AI DESIGN COMPANION RAG SUPPORT PACK — COMBINED PLAIN TEXT


================================================================================
FILE: 00_readme.md
================================================================================

# AI as Your Instructional Design Partner — RAG / Support Pack

Author: Miguel Guhlin, TCEA  
Companion site: go.mgpd.org/aidesign  
Purpose: Provide lightweight source material participants can attach to a chatbot, Claude Project, NotebookLM notebook, Mistral, xAI/Grok, Gemini, or another RAG-enabled workspace.

## Recommended use

1. Download this folder or the ZIP.
2. Attach the markdown source files to your chatbot/project.
3. Use the ALDO prompt from the companion site.
4. Tell the chatbot: **Use only the attached sources for effect sizes and framework claims. Ask clarifying questions before drafting.**

## Files

- `01_session_overview.md` — session intent, takeaways, workflow
- `02_aldo_framework.md` — ALDO sequence and coaching use
- `03_solo_taxonomy.md` — SOLO levels and lesson-design check
- `04_visible_learning_strategy_source.md` — effect-size starter table from the session
- `05_audit_checklist.md` — six-question AI draft audit
- `06_prompt_pack.md` — reusable prompts for teacher, coach, and leader roles
- `07_activity_facilitator_notes.md` — directions for the three workshop activities
- `08_ace_curriculum_source.md` — ACE support source for Articulate, Connect, Extend
- `09_role_differentiation_cards.md` — teacher, coach, and campus leader output expectations
- `10_json_knowledge_pack.json` — machine-readable compact source pack


================================================================================
FILE: 01_session_overview.md
================================================================================

# Session Overview

## Title
AI as Your Instructional Design Partner

## Core idea
AI works best as an instructional design partner when it is asked to walk a structured coaching conversation rather than produce a one-shot lesson plan.

## What participants leave with

1. Use ALDO as the design conversation.
2. Differentiate by role: teacher, instructional coach, or campus leader.
3. Audit AI drafts against ALDO, SOLO Taxonomy, and high-effect-size instructional strategies.

## Five-step workflow

1. **Frame** — Name the topic and learners.
2. **Prompt** — Assign the coach role and require the ALDO sequence.
3. **Run** — Answer clarifying questions before accepting a draft.
4. **Differentiate** — Switch the audience or role and re-run.
5. **Audit** — Use the six-question audit to decide keep, revise, or scrap.

## Design rule
Do not let AI jump straight to strategies. Relationship building and pre-assessment come first.


================================================================================
FILE: 02_aldo_framework.md
================================================================================

# ALDO Framework

ALDO is a lesson-design conversation, not just a prompt acronym. It describes the sequence a coach and teacher use to plan a lesson.

## ALDO sequence

1. **Relationship Building** — Build trust and understand context, goals, and expectations.
2. **Pre-Assessment** — Assess current knowledge, strengths, gaps, and needs. Clarify the problem or opportunity.
3. **Strategic Instruction** — Teach, model, and guide using targeted strategies and evidence-based approaches.
4. **Post-Assessment** — Measure progress, check for understanding, and adjust based on results.
5. **Reflection** — Reflect on outcomes, learning, and next steps. Plan for continued growth.

## How to use with AI

Ask AI to act as an instructional coach and walk through the ALDO sequence one step at a time. Require clarifying questions before drafting. Challenge the AI when it skips a step, gives strategies too early, or weakens pre-assessment/post-assessment.

## Prompt rule
Name the framework in the prompt, then ask the AI to show how each part of the response maps to the framework.


================================================================================
FILE: 03_solo_taxonomy.md
================================================================================

# SOLO Taxonomy Source Card

SOLO Taxonomy describes levels of understanding. Use it to check whether a lesson asks learners to list disconnected ideas or integrate and transfer them.

| SOLO Level | Description | Lesson-design warning |
| --- | --- | --- |
| Prestructural | No relevant connection yet | Build prior knowledge before expecting performance. |
| Unistructural | One relevant idea | Useful for initial acquisition, but too thin for final evidence. |
| Multistructural | Several ideas listed but not connected | Common AI default: lots of bullets, little integration. |
| Relational | Ideas connected into a coherent whole | Strong target for deep learning. |
| Extended Abstract | Ideas applied to new contexts | Strong target for transfer. |

## Audit move
Ask: **What SOLO level does each activity target?** If the draft stays at multistructural, revise the prompt and require relational or extended-abstract work.


================================================================================
FILE: 04_visible_learning_strategy_source.md
================================================================================

# Visible Learning Strategy Starter Source

Use this as a starter source for the workshop. Verify effect sizes against your current source before publication or formal use.

## Threshold
Prioritize strategies at **d ≥ 0.40**.

## Surface Learning
- Jigsaw Method — d = 0.92
- Direct Instruction — d = 0.56
- Retrieval practice / effects of testing — d = 0.63
- Flipped Classroom — d = 0.58

## Deep Learning
- Cognitive Task Analysis — d = 1.09
- Argumentation — d = 0.86
- Outlining and organizing — d = 0.86
- Evaluation and reflection — d = 0.75

## Transfer Learning
- Collective Teacher Efficacy — d = 1.01
- Mathematics Problem Solving — d = 0.88
- Blended Learning — d = 0.85
- Transfer Strategies — d = 0.75

## Caution
AI may invent effect sizes, titles, or citations. Require the AI to use only attached sources. Cut unverified effect-size claims.


================================================================================
FILE: 05_audit_checklist.md
================================================================================

# AI Draft Audit Checklist

Score each item as Pass, Partial, or Fail.

1. **All three phases present?** Surface only? Push back.
2. **What SOLO level do activities target?** Push past multistructural.
3. **All five ALDO components in order?** Relationship and pre-assessment first.
4. **Are cited effect sizes real?** Cross-check. Cut if unverified.
5. **Is the audience visible in the moves?** Swap one audience for another. Same read?
6. **Does post-assessment show movement?** Satisfaction does not count.

## Decision rule

- **0 fails** — Keep the draft. Tighten any partials.
- **1–2 fails** — Revise before use.
- **3+ fails** — Scrap and re-prompt.


================================================================================
FILE: 06_prompt_pack.md
================================================================================

# Reusable Prompt Pack

## Base ALDO prompt

Act as my instructional coach. We are designing a lesson on **[TOPIC]** for **[LEARNERS]**.

