Published using Google Docs
Transcript: Lesson 10 - Opening story
Updated automatically every 5 minutes

BYU-Idaho Online Learning

Video Transcript

Lesson 10 - Correlation/Causation


[One speaker.] 

[President Eyring is sitting in his office at his desk. He is dressed in a suit and is discussing the differences between correlation and causation.]

President Eyring: I have found that for me, the math that has been most valuable is the math that allows me to interrogate a claim or a statistic, to find out what it seems to be saying is really true. Certainly as a business consultant I learned how to analyze things by the numbers. But I also learned the limitation of the numbers. There is a phrase that says, “The things that matter are hard to quantify, and the things that can be quantified don’t matter so much.” Now that’s not entirely true, but there’s a limit to what you can do with numbers. And it's certainly crucial to understand the data and analysis and assumption behind the numbers. One of my favorite books was written in the 1950s; it’s called, “How to Lie with Statistics.” According to the author Darrell Huff, who is just a witty writer, and he tells this story of a pacific island in which the natives assumed that having body lice was healthy. And the more the body lice the healthier you would be. I mean I don’t know if they had lice salesmen on the island, but at any rate they were proud of having lice. I suppose it lead to their taking fewer baths so they’d feel healthier. Well, it turns out that what would happen is that when people died often of fever, their body temperature would rise before they died and the lice would get off because it was too hot. So dead people didn’t have lice, living people did. Therefore, it was assumed you want lice. Well there is a confusion of correlation versus causation. And you don’t have to be a great mathematician to make that assessment. You just ask questions about how did the number, how did you generate that number? Tell me what you did. Who did you talk to? Tell me about the nature not just the number of the people you sampled. But who were they, you know? When you did your analysis of happiness were you at an amusement park or in a jail? Those types of things will make a big difference in terms of the confidence that you’ll put in the numbers that are being used to try to influence your behavior.

[End of activity.]