Using a Large Language Model to Assist with the Development of Rubrics

CIRT ID Team

Faculty members who are committed to enhancing the learning experience for their students often report spending countless hours developing and revising rubrics for the assessments in their courses. These detailed scoring criteria can help guide students’ performance and streamline the grading process; however, developing comprehensive, clear, and fair rubrics can often be a time-consuming task. That’s where a large language model (LLM) like ChatGPT, Google Bard, or Claude 2 can help. 

Why Consider Using an LLM?

Before diving into the how-to’s, let’s explore some of the compelling reasons that make a large language model an asset in rubric creation. Here are some advantages that may resonate with faculty members: 

  • Efficiency: A language model can quickly provide you with a draft or set of options to consider. 
  • Consistency: Automated prompts ensure alignment with your course objectives and instructional goals. 
  • Innovation: Language models can offer fresh perspectives or criteria that might not come to mind otherwise. 
  • Customization: The output is flexible and can be modified to fit your specific needs and subject matter. 

Crafting Effective Prompts

If the use of an LLM seems like an effective option to consider using, the next step is to formulate a prompt that yields a useful output. To get the most out of the technology, keep these guidelines in mind: 

  • Specificity: Make your request detailed. Instead of “a rubric,” ask for a “rubric for evaluating a student essay on climate change for a 300-level environmental science course.” 
  • Context: Provide additional information such as the student level, assignment type, and specific guidelines. 
  • Learning Objectives & Assignment Instructions: Consider including these in your initial prompt to ensure alignment and relevance. 
  • Iterative Approach: The output is flexible and can be modified to fit your specific needs and subject matter. Don’t hesitate to ask follow-up questions to refine the initial output. 

Sample Prompt and Output

Prompt:

The following is an example of a prompt that might be used in an LLM to generate a rubric aligned with specific assessment details: 

“I am going to provide you with instructions for an assignment in an undergraduate college course titled Introduction to Educational Technology for Learning Professionals. I would like you to create a rubric that I might use to evaluate student learning in the assignment. It is worth a total of 20 points. The instructions that students receive for this assignment are as follows:  

‘Reflect on the blended learning resources and examples that you encountered in this module. 

Then create a plan/lesson on how you might design a blended learning experience for your classroom. Recall that the definition of blended learning involves integrating components of both face-to-face learning along with online or digital media.  

At the end of the lesson plan, include a reflection on both the challenges and benefits of a blended learning design.’” 

LLM Output:

Criteria 

Description 

Points 

Understanding of Blended Learning 

The student demonstrates a clear understanding of blended learning by incorporating components of both face-to-face and online learning in their lesson plan.  

4 

Use of Resources 

The student effectively uses the provided resources and examples in the design of their lesson plan. They demonstrate that they have learned from and built upon the ideas in the resources. 

4 

Lesson Plan Design  

The lesson plan is clear, feasible, and includes all necessary components (e.g., objectives, activities, assessments). It is tailored to meet the needs of the student’s specific classroom context.  

4 

Integration of Technology 

The student demonstrates a thoughtful and effective use of technology in their lesson plan. The selected digital media enhances the learning experience and is well integrated with face-to-face components. 

4 

Reflection on Challenges and Benefits 

The student provides a thoughtful reflection on both the challenges and benefits of their blended learning design. The reflection demonstrates a deep understanding of the complexities and potential of blended learning.  

4 

Revision Prompt:

As mentioned previously, consider the output from an LLM flexible; you can easily modify the   results to fit your specific needs by asking follow-up questions. For example, you might ask the following as a follow-up:  

“Can you revise this rubric so that you provide a description for 4 points, 2 points, and 0 points for each criterion?” 

LLM Output:

Criteria  

4 points (Excellent) 

2 points (Satisfactory) 

0 points (Unsatisfactory) 

Understanding of Blended Learning 

The student demonstrates a deep and thorough understanding of blended learning by successfully integrating components of both face-to-face and online learning in their lesson plan. 

The student demonstrates a basic understanding of blended learning. Some components are present but not well integrated. 

The student demonstrates little or no understanding of blended learning, with missing or incorrect integration of face-to-face and online learning components. 

Use of Resources 

The student effectively and creatively uses the provided resources and examples in the design of their lesson plan, clearly building upon the ideas in the resources. 

The student uses the provided resources and examples, but the application to their lesson plan may be superficial or lack depth. 

The student does not effectively use the provided resources and examples in their lesson plan design. 

Lesson Plan Design 

The lesson plan is comprehensive, clear, and feasible, including all necessary components (e.g., objectives, activities, assessments). It is skillfully tailored to meet the needs of the student’s specific classroom context. 

The lesson plan is clear but may lack some necessary components or specifics. The plan may not be fully tailored to the student’s classroom context. 

The lesson plan is unclear, missing key components, not feasible, or does not consider the student’s specific classroom context. 

Integration of Technology 

The student demonstrates a thoughtful and innovative use of technology in their lesson plan. The selected digital media enhances the learning experience and is seamlessly integrated with face-to-face components. 

The student uses technology in their lesson plan, but the digital media may not fully enhance the learning experience or is not well integrated with face-to-face components. 

The student does not effectively use technology in their lesson plan, or the digital media detracts from the learning experience or is poorly integrated with face-to-face components. 

 

 

Reflection on Challenges and Benefits 

The student provides a comprehensive, thoughtful, and insightful reflection on both the challenges and benefits of their blended learning design. The reflection demonstrates a deep understanding of the complexities and potential of blended learning. 

The student provides a reflection on the challenges and benefits of their blended learning design, but it may lack depth or not fully address the complexities and potential of blended learning. 

The student provides a weak or missing reflection on the challenges and benefits of blended learning, demonstrating little understanding of the complexities and potential of this approach. 

Conclusion

While an LLM is not a substitute for the expert insight you bring to your courses, it can be a valuable assistant in rubric development. This frees you to invest more time into the most impactful aspects of education: teaching and mentoring your students.  

This article was written with the assistance of ChatGPT (GTP-4): OpenAI. (2021). ChatGPT, version 4. Retrieved October 13, 2023, from https://chat.openai.com/ 

Brought to You By
CIRT ID Team
Picture of Rob Rose
Rob Rose
CIRT | Senior Instructional Designer
css.php