← Back to PD Hub
Series 03 · GES Grizzlies PD

AI to Analyze Data
& Plan Instruction

How to read MAP Growth-style assessment data, use AI to surface patterns, and translate numbers into differentiated instructional plans — without drowning in spreadsheets.

00Foundation & Data FrameworkBase
01Reading MAP Growth Reports2 hrs
02AI Tools for Data Analysis2 hrs
03Data to Instructional Plans2 hrs
04Data Cycles & FC Integration2 hrs
📖Resources & ToolsLibrary
Series Foundation

From Raw Data to Responsive Teaching

Assessment data is only valuable if it changes instruction. This series gives teachers the literacy to read growth data, the AI tools to analyze it efficiently, and the planning frameworks to act on it.

📊

The Director's Direction

The DoWEA director called for a MAP Growth-style system: vertically scaled, growth-oriented, administered in fall/mid-year/EOY windows. GES doesn't need to wait for the system rollout — this series builds data literacy NOW so teachers are ready when it arrives.

01
📥
Collect
3-window assessment data in fall, mid-year, EOY
02
🤖
Analyze with AI
AI tools surface patterns, clusters, and outliers
03
🎯
Name Groups
Growth tiers: exceeding, on-track, at-risk, intensive
04
📋
Plan Instruction
Differentiated instructional moves per group
05
🔁
Monitor & Adjust
FC meetings as data-response cycles, not compliance
📈

Growth vs. Proficiency

MAP Growth measures how much a student grew — not just where they are. A student at the 30th percentile who grew 2 years in one year is succeeding. A student at the 70th who stagnated is at risk.

🤖

AI as Data Partner

AI tools can analyze a class roster of MAP scores in seconds — identifying clusters, flagging anomalies, comparing to growth norms, and suggesting instructional groupings. Teachers focus on the instructional response, not the arithmetic.

🔁

FC as Data Cycle

Focused Collaboration meetings restructured as monthly data-response cycles: look at the data, name the pattern, plan the instructional response, implement, and return with evidence. Not compliance — precision.

Session 01 · 2 Hours

Reading MAP Growth Reports: Data Literacy for Every Teacher

Before teachers can use data, they must be able to read it. This session builds universal MAP Growth data literacy — what each number means, what "growth" means, and what questions to ask.

SESSION 01 OF 04

MAP Growth Decoded: RIT Scores, Norms & Growth Targets

⏱ 2 Hours👥 Whole Staff
Learning Objectives
  • Explain what a RIT score represents and how it differs from a percentile
  • Interpret a student-level MAP Growth report: growth, projected growth, and met/not met
  • Identify students who are growing faster than expected vs. those who are falling behind the growth norm
  • Ask three productive data questions that lead to instructional action
TimeActivityDescriptionFormat
0:00–0:15LaunchWhat Does This Number Mean?Show a real (anonymized) MAP report. Ask: "What do you see? What do you wonder? What questions would you ask about this student?" Collect responses. Use to diagnose what teachers already know vs. what needs building.Whole group
0:15–0:35TeachMAP Vocabulary: RIT, Growth, NormsExplicit teaching of key terms: RIT score (like a ruler, not a grade), conditional growth norms (how much growth is typical for a student at this starting point?), growth percentile (where does this student's growth fall compared to peers?), projected proficiency. Use NWEA's visual explainers.Whole group
0:35–0:55PracticeRead a Student Report TogetherProvide a model student report. Teacher pairs work through guided questions: What is this student's fall RIT? What was their spring RIT? How much did they grow? What did NWEA project? Did they meet their growth target? What does this student need next?Pairs → share
0:55–1:20ApplyClass Report AnalysisProvide a model class report (anonymized, multi-student). Teachers sort students into four quadrants: High RIT + High Growth / High RIT + Low Growth / Low RIT + High Growth / Low RIT + Low Growth. Each quadrant requires a different instructional response.Grade-band groups
1:20–1:40DiscussWhat Data Tells Us — and Doesn'tCritical discussion: MAP is one data point. What does it NOT tell us? (Why a student grew, what specific skill is missing, whether the test was a good day.) How do we combine MAP data with our classroom observations? What's the danger of over-indexing on one number?Whole group
1:40–2:00PlanThree Questions Per StudentEach teacher selects 3 students from their roster and writes three data-driven questions for each: "What does this student need next?" "What might explain this growth trajectory?" "Who else might be in a similar situation?" These become the FC meeting agenda item before Session 2.Individual
Session 02 · 2 Hours

AI Tools for Data Analysis: Working Smarter, Not Harder

AI doesn't replace a teacher's instructional judgment — it processes data so teachers can focus on students instead of spreadsheets. This session gives every GES teacher a practical AI data workflow.

