Creating Inclusive Grading Structures

Assignments and the associated grades represent one of the most powerful tools in our teaching repertoire. It takes great care and precision to design assessments, and how we measure and translate student achievement into those assessments and into grades. Before delving into specifics of individual assessments, first think about the landscape of your assessments and what they measure (in particular, how those tie into your learning outcomes).

The types of measurement: grading policies

For most students, grading policies are the most impactful component of a course, and for most instructors, assigning grades and responding to questions and/or complaints about grades can be the more frustrating part of teaching. Because grades play such a large role in our education system, your policies are worth critical reflection.

There are three common philosophies when it comes to grading:

• Comparison: In many instances, grades are used as a way to sort students. Who is more deserving of a scholarship, entry into a professional program, or a specific job? Bell curve grading systems are also rooted in this comparison-based structure by assigning certain grades based on ranking within a class.
• Absolute achievement: For many, grades are supposed to represent achievement of a certain set of knowledge, skills, and abilities. This philosophy reflects the ways in which grades may serve as a gateway for entry into higher-level classes or specific opportunities. In these systems, a grade should accurately reflect a students’ accomplishment of learning outcomes through set tasks and measures. Any number of students can earn each specific grade if they have achieved the appropriate level of learning.
• Growth/Development-based: Many people assume grades are a representation of learning. True learning, however, reflects a process of change, growth, and development. For grades to function in this way, they should measure the degree of change from a student’s starting point in the class to their exit from the course

Each of these systems has significant problems and challenges:

Comparison

• Unclear what students have actually learned/achieved, both to them and to others.
• Promotes an atmosphere of competition, which can undermine feelings of relatedness and belonging as well as effective study habits (for most students communal studying is more effective than individual studying).
• Students are often uncertain what grade they will receive until the end of the semester and may not understand why they achieved that grade.

Absolute achievement

• May advantage students based on their opportunities and backgrounds before entering the class.
• It is really difficult to make accurate and unbiased assessments of particular skills and knowledge.
• When do we measure? Absolute achievement system students may not provide a lot of opportunities for students to be able to practice skills and get feedback.

Growth/Development

• Requires highly individualized attention to know where each student starts and ends (as well as individualized attention to design a path for growth and development).
• It is unclear what each student has actually learned or accomplished because the grade reflects a degree of change, but not a starting or ending point.
• May be perceived as overly subjective and/or influenced by bias.

So, what do I do?

None of these systems are perfect. Nor is there consensus around these systems locally or nationally. Most instructors use some type of hybrid system that incorporates at least the first two systems, if not all three. However, these hybrid systems or the philosophy regarding what grades mean in a course are rarely discussed with students, other faculty, administrators, or community and industry partners. Considering that, when constructing your class and its grading system, think about and design based on the degree to which you want each of these systems reflected. Also, communicate this with your students through your syllabus and discussions.

Map your landscape

In designing courses, we can think about our assessments not as separate independent pieces, but a landscape where everything works together toward both supporting and accurately measuring student learning. This perspective also allows us to build in smaller structures that will help learners track their own success and development (often called “metacognition” — their knowledge of their own learning). This approach often means providing more assessments with lower stakes (points toward a final grade) and developmental feedback (which can be automated in some cases). These low stakes assessments can scaffold larger assignments that provide summative assessment of achievement and, therefore, a higher percentage of the course final grade.

As you consider your approach to grading, remember to connect with your learning outcomes. For example, you might assign relative percentages to how each outcome will contribute to the overall or final grade. You might also start thinking about overlap. Cognitive outcomes in some classes overlap quite closely as foundational knowledge goals form a building block toward application and integration-based outcomes. Don’t think of these percentages as set, but rather as guidelines to return to as you design specific assessment tools to maintain balance throughout a course.