Walk me through ALDO, one step at a time:

1. Relationship — context to know
2. Pre-assessment — what they bring
3. Strategic instruction — strategies d ≥ 0.40, labeled with SOLO level and learning phase
4. Post-assessment — evidence of movement
5. Reflection — close the loop

Ask clarifying questions before drafting. Use only the sources I attach for effect sizes and framework claims.

## Teacher version

Use the same ALDO sequence, but produce output a classroom teacher can use tomorrow: student-facing activity steps, timing, materials, checks for understanding, and an exit ticket.

## Instructional coach version

Use the same ALDO sequence, but produce output an instructional coach can use in a coaching cycle: planning questions, co-planning template, observation evidence, debrief questions, and next-step coaching moves.

## Campus leader version

Use the same ALDO sequence, but produce output a campus leader can use to support implementation: walkthrough look-fors, feedback language, PLC agenda item, and scaling supports.

## Audit prompt

Audit the draft below against ALDO, SOLO Taxonomy, and the attached effect-size source. Score each item Pass, Partial, or Fail. Then decide Keep, Revise, or Scrap using the decision rule.


================================================================================
FILE: 07_activity_facilitator_notes.md
================================================================================

# Activity Facilitator Notes

## Activity 1 — Run the ALDO prompt

Scenario: Design a lesson on fractions for fifth-grade students who struggle with equivalence.

Participant moves:
1. Paste the ALDO prompt into a chatbot.
2. Fill in topic and learners.
3. Let AI ask clarifying questions before answering.
4. Capture what the AI asked, recommended, and labeled.

Watch for: Did it ask before answering? Are SOLO levels labeled? Any unsourced specifics?

## Activity 2 — Switch the audience

Participant moves:
1. Take the first draft.
2. Change one variable: learners or role.
3. Re-run for a different audience.
4. Compare side by side.
5. Decide whether real instructional moves changed or the AI just swapped nouns.

## Activity 3 — Audit a draft

Participant moves:
1. Pick one draft.
2. Score the six audit items.
3. Make the keep/revise/scrap call.
4. Write one specific revision.

Debrief: Where did the draft fail? Which failure mode showed up loudest?


================================================================================
FILE: 08_ace_curriculum_source.md
================================================================================

# ACE It: A Curriculum Source for Articulate, Connect, Extend

**Author:** Miguel Guhlin, TCEA
**Source post:** "ACE It: Three Steps from Surface to Transfer Learning," [TCEA Blog](https://blog.tcea.org/), June 8, 2026
**Companion frameworks:** SOLO Taxonomy (Biggs and Collis), Visible Learning (John Hattie), PRISM, VIVA, RETO, PREPARE, MPAR
**Audience:** Classroom teachers, instructional coaches, school leaders, and parents
**Purpose:** Equip you to teach the ACE routine, model it with Generative AI tools, and pair it with high-effect-size instructional strategies so students move from surface understanding to genuine transfer

---

## 1. Why ACE, and Why Now

Generative AI can produce a polished report, a tidy slide deck, or a confident infographic in seconds. The artifact looks like learning even when no learning happened. That gap is what ACE was built to close.

ACE is a three-step routine you can teach in about thirty seconds:

1. **A — Articulate It.** Say what it is in your own words, using something from your life.
2. **C — Connect It.** Show how it fits with what you already know, and explain why it works.
3. **E — Extend It.** Use the idea somewhere new, on a problem nobody handed you.

Each step corresponds to a level of SOLO Taxonomy and to one of John Hattie's Visible Learning phases (Surface, Deep, Transfer). When a student can ACE a concept, they own it, regardless of how the surrounding product was generated.

The ACE routine is not anti-AI. It is a way to make student thinking visible so that you can use AI confidently, knowing the student is the one doing the cognitive work.

---

## 2. The Theoretical Underpinning: SOLO Taxonomy

The **Structure of the Observed Learning Outcome (SOLO) Taxonomy**, developed by John Biggs and Kevin Collis in 1982, describes five levels of increasing complexity in a learner's response. It is the foundation of ACE because each ACE step targets a distinct SOLO transition.

| SOLO Level | What the learner can do | Learning phase | ACE step |
| --- | --- | --- | --- |
| Pre-structural | Has little or no relevant prior knowledge; responses are tangential or missing | Prior knowledge | (before ACE) |
| Uni-structural | Identifies one relevant feature or fact | Surface | Articulate |
| Multi-structural | Lists several relevant features but does not connect them | Surface | Articulate |
| Relational | Integrates the features into a coherent whole; explains how and why | Deep | Connect |
| Extended abstract | Generalizes the idea to new contexts, hypothesizes, transfers | Transfer | Extend |

### Why SOLO matters for AI-saturated classrooms

A Gen AI tool can produce extended-abstract-looking prose at any moment, on demand. SOLO gives you a way to look past the polish of the artifact and listen for the structure of the student's thinking. ACE makes that listening routine.

### Instructional moves by SOLO level

| SOLO Level | Recommended instructional moves |
| --- | --- |
| Pre-structural | Build or activate prior knowledge; AI-generated game/quiz pre-assessment, vocabulary front-loading, anticipation guides |
| Uni-structural | Three-step Jigsaw, vocabulary programs, retrieval practice, basic AI prompts for definitions |
| Multi-structural | Direct instruction, Flipped Classroom, note-taking, summarization, curated AI-summarized resources evaluated with SIFT |
| Relational | Concept mapping, metacognition, self-judgment and reflection, AI as a brainstorming and connection-finding partner |
| Extended abstract | Problem-Based Teaching, Transfer Strategies, Service Learning, AI as a prototyping and innovation partner |

---

## 3. The ACE Steps in Depth

### 3.1 A — Articulate It

> Say what it is in your own words. Use something from your life.

This is the move from uni-structural to multi-structural on SOLO, and the heart of Surface Learning in Hattie's research. You are not asking for a polished essay. You are asking for evidence that the student holds at least one piece of the idea in their own working memory and can name it in plain language.

**Sentence stems**

- "The pattern you see is..."
- "This reminds you of..."
- "Here, you noticed that..."
- "In your own words, this means..."
- "You can describe this as..."
- "An example from your own life is..."

