Improving Decision-Making Through Statistical Models and Calibration Training

19min Snip

00:00

Play full episode

Summary

Transcript

Episode notes

This chapter delves into the importance of measurement, statistical modeling, and replication in science, focusing on the replication crisis in psychology and social sciences. It introduces the Transparent Replications project aimed at enhancing research quality and promoting open science. The discussion compares human intuition to statistical models in various domains, emphasizing how models consistently outperform human judgment due to personal inconsistencies and biases.

Read the full transcript here.

Can we measure everything that matters to us? When is measuring the correlates of a thing pretty much just as good as measuring the thing itself? Why are some people resistant to measuring certain things? What are some things people should be measuring but aren't? What's the connection between measuring things and assigning probabilities to events? How much do we know about how well human intuition performs against "doing the math"? How inconsistent are we at applying our own principles in decision-making? What kinds of calibration training are effective? What is "value of information"? What is the Rule of 5? What are the top three things we can do to improve our decision-making?

Douglas Hubbard is an author, consultant, and recognized expert in decision theory and risk analysis. He has written several books on measurement and measuring risk. His work spans various industries including insurance, finance, pharmaceuticals, aerospace, military, energy, government, tech, and nonprofit organizations. Connect with Doug on LinkedIn, email him at dwhubbard@hubbardresearch.com, or learn more about his work at hubbardresearch.com.

Further reading:

Staff