Two emerging structures to consider

Specifications grading

Specifications grading strives to emphasize student mastery of specific measurable skills and abilities (an absolute achievement orientation). It is done through a series of specifications for what is considered acceptable work (often aligning with a traditional B- grade level), and multiple opportunities for students to demonstrate achievement of those goals. Setting up a specifications system involves significant pre-work to outline transparent goals and measures. It also often involves creating bundles of achievements, so that there is clarity regarding what skills must be demonstrated to pass the class (a grade of C), and what additional skills may be added for higher overall grades. This system provides benefits for courses in a sequence that rely on cumulative skills because it offers clarity that every student passing has demonstrated mastery in the same core set of skills necessary for moving on.

Engaging in a specifications structure places significant emphasis on writing effective learning outcomes and objectives, which are tied to clearly measurable demonstrations of achievement. Everything measured in a specifications structure needs to fall into a pass/not-yet system of measurement. Often this makes grading and feedback more streamlined and transparent (students will know when they have demonstrated mastery or what barriers they still need to overcome to demonstrate mastery). Each of these measures needs to be repeatable at least to a sufficient degree that students can struggle initially and learn from and overcome their challenges in a later assessment of the same skill.

While on the surface, specifications systems are very highly structured, they also have the potential for significantly supporting student autonomy. At a basic level, it provides clarity on different levels of achievement in a class. A student may choose to focus only on skills associated with earning a passing grade and not worry about some of the more complex or complicated topics in the class. Alternatively, bundles may be constructed for students to have options based on future goals. If a class combines students from three majors, there may be three different bundle options for earning an A that align with skills most relevant to those different majors. This balance of autonomy and structure allows an instructor to ensure that every student in the class has the same core set of skills mastered at the end of the class, while also allowing flexibility in ways students can stretch their capabilities to align with their goals and interests.

Student-reflection based ungrading

To address the challenge of measuring growth for many students, many instructors have turned to systems that focus on student development of metacognition and measuring their own learning. While varied, these structures often under the broad umbrella of “ungrading” do not grade individual assignments or exams, but instead ask students to assign a grade to their own work based on reflection about that work and goals the students set for themselves. These self-assigned grades become a point of discussion between the student and instructor ultimately leading to a consensus about what a final grade should be. These reflections and discussions take place at multiple times throughout the course to add clarity around goals and achievements with the aim of transparency.

One of the most common implementations of this student-reflection-based grading schema focuses on class participation and engagement. Students have different levels of comfort speaking in class, and faculty often struggle to track participation in discussions because their primary focus is on facilitating a meaningful discussion. By having students describe their regular discussion habits, set meaningful goals for themselves, and reflect on their progress toward reaching those goals alongside their contributions to the class environment — thus providing a broader range of ways students might make their engagement visible — the process of assigning grades around student participation and engagement becomes fairer.

While specifications grading systems begin with very high structure and instructors must provide opportunities for autonomy to create a balance. Student-reflection based systems experience this challenge from the other side. They begin with far more autonomy in a way that can leave some students feeling like the system is chaotic, thus instructors need to build in sufficient structure support students to develop meaningful goals and in how to gauge progress toward achieving those goals (including valuing the act of setting a very high goal and not quite reaching it).

Other considerations

Joe Feldman, in his 2018 book, Grading for Equity, offers some provocative ideas for considering how our grading systems may undermine their function in measuring student learning and providing equal opportunity for all students in a class. Some of these are included below for your consideration. Note: We understand that the following strategies are historically embedded in assessment and that students often request some of them (e.g., extra credit), however, research often provides mixed evidence of their effectiveness, revealing opportunities for inaccuracy and bias. We present these not as necessary practices to implement and/or avoid, but as thought-provoking opportunities for reconsideration.

Avoiding zeros

Zero points is often the result of an assignment not being turned in or a penalty for not following directions on an assignment. When translated to standard 100-point grading scales, a single 0 can have an outsized effect on a student’s grade average (the most common calculation of final grade). For example, if there are five quizzes and a student earns four As and has a 0 for a missed assignment, the average often works out to a C. Feldman argues that, in demonstrations of achievement, the student has most consistently earned an A (both the mode and median of the set of grades). The answer is not to ignore a missed assignment or quiz entirely, but if we can find opportunities for a grade to be earned to replace the 0, we are more likely to accurately measure student achievement with our final grade.