**Listening for evidence.** You are looking for ownership, not accuracy. A partial, imperfect, lived-in explanation is more useful at this stage than a textbook-perfect one, because it tells you what the student is actually holding onto.

**Worked example, grade four science.** A fourth grader explains photosynthesis by saying, "Plants eat sunlight, kind of like how I eat breakfast." This is simplistic, but it is hers. You now have a foundation. The metaphor of "eating" gives you the next move: introduce the role of chlorophyll, water, and carbon dioxide as the ingredients in the plant's meal.

**Worked example, grade eight ELA.** After reading a short story, an eighth grader articulates the theme as, "The story is about how secrets break trust, kind of like when my older brother said he wasn't going to tell our mom and then he did." The student has named the abstract concept (secrets break trust) and grounded it in lived experience. You can hear that she owns the theme.

**Worked example, high school algebra.** A tenth grader articulates the meaning of slope as, "Slope is how steep a road is. A big slope means a steep hill, and a slope of zero is just flat ground." The student has a uni-structural grasp of slope as steepness. You can build from there toward rate of change.

### 3.2 C — Connect It

> Show how the idea fits with what you already know. Explain why it works.

This is the relational level on SOLO and the bridge from Surface to Deep Learning. The student is no longer listing facts. She is linking them, citing evidence, and explaining mechanisms.

**Sentence stems**

- "This connects to ___ because ___"
- "The reason for this is..."
- "The bigger picture shows..."
- "This works because..."
- "This is similar to ___, but different from ___"
- "The evidence you have for this is..."

**Listening for evidence.** Look for claim, evidence, and reasoning (CER). Look for the student making a connection across two ideas she has not been explicitly told to connect. Look for "because" used correctly.

**Worked example, grade four science continued.** The same fourth grader, after a few days of work, now says, "The basil in our kitchen window doesn't get much sun, so it grows slowly. The leaves stay pale because there isn't enough light to make food." She has linked photosynthesis to an observation in her own home and offered a mechanism (light, food production, leaf color). That is relational thinking.

**Worked example, grade six social studies.** Studying the Texas Revolution, a sixth grader connects: "The settlers wanted Texas to stay independent because the Mexican government changed the rules after they had already moved there. It is like if you started a game with one set of rules, and partway through, someone changed them, you would be upset too." The student has built a bridge from political grievance to a lived analogy, and named the mechanism (rule changes mid-arrangement).

**Worked example, high school algebra continued.** The tenth grader connects slope to rate of change: "Slope is how fast something changes. In the savings account problem, the slope is how many dollars you save each week. A bigger slope means you save faster." She has linked the geometric meaning to the contextual meaning, which is deep learning territory.

### 3.3 E — Extend It

> Use the idea somewhere new. Test it. Try it on a problem nobody handed you.

This is extended abstract on SOLO and Transfer Learning in Hattie's research. You are asking the student to generate a hypothesis, design a test, generalize a principle, or apply the idea to a problem that was not part of the lesson.

**Sentence stems**

- "Another way to think about this is..."
- "You can test this by..."
- "One way to check this is..."
- "If you tried this with ___, then..."
- "You could use this idea to solve..."
- "A new question this raises is..."

**Listening for evidence.** Look for a testable prediction, a new context, or a self-generated question. The student should be doing something the lesson did not directly assign.

**Worked example, grade four science continued.** The fourth grader extends: "If you move the basil to the window with afternoon sun, the leaves should darken within two weeks. You will count new leaves and check the color to see if you are right." She has generated a hypothesis with a falsifiable prediction and a measurement plan. That is transfer.

**Worked example, grade eight ELA continued.** The eighth grader extends: "You wonder if the theme about secrets breaking trust also shows up in news stories about politicians who lie. You will pick two news articles this week and check whether the same pattern is there." The student is transferring a literary theme to a real-world domain, generating an investigation she designed herself.

**Worked example, high school algebra continued.** The tenth grader extends: "If you graph your phone battery percentage by the hour, you can find the slope of how fast your phone drains. Then you can predict when it will hit zero. You will record the percentage every hour tomorrow." She is taking slope into a self-chosen, novel domain with a real measurement plan.

---

## 4. ACE Mapped to Frameworks

| ACE Step | What students do | SOLO level | Learning phase |
| --- | --- | --- | --- |
| Articulate | Say what it is in their own words | Uni-structural to Multi-structural | Surface |
| Connect | Show how it fits and why it works | Relational | Deep |
| Extend | Use it somewhere new | Extended Abstract | Transfer |

ACE is intentionally smaller than PRISM (Patterns, Reasoning, Ideas, Situation, Methods) and lighter than VIVA (Voice, Insight, Verification, Application). Both PRISM and VIVA remain useful for older students and structured coaching. ACE is the routine you teach in thirty seconds so that students, teachers, and parents share a common shorthand.

---

## 5. Effect Sizes for Instructional Strategies

Pair each ACE step with strategies that have a Visible Learning effect size of d ≥ 0.40, the "hinge point" for one year's growth in one school year. Below d = 0.40, a strategy is unlikely to outperform what students would learn anyway. The strategies listed are drawn from the Visible Learning MetaX database (February 2025 snapshot).

### 5.1 Surface Learning strategies (pair with Articulate)

| Influence | Effect size (d) | Impact |
| --- | --- | --- |
| Jigsaw Method | 0.92 | Potential to considerably accelerate |
| Feedback (corrective, reinforcement, and cues) | 0.92 | Potential to considerably accelerate |
| Captions and subtitles | 0.91 | Potential to considerably accelerate |
| Augmented Reality | 0.63 | Potential to considerably accelerate |
| Effects of testing (retrieval practice) | 0.63 | Potential to accelerate |
| Differentiation | 0.58 | Potential to accelerate |
| Flipped Classroom | 0.58 | Potential to considerably accelerate |
| Direct Instruction | 0.56 | Potential to considerably accelerate |
| Inquiry-based teaching | 0.49 | Potential to accelerate |
| Exposure to reading | 0.48 | Potential to accelerate |
| Reading strategies | 0.47 | Potential to accelerate |
| Feedback from student re quality | 0.47 | Potential to accelerate |
| Feedback from tests | 0.41 | Potential to accelerate |

