Version 5 of the rubric generation prompt builds upon v4’s systematic framework by specifically optimizing the rubric creation process for AI compatibility while maintaining human usability. The prompt introduces three new core principles - Conciseness, Action Verbs, and Response Specificity - that guide the creation of machine-readable yet pedagogically sound rubrics. Each principle includes specific guidelines, such as using clear, simple language, employing precise verbs that reflect cognitive levels, and defining explicit requirements for partial versus full credit.
While v4 focused on creating comprehensive rubrics through a structured process, v5 adds an additional layer of technical precision by emphasizing AI-compatible language patterns. It maintains the five-stage development process but enhances each stage with specific guidelines for creating rubrics that can be consistently interpreted by both AI and human graders. The prompt also places greater emphasis on defining exact terms, acceptable variations, and explicit criteria for different levels of credit, making it particularly well-suited for automated grading systems while preserving the pedagogical depth established in v4.
Prompt
You are a professor tasked with creating comprehensive, AI-compatible rubrics for grading an exam question. Your rubrics must be clear, effective, and suitable for both AI and human graders.
First, let's review the essential information for this task:
The exam question:
<exam_question>
{{EXAM_QUESTION}}
</exam_question>
Your area of expertise:
<topic_domain>
{{topic_domain}}
</topic_domain>
<topic_specific>
{{topic_specific}}
</topic_specific>
Maximum points for this question:
<max_points>
{{MAX_POINTS}}
</max_points>
Example solutions that would earn maximum points:
<example_solutions>
{{EXAMPLE_SOLUTIONS}}
</example_solutions>
Grading hints or guidelines provided by the instructor:
<grading_hints>
{{GRADING_HINTS}}
</grading_hints>
Initial rubric drafts to be expanded:
<rubric_drafts>
{{RUBRIC_DRAFTS}}
</rubric_drafts>
Example student responses (not necessarily correct):
<example_answers>
{{EXAMPLE_ANSWERS}}
</example_answers>
Now, let's proceed with creating AI-compatible rubrics. Follow these guidelines:
1. Conciseness:
- Use clear, simple language
- Break complex criteria into separate rubric items
- Keep each rubric to one sentence when possible
- Acknowledge multiple valid solution approaches
2. Action Verbs:
- Use precise verbs reflecting the required cognitive level
- Use "explicitly mentions" for required specific terms
- Use "implies" when indirect demonstration is acceptable
- Align verbs with the depth of understanding being assessed
3. Response Specificity:
- Define expected depth of responses
- Specify when exact terms or concepts must be named
- Detail requirements for partial vs. full credit
- Include examples of acceptable variations
Develop the rubrics through this process:
1. Analyze the question:
- Break it down into core components
- Identify key concepts and skills
- Determine cognitive level using Bloom's taxonomy
- Note explicit requirements or constraints
2. Assess learning objectives:
- Map the question to specific objectives
- Identify essential knowledge and skills
- Consider different levels of understanding
- Determine partial versus complete understanding criteria
3. Analyze reference material:
- Study example solutions
- Review instructor's grading hints
- Expand initial rubric drafts
- Analyze student example answers
4. Develop rubric strategy:
- Design a fair point allocation scheme
- Create detailed, objective criteria
- Include guidelines for full, partial, and zero credit
- Consider special cases or edge cases
5. Validate rubrics:
- Review against AI compatibility guidelines
- Ensure all aspects of the question are covered
- Verify fair and logical point allocations
- Check for clear and unambiguous language
Document your thought process inside <rubric_development> tags. In this analysis:
1. List the key concepts, skills, and cognitive levels required by the exam question.
2. Outline a detailed point allocation strategy, explaining how you'll distribute points across different aspects of the answer.
3. Identify potential edge cases or special scenarios that the rubric should account for.
4. For each main component of the rubric, provide a brief justification for its inclusion and point value.
It's OK for this section to be quite long.
Then provide the final rubrics in <rubrics> tags. Use this format for each rubric item:
<rubrics>
<rubric_item>
[Detailed rubric item following AI compatibility guidelines]
[X] points
</rubric_item>
</rubrics>
Ensure that:
1. Each rubric follows conciseness, action verb, and specificity guidelines
2. Total points add up to the specified maximum
3. Each item is clear, measurable, and objective
4. Rubrics account for different levels of understanding
5. Partial credit possibilities are clearly defined
6. Criteria can be consistently applied by AI and human graders
7. Multiple valid approaches are accommodated where appropriate
Begin your response by documenting your thought process, then provide the final rubrics.