Course Description Assessment¶
Generated by the Course Description Analyzer skill (v0.03) for AI Strategy for Education.
1. Overall Score¶
97 / 100
2. Quality Rating¶
Excellent — Ready for learning graph generation. (90–100)
The course description is complete across all required elements, with a specific multi-segment audience, explicit prerequisites and exclusions, a comprehensive topic list, and three or more concrete, action-verb outcomes at every level of the 2001 Bloom's Taxonomy. It is well above the 85-point threshold for proceeding to the learning-graph generator.
3. Detailed Scoring Breakdown¶
| Element | Points | Earned | Notes |
|---|---|---|---|
| Title | 5 | 5 | "AI Strategy for Education" — clear and matches site_name. |
| Target Audience | 5 | 5 | Specific across K-12 and higher ed (administrators, teachers, parents, board/trustees), with "no technical background assumed." |
| Prerequisites | 5 | 5 | Explicitly "None," with a helpful note on what makes the exercises richer. |
| Main Topics Covered | 10 | 9 | 14 substantive topics. Minor: a few items overlap (the idea-funnel stages appear both as one bundled topic and as separate topics) and could be consolidated. |
| Topics Excluded | 5 | 5 | Clear "Topics Not Covered" section sets strong boundaries (no model-building, no coding, no vendor reviews, no legal advice). |
| Learning Outcomes Header | 5 | 5 | "After completing this course, participants will be able to:" present. |
| Remember | 10 | 10 | 4 specific recall/identify/list outcomes. |
| Understand | 10 | 10 | 5 explain/describe/summarize/interpret outcomes. |
| Apply | 10 | 10 | 4 run/populate/apply/draft outcomes tied to the learner's own institution. |
| Analyze | 10 | 10 | 4 compare/break-down/distinguish outcomes. |
| Evaluate | 10 | 10 | 4 judge/critique/assess/weigh outcomes, including the ethics of denying AI access. |
| Create | 8 | 8 | Strong board-ready capstone plus 2 design/produce outcomes — meets the bar; would benefit from one more distinct "Create" outcome and from splitting the compound capstone into named deliverables. |
| Descriptive Context | 5 | 5 | Rich overview grounds the course in the METR doubling data, the idea-funnel workflow, and the four near-term assumptions. |
| Total | 100 | 97 |
4. Gap Analysis¶
The description has no missing elements. The only items scoring below full marks are minor polish opportunities:
- Topic overlap (Main Topics, −1). The six idea-funnel stages are listed both as a single "operating system" topic and again as individual topics. This is fine for emphasis but slightly inflates the list; consider one canonical treatment.
- Create level breadth (Create, −2). The level meets the three-outcome minimum and the capstone is excellent, but it is also the level most likely to seed project-type concepts in the learning graph. One or two additional distinct "Create" outcomes (e.g., design a one-hour idea-generation workshop agenda, compose a parent/community AI-communication plan) would deepen that part of the graph.
5. Improvement Suggestions (Prioritized)¶
- Optional — add 1–2 Create outcomes. Highest leverage for richer capstone/project concepts in the learning graph. Suggested verbs: design, compose, assemble, formulate.
- Optional — de-duplicate the topic list. Either present the idea funnel once as a bundled topic, or list only its six stages — not both.
- Optional — split compound outcomes. A few Understand/Evaluate bullets bundle two ideas; splitting them improves measurability and yields cleaner concept boundaries.
None of these block progress — they are refinements, not gaps.
6. Next Steps¶
The score (97) is well above 85. The course description is ready for the
learning-graph-generator skill to enumerate ~200 concepts with dependencies and a taxonomy.
7. Concept Generation Readiness¶
Strong. The description should comfortably support 200+ concepts:
- Topic breadth spans capability measurement (METR, doubling, benchmarks), the six-stage idea funnel, the intelligent-textbook/xAPI/LRS/AI-LMS stack, the Alpha School operating model, balanced risk/reward (privacy, FERPA, bias, equity, integrity, lock-in), SWOT and strategic-planning method, and governance/change-management/ethics.
- Concept diversity is high: the funnel alone yields process concepts (gather → register → evaluate → select → resource → evaluate); the technology stack yields system concepts (xAPI, LRS, recommendation engine, learning plan); the risk section yields a full risk taxonomy; and the SWOT companion (swot-case-studies.md) supplies twelve concrete institutional archetypes spanning the resource spectrum.
- Bloom coverage reaches Create with a board-ready capstone, ensuring the graph terminates in synthesis/project concepts rather than only recall.
Estimate: 200–260 concepts are readily derivable from the current description.