Improved Grading Makes Classrooms More Equitable

Ask any faculty member about how they grade their students, and they will probably explain the precise weights they give quizzes, tests, papers, labs and other factors -- as well as how they average student results over the term to determine a final grade.

Even though the scholarship, technology and pedagogy of postsecondary courses have significantly evolved in the last century, the ways students are graded has remained unchanged. This should come as no surprise, considering that most college and university faculty members receive no training in how to grade, either in graduate school or professional development on the job, and so most typically grade as they were graded. Plus, because faculty members rarely receive support to examine and learn about grading, each professor’s grading policies are filtered through their own individual beliefs about how students learn, how to motivate them and how best to describe student achievement.

As a result, grades often vary within a department and even within a course taught by different instructors. That is particularly true at community colleges, which depend heavily on part-time faculty who are rarely involved in any deep way with the department in which they teach, but it is also often the case in research institutions, where grading is often the responsibility of teaching assistants, who rarely discuss grading practice with faculty members or department chairs.

While faculty members believe that their grading practices are fair and objective, a closer look reveals that they are anything but. And while employers and other institutions rely on those grades as a reliable marker of student achievement, it might shock them to know how much grading practices reflect the idiosyncratic preferences of individual faculty members.

Two examples:

• Frequently, faculty members incorporate into a student’s grade many highly subjective criteria -- such as a student’s “effort,” “participation” and “engagement” -- behaviors which the professor subjectively witnesses, interprets and judges through a culturally specific and biased lens.
• Many faculty members grade on a curve, which makes grades dependent on the particular students in that particular classroom in that particular term. It unhelpfully describes student achievement not based on what the student learned but rather on how well they did relative to others in the class. Plus, this method translates learning into a competition, which adds stress that undermines collaboration and has been found to inhibit learning.

But beyond the fact that a grade can be more reflective of how a professor grades than what a student learned, many common grading practices have a more profound consequence. As colleges and universities become more committed to enrolling and retaining historically underserved students, they rarely recognize how traditional approaches to grading frequently result in perpetuating achievement disparities -- advantaging higher-income and privileged students with inflated grades, while lowering grades for underserved students and painting a misleading picture of performance. For example, professors typically calculate grades by averaging a student’s performance over time, which actually punishes those who enter the class with less prior knowledge, even if they ultimately master the content, and hides their growth.

In advising faculty members at a variety of institutions -- including the College of Marin, Stanford University, the State University of New York at Cobleskill and the University of California, Berkeley -- I have found that educators at all levels of schooling, including college faculty, are motivated to try alternative, more equitable grading practices once they learn about the harms caused by our century-old approaches to grading.

These improved grading practices:

• Are mathematically accurate to validly describe a student’s level of mastery. They apply a more proportionately structured 0-4 scale instead of the 0-100 scale, which is mathematically oriented toward failure. They also use sound mathematical principles that reflect recent performance and growth instead of averaging performance over time.
• Evaluate students based on their knowledge, not their environment, history or behavior. They exclude subjectively interpreted behaviors, such as a student's "effort" or "participation." They determine grades based on a student’s demonstration of course content and not homework completion, and they don’t use grades to reward compliance.
• Support hope and a growth mind-set. They allow test/project retakes to emphasize and reward learning rather than penalize it, and they override previous scores with current scores that build learning persistence.
• “Lift the veil” on how to succeed. They create explicit descriptions of what constitutes demonstration of content mastery through rubrics or proficiency scales. In addition, they simplify grade books and expand the methods of assessments to generate more accurate feedback and reporting about each student’s learning relative to the expected outcomes.

Faculty members, rightly so, need to own their grading, so colleges that want to explore new approaches must support learning communities where professors can critically examine current practices and try more equitable approaches. One model of this type of learning incorporates a research cycle in which faculty members try new practices in their own classes, examine changes in their students’ performance and motivation, share results with colleagues, repeat the cycle to continue integrate new approaches over time, and ultimately build a body of evidence that supports common and coherent grading practices.

Professors who have gone through this kind of process become comfortable and confident with equitable grading, and they report that it provides them an opportunity to uncover and reimagine what effective learning could look like. They find that their students experience less stress in learning and shift their focus from earning scores and behaving in ways that appeal to the teacher’s cultural expectations of a “good student” to learning and achieving demonstrations of course content mastery.

Independent researchers have also found that, in secondary schools, such practices result in fewer D’s and F’s, particularly for underserved students, and lower costs for remediation. They also improve engagement and student-teacher collaboration, reduce grade inflation, and more closely align grades with test scores. As increasing numbers of colleges and universities pursue this work, researchers are also conducting similar investigations at the postsecondary level. In their preliminary feedback, faculty report similar results in their own classrooms with their own students.

If we are serious about eliminating achievement disparities, we have to be willing to tackle grading, an important but largely unaddressed aspect of higher education teaching and student success. With college faculty leading this work, postsecondary institutions can grade in ways that more consistently and accurately reflect student learning and counteract long-standing inequities. We can foster a conversation about why grades are important, what we want to measure and not measure, and how to ensure that our grading practices reinforce -- rather than undermine -- our commitment to excellence and equity.