Developing Rubrics for Ambiguity
Our second Rubric Roundtable was a resounding success, bringing together lecturers, educational developers, and programme leaders from King’s College London, Nottingham Trent University, Greenwich University, and the University of Westminster. The session focused on developing rubrics for ambiguity and defining descriptors, particularly in the context of AI integration in education. This roundtable provided a valuable platform for sharing best practices, discussing common challenges, and exploring practical solutions. Here are the key takeaways from our discussion:
Understanding and Managing Ambiguity in Rubrics
One of the primary sources of ambiguity in rubrics is the use of vague terminology such as “excellent” and “good.” These terms can be interpreted differently by students and assessors, leading to inconsistencies in assessment. Participants suggested several ways to manage this ambiguity, including breaking down complex skills into specific components and being explicit about the proxies used for assessment. It was also emphasized that ambiguity is not necessarily a bad thing; students need to understand the subjectivity of marker judgment and the interpretative scope that comes with mastery. We don’t want to be overly prescriptive, as constraining exploration with rigid criteria can stifle creativity and deeper understanding.
Assessing Reflection and Navigating Challenges
A significant challenge discussed was how to assess students’ reflection on their learning processes and the challenges they face. Traditional rubrics can sometimes encourage students to fabricate experiences to meet the criteria. Innovative approaches like processfolios were suggested as a solution, where students depict their learning process through various artifacts, ensuring a more genuine representation of their efforts.
Subverting the Hidden Curriculum with Rubrics
Rubrics can play a crucial role in making the hidden curriculum explicit. By clearly outlining the behaviors and skills expected from students, rubrics can help demystify academic expectations and support students in achieving their goals. This is particularly important when considering the impact of AI on assessment practices.
Impact of AI on Rubric Standards
The integration of AI in student work raises questions about how our perception of quality changes when AI tools are used. While the rubric criteria may not need to change, the standards of what constitutes quality work might. We need to develop clear standards that reflect the capabilities and limitations of AI, ensuring that students are assessed fairly and accurately.
Developing Standards in the AI Age
As AI becomes more prevalent in education, there is a growing need to establish standards that address its use in assessments. This includes finding examples of rubrics where AI has been integrated and specifying how AI-generated work should be evaluated. Understanding the link between the prompts students use and the quality of their AI-assisted work is also crucial.
Using Exemplars Effectively
The use of exemplars can help students understand what is expected of them. However, it’s important to provide a range of examples to avoid limiting students’ creativity. Exemplars should serve as a guide, not a prescriptive model, allowing students to innovate and demonstrate their unique understanding and skills. Some participants suggested using anonymized student work from previous years as examples, and others advocated for showing a variety of high-quality work without attaching specific grades to encourage a broader range of acceptable outcomes.
Future Roundtables
We are excited to continue these discussions in future roundtables, with the next session focusing on developing rubrics for dissertations and projects. This session will explore the unique challenges of assessing broad, complex assignments and explore strategies for creating effective rubrics that can guide students through their dissertations.
Leave a Reply