted to students with learning disabilities to allow them to demonstrate their knowledge and skills, provided that the adaptations do not jeopardize the integrity or content of the assessment” (Manitoba Ministry of Education, 2006, as cited in in O’Connor, 2018; our italics for emphasis). It is particularly important to note that the second sentence means that fairness requires that all students, not just students with disabilities, have access to accommodations that don’t change the what or the how well of the learning goals being assessed.
Because students, families, and educators make so many important decisions about students on the basis of their grades, leaders must ensure that the grades students receive in their school are an accurate representation of their achievement and that behaviors are reported separately. As noted above, traditional grades have often reflected not just achievement, but also behavior, compliance, and test-taking skills. The result is grades that have no meaning because they are an uncertain mix of achievement and behavior. What is needed is a report card like the provincial report card in Ontario, Canada, that provides information on six “Learning Skills and Work Habits” (Responsibility, Works Independently, Collaboration, Organization, Self-Regulation, and Initiative) and a grade that accurately summarizes achievement of the learning goals/standards.
Evidence used to determine grades should come from a limited number of high-quality assessments of learning that allow students to demonstrate a full range of performance, not a large number of assessments (many low quality) that are limited in scope, and certainly not from assessments for learning (i.e., formative assessments). Assessment is a sampling procedure, so teachers will need to accumulate enough evidence to make valid inferences about achievement, but the more assessments there are, the more likely they are to be of low quality and the less each one will matter.
Accuracy also requires that grades be based on more recent evidence, not all of the evidence over a school year, and that students have multiple opportunities to show what they know, instead of having to rely on one-shot assessments. In addition, leaders must support their teachers’ professional judgment, not just calculation of the mean, when determining final grades. Because the mean is calculated using scores from across the grading period, it is often not an accurate representation of current achievement (see Wormeli, n.d.).
Leaders must require the elimination of common practices that compromise the accuracy of grades. These include penalties for late work, academic dishonesty, absences, and inappropriate behaviors; extra credit that awards points for behaviors that are unrelated to the standards; group scores; any semblance of a bell curve; zeros on the 101-point scale; and almost all homework. (To be clear, we certainly believe in having students practice, sometimes at home, but effective practice includes feedback, differentiated work just outside the student’s comfort zone, and the flexibility to proceed to the next level of difficulty — properties almost universally absent from traditional homework, especially when it’s made part of a student’s grade.) Because the grade should include only the data that accurately represent the student’s current performance, it is not necessary for teachers to devote inordinate amounts of time entering all these data into a master database.
One of the most common sources of inaccuracy in grading, particularly in schools with diverse student populations, is the tendency to confuse proficiency in English with subject-matter competence. A student may be able to demonstrate mastery of number operations when the questions are presented with numbers and symbols, but fail the same subject when the questions are embedded in unfamiliar language. For that student, an accurate grade for Mathematical Operations would be an A, with a C for Mathematical Communication. While both of these proficiencies are important, it would not be accurate to award a grade of C in math, which the student has largely mastered, because the student lacks advanced proficiency in English.
The role of the teacher is to develop, encourage, and extend learning. To accomplish this, we must be timely in our feedback.
Specific grades are grades based on standards and learning goals, not assessment methods, and grades based on clear descriptions of a limited number of levels, not points and percentages.
For a wonderful example of the principle of specificity, visit a music class or watch a practice for an individual or team sport. You will never see the chorus conductor or track coach making notes about every mistake and then, at the end of a musical performance or athletic season, announce, “Well, I guess that was about a B-plus.” Although classes in music and physical education do often result in a letter grade, the mark on the transcript is entirely secondary to the moment-to-moment highly specific feedback that directors and coaches provide. In these disciplines, the grade of A does not reflect perfection — the absence of wrong notes, failed goals, and missed defensive moves. Instead, it reflects how students respond to mistakes.
Contrast this to the typical system in which the final grade is a mechanical and mindless calculation that reflects not the students’ progress, but punishment for every missed homework assignment and wrong answer along the way. We argue that the principle of specificity allows teachers to both point out errors and praise the learning that results from the errors. Indeed, an error-free class might as well be called a learning-free class.
The principle of timeliness is directly related to a fundamental purpose of classroom assessment — to provide information that improves teaching and learning. Even if grades are based on psychometrically perfect assessments and calculated to the ten-thousandths of a point, they are of no value if they are not presented in time for students and teachers to analyze the root causes and correct them. Almost every reader of this article knows at this very moment which of their students are in danger of low grades at the end of this reporting period. We should take scant comfort in the fact that our predictions may be accurate. The question is, “What are we and our students going to do about it?”
Certainly, all educators understand that the more quickly information is given, the more quickly students and educators can work to make adjustments and improve. The role of the teacher is to develop, encourage, and extend learning. To accomplish this, we must be timely in our feedback. If given feedback in time, a writing student is able to improve low-quality drafts of a paper before creating a final portfolio, or a student coder is able to correct bugs before a final project is due.
We see the primary purpose of grades as communicating current student achievement to whomever has the need and right to know that information.
But measuring learning and communicating in a timely fashion can present real challenges for educators of all levels. Secondary-level teachers have hundreds of students, and elementary teachers have to assess learning for every subject area. With this in mind, one of the most important questions educators can ask is, “Am I measuring what matters?” If we have more checklists, quizzes, and assignments than we have time for, it may be best to reduce the quantity of assessments in order to increase the quality. Thoughtfully assessing students’ performance on a single project that showcases their skills authentically across multiple standards may be a better choice than marking many quizzes that provide little fuel for reflection and improvement. Leaders should grant permission to their faculty to “cut the fat” and keep their assessment efforts lean.
