Interactive Textbook Module Syllabus

Sampling & Data Collection

Probability vs. non-probability sampling, sample size determination, non-response bias, data collection methods, AI-era sampling challenges, data quality screening, and ethical considerations in sampling.

⏱️ ~39 min reading & interactives 📑 9 sections 🎓 Level: Core undergraduate 📐 Math: Interpret output 📚 Courses: Marketing Research Methods 🏷️ Topics: Sampling 🔗 Prerequisite: Survey Design & Measurement

Chapter Section Outline (9 Sections)

Topics and managerial frameworks covered in Sampling & Data Collection:

1. Why Sampling Matters More Than You Think
2. Probability vs. Non-Probability Sampling
3. Sample Size: How Many Is Enough?
4. Non-Response: The Silent Killer of Survey Quality
5. Data Collection Methods: Beyond the Online Survey
6. Sampling Bias in the Age of AI
7. Ethical Issues in Sampling & Data Collection
8. Case Application — MapleRoast Coffee Expansion Sampling Strategy
9. Putting It All Together: From Sampling Plan to Clean Dataset

Interactive Concept Mini-Widgets Embedded in This Chapter (4)

Students manipulate these interactive visual simulations directly inside the chapter reading before answering gated check-in questions:

Stratified Allocator / Niche-MOE Blow-Up

A population split into 3 strata (Mainstream 80% / Enthusiasts 15% / VIP niche 5%). The student sets total n and allocation mode (Proportionate vs.

Teaches: a 1,000-person study can be useless for a niche that only got 50 of them; you fix it by deliberately oversampling the small-but-important stratum and weighting back — not by inflating the total. Precision rides on the SUBGROUP's n, not the grand total.

Sampling Error Visualizer

Adjust sample size and see how margin of error changes. Demonstrates diminishing returns visually with a Plotly chart.

Response Rate Calculator

Set target completed responses, choose audience type and method, see estimated invitations needed.

Weight It Back

A 4-cell cross-tab (gender x age) where the sample over-represents older women (happy) and under-represents younger men (unhappy). Each cell shows sample %, known population %, and a satisfaction score.

Teaches: the same raw data tells a much worse (and more honest) story once made representative; weighting is the correction, and the 7.4 -> 4.6 gap is the cost of an unrepresentative sample. Weights only fix what you adjust for.

Paired Browser-Based Analytics Calculators (3)

This chapter connects directly to these full MktPraxis analytical apps (with pre-loaded business case scenarios, CSV upload, and reproducible Python export):