Beyond Traditional Grades: Alternative Assessment Methods

As educators navigate the challenges and opportunities presented by generative AI (genAI), many are reconsidering traditional assessment approaches. Alternative assessment methods offer innovative ways to evaluate student learning that go beyond conventional grading systems, focusing on authentic learning, skill development, and meaningful engagement. These approaches not only address the challenges posed by AI but also align with research on effective learning and motivation (Furze, 2023; Pitts Donahoe, 2023).

Research has consistently shown that traditional grading can diminish students' interest in learning, encourage a preference for easier tasks, and reduce the quality of students' thinking (Kohn, 2011). Grades often lead students to focus more on achieving high marks than on genuine understanding, resulting in surface-level learning strategies such as memorizing facts for tests rather than engaging deeply with the material. Furthermore, a grade-oriented environment has been associated with increased cheating and a fear of failure, even among high-achieving students (Anderman & Murdock, 2007; Pulfrey et al., 2011).

Several alternative assessment approaches have emerged as promising solutions:

Ungrading shifts focus from traditional grades to a more holistic, feedback-centric model of evaluation (Stommel, 2023). Instead of assigning numerical scores or letter grades, educators provide detailed, constructive feedback on students' work, encouraging self-reflection and peer review. This method aims to foster a deeper engagement with learning, emphasizing the process of growth and improvement rather than final outcomes. Ungrading promotes collaboration and reduces the fear and competition often associated with traditional grading systems, aligning more closely with real-world skill development and lifelong learning practices (Blum, 2020). In the context of genAI, ungrading offers a particularly useful strategy to address challenges in academic integrity and authentic learning (Furze, 2023). By de-emphasizing final grades, ungrading reduces the incentive for students to misuse AI tools for the sole purpose of achieving higher marks. Instead, it encourages students to view genAI as a learning aid rather than a shortcut, promoting meaningful discussions about the appropriate use of technology in education.

Standards-based grading (SBG) focuses on proficiency in specific learning objectives, with grades indicating a student’s level of proficiency on each standard (Iamarino, 2014). This assessment approach shifts the focus away from cumulative point totals to what students know and can do. It allows educators to identify specific areas where students excel or need improvement, making learning more targeted and meaningful. In the age of genAI, SBG is particularly useful because it emphasizes deep understanding and skill acquisition rather than rote memorization or superficial learning. GenAI tools can assist in creating personalized learning paths and assessments, helping students reach proficiency in each standard at their own pace.

Specifications-based grading (specs grading) involves evaluating student work against detailed, binary pass/fail criteria, where assignments must meet all specified requirements to earn credit (Earl, 2022; Leslie & Lundblom, 2020). While initially developed and widely adopted in STEM fields, particularly computer science and engineering, specs grading has since been successfully implemented across diverse disciplines including writing composition, professional programs like nursing, and business courses. Specs grading shifts the focus away from assigning traditional letter grades or point values, requiring students to achieve a specified level of performance for their work to be considered satisfactory. The course's final grade is determined by the number and type of satisfactory assignments or assessments the student has successfully completed (Elkins, 2016). Specs grading often allows for student choice in assignment bundles to determine their desired grade. This assessment approach encourages high-quality work and reduces ambiguity in grading, allowing students to focus on meeting clear expectations (Graves, 2023). In the context of genAI, specs grading is advantageous because it helps students develop precision and accountability in their work. AI tools can assist in refining these specifications and providing immediate feedback, making the process of revision and resubmission more efficient and educational.

Labor-based grading contracts assess students based on the effort and time they invest in their work, rather than on traditional measures like correctness or proficiency (Inoue, 2023). Grades are determined by the completion of tasks agreed upon in a contract at the start of the course, emphasizing the learning process over the final product. This approach is particularly useful with the rise of genAI, as it shifts the focus from the end product—which AI can increasingly produce—to the student's engagement and effort. By valuing the learning process, labor-based grading encourages students to develop their skills and knowledge (Mena & Stevenson, 2022), motivating them to be actively involved in their education rather than relying on AI for content creation.

Competency-based education (CBE) is an educational approach that focuses on students demonstrating attainment of specific skills or competencies rather than progressing through a course based on time spent in class (Burnette, 2016). CBE allows students to move at their own pace, advancing as soon as they have proven their understanding, making learning more personalized and efficient. Even in traditionally-paced courses, educators can incorporate competency-based assessments (Competency-Based Education Network [C-BEN], 2021) to support students’ career readiness. In the age of genAI, CBE is particularly valuable because it emphasizes the actual acquisition of skills and knowledge, rather than merely completing assignments or passing tests. As AI tools become more adept at generating content, CBE ensures that students are truly competent in their areas of study, requiring them to demonstrate their abilities in meaningful and practical ways. This approach helps maintain academic integrity and ensures that students develop the critical skills needed in their fields. AI can support CBE by offering personalized learning paths, identifying gaps in student knowledge, and providing targeted practice. In assessments, AI can help students demonstrate competencies through adaptive testing or by simulating real-world tasks (Padovano & Cardamone, 2024).

Alternative assessment methods offer promising solutions for education in an AI-mediated world. These approaches not only address concerns about AI misuse but also promote deeper learning, intrinsic motivation, and authentic skill development. By moving beyond traditional grading systems, educators can create learning environments that better prepare students for a future where creative thinking, adaptability, and genuine competency are paramount.

Anderman, E. M., & Murdock, T. B. (2007). Psychology of academic cheating. Elsevier Academic Press.

Blum, S. D. (2020). Ungrading: Why rating students undermines learning (and what to do instead). West Virginia University Press.

Burnette, D. M. (2016). The renewal of competency-based education: A review of the literature. The Journal of Continuing Higher Education, 64(2), 84–93.

Competency-Based Education Network (C-BEN). (2021). Hallmark practices in CBE assessment.

Earl, D. (2022). Two years of specifications grading in philosophy. Teaching Philosophy, 45(1), 23–64.

Elkins, D. M. (2016). Grading to learn: An analysis of the importance and application of specifications grading in a communication course. Kentucky Journal of Communication, 35(2), 26–48.

Furze, L. (2023, November 1). Rethinking assessment for generative AI: Ungrading.

Graves, B. C. (2023). Specifications grading to promote student engagement, motivation and learning: Possibilities and cautions. Assessing Writing, 57(1), Article 100754.

Iamarino, D. L. (2014). The benefits of standards-based grading: A critical evaluation of modern grading practices. Current Issues in Education, 17(2).

Inoue, A. B. (2023). Labor-based grading contracts: Building equity and inclusion in the compassionate writing classroom (2nd ed.). WAC Clearinghouse.

Kohn, A. (2011). The case against grades. Educational Leadership, 69(3), 28–33.

Leslie, P., & Lundblom, E. (2020). Specifications grading: What it is, and lessons learned. Seminars in Speech and Language, 41(4), 298–309.

Mena, J. A., & Stevenson, J. R. (2022). The promise of labor-based grading contracts for the teaching of psychology and neuroscience. Teaching of Psychology, 51(4), 466–471.

Padovano, A., & Cardamone, M. (2024). Towards human-AI collaboration in the competency-based curriculum development process: The case of industrial engineering and management education. Computers and Education: Artificial Intelligence, 7, Article 100256.

Pitts Donahoe, E. (2023, September 29). The rise of generative AI calls for new approaches to grading. Unmaking the Grade.

Pulfrey, C., Buchs, C., & Butera, F. (2011). Why grades engender performance-avoidance goals: The mediating role of autonomous motivation. Journal of Educational Psychology, 103, 683–700.

Stommel, J. (2023, April 6). What is ungrading?: A Q&A.