The subtle art of misleading with Statistics
summary
Shows concrete tricks and real numbers: switching between arithmetic vs harmonic mean turns 1.56 hours into 0.53 hours; two samples with SD 3.3 ms and 44.8 ms look similar until you see ranges; surveys of five people produce “4 of 5 doctors” or a 50/50 Terraform/OpenTofu claim; charts use y-axis truncation and one‑dimensional images to amplify a 71→82 change; examples include turned‑off QA tests, pie charts from multi‑select data and data dredging that yields spurious correlations.
AI-written summary · about
chapters (12)
- 0:00Introduction to misleading statistics
- 1:03Averages and misleading metrics
- 3:54Standard deviation and data spread
- 5:19Sampling size and bias
- 7:27Types of survey bias
- 11:46Manipulating graphs and charts
- 16:22Pie charts and pie chart flaws
- 17:25Correlation vs causation
- 20:54Data dredging and p-hacking
- 23:42Political misuse of statistics
- 26:39Responsible data presentation
- 28:03Q&A on statistical scales
