Data Misleading Detection

Format: Identify misleading techniques in data presentations.

Common Techniques:

1. Truncated Y-axis: Not starting from 0, making small changes look like dramatic fluctuations
2. Selective time range: Only showing the time period where data looks good
3. Correlation ≠ Causation: Ice cream sales and drowning rates rise together (both caused by summer)
4. Survivorship bias: Only looking at success cases, ignoring failures
5. Percentage vs Absolute numbers: "100% growth" might just be going from 1 user to 2

Exercise: Find data presentations in 3 product/company marketing materials and check for the misleading techniques above.