Linda Nilson argues that our current grading system is fundamentally broken—it fails to reflect what students actually learn, burdens faculty with time-consuming and subjective decisions, and fuels stress and grade disputes. In Specifications Grading (2014), she proposes a compelling alternative: Specifications (specs) grading, a system built on clear, pass/fail criteria tied directly to learning outcomes.
By eliminating partial credit, streamlining grading with one-level rubrics, and incorporating flexible tools like revision tokens, specs grading restores academic rigor, motivates students to do higher-quality work, and significantly reduces grading time and conflict. As explained in Chapter 5 of Specifications Grading, its core elements include:
- Binary Grading System (Pass or Fail). Specs grading simplifies the evaluation process: students either meet all the stated requirements or they don’t. There’s no partial credit. If an assignment meets the “specs”—which are aligned with course learning outcomes—it earns full credit. If not, it earns none. This encourages students to focus on doing work that actually meets academic standards.
- Clear, Transparent Expectations. Specs grading demands precision from instructors. Assignments come with detailed, unambiguous instructions. These specs might include format, length, quality benchmarks, required elements, and examples of acceptable work. Students know exactly what success looks like before they start.
- Alignment with Learning Outcomes. Each assignment is designed to assess whether a student has achieved one or more course outcomes.
- Use of One-Level Rubrics. Specs grading uses “one-level rubrics,” where the assignment is either completed to standard or not. These rubrics consolidate what would traditionally be a “B-level or better” performance into a single description of satisfactory work, which must be met fully to receive credit.
- Flexibility Through Tokens and Revisions. To reduce student stress and encourage perseverance, many instructors incorporate “tokens”—a kind of currency students can exchange for second chances, such as revising a failed assignment or getting an extension. This system acknowledges that learning is a process and supports improvement without diluting standards.
UChicago Instructors Explore Specifications Grading in their Teaching
At the University of Chicago, many professors are already applying the specifications grading in innovative and practical ways. During an event held by one of the Exploratory Teaching Groups sponsored by the Chicago Center for Teaching and Learning in the 2024-2025 academic year, four faculty members shared their experiences with implementing this approach.
Borja Sotomayor, Senior Instructional Professor, Computer Science
“Our experience with Specifications Grading has taught us that adopting it can shift students’ focus from their grades to their learning, but that incorporating it into a well-designed course is complex and often requires revision,” notes Sotomayor. In his Foundations of Computer Networks class, Sotomayor implements ten “checkpoint exams” throughout the quarter, which are graded on an S/N/U scale:
- Satisfactory (S): The student demonstrates sufficient mastery of the material.
- Needs Improvement (N): The student has put in a good-faith effort to complete the work, but revealed a lack of mastery in the material that can be addressed via concrete feedback. The work could become Satisfactory with revisions.
- Unsatisfactory (U): The student did not submit any work, or did not complete a sufficient portion of the work (e.g., completed less than half the work that was assigned).
Sotomayor’s syllabus explains how the final exam works: “[It] serves only to provide a second chance to improve your score on the checkpoint exams. This means that, if you earn an S on all the checkpoint exams, you can skip the final exam. It also means you could skip all the checkpoint exams, and only take the final exam.”
Jean Clipperton, Associate Senior Instructional Professor, Computational Social Science
In her Data Visualization course, Clipperton uses a qualitative rubric to evaluate homework assignments across six criteria—truthfulness, functionality, design, insight, technical execution, and reproducibility—using a three-tier system:
- Excellent (✓+)
- Satisfactory (✓)
- Needs Improvement (✓–)
Final letter grades are based on overall consistency across these categories. All ✓s typically correspond to a B+, a mix of ✓ and ✓+ may result in an A– or A, while a mix of ✓ and ✓– may result in a B or lower. Incomplete or missing work receives no credit. This model aims to reduce grading anxiety and enhance the quality of feedback students receive.
Crystal Beiersdorfer, Lecturer in Media Arts and Design
Beiersdorfer applies specs grading in her Media Arts And Design Practice course to foster creativity, iteration, and student autonomy. Work is evaluated holistically based on engagement, willingness to experiment, and responsiveness to feedback. Students receive one of three marks:
- “Complete” (C) for strong, thoughtful, and creative submissions that meet or exceed expectations
- “In Progress” (I) for promising work needing more development
- “Not Submitted” (N) for missing or improperly submitted assignments
Students with an “I” may revise and resubmit with a brief reflection (text, audio, or video) explaining their changes. The approach embraces iteration, encouraging students to view unfinished ideas as valuable starting points for growth. More details can be found in her course webpage.
