csis1/IMPLEMENTATION-PLAN.md

201 lines
9.3 KiB
Markdown
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

# CSIS 1 Summer 2026 — Research-Backed Implementation Plan
Source: `projects/gav-outreach/research/student-success-proposal.md` (Ranks 19)
This plan applies the top 9 evidence-based practices to CSIS 1, a 6-week compressed summer course. The course is fully online, asynchronous, with Monday evening Zoom demos.
---
## Practice → CSIS 1 Application Map
### Rank 1: Structured Weekly Modules with Consistent Layout
**Evidence:** DFW dropped 48% → 24% with consistent module structure.
**CSIS 1 Implementation:**
Every week follows the exact same sequence:
1. **Welcome Announcement** (Monday morning) — what's this week about, what to do first
2. **Learning Objectives** — 35 measurable outcomes per week
3. **Readings** — OpenStax chapters + supplementary links (already in syllabus)
4. **Monday Demo Recording** — posted Tuesday if students miss live session
5. **Practice Quiz** (low-stakes, 10 questions, unlimited attempts)
6. **Discussion** — one prompt per week, metacognitive where possible
7. **Real-World Task** — hands-on assignment due Sunday 11:59 PM
8. **Weekly Check-In Quiz** (graded, 10 questions, 2 attempts)
9. **Friday Wrap-Up Announcement** — what went well, common mistakes, preview of next week
Module prerequisites: must complete Week N quiz before Week N+1 unlocks.
---
### Rank 2: Frequent Low-Stakes Quizzing (Retrieval Practice)
**Evidence:** Meta-analysis d=0.420.51 effect size. Quiz submission rate predicts lower DFW.
**CSIS 1 Implementation:**
- **Practice Quiz** (per week): 10 questions, unlimited attempts, not graded (or 5% total). Purpose: retrieval practice before graded quiz.
- **Weekly Check-In Quiz** (per week): 10 questions, 2 attempts, graded. Mix of factual recall and applied scenarios.
- **Final Exam**: 30 questions drawn from a bank of 60+ (all weekly questions reshuffled + new synthesis questions).
- Questions drawn from OpenStax chapter content, GCFGlobal modules, and the imported Summer 2025 course quiz bank.
- Target: students encounter 20+ quiz questions per week across practice + graded.
**Quiz Design Principles:**
- Mix question types: factual recall (30%), applied scenarios (40%), "which of these is the best practice" (30%)
- Include "all of the above" and "none of the above" sparingly
- Write wrong answers that are plausible (common misconceptions), not absurd
- For security/scam questions: use real-world scenarios
---
### Rank 3: Regular Announcements (23 per Week)
**Evidence:** Optimal frequency 23/week. Beyond 3/week shows diminishing returns.
**CSIS 1 Implementation:**
3 announcements per week, consistent schedule:
- **Monday AM: "This Week in CSIS 1"** — overview, objectives, what to do first, Zoom demo reminder
- **Wednesday: Mid-Week Check-In** — tips, common questions from Canvas inbox, encouragement, "here's what I'm seeing so far"
- **Friday: Week Wrap-Up** — what to finish by Sunday, common mistakes on the quiz, preview of next week
Plus situational announcements: assignment clarifications, demo recording posted, grade updates.
**Tone:** Conversational, not corporate. Short paragraphs. Direct. Peter's voice.
---
### Rank 4: Transparent Gradebook with Visible Rubrics
**Evidence:** Gradebook use correlated with lower DFW (F=7.7, p=0.005).
**CSIS 1 Implementation:**
- Rubrics attached and visible on every Real-World Task assignment
- Simple 4-level rubrics: Excellent / Good / Needs Work / Missing
- Criteria categories: Completeness, Accuracy, Formatting/Presentation, Critical Thinking
- Gradebook updated within 48 hours of due date (see Rank 5)
- Grade breakdown visible in syllabus: Quizzes 30%, Discussions 15%, Real-World Tasks 30%, Final 20%, Participation/Practice 5%
---
### Rank 5: Prompt Feedback Turnaround (≤48h Communication, ≤7 Days Grading)
**Evidence:** Slow grading is consistently cited as a withdrawal driver.
**CSIS 1 Implementation:**
- Canvas messages answered within 24 hours (weekdays), 48 hours (weekends)
- Weekly quizzes auto-graded immediately
- Discussion posts graded within 3 days
- Real-World Tasks graded within 5 days (before next week's task is due)
- Stated in syllabus as a commitment, not just a goal
---
### Rank 6: Embedded Metacognitive Prompts
**Evidence:** +6.1 and +4.2 points on successive exams. Explains 25.3% of variance in final exam scores.
**CSIS 1 Implementation:**
Metacognition woven into discussions and assignments:
- **Week 1 Discussion:** "What does being 'digitally literate' mean to you? What do you already know vs. what feels unfamiliar?" (self-assessment)
- **Week 2 Discussion:** "What was hardest about the document/spreadsheet work this week? What strategy did you use to figure it out?" (process reflection)
- **Week 3 Discussion:** "Look at your quiz scores so far. What topics do you feel confident about? What do you need to review?" (self-monitoring)
- **Week 4 Discussion:** "How do you decide if a website is trustworthy? Walk us through your actual process." (metacognitive modeling)
- **Week 5 Discussion:** "After the security audit, what surprised you most about your own digital habits? What will you change?" (transfer)
- **Week 6 Discussion:** "You just used an AI tool. How did you decide what to trust and what to verify? What would you do differently next time?" (critical evaluation)
Every Real-World Task includes a short reflection paragraph: "What was the hardest part? What would you do differently?"