SESSION 02 OF 04

Practical AI Data Analysis: Prompts, Patterns & Pitfalls

⏱ 2 Hours💻 Hands-On
Learning Objectives
  • Use an approved AI tool to analyze a class dataset and identify growth patterns
  • Write effective prompts that produce actionable data summaries from AI
  • Interpret AI-generated data analysis critically — knowing when to trust and when to question
  • Maintain student privacy when using AI tools for data work
🔒

Privacy First — Always

Before any AI data work: student names and identifying information are NEVER entered into AI tools. Teachers use initials, student numbers, or anonymized data only. This is non-negotiable and must be established before the hands-on portion begins. Principal reviews DoDEA-approved AI tools before this session.

TimeActivityDescriptionFormat
0:00–0:15LaunchThe Spreadsheet ProblemDiscussion: "How long does it take you to go from receiving MAP data to having a usable instructional plan?" Surface the answer: too long. That's the problem AI solves. Show a 2-minute demo of AI analyzing 24 student RIT scores and generating a grouping recommendation.Whole group
0:15–0:30TeachThe AI Data WorkflowIntroduce the 4-step process: (1) Anonymize data, (2) Structure the prompt, (3) Review AI output critically, (4) Add your professional judgment. Walk through each step with a live demonstration using an approved tool (Claude, Khanmigo, or similar DoDEA-approved option).Whole group
0:30–1:00PracticePrompt Engineering for DataTeachers practice writing AI prompts using a structured template: "I have [X] students with MAP Reading RIT scores ranging from [low] to [high]. The class growth target is [X] RIT points. Please: (1) identify students who need urgent support, (2) suggest 3 groupings for differentiated instruction, (3) identify the top 2 skills to prioritize based on these RIT scores." Iterate prompts together.Pairs → share
1:00–1:25ApplyLive AI Analysis with Class DataEach teacher (or team) enters their anonymized class data into the AI tool using their refined prompt. AI generates: growth groupings, skill gap summary, instructional recommendations. Teachers annotate: "This matches my observation" / "This surprises me" / "I'd adjust this because..."Individual/pairs
1:25–1:45CritiqueWhen AI Gets It WrongShow examples of AI data analysis errors: misidentifying a student's greatest need, overfitting to one data point, ignoring context. Discussion: AI is your assistant, not your expert. What does YOUR professional knowledge add that AI cannot provide? Establish the "always verify" habit.Whole group
1:45–2:00CommitAI Data Analysis Workflow CardTeachers complete their personal AI data workflow card: which tool, which prompt template, which privacy rules, and how they'll verify AI output. This card goes on their desk — it is their standing protocol for the rest of the year.Individual
🔴 Kinder Data Focus
  • Key K data: CKLA phonics screener, phoneme segmentation fluency, letter-naming fluency
  • MAP in K: use with caution — young children's test behavior adds noise
  • AI prompt: "These are my K students' phoneme segmentation scores: [data]. Which students are below benchmark? What PA skills should I prioritize?"
  • Always pair quantitative data with observation notes
🟣 Primary Data Focus
  • Key data: MAP Reading RIT + CKLA phonics assessments + fluency CBM
  • Look for: strong decoders with low comprehension (SVR LC gap) — these need vocabulary and knowledge-building, not more phonics
  • AI prompt: "Group these 1st graders [RIT scores] into 3 differentiated reading groups with instructional focus for each"
🟢 Intermediate Data Focus
  • Key data: MAP Reading + MAP Math + writing samples + content-area assessments
  • Look for: students with reading RIT gaps who are masking with content knowledge — or vice versa
  • AI prompt: "Analyze these 4th-grade MAP scores across reading and math. Which students show divergent performance? What might explain the gap?"
Session 03 · 2 Hours

From Data to Instructional Plans: Closing the Loop

Data is only valuable if it drives instruction. This session builds the specific instructional responses to each data pattern — so teachers leave with actual lesson adjustments, not just groupings.

