Specifications grading is designed to increase instructor transparency, allow for student autonomy, and provide opportunities for practice and improvement. Unlike traditional grading systems where each individual assignment is graded for partial credit out of a total possible number of points, in specifications (or “specs”) grading all assignments are simply graded as either meeting a specified set of criteria or not. Key to this approach is that the instructor creates comprehensive rubrics defining the competency threshold for each assignment and shares them with students ahead of the assignment. Often, specs grading systems set the threshold for satisfactory work at approximately 80% or B-level, with the goal of encouraging a higher level of effort and quality. 

Where possible, it is recommended that students be given opportunities to revise and resubmit an assignment if they did not meet the specifications on the first attempt. This encourages a formative approach to learning. A set number of revision opportunities may be allocated at the start of the semester for eligible assignments, or students may earn these opportunities by completing additional tasks throughout the course.

Final course grades are earned by satisfactory completion of a pre-established group of assignments defined on the course syllabus. This means that expectations are concrete and transparent, allowing students to choose which assignments they need to complete to earn the grade they want to target.

Examples of Spec Grading:

Specs Grading Scheme From a CMU Business Course

Specs Grading Scheme From a CMU Modern Languages Course

Keep in mind that these are just a few examples of how instructors might implement specs grading. For more specifications grading procedures, tips, and examples see Linda Nilson’s book Specifications Grading: Restoring Rigor, Motivating Students, and Saving Faculty Time (2014).

With careful course design, specs grading can support the following common instructor values and/or needs:

Making expectations clear

• Highly detailed expectations are laid out transparently, both in the specifications of what satisfactory work is for each assignment and in the requirements for course grades.

Upholding high academic standards

• A high bar for satisfactory work holds students to high standards. 
• The instructor can encourage students to strive for higher levels than they might initially choose (particularly true for at-risk students who may target lower levels due to preconceived beliefs about their abilities).

Increasing equity    

• Clear rubrics reduce opportunities for implicit bias, so that all students are held to the same standards on each assignment.
• Opportunities for revision make it possible to have a rigorous course while still allowing students with varying amounts of prior knowledge to succeed.  

Giving students feedback they can use    

• Clear indications where work does not yet meet the specifications can guide revisions, and the ease of giving rubric-based feedback allows this to be done quickly.

Reflecting student learning outcomes

• When criteria for final letter grades are built around learning objectives rather than the accumulation of a certain number of points, the differences between letter grades can represent which objectives have been mastered.

Motivating students to learn  

• Specs grading provides a structured reward mechanism.
• Rubrics make it clear what students have learned and where they need more work, and resubmissions allow them to improve. 

Discouraging cheating

• Having opportunities to iterate lowers the stakes and reduces the pressure for students to achieve perfection, which often drives students to cheat.

Making students feel responsible for their grades 

• Students can choose the sets of assignments and/or difficulty level to attempt.
• Ideally, there is a clear connection between effort, amount of work, and grade results.

Minimizing conflict between instructors and students

• Clear criteria means there is less room to argue over grades.
• Having only satisfactory/not yet satisfactory levels on assignments reduces the importance of small numbers of points in determining grades.

Who is using specs grading here at CMU?

Although individual instructors have had success with specifications grading, there has not yet been a substantial amount of research testing its impacts. There are still a lot of unanswered questions! If you are interested in discussing whether specifications grading may be a good fit for your course and potentially assessing the results, please contact eberly-assist@andrew.cmu.edu to set up a consultation.

Ungrading places priority on individual student growth, often centered around providing multiple avenues for personalized feedback, improvement, and reflection. No letter grades or marks are awarded on individual assignments and formative feedback is the primary assessment mechanism. Although ungrading can take many forms, most rely heavily on student reflection on their learning process, in-depth conversations between students and instructors to discuss feedback on student work, and student autonomy in defining their own learning goals, objectives, and trajectory.

There are many ways to incorporate elements of ungrading without removing all grades from an entire course. Some strategies that support this goal include:

• offering certain assignments or portions of the course that will be ungraded, but relying on traditional grades for others,
• using self and/or peer assessment, 
• using written or spoken annotation, narration, or description of the learning process, highlighting representative examples of work (process letters),
• negotiating grading contracts with students at the start of the course,
• evaluating work holistically through portfolios,
• and inviting students to create or co-create rubrics.

(How to Ungrade: A FAQ, Jesse Stommel).

Even if course assignments are ungraded, instructors may still need to submit final course grades, per university policy. To reconcile this, many ungrading proponents encourage students to choose and provide justification for their own course grade, although instructors vary on the amount of “veto power,” if any, they choose to wield. Instructors should exercise caution and considerate judgment when exercising their authority to veto grades. Disagreements over final grades may arise. However, most instructors report that this is rare, that students still earn a range of course grades from ‘A’ to ‘F,’ and that students often grade themselves lower than the instructor would. In particular, students who have low self-efficacy and those who suffer from impostor syndrome (disproportionately, those from underrepresented groups in both academia and in their fields of study) may be more inclined to undervalue their contributions. To account for this inequity, some instructors reserve the power to only raise the student’s requested grade but not lower it.

For more specific ungrading procedures, tips and systems see Blum, S. Ungrading: Why Rating Students Undermines Learning (and What to Do Instead) (2020).

With careful course design, ungrading can support the following common instructor values and/or needs:

Reflecting student learning outcomes.

• Learning objectives should guide assignments, rubrics, and feedback or self-assessment.

Sharing power and increasing equity.

• Ungrading reduces the power imbalance between instructor and student.
• The instructor should be aware that some groups of students tend to undervalue their own work, and consider mechanisms to address and correct this (for example grade conferences, and veto power).

Making students feel responsible for their grades  

• Ungrading embraces subjectivity and puts students in charge of their own learning. 
• Students self-evaluate through reflections, peer review, and choosing their own grades.

Motivating students to learn  

• Removing the external motivator of grades can foster intrinsic motivation. 

Discouraging cheating  

• Eliminating numerical or letter grades on assignments alleviates the burden of striving for perfection, which frequently leads students to engage in cheating.
• Emphasizing authentic, personalized work that receives feedback instead of a grade removes the incentives for academic dishonesty.

Giving students personalized feedback 

• Extensive feedback and iteration on deliverables is built-in. Feedback may include conferences with the instructor, personal reflections, and/or peer feedback

Fostering higher-order cognitive development and creativity 

• Removing the emphasis on scores and the pursuit of "perfection" allows students to be more creative.

Reducing conflict between instructors and students

• The occurrence of "grade-grubbing" incidents may decrease or be eliminated when there is no emphasis on points or grades.
• Removing grades makes the instructor-student relationship less “transactional.” 

Who is using ungrading here at CMU?

To learn more about ungrading, discuss whether it may be a good fit for your course, and receive support with implementation, please contact eberly-assist@andrew.cmu.edu to set up a consultation.