### 5.2 Deep Learning strategies (pair with Connect)

| Influence | Effect size (d) | Impact |
| --- | --- | --- |
| Cognitive task analysis | 1.09 | Potential to considerably accelerate |
| Jigsaw Method | 0.92 | Potential to considerably accelerate |
| Constructivist teaching | 0.90 | Potential to considerably accelerate |
| Feedback timing | 0.89 | Potential to considerably accelerate |
| Argumentation | 0.86 | Potential to considerably accelerate |
| Outlining and organizing | 0.86 | Potential to considerably accelerate |
| Effort management | 0.77 | Potential to considerably accelerate |
| Evaluation and reflection | 0.75 | Potential to considerably accelerate |
| Elaboration and organization | 0.72 | Potential to considerably accelerate |
| Learning strategies | 0.67 | Potential to considerably accelerate |
| Alternative assessment methods | 0.66 | Potential to accelerate |
| Feedback tasks and processes | 0.63 | Potential to considerably accelerate |
| Inductive teaching | 0.60 | Potential to accelerate |
| Design thinking | 0.51 | Potential to accelerate |

### 5.3 Transfer Learning strategies (pair with Extend)

| Influence | Effect size (d) | Impact |
| --- | --- | --- |
| Collective teacher efficacy | 1.01 | Potential to considerably accelerate |
| Virtual Reality in Languages | 0.93 | Potential to considerably accelerate |
| Jigsaw Method | 0.92 | Potential to considerably accelerate |
| Mathematics problem solving | 0.88 | Potential to considerably accelerate |
| Blended learning | 0.85 | Potential to considerably accelerate |
| Virtual Reality in Science | 0.80 | Potential to considerably accelerate |
| Transfer strategies | 0.75 | Potential to considerably accelerate |
| Acceleration programs | 0.55 | Potential to considerably accelerate |

### 5.4 How to read effect sizes

- **d < 0.20** — developmental effect; probably not worth class time
- **d = 0.20 to 0.39** — teacher effect, typical of any instruction
- **d = 0.40 to 0.59** — zone of desired effects; one year's growth
- **d = 0.60 to 0.79** — potential to accelerate
- **d ≥ 0.80** — potential to considerably accelerate

The hinge point of d = 0.40 is the threshold below which a strategy is doing less than business-as-usual instruction would do anyway. When you have a choice between two strategies, pick the one with the higher effect size and the lower implementation cost, then verify the result with your own students.

### 5.5 The Jigsaw Method, all three phases

Notice that the Jigsaw Method (d = 0.92) appears in all three phases. That is unusual and worth noting. When you build a lesson around expert groups teaching each other and then returning to home groups to integrate, you give every student a turn to articulate, connect, and extend within a single class period. Jigsaw is one of the most efficient ACE-aligned routines you can run.

---

## 6. Best Practices for Prompt Development (with Generative AI)

Teaching ACE in an AI-rich classroom means modeling how to prompt AI well. A student who can write a precise prompt and then ACE the result has used AI as a thinking partner, not a substitute. Below is a synthesis of the best practices Miguel Guhlin teaches in PRISM, RETO, PREPARE, and the Meta-Prompt Analysis Rubric (MPAR).

### 6.1 The twelve components of a strong prompt (MPAR)

A prompt earns the highest score on the MPAR when it specifies each of these:

1. **Role assignment.** Define the AI's persona or expertise. *Example:* "As an experienced fourth-grade science teacher..."
2. **Goal setting.** State a clear objective. *Example:* "...write three sentence stems that help a student articulate photosynthesis in their own words."
3. **Background information.** Provide necessary context. *Example:* "The students have just finished a unit on plant parts and are weak on the role of chlorophyll."
4. **Clarity.** Make the request unambiguous. Cut every word that could be read two ways.
5. **Task breakdown.** Divide a complex request into subtasks. *Example:* "First, list the stems. Second, rate each by accessibility. Third, suggest a follow-up question."
6. **Boundaries.** State what to include and exclude. *Example:* "Do not use jargon above grade-four reading level. Do not introduce light wavelengths."
7. **Output structure.** Specify the format. *Example:* "Return a markdown table with columns: Stem, Reading Level, Follow-Up Question."
8. **Scope.** Indicate response length. *Example:* "Limit to six stems."
9. **Exemplification.** Provide examples or samples. *Example:* "Here is one stem to model the style: 'In your own words, photosynthesis is...'"
10. **Flexibility.** Leave room for creativity where appropriate. *Example:* "Feel free to suggest a metaphor I have not used."
11. **Feedback loop.** Ask for output evaluation. *Example:* "After listing the stems, rate them on a one-to-five scale for grade-four readability."
12. **Troubleshooting.** Guide error handling. *Example:* "If you cannot find a developmentally appropriate metaphor, say so and ask for more context."

### 6.2 Prompt patterns aligned to ACE

The ACE structure itself is a prompt pattern you can teach students to use with any AI tool.

**Articulate pattern.** Ask the AI to explain something at your level, then check whether you can re-explain it without the screen.

> "Explain ___ in three sentences a sixth grader could understand. Use one everyday example. Then ask me a question to check whether I understood."

**Connect pattern.** Ask the AI to help you connect a new idea to something you already know.

> "You already know ___. You are learning ___. List three ways the new idea connects to what you already know, and one way it is different. Then ask you which connection is strongest."

**Extend pattern.** Ask the AI to help you generate a transfer task you can actually carry out.

> "You learned ___ in class today. Suggest three problems from your own life or community where you could test or apply this idea. For each, list what you would measure and what would count as evidence."

### 6.3 The PREPARE framework for educator prompts

When you write a prompt for your own teacher work (lesson plans, rubrics, parent letters), use PREPARE:

- **P — Persona.** Who should the AI be?
- **R — Role of the user.** Who are you, and what is your context?
- **E — Expected output.** What format, length, and style?
- **P — Parameters.** What constraints, exclusions, or requirements?
- **A — Audience.** Who will read or use the output?
- **R — References.** What sources, standards, or examples should ground the response? (For Texas, this is usually TEKS, STAAR, or SBEC.)
- **E — Evaluation.** How will you check the quality of the output before using it?

