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


  • Define the term “confidence interval” and explain its role in the AIE process.
  • Describe two extremes of biases in subjective confidence.
  • Evaluate your own ability to calibrate probabilities.
  • Describe methods used to calibrate people.
  • Explain some of the conceptual obstacles to accurate calibration.
  • Describe the performance of individuals in calibration training research.


Chapter NaN explains how a person can describe one's own uncertainty quantitatively. Most research suggests that, due to things like overconfidence and underconfidence, people are not very good estimators of probability. However, there are ways to improve on this with training.

In these exercises, participants assess their confidence about a series of statements (either 90% confidence interval or binary true/false) and are then shown the correct answers, giving them immediate feedback showing whether their stated probabilities were realistic. A variety of other means for improving your calibration are also recommended, and further insight into the variation in individual performance is provided.


  1. Suppose you are asked how many leads you expect will turn into sales, and you give a 90% confidence interval of 11 to 23. What does this mean?
    1. You have very little information about how many leads will turn into sales.
    2. You believe there is a 90% chance the actual number will be between 11 and 23.
    3. You are ...

Get How to Measure Anything Workbook: Finding the Value of Intangibles in Business now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.