← Back to Session 7

Session 7 Study Guide: When Data Tricks You

Data Science for Young Minds — Grade 3

Key Topics

TopicDetails
Truncated y-axisTruncated y-axis: starting the scale at a number other than zero
Unequal bar widths or stretched scalesUnequal bar widths or stretched scales
Missing labels and titles that hide impoMissing labels and titles that hide important information
ActivityActivity: spot the trick in 5 misleading graphs
Why 3 people is not enough to make conclWhy 3 people is not enough to make conclusions about 300
How sample size affects reliabilityHow sample size affects reliability
The coin flip exampleThe coin flip example: small samples give weird results
When is a sample big enough?When is a sample big enough?
What bias meansWhat bias means: a built-in unfairness that skews results
Selection biasSelection bias: who you chose to ask
Question biasQuestion bias: how you worded the question
Confirmation biasConfirmation bias: only seeing data that supports what you already believe
Questions every data detective asksQuestions every data detective asks: Who collected this? Why? How? How many?
Red flagsRed flags: no source, tiny sample, emotional language, missing context
The difference between data that informsThe difference between data that informs and data that persuades
ActivityActivity: evaluate 3 real-world data claims

Lesson Summaries

Lesson 1: Misleading Graphs

Learn to spot graphs that look convincing but actually distort the truth.

Lesson 2: The Problem With Small Samples

Learn why asking just a few people can give completely wrong results.

Lesson 3: Biased Data

Explore how the way data is collected can build in unfairness from the start.

Lesson 4: Be a Data Detective

Put your critical thinking skills together. Evaluate real-world data claims like a detective.

Review Questions

  1. What is a truncated y-axis?
  2. How can bar widths mislead?
  3. Why are missing labels a problem?
  4. How can you protect yourself from misleading graphs?
  5. Why are small samples unreliable?
  6. What is the coin flip example?
  7. When is a sample big enough?
  8. Can a large sample still be wrong?
  9. What is bias in data?
  10. What is selection bias?
  11. What is confirmation bias?
  12. How can you reduce bias in your data?
  13. What questions should a data detective ask?
  14. What are red flags in data claims?
  15. What is the difference between informing and persuading with data?
  16. Can you trust data in advertisements?