Re-weighting or replacing older grades

When it comes to measuring achievement, Feldman suggests weighting recent performances related to a particular outcome or objective more highly than early work. If a final grade is supposed to measure achievement at the end of the course, it seems disingenuous to weigh a student’s performance related to an outcome at the start of the semester equally with their performance at the end. By extension, our assessment will be more valid when it is derived from multiple opportunities for demonstration as well as multiple modalities of demonstration. To achieve this goal, it is important to have a clear mapping so we can replace grades based on particular course objectives or outcomes (often represented by sections of a test rather than the whole thing).

Assigning individual grades for group projects

One assessment that is often embedded in course outcomes is performance within groups. Here too, Feldman emphasizes assessment that reflects individual achievement rather than assigning group grades. The practice of assigning grades to groups is typically viewed as a way to enhance motivation. In reality, grades assigned to groups may lead to inaccuracies in measuring an individual group member’s contribution to the group and demonstration of their learning. Collaborative learning is highly effective, but assessments in alignment with learning outcomes and objectives can be tied to a student’s contributions to making the group function well as a team.

Valuing knowledge and skills, not behavior

Feldman distinguishes between assessing the mastery of knowledge and skills and measuring nonacademic behaviors. The latter invites unintended bias because it calls upon an instructor to interpret behavior without complete information, and assume all students come from similar environments, have access to the same resources beyond course requirements, and can exercise control over their circumstances. Including measurements of nonacademic behaviors may inflate or deflate a course grade meant to reflect student achievement of learning outcomes. It may result in an atmosphere that students experience as controlling and punitive rather than supporting their learning. While many of these strategies come from good intentions to support students, they do not provide valid evidence of a student’s content knowledge or skills achievement. 

Avoid extra credit by incorporating activities into the class

Students often request extra credit assignments, and instructors create fun activities that may or may not be tied to course learning outcomes. These extras often involve participation in activities, knowledge, or skills that are not part of the class and/or are outside class time, which means they may not be equally accessible for all students in the class. Even when directly tied to learning goals, we need to ask what differentiates it as extra credit. Feldman advises: “If the work is important, require it; if it’s not, don’t include it in the grade.” Often the concern in making something extra credit instead of an assignment may involve the same concerns about accessibility. For example, if the extra credit assignment involves attending a presentation by a guest speaker. One might also offer an option to watch a recording of a presentation by the same or a comparable speaker and sharing a reflection on it. In class time, having students engage with multiple presentations by the same scholar can lead to fruitful discussions including exploring how a scholar’s research can develop over time.

Grade work, not timing of work

One of the most significant ways grades become about nonacademic behavior rather than achievement is through timing – frequently expressed through “points off” for late assignments. These penalties vary wildly from class to class, from no penalty if excused documentation is provided to percentages lost per day (5-40% is not uncommon), to instant zeros for missed assignments. Whatever method used, subtracting points for tardiness means that the grade no longer reflects a student’s knowledge or skills, which undermines the entire construct of grades measuring the achievement of outcomes (unless, of course, the outcome is turning the work in on time). These penalties also tend to affect students disproportionately because of the variety of situations that may interfere with the ability to turn something in “on time.”

Certainly, there need to be limits to when assignments may be submitted, but rationale for these expectations can be made transparent to students from the beginning. And when students do miss a standard deadline, we can view this as an opportunity to check in with that student and see what they need to accurately demonstrate their knowledge and skills.

As a corollary to not punishing late work, we can also identify alternative non-grade consequences for cheating and violations of academic integrity. Grade penalties stem from views of punishment through deterrence (we want to prevent this from happening through the threat of punishment) and retribution (the person who committed a bad act has harmed the class/university/society and therefore must be harmed back). But, instead, we may approach this through the lens of rehabilitation (the person who committed a violation must learn to correct their behavior and contribute positively to the class/university/society). One core element of education and learning is the ability to learn from mistakes, and this is precisely what a rehabilitation perspective seeks to do.

---

Next Steps

Activity

Using the Alternative Grading worksheet, reflect on your practices and opportunities for your class. This worksheet will guide you through thinking about how many of the elements discussed above relate to your class and prepare for you want to approach grading and assessment.