Omid Gheysar 

Cohort 1

Public Cohort 

Cohort 2

Use this space to describe why someone will want to download your resource. You can write as much as you'd like about the "why" behind the resource and what people can expect to get out of it.

Test

Product managers who drive strategy, development, and launch of products.

Data scientists who apply statistical and ML techniques to extract insights from data.

Using data and metrics to measure performance, run experiments, and inform decisions.

Build analytics for human-behavior surveys: clean data, test claims, train logistic and trees, present a decision brief.

Omid Gheysar

Turn Survey Data Into Decisions: Python and R with AI Assist, no coding required

UX Researcher

User Researcher

Most orgs collect surveys but stop at pretty charts. Decisions need&nbsp;tested claims and interpretable models: “Does stress at work predict care-seeking?” “Which factors most move satisfaction?” This session shows a&nbsp;repeatable path&nbsp;from messy responses to evidence you can present with confidence—fast enough for busy analysts and PMs, rigorous enough for research reviews. ✅ No prior coding required—start at true zero🤖 AI-assisted coding—prompt → runnable code🧩 Guided prompts—what to ask AI and why🧪 Human-behavior datasets—clean synthetic core + optional real-survey snapshot🛠️ Reusable templates—notebooks you can keep Work with a real-world-style survey (Sleep Health &amp; Lifestyle) to turn messy data into decisions—without a coding background. You’ll guide AI to draft clean R/Python notebooks, test claims correctly, and build practical models (logistic + decision tree). We focus on evaluation, thresholds, fairness, and clear reporting—so you leave with a&nbsp;5-slide Decision Brief&nbsp;and a reproducible notebook. Objective:&nbsp;build a repeatable, AI-assisted, evidence-based workflow for human-behavior data—prompting and verifying code, diagnosing errors, and communicating results clearly. (Optional Week-1 add-on: a tiny OSMI snapshot for real-survey “reality check.”)📚 What we’ll cover (from messy to actionable) 📦&nbsp;Week 1 — Prep &amp; Theory (Async + Office Hours)Light, self-paced foundations + two live setup clinics so everyone is bootcamp-ready. 🧩 Module A — Environment &amp; PreflightUse&nbsp;Google Colab&nbsp;(Python) and&nbsp;RStudio on Posit Cloud&nbsp;(R); organize project foldersStarter notebook tour and ✅&nbsp;Readiness check&nbsp;(imports, data access via Drive / Posit Cloud) 📊 Module B — Quantifying One VariableCenter (mean/median), spread (SD/IQR), shape (skew/kurtosis), percentilesPick the right plot: histogram/box/violin/barMissingness &amp; data types; tidy columns 🧭 Module C — Relationships &amp; CorrelationAssociation ≠ causation; Simpson’s paradox (quick)Pearson vs Spearman; nonlinearity/outliersRight plots; stratify by BMI/Stress tier 🧪 Module D — Hypothesis Tests &amp; EffectsChoose χ², t-test, or ANOVA (assumptions)Effect sizes (risk diff, odds ratio) + CIs“Claim → test → effect → limitation” memo 🧠 Module E — Logistic Regression (concepts)Intuition: log-odds → probabilities (S-curve)Target =&nbsp;HasSleepCondition; leakage pitfalls; one-hot BMI/GenderInterpret coefficients as&nbsp;odds ratios; ROC/AUC, PR, calibration 🌳 Module F — Decision Trees (concepts)Splits/impurity; depth caps; overfittingInterpreting path rules; when trees vs logistic Delivery format:&nbsp;Live coding in Python on Google Colab. R provided as recorded RStudio tutorials with full project files and a&nbsp;Posit Cloud fallback&nbsp;(RStudio in your browser—no install, just click the link and make a copy).&nbsp;We will use ChatGPT; other assistants like Copilot, Codeium, or Tabnine are also welcome. Please verify outputs in Colab. 📌 Week-1 Deliverables (due Fri):Preflight pass&nbsp;(screenshot)EDA one-pager:&nbsp;2 plots + 3 insightsClaim &amp; Check&nbsp;memo (≤150 words)(Optional): first logistic + decision tree run using provided cells) 🛠️ Live Office Hours (setup clinics):📅&nbsp;12:00–1:00 PM (PST):&nbsp;Setup &amp; environment troubleshooting📅&nbsp;12:00–1:00 PM (PST):&nbsp;Preflight, plotting help, claim selection 🧑‍💻&nbsp;Week 2 — Weekend Bootcamp (Live, 2 days)From one-variable to two-variable to baseline models → thresholded decision. 📅 Day 1 (Sat)AM (9:00–12:00 PST) — 🔍 One-Variable → Two-VariableFinalize EDA + tested claim; stratified visuals; peer feedback PM (1:00–3:00 PST) — 🧮 Logistic BaselineTrain/validation split; odds-ratio table; ROC/AUC; quick calibrationSave mini&nbsp;📇 model card&nbsp;(data, target, metric, assumptions, risks) 📅 Day 2 (Sun)AM (9:00–12:00 PST) — 🌲 Decision Tree + ComparisonDepth-capped tree (≤4), path interpretation; compare to logisticError analysis &amp; subgroup checks (Gender, Age bands, BMI) PM (1:00–3:00 PST) — 🎛️ Thresholds → 🧾 Decision BriefTune threshold (PR/ROC) → action rule (e.g., sleep-hygiene nudge)Build &amp; present your&nbsp;5-slide Decision Brief:Problem &amp; dataKey chart (from Week 1)Model results (logistic + tree)Decision rule &amp; trade-offsRisks &amp; next steps 🏁 Final Deliverables:&nbsp;two-model baseline +&nbsp;Decision Brief&nbsp;+ reproducible notebook 🧰&nbsp;What you will practiceData:&nbsp;Sleep Duration, Quality of Sleep, Stress Level, Physical Activity, BMI Category, Systolic/Diastolic BP (parsed), Heart Rate, Daily Steps, Sleep Disorder (+ derived flags/tiers).Languages:&nbsp;R&nbsp;and&nbsp;Python (parallel materials).Skills:&nbsp;disciplined EDA; correct test selection; odds-ratio interpretation; model evaluation (ROC/AUC, PR, calibration); error analysis; thresholding; fairness &amp; leakage checks; decision reporting. 🧳 What you will leave withA methodical workflow:&nbsp;question → data → model → evaluation → decisionA concise portfolio on the Sleep dataset:&nbsp;EDA one-pager,&nbsp;Claim &amp; Check,&nbsp;two-model baseline, and a&nbsp;Decision BriefThe confidence to state assumptions, avoid leakage, and defend recommendations 🎁 What’s included🎥 Live sessions:&nbsp;interactive instruction with Omid🛟 Office hours:&nbsp;two Week-1 setup clinics + optional follow-up hour📼 Lifetime access:&nbsp;recordings &amp; materials🤝 Community:&nbsp;peer feedback &amp; shared project space📜 Certificate of completion✅ Maven Guarantee:&nbsp;full refund up to the halfway point