### 6.4 A worked prompt-development example (RETO style)

**Weak prompt:** "Make me a photosynthesis lesson."

**Stronger prompt (RETO — Role, Expectation, Task, Output):**

> "As a fourth-grade Texas science teacher aligned to TEKS 4.10A, design a forty-five-minute lesson that brings students from a uni-structural to a multi-structural understanding of photosynthesis. Open with a three-minute Articulate prompt using the sentence stem 'In your own words, plants eat sunlight by...' Include a ten-minute Direct Instruction segment (d = 0.56) and a fifteen-minute Jigsaw Method activity (d = 0.92). Close with an exit ticket that asks each student to connect photosynthesis to one plant in their own home. Output the lesson as a markdown document with sections for Objective, Materials, Procedure, Assessment, and Differentiation. Cap the lesson at 600 words."

This prompt is grounded in standards, names two high-effect strategies, anchors them to ACE, and constrains output. The teacher receives a usable artifact and can then evaluate it against the MPAR or the TCEA style guide.

### 6.5 Verifying AI output before classroom use

Always check AI output against three filters before classroom use:

1. **Accuracy.** Does the content match the standard you are teaching? Does it match what reliable sources say? Spot-check facts.
2. **Alignment.** Does it match the SOLO level you intended? An AI will often default to multi-structural prose even when you asked for uni-structural simplicity.
3. **Appropriateness.** Is it readable at the right grade level, bias-free, and consistent with district policy?

---

## 7. Classroom Protocol: How to Run ACE in Thirty Seconds

The whole point of ACE is that you can teach the routine in half a minute. Here is the script.

**Setup, twenty seconds.** "When you finish anything in this class, you are going to ACE it. A is articulate. Say what it is in your own words. C is connect. Show how it fits with something you already know. E is extend. Use it on something new."

**First model, ten seconds.** Pick any concept you taught in the last week. Articulate it in your own words. Connect it to something the class already knows. Extend it to a question you are still curious about. Done.

**Run ACE at three moments.**

1. **Exit ticket.** Three lines, one per ACE step.
2. **Conferring.** During independent work, sit beside one student and ask the three questions out loud.
3. **Parent check-in.** Send the script home so parents can ask the same three questions at homework time.

**The thirty-second variant for parents.** "When your child shows you what they did, ask them three things. Tell me what this is in your own words. How does this connect to something you already know? Where else could you use this? If they can answer all three, they own the work, regardless of whether AI helped them produce it."

---

## 8. Worked Lessons Across Subjects

### 8.1 Grade three reading

**Concept:** Cause and effect in a short story.

- **Articulate.** "In your own words, what happened in the story, and what caused it?"
- **Connect.** "When have you seen one thing cause another thing in your own life? How is that like the story?"
- **Extend.** "Make up a new ending. What would change if the main character had done one thing differently?"

**Strategies to pair.** Direct Instruction (d = 0.56), reading strategies (d = 0.47) for Articulate; argumentation (d = 0.86) for Connect; transfer strategies (d = 0.75) for Extend.

### 8.2 Grade five mathematics

**Concept:** Fractions as equal parts of a whole.

- **Articulate.** "Describe a fraction using something in your kitchen."
- **Connect.** "How is one-half of a pizza the same as four-eighths? Why does that work?"
- **Extend.** "Design a snack you would split fairly among five friends. Show the fractions and prove that each share is equal."

**Strategies to pair.** Jigsaw Method (d = 0.92), retrieval practice (d = 0.63) for Articulate; cognitive task analysis (d = 1.09) for Connect; mathematics problem solving (d = 0.88) for Extend.

### 8.3 Grade seven Texas history

**Concept:** The causes of the Texas Revolution.

- **Articulate.** "Name one reason settlers wanted independence, and put it in your own words."
- **Connect.** "How did the Mexican government's policy changes connect to the settlers' grievances? What is the bigger picture?"
- **Extend.** "Find a current news story where a group is upset because the rules changed mid-stream. Compare it to the Texas Revolution."

**Strategies to pair.** Flipped Classroom (d = 0.58) for Articulate; argumentation (d = 0.86), constructivist teaching (d = 0.90) for Connect; transfer strategies (d = 0.75) for Extend.

### 8.4 High school biology

**Concept:** Natural selection.

- **Articulate.** "Explain natural selection in your own words. Use one example from an animal you have seen this week."
- **Connect.** "How does natural selection connect to antibiotic resistance you have read about in the news? Why does that work?"
- **Extend.** "Design a thought experiment to predict what would happen to a peppered moth population in a city that switched from coal heating to solar power. What would you measure?"

**Strategies to pair.** Direct Instruction (d = 0.56) for Articulate; constructivist teaching (d = 0.90), argumentation (d = 0.86) for Connect; design thinking (d = 0.51), transfer strategies (d = 0.75) for Extend.

### 8.5 High school English Language Arts

**Concept:** Rhetorical appeals (ethos, pathos, logos).

- **Articulate.** "In your own words, what is each appeal? Give one example from a TikTok or commercial you saw this week."
- **Connect.** "How do speakers combine ethos, pathos, and logos to be persuasive? Pick a speech and find all three."
- **Extend.** "Write a one-minute persuasive video script for a cause you care about. Mark where each appeal appears, and explain why you used it there."

**Strategies to pair.** Direct Instruction (d = 0.56) for Articulate; argumentation (d = 0.86) for Connect; transfer strategies (d = 0.75), design thinking (d = 0.51) for Extend.

### 8.6 Career and Technical Education (CTE), high school

**Concept:** Customer service in a retail simulation.

- **Articulate.** "In your own words, what is good customer service? Give one example from a place you have shopped."
- **Connect.** "How does good customer service connect to brand loyalty and business outcomes? Why does that work?"
- **Extend.** "Roleplay a difficult customer interaction your team has not practiced. Apply two specific techniques and reflect on what worked."

**Strategies to pair.** Flipped Classroom (d = 0.58), Direct Instruction (d = 0.56) for Articulate; constructivist teaching (d = 0.90) for Connect; transfer strategies (d = 0.75), blended learning (d = 0.85) for Extend.

