The Model Thinker audiobook cover - What You Need to Know to Make Data Work for You

The Model Thinker

What You Need to Know to Make Data Work for You

Scott E. Page

3.9 / 5(161 ratings)
Start ListeningDownloadQR code that opens AudiobookHub on the App StoreTry free on iPhoneScan to start in 5 seconds

If You're Curious About These Questions...

You should listen to this audiobook

Listen to The Model Thinker — Free Audiobook

Loading player...

Key Takeaways from The Model Thinker

Learning Tools

Reinforce what you learned from The Model Thinker

Mind Map

The Model Thinker
Core Purpose of Models+
The Many-Model Approach+
Distribution Models+
Linear & Nonlinear Models+
Modeling Human Behavior+

Quiz — Test Your Understanding

Question 1 of 7
According to the text, what are the three primary functions that models help us achieve?
  • A. To memorize, categorize, and archive historical data.
  • B. To explain, design, and predict.
  • C. To manipulate, control, and profit from human behavior.
  • D. To prove causality, establish universal laws, and eliminate variables.
Question 2 of 7
How does Condorcet’s jury theorem apply to the use of multiple models?
  • A. It proves that using a single, highly complex model is more accurate than using several simple ones.
  • B. It guarantees that human error can be completely eliminated from statistical polling.
  • C. It suggests that if each model is mostly correct, combining multiple models reduces the overall chance of making an error.
  • D. It demonstrates that models must strictly categorize citizens by overlapping traits to be legally valid.
Question 3 of 7
Which of the following best describes a system that follows a normal distribution?
  • A. A small number of extreme outliers contain the vast majority of the values.
  • B. Growth accelerates exponentially over time because 'more leads to more'.
  • C. Values form a straight, upward-sloping line when plotted on a graph.
  • D. Most values cluster symmetrically around a central mean, making extreme outliers rare.
Question 4 of 7
The 'preferential attachment model' is used to explain which type of system?
  • A. A power-law distribution where certain things grow at rates relative to their proportions.
  • B. A normal distribution where outcomes naturally regress to the mean.
  • C. A linear regression where two variables share a negative association.
  • D. A fixed rule-based system where human behavior remains static.
Question 5 of 7
What is an important caveat to remember when using linear regression to find an association between two variables?
  • A. It only works for variables that follow a power-law distribution.
  • B. Identifying a correlation does not prove causality between the variables.
  • C. The values must form a perfect bell curve for the regression to be valid.
  • D. It can only identify positive associations, not negative ones.
Question 6 of 7
Why does the author use the example of eating a large pizza to illustrate a concave function?
  • A. Because the enjoyment of the pizza increases exponentially with every slice eaten.
  • B. Because the enjoyment remains constant throughout the entire meal, forming a straight line.
  • C. Because the enjoyment peaks early on and then gradually curves downward as hunger abates.
  • D. Because the cost of producing the pizza increases as more slices are consumed.
Question 7 of 7
When modeling human behavior, how does the 'rational-actor' model differ from 'rule-based' models?
  • A. The rational-actor model assumes people calculate optimal outcomes, while rule-based models assume people follow fixed or adaptive guidelines.
  • B. The rational-actor model is used exclusively for low-stakes decisions, while rule-based models are used for complex negotiations.
  • C. The rational-actor model assumes humans act entirely randomly, while rule-based models assume perfect predictability.
  • D. The rational-actor model only applies to fixed rules, while rule-based models only apply to adaptive rules.

The Model Thinker — Full Chapter Overview

The Model Thinker Summary & Overview

The Model Thinker (2018) is a guide to using models to make data talk. In a world inundated with information, it sheds some much-needed light on the patterns underlying the noise – and points us toward the ways we can reveal those patterns for ourselves.

Who Should Listen to The Model Thinker?

  • Modeling novices interested in making sense of data
  • Future-focused leaders interested in predicting the next big thing
  • Anyone who wants to sound a little smarter at dinner parties

About the Author: Scott E. Page

Scott E. Page is an American social scientist based at the University of Michigan, where he’s John Seely Brown Distinguished University Professor of Complexity, Social Science, and Management. He’s the author of several other titles, including The Diversity Bonus and The Difference.

🎧
Listen in the AppOffline playback & background play
Get App