🚀 From Zero to Hypotheses, Multivariable Models, and Decisions

Course syllabus

but basic stats literacy helps: mean/median, percentages, what a correlation is.

No coding required

for Colab; ability to open shared links and save a copy to Drive.

Google account

We’ll use ChatGPT; Copilot/Codeium/Tabnine also fine

Willingness to use an AI assistant

<ul><li>Mon: 2 sessions release (videos)</li><li>Wednesday: 1 session release (video)</li><li>Saturday: 2 session release (videos)</li><li>Office Hours: Saturday, 12:00–1:00 PM (PST); </li><li>Week-1 deliverables due: Fri, 11:59 PM (local)</li></ul>

Week 1 (Async)

Weekend Bootcamp (Week 2)
 

<ul><li>EDA one-pager; Claim &amp; Check; two-model baseline; Decision Brief</li></ul>

Weekly projects

Course schedule

Analysts, researchers, and Product Managers who work with human-behavior survey data and want practical ML results fast.

Policy, public health, and government staff who need evidence-based briefs without heavy coding.

Data-curious professionals (UX, ops, HR, marketing) ready to move from charts to testable claims and simple models.

Who is this course for

What you’ll get out of this course

Train a logistic regression and a decision tree on human behavior dataset; report AUC/PR, calibration, and a confusion matrix with a chosen threshold.

Ship two working models by Sunday.

Produce an EDA one-pager (2 plots + 3 insights) and a Claim &amp; Check memo using the right test (χ² or t-test/ANOVA) with effect size and a confidence interval.

Turn messy survey data into insight.

Build a 5-slide Decision Brief: problem, key chart, model results, decision rule with trade-offs, and risks/next steps.

Present a decision your team can act on.

Follow prompt templates to generate code, then verify and debug in Colab; learn leakage checks, seeds/splits, and plain-English reporting.

Use AI effectively and safely.

Keep the polished Python (Colab) notebook and the recorded RStudio materials (with Posit Cloud fallback) so you can repeat the workflow at work.

Learn directly from Omid Gheysar

Omid Gheysar

Courses

Turn Survey Data Into Decisions: Python and R with AI Assist, no coding required

Lightning Lessons

Turn Survey Data Into Decisions: Python/R + AI Assist

Resources

Test