Set Up Specs Grading in Canvas
Learning management systems like Canvas are primarily designed for traditional letter grading and rely heavily on point-based calculations. This can pose difficulties for instructors adopting specifications grading, which typically avoids points and partial credit altogether. However, Canvas can still be adapted to support specs grading—by using points not as a measure of partial performance, but as signals of discrete outcomes.
The example below outlines a simple way to do this, using three possible marks for student work: Complete, Needs Revision, and Missing. While not a perfect system, this method allows instructors to use Canvas to track student progress and guide final grade decisions in a specs-based course.
Step 1: Define Your Grading Criteria
Before customizing Canvas, instructors must define how specifications grading will translate into final letter grades. This means clearly stating how many assignments must be marked Complete for each grade level, and how assignments marked Needs Revision factor in.
Here is an example grading policy:
| To earn a… | Students must complete: |
| A | 3 practice problems, 3 group projects, 5 homeworks |
| B | 2 practice problems, 3 group projects, 4 homeworks |
| C | 2 practice problems, 2 group projects, 3 homeworks |
| D | 1 practice problem, 1 group project, 2 homeworks |
This framework helps both instructors and students clearly understand expectations.
Step 2: Set Up a Custom Grading Scheme
Start by creating a new grading scheme in Canvas. In this example, we’ll assign a specific point value that reflects a status:
- 1 for Complete
- 0.1 for Needs Revision
- 0 for Missing
Although Canvas will interpret these as percentages or point values, here they serve a different purpose: They communicate clear, binary (or near-binary) outcomes aligned with specifications. In this example, a “1” indicates the student has fully met expectations; “0.1” signals that revision is needed; and “0” means the assignment was not submitted or failed to meet minimum criteria.
Even though this example uses percentage grading, the logic works the same with raw points. What matters is not the total percentage, but the sum of these values—used later to determine whether students have met the required number of completions for a given grade.
Step 3: Organize Assignments into Groups
Next, create assignment groups to reflect course modules or by type of assignment. For instance, each group could represent a type of assignment (e.g., “Practice Problems”, “Group Projects”, etc.). This structure helps organize assignments and makes it easier to calculate totals when translating specs grades into final letter grades.
In each group title, you can include the number of assignments required (e.g., Group Projects – Complete 3 of 3). Accompany this with a rubric and clear explanation of how work will be evaluated and how final grades are determined.
Step 4: Evaluate Student Submissions
Once students begin submitting work, the instructor will assign each assignment its corresponding score (1, 0.1, 0). For example, if a student:
- Completes one assignment (1)
- Submits one that needs revision (0.1), and
- Misses another (0)
Then the group total will display as 1.01/3.
Important: Ignore the points displayed in Canvas. Instead, refer to the numerical total shown in the black box that appears when clicking the assignment group’s total. This reflects the actual points used in the specs grading logic.
To monitor student progress more holistically, instructors can also use the Learning Mastery Gradebook to view outcome-level performance. This tool makes it easier to identify patterns, provide targeted support, and ensure alignment with learning goals. One major advantage is that it allows instructors to view all students’ outcome progress at the same time, making it especially useful for spotting trends across the class.
Step 5: Determine Final Grades
To assign final grades, return to the grading criteria defined in Step 1. Use the Gradebook to count the number of Complete, Needs Revision, and Missing marks for each assignment type.
In the example above, the student would earn a B, having completed 2 practice problems, 3 group projects, and 4 homework assignments. Instructors may also define specific policies for how a Needs Revision submission can be updated to count as Complete, and how such revisions factor into the final grade calculation.
Further Resources
Subscribe to ATS’ blog and newsletter for updates on new resources to support you in using these tools. For individual assistance, you can visit our office hours, book a consultation with an instructional designer, or email academictech@uchicago.edu. For a list of our upcoming ATS workshops, please visit our workshop schedule for events that fit your schedule.