SESSION 03 OF 04

Data-Driven Differentiation: Four Groups, Four Responses

⏱ 2 Hours📋 Planning-Heavy
Learning Objectives
  • Design differentiated instructional responses for each of the four MAP growth quadrants
  • Use AI to generate differentiated lesson adaptations for specific student clusters
  • Distinguish between instructional responses that require re-teaching vs. enrichment vs. acceleration
  • Build a 2-week instructional response plan for a specific student group
TimeActivityDescriptionFormat
0:00–0:15DebriefImplementation SharePairs share: "I analyzed my data using AI. Here's what I found. Here's one thing that surprised me." Principal listens for patterns — any AI tools concerns, privacy issues, or misconceptions to address.Pairs → whole
0:15–0:35TeachThe Four Quadrant Response FrameworkPresent the instructional response for each data quadrant: (1) High RIT + High Growth → Acceleration, enrichment, AI-extended challenges; (2) High RIT + Low Growth → Engagement, monitoring, goal-setting conversations; (3) Low RIT + High Growth → Celebrate, sustain, vocabulary-building; (4) Low RIT + Low Growth → Intensive re-teach, small group, progress monitoring.Whole group
0:35–1:00AI ApplyGenerate Differentiated Lesson AdaptationsTeachers use AI to generate 3–4 differentiated versions of an upcoming lesson. Prompt model: "I am teaching [skill] to a [grade] class. I have 4 students who need acceleration, 8 who are on track, 6 who need additional scaffolding, and 3 who need intensive support. Generate differentiated practice tasks for each group." Review and refine AI output together.Subject groups
1:00–1:30Build2-Week Response PlanEach teacher builds a 2-week instructional response plan for their lowest-growth student cluster: specific instructional moves, AI tools to deploy, progress monitoring checkpoints, and a success indicator. Plans are shared across pairs for feedback.Individual + pairs
1:30–1:45DiscussWhat About the High Flyers?The director specifically named agentic AI for Tier 2 gifted learners. Discussion: how do we use AI to extend and challenge high-growth students without losing them to busywork? Preview Session 4 of the AI Instruction series. Teachers identify their highest-growth students by name.Whole group
1:45–2:00CommitImplementation + FC PreviewTeachers commit to implementing their 2-week plan. Principal previews how the next FC meeting will function as a data-response cycle — teachers will bring evidence of how the plan worked.Whole group
Session 04 · 2 Hours

Data Cycles & FC Integration: Making Data Part of the Culture

The final session embeds the data analysis workflow into GES's ongoing FC meeting structure — so data-driven instruction becomes the default, not the exception.

SESSION 04 OF 04

The GES Data Cycle: FC Meetings as Precision Instructional Planning

⏱ 2 Hours🏁 Culminating
Learning Objectives
  • Design a monthly FC meeting agenda structured as a data-response cycle
  • Use AI to prepare data summaries for FC meetings — reducing prep time and increasing focus
  • Describe GES's 3-window data calendar and what happens at each window
  • Commit to a year-long data culture practice grounded in growth, not compliance
TimeActivityDescriptionFormat
0:00–0:20ModelA Model FC Data MeetingPrincipal and one volunteer teacher model a 15-minute data FC cycle: teacher brings anonymized data, AI-generated summary, and a specific student cluster. Principal asks: "What does the data show? What's your hypothesis? What's your instructional response? What evidence will you bring back?" Class observes and debriefs.Model + debrief
0:20–0:45PracticeRun a Grade-Band Data FCEach grade band runs their own 20-minute data FC using real (anonymized) data. Roles: data presenter, questioner, recorder. Rotate through. Principal circulates. Each group leaves with one agreed instructional priority and a monitoring plan.Band groups
0:45–1:05DesignThe GES 3-Window CalendarTogether, map GES's assessment calendar: fall window, mid-year window, EOY window. What happens at each window? What AI analysis runs at each? What does the FC meeting look like after each window? Build the shared calendar that goes into every teacher's planning system.Whole group
1:05–1:25BuildAI FC Prep ProtocolDesign the standard AI prompt teachers will run before every data FC: "Here is my class's growth data from [window]. Prepare a 5-bullet summary of key patterns, identify 3 students who need urgent attention, and suggest 2 instructional priorities for our next 4-week planning cycle." Practice running it live.Individual + pairs
1:25–1:45ReflectData Culture Self-AssessmentTeachers rate current data practice: Do I look at data before or after making instructional decisions? Do I use data to celebrate growth or just flag failure? Do my FC meetings produce instructional plans or compliance reports? Set one data culture commitment for the next 90 days.Individual
1:45–2:00CloseGrowth Is the StoryPrincipal closes: "Every student at GES will have a growth story this year. Our job is to know that story before a parent asks — and to be the ones who made it happen." Connect to the GES data identity: we measure growth, not just achievement. We celebrate trajectory.Whole group
Research Library

AI + Data Resources