---

## 9. Assessment Rubric: Did the Student ACE It?

Use this rubric to judge whether a student owns the concept, regardless of how the surrounding product was made.

| Criterion | 1 — Surface only | 2 — Approaching | 3 — Owns it |
| --- | --- | --- | --- |
| Articulate | Restates definition word-for-word; cannot rephrase | Rephrases partially; uses one own-life example | Explains clearly in their own words; uses a personal example that fits |
| Connect | Lists facts; no linking | Links two facts but cannot say why | Explains why the connection works; cites evidence or mechanism |
| Extend | Cannot apply outside the lesson | Applies to a similar problem with prompting | Applies to a novel, self-chosen problem; generates a testable prediction |

**Interpretation.** If a student scores three on all three rows, they own the concept. If their product (essay, slide deck, infographic, video) is qualitatively above their ACE score, Generative AI did more of the cognitive work than they did. The instructional response is to scaffold the weakest stage, not to accuse the student of cheating.

---

## 10. Pacing Guide: A Two-Week Introduction

If you want to introduce ACE intentionally rather than on the fly, here is a two-week pacing guide.

### Week one: Build the routine

- **Day one.** Teach ACE in thirty seconds. Model with one concept from any subject.
- **Day two.** Run an Articulate-only exit ticket. Read aloud three strong examples the next morning.
- **Day three.** Add Connect. Use the sentence stems on the wall.
- **Day four.** Add Extend. Cap student responses at one sentence each so the routine feels easy.
- **Day five.** Run a full ACE exit ticket. Score one class set against the rubric. Identify which stage needs the most scaffolding next week.

### Week two: Pair with high-effect strategies

- **Day six.** Use the Jigsaw Method (d = 0.92) for a concept you want students to Articulate and Connect in the same session.
- **Day seven.** Use cognitive task analysis (d = 1.09) as a Connect scaffold. Break the concept into its operational steps and have students re-narrate them.
- **Day eight.** Use argumentation (d = 0.86) to push Connect into Extend. Students take a position and defend it with evidence.
- **Day nine.** Use a transfer task (d = 0.75) for Extend. The task must be one the lesson did not assign.
- **Day ten.** Score a class set against the rubric. Send the parent script home for the weekend.

---

## 11. Parent Script (Send Home)

> Your child is learning a routine called ACE. It stands for Articulate, Connect, Extend.
>
> At homework time, you can use it in three questions. They take less than a minute.
>
> 1. **Tell me what this is in your own words.** Use something from your life.
> 2. **How does this connect to something you already know?** Why does that work?
> 3. **Where else could you use this idea?** Try it on something new.
>
> If your child can answer all three, they own the work, regardless of whether AI helped them produce it. If they stall on one question, that is where they need more practice. You do not need to be the expert on the subject. You only need to ask the three questions.
>
> Thank you for partnering with us.

---

## 12. Differentiation Notes

**For multilingual learners.** Allow articulation in the home language first, then in English. Provide sentence stems in both languages. Accept drawings, photos, or voice recordings as articulation evidence; the goal is to capture thinking, not language production. Pair captions and subtitles (d = 0.91) with any video used to introduce the concept.

**For students with IEPs.** Reduce the ACE step count if needed. Run Articulate-only for a week before adding Connect. Use visual organizers (concept maps for relational; comparative maps such as Venn diagrams for extended abstract). Provide AI-generated examples at the right level as scaffolds, then have the student rephrase them.

**For advanced learners.** Push Extend harder. Require a testable prediction, a measurement plan, or a real-world product. Pair with design thinking (d = 0.51) and transfer strategies (d = 0.75). Ask "what would falsify your idea?" as the next question after Extend.

**For Gen-AI-confident students.** Require them to show the prompt they used, the output they received, and an ACE of the underlying concept. This is the "prompt-and-receipts" practice. It treats AI as a tool the student is accountable for.

---

## 13. Do and Don't List

**Do:**

- Teach ACE in thirty seconds, then use it every day for a week
- Pair each step with strategies at d ≥ 0.40
- Use ACE conversationally during conferring, not only as written exit tickets
- Send the parent script home and model it once at back-to-school night
- Score student responses against the ACE rubric, not against the polish of the artifact

**Don't:**

- Accuse students of using AI based on artifact quality alone; let ACE surface the gap instead
- Skip Articulate to save time; it is the diagnostic step
- Pair ACE with low-effect strategies (d < 0.40) and expect a year's growth
- Default to AI-generated stems and forget the ones your students have already practiced with
- Treat the rubric as a gotcha; treat it as a map of where to scaffold next

---

## 14. Frequently Asked Questions

**Is ACE a replacement for SOLO Taxonomy or Visible Learning?**

No. ACE is a shorthand routine that sits on top of them. SOLO and Visible Learning remain the underlying research base. ACE is the version you teach in thirty seconds and the version a parent can run at the kitchen table.

**Does ACE work in every subject?**

Yes. It works any time you want to know whether a student owns an idea. It does not depend on subject area, grade level, or whether AI was used in producing the artifact.

**Do you need AI to use ACE?**

No. ACE predates and outlasts any specific Generative AI tool. It is a sense-making routine that happens to be especially useful in AI-rich classrooms because it surfaces what a student actually understands.

**How long until students internalize ACE?**

Most classes have the vocabulary in a week and the habit in three weeks. After that, students start ACE-ing each other in peer review without prompting. That is the signal that the routine has taken root.

**What if a student articulates well but cannot extend?**

That is useful information. It means the concept has reached relational status (Deep Learning) but has not transferred. Pair with transfer strategies (d = 0.75), problem-based teaching, and design thinking (d = 0.51). Provide a novel problem the student selects.

**What if AI generates strong ACE-shaped responses?**

It will. Move the routine to live conversation. A student who can ACE a concept verbally, on the spot, with their own examples, is not relying on an AI assist. That is the gold standard.

---

## 15. Sources and Further Reading

- Guhlin, M. (2026, June 8). **ACE It: Three Steps from Surface to Transfer Learning**. TCEA Blog.
- Biggs, J. and Collis, K. (1982). **Evaluating the Quality of Learning: The SOLO Taxonomy**. Academic Press.
- Hattie, J. **Visible Learning** synthesis and the **MetaX** database, [Visible Learning Meta-X](https://www.visiblelearningmetax.com/).
- Guhlin, M. **AI and SOLO Taxonomy: A Path to Deeper Learning**, TCEA Blog.
- Guhlin, M. **The PRISM Framework** posts, TCEA Blog.
- Guhlin, M. **Teaching Oral Assessment with VIVA**, TCEA Blog.
- TCEA **Visible Learning with Ed Tech** course (twelve CPE hours, self-paced).
- TCEA Style Guide (this vault, `/style_guide.md`).

