How to Measure Anything Workbook: Finding the Value of Intangibles in Business

CHAPTER 5 Calibrated Estimates: How Much Do You Know Now?

1. B
2. False. This represents your uncertainty about the bugs, not the actual average number of bugs.
3. False. Since we are focused on decision analysis, we require the subjective, Bayesian interpretation of probability. Under the frequentist’s interpretation, probability cannot be applied to future, one-off events.
4. A
5. D
6. 50%
7. True. Being lucky enough to get 9 out of 10 is far more likely than being lucky enough to get 90 out of 100 if a person actually has a much less than 90% chance of the answers between their stated bounds.
8. False. While based on trivia questions, calibration exercises are really a measure of how well people evaluate their own uncertainty as opposed to how much they know about trivia.
9. A
10. C
11. C
12. True
13. B. De Finetti used the market price of a contract that would have a chance of paying a given amount as the “operational” definition of probability.
14. A
15. B
16. D
17. True
18. False
19. B
20. D
21. A
22. True. The research conducted by the author shows that after training with trivia questions, analysts were better at applying probabilities to predictions about real-world events than executives who have not had the training.
23. B. The frequency of seemingly calibrated results on 10 question tests is almost the same as the proportion of individuals from an uncalibrated population getting lucky with the test.
24. When the person claims to be 100% confident, every one of such responses must turn out to be correct or the person is overconfident.
25. A. The answer ...