You’re offline. Practice still works and is saved on this device; videos need a connection.
Project 1: Sibling Interviews · Lesson 1.0
Watch Lesson 1.0 first. These questions use new examples, so you can check that the ideas carry over. Ask for a hint whenever you want one.
Population is the pot, sample is the spoonful, and the estimate is the judgment we make from the spoonful.
A school cafeteria asks 30 students which lunch they liked best this month, then plans next month’s menu for the whole school. Sort each part.
The population is everyone the menu is for. The sample is the 30 students they asked. The estimate is the educated guess about everyone, made from those 30.
Now the paleontologists from the lesson. Sort each part.
The population is every T. rex that ever lived, most of which turned to dust long ago. The fossils found so far are the sample, and the number of T. rex that once lived is estimated from them.
What do we call the process of using a sample to understand a population?
Statistical inference: using what we can see (the sample) to understand what we cannot (the population).
Which description fits an estimate?
An estimate is an educated guess about the population, made from what we observe in the sample. The governor’s verdict on the whole pot is one.
You might ask: with all the data we have today, do we still need samples? The lesson’s answer is yes, and it is not only about size.
Why can’t a weather forecaster observe the whole population of weather they care about?
The population includes both past and future weather, and we can’t observe tomorrow today. So we use the past as a sample to estimate the part that hasn’t happened yet.
True or false: we only take because populations are too big to count.
Size is only part of the story. A population can be in the future, destroyed by the test, or mostly gone. Statistics is a set of tools for understanding what we cannot see, using what we can.
Why does each of these need a ? Pick the main reason.
Some populations are just too big to reach, but others can’t be observed at all: a tested firework is gone, next week hasn’t happened, and most dinosaur bones turned to dust long ago.
The soup analogy goes further than population and sample. Each of these cases has a proper name in statistics, which we will get to in later lessons.
The soup is chunky, full of meatballs. What should the governor do to judge it fairly?
With a chunky soup, one spoonful may not hold a fair share of everything in the pot, so we need a bigger sample: a bowl.
The soup is fresh, and the flavors haven’t had time to mingle. What should the governor do before tasting?
A spoonful only stands for the whole pot if the pot is well mixed. Stirring first makes any spoonful a fair taste of the rest.
Our data collection method is an interview. Small changes in wording decide whether the answers can be compared.
You ask 18 people the same set of questions, face to face. What is that called?
Asking people the same set of questions is a structured interview. Sending those questions out on a mass scale, without facing anyone, is a survey.
Why might “How many hours do you sleep?” get answers that can’t be compared?
Some people will think of a school night and others of a weekend. Saying which one you mean (or asking about a typical night) gets answers that mean the same thing.
Which version of the cousins question will get the most consistent answers?
Spelling out first cousins, and leaving out siblings and the person answering, means every answer counts the same thing.
You ask someone their favorite color and they say “teal.” What does our rule from the lesson say to do?
We keep colors to the rainbow plus black and white, so answers can be grouped and compared. The person still chooses.
You are interviewing classmates, and nearly all of them grew up in the same state. Which question will tell you more?
Pick the level that varies from person to person in your population. For classmates from one state, that’s the town; for people from many countries, the country may be the better question.
These are about your own project, so they are not graded. Write your answer, then compare it with ours.
Who are your interviews a spoonful of? Write your in one sentence.
Keep this sentence. When you make an estimate later, it tells you who the estimate is about.
Choose one interview question (or your own unique question) and rewrite it so nobody could read it two ways.
A good test: would two people with the same life give the same answer? If not, the question still has room for two readings.
This lesson has no dataset yet: your interviews become one in Lesson 1.1. The examples here are made up for practice, except the ones from the lesson itself (the soup, weather forecasts, fireworks testing and T. rex fossils).