---
### Rank 7: Proactive Early Alert Outreach
**Evidence:** RCT: +0.33 GPA, 80% higher persistence.
**CSIS 1 Implementation:**
- After Week 1: personal Canvas message to anyone who hasn't submitted the quiz or discussion
- After Week 2: message to anyone with <60% quiz average offer help, Zoom office hours
- After Week 3 (midpoint): mid-course check-in to entire class; personal outreach to anyone missing 2+ assignments
- Template messages drafted in advance, personalized with name and specific missing items
- Tone: "I noticed you haven't turned in X yet I want to make sure you're okay. Can I help?"
---
### Rank 8: Authentic Engagement Over Performative Discussion Boards
**Evidence:** "Post once, reply twice" universally despised. Instructor must participate visibly.
**CSIS 1 Implementation:**
- NO "post once, reply twice" requirement
- Instead: "Post your response (150250 words). Then read at least 2 classmates' posts and leave a meaningful comment ask a question, share a related experience, or respectfully disagree."
- Instructor (Peter) replies to 35 posts per discussion with substantive comments
- Prompts designed to invite genuine sharing, not regurgitation (see metacognitive prompts above)
- Some discussions are scenario-based: "Your grandmother calls and says Microsoft told her to buy gift cards. What do you say?"
- Grading: Complete/Incomplete based on effort and substance, not word count
---
### Rank 9: Multimedia Integration with Chunked Content
**Evidence:** Chunked content with interactive elements: DFW 16% 7% in developmental math.
**CSIS 1 Implementation:**
- Monday demo recordings chunked into segments (≤15 min each) with chapter markers
- Reading assignments broken into specific sections (not "read chapters 12" but specific URLs to sections)
- GCFGlobal tutorials used as interactive supplements (they're already chunked and interactive)
- Code.org video series (Week 4) is 6 short videos naturally chunked
- Each module page lists content in order with time estimates: "📖 Reading (~30 min) 🎥 Video (~15 min) Practice Quiz (~10 min)"
---
## Deliverables to Build
| Deliverable | Agent | Output Location |
|-------------|-------|-----------------|
| Weekly quiz questions (practice + graded, 6 weeks × 20 Qs) | Agent 1: Quiz Writer | `projects/csis1/content/quizzes/` |
| Weekly announcements (3 per week × 6 weeks = 18) | Agent 2: Announcement Writer | `projects/csis1/content/announcements/` |
| Discussion prompts with rubrics (6 weeks) | Agent 3: Discussion Writer | `projects/csis1/content/discussions/` |
| Early alert message templates | Charlie (inline) | `projects/csis1/content/early-alerts.md` |
| Assignment rubrics (6 Real-World Tasks) | Agent 3 (or Charlie) | `projects/csis1/content/rubrics/` |
---
## Grade Breakdown
| Category | Weight | Items |
|----------|--------|-------|
| Weekly Check-In Quizzes (graded) | 30% | 6 quizzes, drop lowest |
| Discussions | 15% | 6 discussions, complete/incomplete |
| Real-World Tasks | 30% | 6 assignments with rubrics |
| Final Exam | 20% | 30 questions from bank |
| Practice Quizzes + Participation | 5% | Completion credit |
---
## Module Template (Every Week)
```
Module N: [Topic] (Date Range)
📋 Learning Objectives
- Students will be able to...
- Students will be able to...
- Students will be able to...
📖 Readings & Resources
- [OpenStax chapter link]
- [Supplementary resource]
- Estimated time: X minutes
🎥 Monday Demo
- Recording link (posted Tuesday)
- Segments: [topic 1] (0:0012:00), [topic 2] (12:0025:00)
✅ Practice Quiz (unlimited attempts)
- 10 questions, not graded
- Purpose: check your understanding before the real quiz
💬 Discussion
- [Prompt]
- Due: Wednesday 11:59 PM (post) / Friday 11:59 PM (replies)
🛠 Real-World Task
- [Assignment description]
- Rubric attached
- Due: Sunday 11:59 PM
📝 Weekly Check-In Quiz (2 attempts)
- 10 questions, graded
- Covers this week's readings + demo
- Due: Sunday 11:59 PM
```