Leading the change
It is up to building and district leaders to gear up for the urgent change that is needed by putting in place the supports and the pressure to move assessment and grading in this positive direction. As we move forward with change, we must recognize that requiring complete buy-in is a misguided approach. At best, it supplants professional engagement with superficial acquiescence. At worst, it drives opposition away from potentially useful dialogue and into the parking lots, streets, back fences, board meetings, and newspaper editorial rooms. We need to replace the calls for buy-in with a call to test hypotheses. In other words, leaders need not ask faculty to buy in or agree; what they need is a commitment to try these ideas and then carefully measure the results.
Remember that change requires action, not necessarily agreement. We must reject the “pep rally” model that attempts to garner universal excitement for change. Rather, we can use what researchers have called the “nudge” factor (Thaler & Sunstein, 2008) to create circumstances where the FAST approach is more appealing and easier to implement. For example, if the leadership changes the grading scale from a percentage-based, 100-point scale to an integer-based, five-point scale, the computer-generated average percentage is eliminated, nudging teachers toward assessing proficiency. Nudges can be applied through mandatory professional learning experiences, changes to policy and procedures, and revised report cards. The latter two are ultimately essential but are best developed through ongoing dialogue and constant communication with all stakeholders.
Even though we’re not asking every stakeholder to agree with grading reform, it is important that we build common ground (i.e., a shared understanding of purposes and principles). Parents, communities, unions, and thought leaders may never agree on a specific grading procedure, but it’s probable that they will agree on values, such as the desire to build personal responsibility and preparedness for the world beyond school. This work to build common ground is the heavy lifting that must occur for change in procedures to be successful. The social and cultural context has been missing in many change models, but it has particularly undermined proposed changes in grading policies. Parents especially need to understand that high grades alone won’t get their children into the elite college of their choice. As a recent New York Times article pointed out, “Like it or not, colleges aren’t looking to reel in the greatest number of straight-A students who’ve taken seven or more Advanced Placement courses. A rejection isn’t really about you; it’s about a maddening mishmash of competing objectives” (Hoover, 2017).
To make change, we must understand the rationale for it. If a school building is on fire, if a bus is unsafe, or if there are guns in the hallway, we don’t deliberate about whether it’s a good idea to keep students safe. We act immediately to safeguard our students. Likewise, if fundamental values, such as equity, access to education, and fair treatment for all students, are threatened, we must act decisively to promote and protect those values. Grading practices that have the potential to reduce failure, reduce dropouts, and improve school safety are, indeed, urgent.
Getting past tradition
Much of the defense of traditional trading systems is based on the claim that, “It’s not our fault that we use toxic grading practices, because we have to get them ready for college!” However, a growing number of colleges, including MIT and Wellesley, are substantially reforming their grading systems, to the point that they provide rich feedback, but no letter grades, to first-year students. And while some colleges and universities do have awful grading practices, terrible pedagogy, and hazing rituals by social groups, wise K-12 educators do not allow those practices in higher education to justify unacceptable practices for younger students.
Most schools that have moved to more effective grading practices continue to provide grades and traditional transcripts, but the Mastery Transcript Consortium, a group initially composed of over 50 leading American independent schools that has now grown to over 170 schools, is proposing a drastically different transcript that would emphasize proficiencies developed over a student’s years in high school and would not provide grades for subjects (see www.mastery.org).
Many districts have been successful in moving to more effective grading practices, but far too many have not begun the journey. Others have tried and failed, perhaps because they took the wrong approach to change. Effective change requires a sense of urgency, common ground, and action, but it doesn’t necessarily require universal agreement. Change is best achieved through a judicious balance of pressure and support. If we only operate at the pressure end of the continuum, it may appear that change is occurring but it will only be surface change and there may be lots of subversion. On the other hand, if we only operate at the support end of the continuum, there is very little real change. Creating circumstances that “nudge” teachers and other stakeholders toward the ultimate goal may be the best way to start.
Hoover, E. (2017, November 1). What colleges want in an applicant (Everything). New York Times.
Marklein, M.B. (2013, February 27). Grades pointless? Some colleges don’t care about GPAs. USA Today. www.usatoday.com/story/news/nation/2013/02/27/college-grade-point-averages/1947415
O’Connor, K. (2018) How to grade for learning. Thousand Oaks, CA: Corwin.
Patterson, W. (2003). Breaking out of our boxes. Phi Delta Kappan, 84 (8), 568-574.
Peterson’s Staff. (2018, January 29). College admissions requirements and your GPA. Peterson’s. www.petersons.com/blog/college-admission-requirements-and-your-gpa
Thaler, R.H. & Sunstein. C. (2008). Nudge: Improving decisions about health, wealth, and happiness. New Haven, CT: Yale University Press.
Wormeli, R. (n.d.) It’s time to stop averaging grades. Association for Middle Level Education. www.amle.org/BrowsebyTopic/WhatsNew/WNDet/TabId/270/ArtMID/888/ArticleID/21/Its-Time-to-Stop-Averaging-Grades.aspx
Originally published in May 2018 Phi Delta Kappan 99 (8), 67-71. © 2018 Phi Delta Kappa International. All rights reserved.