---

## 16. Quick Reference Card

Print this and hand it to teachers, parents, or students.

> **ACE It.**
>
> **A — Articulate.** Say it in your own words. Use something from your life.
>
> **C — Connect.** Show how it fits with what you already know. Explain why.
>
> **E — Extend.** Use it somewhere new. Try a problem nobody handed you.
>
> If you can ACE it, you own it.


================================================================================
FILE: 09_role_differentiation_cards.md
================================================================================

# Role Differentiation Cards

## Classroom Teacher

Unit of work: the classroom lesson.

Expected outputs:
- What I do with students tomorrow
- Strategy demonstrations
- Lesson-level moves
- Exit ticket or student evidence

## Instructional Coach

Unit of work: the coaching cycle.

Expected outputs:
- Coaching questions
- Co-planning template
- Evidence protocols
- Debrief and next-step prompts

## Campus Leader

Unit of work: the campus system.

Expected outputs:
- Walkthrough look-fors
- Feedback structures
- PLC or faculty-meeting agenda item
- Scaling and support plan


================================================================================
FILE: 10_json_knowledge_pack.json
================================================================================

{
  "title": "AI as Your Instructional Design Partner Knowledge Pack",
  "author": "Miguel Guhlin, TCEA",
  "frameworks": [
    "ALDO",
    "SOLO Taxonomy",
    "Visible Learning phases",
    "ACE"
  ],
  "workflow": [
    "Frame topic and learners",
    "Prompt coach role + ALDO",
    "Run and answer questions",
    "Differentiate by audience",
    "Audit using six questions"
  ],
  "audit_questions": [
    "All three phases present?",
    "What SOLO level do activities target?",
    "All five ALDO components in order?",
    "Are cited effect sizes real?",
    "Is the audience visible in the moves?",
    "Does post-assessment show movement?"
  ],
  "decision_rule": {
    "0_fail": "Keep",
    "1_2_fails": "Revise",
    "3_plus_fails": "Scrap and re-prompt"
  },
  "chatbot_links": {
    "ChatGPT": "https://chatgpt.com/",
    "Claude": "https://claude.ai/new",
    "Gemini": "https://gemini.google.com/app",
    "xAI Grok": "https://grok.com/",
    "Mistral Le Chat": "https://chat.mistral.ai/chat"
  }
}

================================================================================
FILE: 11_project_instructions.md
================================================================================

# AI Design Companion — Project Instructions

Use these instructions in a Custom GPT, Claude Project, Gemini Gem, NotebookLM notebook, or any chatbot project that allows attached knowledge/source files.

## Role

You are the AI Design Companion, an instructional design partner for K-12 educators, instructional coaches, librarians, and campus leaders.

Your job is to help users design, revise, and audit instruction using the attached source files. Prioritize the source files over general knowledge. When the source files do not contain enough information, say so and ask the user for the missing context or source.

## Core Purpose

Help users:

1. Use ALDO as a lesson design conversation, not a one-shot lesson generator.
2. Align instruction to SOLO Taxonomy, Visible Learning phases, and high-effect-size strategies.
3. Create TEKS-aligned lesson, coaching, and leadership outputs.
4. Audit AI-generated drafts for quality, accuracy, phase alignment, and classroom usefulness.
5. Use source files responsibly so recommendations are grounded, current, and verifiable.

## Source Priority

Use the attached RAG/source files in this order:

1. Strategy Reference / Visible Learning source files
   - Use these for learning phases, effect sizes, strategy selection, SOLO alignment, ALDO, EIIR, RISE, LEARNS, SIFT, FLOATER, PRISM, and AI + SOLO guidance.

2. ALDO Conversation Prompt / AI Instructional Design Partner source files
   - Use these for the main workflow: frame, prompt, run, differentiate, audit.
   - ALDO should guide the instructional conversation in this order: relationship building, pre-assessment, strategic instruction, post-assessment, and reflection.
   - The AI should act as a coach, not simply generate a lesson plan.

3. TEKS-Aligned Prompt Library
   - Use this for prompt examples by grade band, role, subject, and use case.
   - Do not invent TEKS numbers.
   - Ask the user to paste or attach the exact TEKS expectation when standard-level precision is required.

4. ACE / SOLO / Visible Learning support files
   - Use these when users need a student-facing learning check, AI-use accountability routine, oral assessment, exit ticket, parent script, or surface-deep-transfer scaffold.
   - ACE maps Articulate to surface/SOLO uni- or multistructural, Connect to deep/SOLO relational, and Extend to transfer/SOLO extended abstract.

## Required Behavior

Always identify the user's role and instructional context if it is missing:

- Role: teacher, instructional coach, instructional leader, librarian, or other
- Grade level or grade band
- Subject
- Topic or unit
- Learners
- Desired output
- Exact TEKS, if TEKS alignment is requested

If the user asks for a lesson or learning activity, do not immediately produce a full lesson unless they ask for a finished draft. First, offer to walk them through ALDO.

## ALDO Workflow

Use ALDO as the default design sequence:

1. Relationship Building
   - Ask what matters about the learners, classroom context, trust, motivation, prior experiences, and learner identities.

2. Pre-Assessment
   - Ask what students already know, what misconceptions are likely, and what quick evidence can reveal their current SOLO level.

3. Strategic Instruction
   - Recommend strategies only after the pre-assessment context is known.
   - Choose strategies with effect sizes at or above d = 0.40 when supported by the source files.
   - Label each strategy by learning phase: Surface, Deep, or Transfer.

4. Post-Assessment
   - Design evidence of movement, not satisfaction.
   - The post-assessment should show how learners moved from one SOLO level or phase to another.

5. Reflection
   - Prompt the teacher, coach, or leader to identify what worked, what needs adjustment, and what evidence supports that decision.

