Education & Learning / Curriculum Design
Map assessment items to learning objectives and cognitive levels — test alignment for valid measurement.
Why This Prompt Exists
Most tests measure something different from what was taught. Students fail not because they don’t know the material, but because the test doesn’t match the instruction.
You get:
- assessments that don’t align with learning objectives
- too many low-level questions (recall only)
- no balance across cognitive levels
- students surprised by test content
- invalid conclusions about student learning
But assessment blueprints ensure alignment:
- objectives: what was taught
- cognitive level: Bloom’s taxonomy level
- item type: multiple choice, short answer, essay, performance
- point value: weighting per objective
- item count: sufficient to measure each objective
Without blueprint, tests are invalid.
This prompt maps assessment items to learning objectives and cognitive levels.
The Prompt
Assume the role of an assessment designer who creates test blueprints. Your task is to map assessment items to learning objectives and cognitive levels. Generate: 1. ASSESSMENT PARAMETERS - Assessment title: [name] - Duration: [X minutes] - Total points: [X] - Purpose: [formative / summative / diagnostic] 2. LEARNING OBJECTIVES (from CD-01) | ID | Objective | Bloom's Level | Priority (H/M/L) | |----|-----------|---------------|------------------| | LO1 | [text] | [level] | [priority] | | LO2 | [text] | [level] | [priority] | 3. BLUEPRINT TABLE | Objective | Bloom's Level | Points | % of Test | Item Type | Item Count | |-----------|---------------|--------|-----------|-----------|------------| | LO1 | Remember | X | X% | MC, T/F | X | | LO2 | Understand | X | X% | Short answer | X | | LO3 | Apply | X | X% | Problem | X | | LO4 | Analyze | X | X% | Essay | X | 4. ITEM SPECIFICATIONS (per objective) **Objective LO1: [description]** - Cognitive level: Remember - Item types allowed: Multiple choice, matching, true/false, fill-in-blank - Sample item: [example question] - Difficulty target: 80-90% correct - Point value: [X] **Objective LO2: [description]** - Cognitive level: Understand - Item types allowed: Short answer, diagram labeling, concept map - Sample item: [example question] - Difficulty target: 70-80% correct - Point value: [X] 5. TEST BALANCE VERIFICATION | Bloom's Level | Points | % of Test | Target % | Acceptable? | |---------------|--------|-----------|----------|-------------| | Remember | X | X% | 10-30% | Yes/No | | Understand | X | X% | 20-40% | Yes/No | | Apply | X | X% | 20-40% | Yes/No | | Analyze | X | X% | 10-20% | Yes/No | | Evaluate | X | X% | 5-15% | Yes/No | | Create | X | X% | 0-10% | Yes/No | 6. ASSESSMENT VALIDITY CHECKLIST - [ ] Every learning objective has at least one assessment item - [ ] Higher-priority objectives have more points - [ ] Bloom's levels match the objective (not testing recall if objective was analysis) - [ ] Point distribution matches time allocation during instruction - [ ] Item types are appropriate for the cognitive level - [ ] Total test length fits within available time 7. COMMON BLUEPRINT MISTAKES | Mistake | Why It Fails | Correct Approach | |---------|--------------|------------------| | All items at recall level | Doesn't measure higher-order thinking | Distribute across Bloom's levels | | No objective-item map | Can't tell what students actually know | Map each item to an objective | | Too few items per objective | Unreliable measurement | 2-4 items per low-stakes, 5-10 for high-stakes | | Mismatched level | Tests something not taught | Match item level to objective level | | Uneven point distribution | Misrepresents importance | Weight by priority and instructional time | INPUTS: Learning objectives (from CD-01): [PASTE OBJECTIVES WITH BLOOM'S LEVELS] Assessment duration: [E.G., "60 minutes"] Total points: [E.G., "100"] Assessment purpose: [FORMATIVE / SUMMATIVE / DIAGNOSTIC] Item type preferences (optional): [E.G., "Mostly multiple choice with 2-3 short answer"] RULES: - Every learning objective must have at least one assessment item - Higher-priority objectives should have more points - Match item cognitive level to objective Bloom's level (not recall for analysis objectives) - Distribute points proportionally to instructional time spent on each objective - Use 2-4 items per objective for low-stakes assessments - Use 5-10 items per objective for high-stakes assessments - Total test length should fit within the available time (1 minute per point is typical)
How To Use It
- Every learning objective must have at least one assessment item — otherwise, why teach it?
- Higher-priority objectives should have more points — weight by importance, not convenience.
- Match item cognitive level to objective Bloom’s level — don’t test recall if the objective was analysis.
- Distribute points proportionally to instructional time spent on each objective — teach what you test, test what you teach.
- Use 2-4 items per objective for low-stakes assessments — enough for reliability, not too many.
- Use 5-10 items per objective for high-stakes assessments — more items, more reliable measurement.
- Total test length should fit within the available time — 1 minute per point is typical for multiple choice; more for essays.
Example Input
Learning objectives:
“LO1: Define the three branches of government (Remember)”
“LO2: Explain the system of checks and balances (Understand)”
“LO3: Analyze how a bill becomes a law (Analyze)”
“LO4: Compare federalist and anti-federalist positions (Analyze)”
Assessment duration: “45 minutes”
Total points: “50”
Assessment purpose: “SUMMATIVE (end of unit)”
Item type preferences: “Mix of multiple choice and short answer”
Why It Works
Most tests measure something different from what was taught. Students fail not because they lack knowledge, but because the test doesn’t match the instruction.
This framework improves outcomes by forcing:
- objective-to-item mapping (each objective has at least one item)
- Bloom’s level alignment (item level matches objective level)
- point distribution by priority (more important objectives get more points)
- test balance verification (distribution across cognitive levels)
- validity checklist (ensuring test measures what it claims)
Failure modes this prevents:
- assessments that don’t align with learning objectives
- too many low-level questions (recall only)
- students surprised by test content
- invalid conclusions about student learning
This improves on: Unplanned test design. Assessment blueprints ensure test validity and alignment.
Related to: CD-01 (Learning Objectives) for outcomes; CD-03 (Unit Plan) for instruction alignment; QZ-04 (Rubric) for scoring.
Build Better AI Systems
Subscribe for advanced prompt engineering, AI coding tools, debugging frameworks, and practical strategies for developers and engineers.