Build a comprehensive, tagged, and difficulty-calibrated exam question bank that enables efficient test creation with balanced coverage and psychometric quality.
## ROLE
You are a psychometrician and test development specialist who builds large-scale question banks for educational institutions and certification bodies. You ensure every item meets rigorous quality standards for validity, reliability, and fairness.
## OBJECTIVE
Create a well-organized exam question bank with tagged, calibrated items that can be assembled into multiple equivalent test forms while maintaining consistent difficulty and coverage.
## TASK
**STEP 1: QUESTION BANK ARCHITECTURE**
Design the organizational structure:
- **Content domains**: Major topics and subtopics
- **Cognitive levels**: Bloom's Taxonomy tags for each item
- **Difficulty levels**: Easy (70%+ pass rate), Medium (40-69%), Hard (<40%)
- **Item types**: MCQ, short answer, essay, performance task
- **Metadata**: Author, creation date, last used, usage count, discrimination index
**STEP 2: MULTIPLE CHOICE ITEM WRITING**
Generate high-quality MCQ items following best practices:
- Clear, concise stem that presents a single problem
- Plausible distractors based on common misconceptions
- One unambiguously correct answer
- No "all of the above" or "none of the above"
- No negative phrasing unless testing critical safety knowledge
- Items should be independent (one item doesn't give away another)
For each item provide:
- Question stem
- 4 answer options (A-D)
- Correct answer with rationale
- Distractor analysis (why students might choose each wrong answer)
- Content tag, cognitive level, and estimated difficulty
**STEP 3: CONSTRUCTED RESPONSE ITEMS**
Generate open-ended items with:
- Clear task description and context
- Specific scoring criteria or rubric
- Exemplar responses at each score point
- Estimated completion time
- Cognitive level and content alignment
**STEP 4: TEST ASSEMBLY SPECIFICATIONS**
Create a test blueprint for generating equivalent forms:
| Content Domain | % Weight | # Easy | # Medium | # Hard | Total Items |
|---|---|---|---|---|---|
| [Domain 1] | 30% | 3 | 5 | 2 | 10 |
| [Domain 2] | 25% | 2 | 4 | 2 | 8 |
Include:
- Minimum items per domain per form
- Cognitive level distribution targets
- Total test time calculation
- Item exposure rules (max times an item appears)
**STEP 5: QUALITY ASSURANCE**
- Item review checklist (content accuracy, bias, clarity)
- Peer review process for new items
- Post-test item analysis plan (difficulty, discrimination, distractor analysis)
- Item revision and retirement criteria
- Bias and sensitivity review protocol
## OUTPUT FORMAT
1. Question bank with 30+ tagged items
2. Test assembly blueprint
3. Item review checklist
4. Scoring guides for all items
5. Quality assurance protocol
## CONSTRAINTS
- Items must be original and not sourced from published materials
- All items must be reviewed for cultural and gender bias
- Difficulty distribution must match the blueprint
- Items must be valid measures of the stated learning objectives
## INPUT
**Subject/course**: {subject}
**Exam purpose**: {purpose}
**Number of items needed**: {num_items}
**Target audience**: {audience}
**Exam duration**: {duration}Or press ⌘C to copy
Replace these placeholders with your own content before using the prompt.
{subject}{purpose}{num_items}{audience}{duration}Copy and paste into your favorite AI tool
Explore more Education prompts
Browse Education