Speak Uncertainty with Confidence
summary
Argument: averages hide decision‑critical uncertainty; replace single numbers with 90% confidence intervals, include a median inside interval, and write them as brackets like 2,000–7,000 (90%). Practical claim: textual intervals plus median let a budget decision (will it exceed 5,000?) and group comparisons become actionable. Tools: simple notation, interpretation rules, and Monte Carlo simulations to combine or compare estimates. Outcome: wider intervals show certainties, reduce misinterpretation, and make intuition usable alongside data.
AI-written summary · about
chapters (17)
- 0:00Introduction
- 1:50Data projects often fail
- 3:39Data distributions matter
- 5:08Problems with averages
- 7:18Averages are ambiguous
- 9:27Skewness affects averages
- 11:59Hidden uncertainty in averages
- 14:25Averages overused in communication
- 15:08Confidence intervals introduction
- 16:34What is a confidence interval
- 17:40Communicating confidence intervals
- 19:08Why 90% confidence intervals
- 21:36Practicing confidence intervals
- 23:00Confidence intervals vs averages
- 26:11Benefits of confidence intervals
- 27:56Key takeaways
- 29:02Closing
related talks
40mHow to Lie With Stats and Charts
28mUC2 Risk Ruler for CVSS 4.0: Visualizing Vulnerability Severity and Data Confidence
41mCommunicating Risk Reactively
43mHow to look like a statistician (...)
20mTrust, but Verify: Optimistic Visualizations of Approximate Queries for Exploring Big Data
01h 01m'Scenarios for Uncertainty in Business Analytics as Your Crystal Ball
11mEverything That You Wanted To Know About P-Values But Were Afraid To Ask
10mTurning data into leverage