Project 1: Sibling Interviews · Grades 5–8
1.0 Collecting Interviews
On YouTube as Foundations of All Statistics: Population, Sample, Estimate
To judge a soup, you only need a spoonful.
If you have played Stardew Valley, you know there is a festival every season. We thought the summer festival, the Luau, would be a great analogy for this lesson. During the Luau, everyone brings an ingredient to contribute to the communal soup. And the governor takes a spoonful to judge whether the soup tastes any good.
What you will be able to do
- Tell the difference between a population and a sample, and say which is which for a question of your own.
- Explain what an estimate is, and why a careful one beats a confident one.
- Understand why we study a sample even when the population is small, which turns out not to be only about size.
- Walk away with your own interview data in a notebook, ready to become a dataset in 1.1.
What you need
A pen and a physical notebook. Nothing else, and no spreadsheet yet. Nicknames instead of real names keep the project safe to run in a classroom, and cost nothing analytically, because the name was never a variable we use.
What the lesson covers
- How one spoonful can be informative about a whole pot: statistical inference and estimation, and the conditions that make this judgment reliable, like representative and i.i.d. sampling.
- What happens when the conditions are not met. A chunky soup needs a bowl rather than a spoon, oil sitting on top means you need to take a taste from several places, and a fresh pot needs stirring first. Each of those has a formal name in statistics (bias, cluster, strata, heterogeneity…), and we will cover them in later lessons and projects.
- Can't we just calculate population characteristics instead of estimating them? If we can collect lots of data on the whole population, do we still need statistical inference, and if so, why?
- Choosing our interview questions and avoiding the ambiguities that may come up: whether "how many siblings" includes the person answering, whether cousins count as siblings, whether the sleep question refers to weekdays or weekends, and whether hometown should be a home state or a home country instead. Should we or should we not restrict the answers to certain questions, like favorite color?
Concepts introduced
- Population
- Everything we want to know about: the whole group our question is about.
- Sample
- The part of the population we actually observe.
- Estimate
- An educated guess about the population, made from the sample.
Field task
Interview 15 to 20 people and write their answers in a physical notebook. Your subjects can be classmates, family, friends, neighbors, or anyone else you can reach. Keep every respondent anonymous, using nicknames or numbers rather than real names. Make sure you remember or keep a note of their nicknames in case you need to ask them clarifying questions later.
The field task is not extra credit, it is the curriculum. The dataset the class analyzes later is the one you build here.