## Strategy Selection Rules

When recommending instructional strategies:

- Match the strategy to the learner’s current phase: Surface, Deep, or Transfer.
- Do not treat high effect size as the only decision factor. Also consider timing, readiness, task complexity, and implementation cost.
- Explain why the strategy fits the phase.
- Warn users when a strategy may be mistimed.
- Avoid fabricated effect sizes. If an effect size is not in the attached sources, say: “I do not see that effect size in the provided sources.”

## SOLO Taxonomy Rules

Use SOLO Taxonomy to describe the depth of learning:

- Prestructural: no meaningful connection yet
- Unistructural: one relevant idea
- Multistructural: several ideas, not connected
- Relational: connected understanding
- Extended abstract: transfer to new contexts

When reviewing a draft, identify whether the task is mostly surface, deep, or transfer. AI drafts often over-index on multistructural output, so push toward relational and extended abstract when appropriate.

## TEKS Alignment Rules

When TEKS alignment is requested:

- Ask the user to paste the exact TEKS expectation or attach the current TEKS source.
- Do not invent TEKS codes, student expectations, or standard language.
- If the user provides only a topic, use the topic to draft a TEKS-ready structure but label it as “pending exact TEKS confirmation.”
- Keep the output easy for Texas educators to adapt.

## Prompt Library Use

When the user asks for a prompt:

- Ask which role the prompt should serve: teacher, coach, or instructional leader.
- Ask for grade band: K-2, 3-5, 6-8, or 9-12.
- Ask for subject and topic.
- Ask whether they want the output to be a lesson, coaching plan, leadership agenda, assessment, rubric, student task, or audit checklist.
- Use the Prompt Library style: role-specific, TEKS-aware, ALDO-structured, and source-grounded.

Every generated prompt should include:

- Role
- Topic
- Learners
- TEKS placeholder or exact TEKS if provided
- ALDO sequence
- SOLO target
- Visible Learning phase
- Source-use instruction
- Output format
- Audit step

## AI Draft Audit

When asked to evaluate an AI-generated lesson, prompt, or plan, use this six-question audit:

1. Are all three learning phases present where appropriate?
2. What SOLO level do the activities target?
3. Are all five ALDO components present and in order?
4. Are cited effect sizes real and supported by the attached sources?
5. Is the intended audience visible in the actual moves?
6. Does the post-assessment show evidence of learning movement?

Use this decision rule:

- 0 fails: Keep and refine.
- 1-2 fails: Revise before use.
- 3 or more fails: Scrap and re-prompt.

Watch especially for generic output, wrong sequencing, invented citations, and no phase awareness. These are known AI failure modes in the companion materials.

## ACE Use

Use ACE when the user wants a quick student-facing learning check, AI-accountability routine, parent script, or exit ticket.

ACE means:

- Articulate: explain it in your own words.
- Connect: show how it fits with what you already know.
- Extend: use it somewhere new.

Use ACE to make student thinking visible, especially when students use GenAI. Do not treat polished AI-assisted artifacts as proof of understanding.

## Output Style

Default to practical, classroom-ready outputs.

Use:

- Clear headings
- Short paragraphs
- Tables when they improve readability
- Copy-ready prompts
- Teacher-friendly language
- Specific next steps

Avoid:

- Long theory dumps
- Unsupported claims
- Fake citations
- Invented TEKS codes
- Overly generic lesson plans
- Strategies before context
- AI hype language

## Privacy and Responsible Use

Do not ask users to provide student personally identifiable information. If a user includes student names, ID numbers, disability details, or sensitive information, suggest replacing it with non-identifying descriptors.

Use phrases such as:

- “Use a non-identifying learner profile instead of student names.”
- “Describe the learning need without including private student information.”

## Default Response Pattern

When the user asks for help designing instruction, respond in this order:

1. Briefly state what you can help create.
2. Ask for any missing essentials.
3. If enough information is present, produce a draft using ALDO.
4. Label strategies by learning phase and SOLO level.
5. Include an audit checklist or revision step.
6. Note where the user should attach or verify sources, especially TEKS and effect sizes.

## Example First Response

“I can build this as an ALDO-guided design conversation. I need four things first: grade level, subject/topic, learner context, and the exact TEKS expectation if you want standard-level alignment. Once you provide those, I’ll walk through relationship building, pre-assessment, strategic instruction, post-assessment, and reflection, then audit the draft against SOLO and Visible Learning.”


================================================================================
FILE: README_USE_THESE_FILES.md
================================================================================

# How to Use This RAG Support Pack

This zip file contains source files for the AI Design Companion website and chatbot/RAG setup.

## What is included

- Markdown source files for the main frameworks and companion workflow
- A JSON knowledge pack for systems that prefer structured data
- Project instructions for configuring a chatbot assistant
- Plain-text copies of every source file for platforms that struggle with Markdown or JSON
- One combined plain-text file containing all RAG source content

## Recommended setup

### Option 1: Custom GPT / Claude Project / Gemini Gem

1. Upload all files in the `rag-support/` folder.
2. Paste the contents of `11_project_instructions.md` into the project/system instructions field.
3. Tell the chatbot to answer from the uploaded files first.
4. Ask users to paste exact TEKS expectations when TEKS-level accuracy is needed.

### Option 2: NotebookLM / Gemini with Google Drive sources

1. Place the source files in a Google Drive folder.
2. Add them as NotebookLM sources.
3. Keep source files updated in Drive.
4. Use the project instructions as the notebook guide or custom instruction text.

### Option 3: Plain-text fallback

If the receiving chatbot cannot read Markdown or JSON reliably, upload the files in the `plain-text/` folder instead.

Use `ALL_RAG_FILES_PLAIN_TEXT.txt` when the chatbot allows only one source file.

## Important use notes

- Do not let the chatbot invent TEKS codes or student expectations.
- Do not let the chatbot invent effect sizes.
- Ask users to attach or paste the exact TEKS and source materials when precision matters.
- Use ALDO as a conversation sequence, not a generic lesson-plan template.
- Use SOLO and Visible Learning to audit depth and strategy timing.
- Avoid student personally identifiable information.

## Suggested first chatbot instruction

Use the attached files as your knowledge base. Act as an instructional design partner. Use ALDO, SOLO Taxonomy, Visible Learning phases, TEKS-aware prompting, and the audit checklist. If the source files do not support a claim, say so and ask for the missing source.
