Studio Aletheia · The Center for Quantitative Studies
6.DPSR.1.1SC 6th Grade Math
Sampling Like a Data Analyst
Lesson 04 · Application, Identify Sample Size
Companies make real decisions, what to sell, what to fix, who to hire, based on samples, never the whole population. Today you'll step into a data analyst's shoes and design a sampling plan you could actually defend to a boss.
Learning Targets and Success Criteria
Real companies pay data analysts to answer one question well: is this sample good enough to act on?
What I will learn
- I can design a sampling plan for a real-world decision.
- I can justify a sample size using the size of the population it represents.
- I can identify how a data analyst's sampling choice affects a company's decision.
- I can critique a real sampling plan and suggest an improvement.
What success looks like
- I design a sampling plan with a clearly stated sample size and population.
- I justify my sample size with a specific, reasoned explanation.
- I connect my sampling choice to the decision a company would make from it.
- I critique a flawed real-world sample using sample size and population language.
Design and defend a sampling plan the way a data analyst would for a real decision.
The words we'll use in today's lesson.
These terms will carry through today's mini-lesson, Data Lab, hands-on activity, journal, and exit challenge, and they'll keep coming back all year.
Sample Size
The number of data values collected from a population to represent it in a data set.
Population
The entire group being studied, from which a smaller sample may be collected.
Where this shows up in the real world.
Thinking like a mathematician is not just a school skill. It's what people get paid to do every day, in jobs you may not have heard of yet.
Sizing Up a Customer Survey
A sneaker company wants to know if customers will like a new shoe before making 50,000 pairs. A data analyst can't ask all 2 million customers, so they design a sample, maybe 800 customers chosen from every region and age group, and use their sample size to decide how confident the company can be before spending millions on production.
When a Sample Goes Wrong
In 1936, a magazine predicted a U.S. election winner by surveying 2 million people, a huge sample size, but got the result completely wrong, because everyone surveyed came from phone books and car registries, groups that left out most of the population. It's a famous reminder that a data scientist checks who's missing, not just how many are counted.
A Sampling Plan a Boss Would Trust
Real companies pay data analysts to answer one question well: is this sample good enough to act on?
A data analyst rarely gets to study an entire populationThe entire group being studied, from which a smaller sample may be collected.. A streaming company with 40 million subscribers can't survey all of them before deciding whether to renew a show, so instead an analyst picks a sample sizeThe number of data values collected from a population to represent it in a data set. that's large enough, and chosen carefully enough, to stand in for the whole 40 million. Choosing that sample size is a real decision with real money attached: too small, and the company might cancel a show people actually love; too narrow, and the sample might only reflect one type of viewer.
That's why analysts document their sampling plan before collecting a single value: how many people, chosen how, from which parts of the population. Today you'll write a sampling plan like that yourself, for a decision a real company might actually need to make.
Adapted from Studio Aletheia's The Center for Quantitative Studies curriculum library, drawing on mathematical resources and the SC CCR Mathematics Standards.
Materials for the Data Lab.
- A. Company Scenario Cards
- B. Sampling Plan Template
- C. Sticky notes
- D. Your Data Journal
- E. A calculator (optional)
Every sampling plan names its population, states its sample size, and explains why that size and selection method are trustworthy enough to act on.
Data Analyst Data Lab
You'll build a sampling plan for a company scenario, the way a junior data analyst would for their first assignment.
Write a Sampling Plan Memo
You'll write a short memo a real data analyst could hand to a manager, recommending a sampling plan.
Data Journal Entry
Accountability Checklist
Would You Approve This Plan?
A company with 10,000 employees wants to know if people like the new cafeteria menu, so it surveys 40 employees, all from the same department. As a data analyst, would you approve this sample size as representing the population? Explain your recommendation.
The Center for Quantitative Studies · Course Navigation
Move between lessons or return to the larger Studio Aletheia ecosystem.