Data Science for Young Minds — Grade 3
| Topic | Details |
|---|---|
| Why sampling is necessary | Why sampling is necessary: you cannot survey 8 billion people |
| What a population is | What a population is: the entire group you want to learn about |
| What a sample is | What a sample is: the smaller group you actually study |
| The goal | The goal: make your sample look like the population |
| Random sampling | Random sampling: everyone has an equal chance of being selected |
| Stratified sampling | Stratified sampling: ensuring important subgroups are included |
| Why convenience sampling is tempting but | Why convenience sampling is tempting but dangerous |
| Activity | Activity: design 3 sampling strategies for the same research question |
| Selection bias | Selection bias: your sample systematically excludes certain people |
| Self-selection bias | Self-selection bias: only people with strong opinions respond |
| Survivorship bias | Survivorship bias: you only see the successes, not the failures |
| Famous sampling failures | Famous sampling failures: the 1936 Literary Digest poll |
| Questions to ask | Questions to ask: who was surveyed, how many, how were they selected? |
| Red flags | Red flags: tiny samples, convenience sampling, self-selected respondents |
| How news reports polls | How news reports polls: what gets left out |
| Activity | Activity: evaluate 3 real polls or surveys for sampling quality |
You cannot ask everyone. Sampling lets you learn about a large group by studying a smaller one.
Learn how to select a sample that looks like the population you are studying.
Explore real-world examples of biased samples that led to wrong conclusions.
Practice evaluating real-world polls and surveys. Are the samples representative?