Goodreads helps you keep track of books you want to read.

Start by marking “The Model Thinker: What You Need to Know to Make Data Work for You” as Want to Read:

Other editions

Enlarge cover

Rate this book

Clear rating

1 of 5 stars 2 of 5 stars 3 of 5 stars 4 of 5 stars 5 of 5 stars

Open Preview

The Model Thinker: What You Need to Know to Make Data Work for You

Scott E. Page

3.87 · Rating details · 216 ratings · 29 reviews

How anyone can become a data ninja

From the stock market to genomics laboratories, census figures to marketing email blasts, we are awash with data. But as anyone who has ever opened up a spreadsheet packed with seemingly infinite lines of data knows, numbers aren't enough: we need to know how to make those numbers talk. In The Model Thinker, social scientist Scott E. Page ...more

Get A Copy

Hardcover, 448 pages

Published November 27th 2018 by Basic Books

More Details... Edit Details

Friend Reviews

To see what your friends thought of this book, please sign up.

Reader Q&A

To ask other readers questions about The Model Thinker, please sign up.

Be the first to ask a question about The Model Thinker

Lists with This Book

21 Lessons for the 21st Century by Yuval Noah Harari

teamNOread

22 books — 1 voter

Notes from Underground by Fyodor Dostoyevsky

INTP Books - Personality Type

56 books — 10 voters

More lists with this book...

Community Reviews

Showing 1-30

Average rating 3.87 ·

Rating details

· 216 ratings · 29 reviews

More filters

Sort order

Start your review of The Model Thinker: What You Need to Know to Make Data Work for You

Apr 17, 2019 Mel (Epic Reading) rated it it was amazing

Shelves: non-read-challenge, arc-netgalley

As a professional Business Intelligence Analyst (BIA) this is the perfect non-fiction book for my desk. I do data aggregation, reporting, and analysis at my day job. And we are constantly trying to determine what the best correlation, representation or model is for analysis the data available to us.
Scott E. Page starts us off talking about WHY. This is an often overlooked piece of any business work. The why. Why do we do something? Why do we care? Why use X over Y? And so on...
In this case As a professional Business Intelligence Analyst (BIA) this is the perfect non-fiction book for my desk. I do data aggregation, reporting, and analysis at my day job. And we are constantly trying to determine what the best correlation, representation or model is for analysis the data available to us.
Scott E. Page starts us off talking about WHY. This is an often overlooked piece of any business work. The why. Why do we do something? Why do we care? Why use X over Y? And so on...
In this case Page is able to eloquently argue that in today's world of "big data" we need to be more aware of what options are available to us for analysis. It's no longer appropriate to use one model to analysis a problem. Instead we need to leverage the multi-model approach and look at complex issues, problems or phenomena (as he calls them) from many angles. As the world has gotten more complex, our data has gotten larger and more granular; which means we need to look at it from many different perspectives.

Seriously Detailed
I'm on a team of BIA's, many more senior than I, and so I chatted with them about the core concepts in Page's book and we all agreed on one thing. It's comprehensive! If there is a major model not represented in The Model Thinker I'd be shocked. Page does a great job of touching on 30+ models and giving three very specific things for each:
1) The definition and general usage
2) The actual mathematical breakdown
3) A real-life, relevant example of using the model.
The best part of reading any portion of Page's epic selection of models is easily the examples. From health care to criminal form to food quality to the stock market to population growth; we are given applicable scenarios to understand the nuances of each model and why it's the best choice. Plus there are tons of little tidbits and facts that are fun to share at cocktail parties (or if you're me, in random elevator conversations) in here!

Creativity
There is one thing that really surprised me in The Model Thinker and that is Page's emphasis on creativity. As someone who has a BA in Communications, Marketing & Design, and used to work in the magazine industry, I was surprised to find that my current Analytical career had such a basis in creativity. Reflecting upon Page's statements and my job I realized that he is right. When we program/code, develop visualizations or infer outcomes from data; we are looking at something and creatively manipulating it. Perhaps this explains how I went from an Art Director to a Analyst in one lifetime.

Overall
This is a book that I will be purchasing for my office shelf. I have already gone back to my eARC copy multiple times to look things up and to continue learning the models. It's not a book you will likely read cover to cover at any given time. But the first 50 pages of introduction and concepts are superb and well worth reading in order. After that you can jump around to the models you are most interested in, or if you're looking for the right solution for data crunching, read the intros to each chapter to determine if there might be applicable use to your situation. I know that The Model Thinker has already been picked up by some mainstream Universities and Colleges as a required textbook and certainly I can see why. In one book you gain the knowledge of hundreds of years worth of calculations and analyzation. Whether you currently work as a Data or Business Analyst, have a desire to learn to use big data, are a programmer or just a geek that loves graphs; The Model Thinker is likely to fill a void, you didn't even know existed, by giving you more models, examples and calculations than you will ever need.

To read this and more of my reviews visit my blog at Epic Reading

Please note: I received an eARC of this book from the publisher via NetGalley. This is an honest and unbiased review. ...more

flag 15 likes · Like · see review

Nov 25, 2018 Greg rated it really liked it

Shelves: self-help, business, reference

I really like the subject of this book. Model thinking is one of the best subjects I have taken in Coursera. The concepts are really useful and practical. It helps you to frame your thinking about numerous things in our world.

I would not give this book a 5 stars because I think this subject is best convey through other format like videos, lectures, or course. The topic is a bit complex especially for those who ate not comfortable with numerical reasoning. By putting it in book format it limited ...more

flag 6 likes · Like · see review

Oct 26, 2018 Rachel rated it liked it

Shelves: arc, data, economics, internet, non-fiction, politics, research, technology, net-galley, dnf

Professor Page certainly introduces some interesting concepts in The Model Thinker. His overall aim here is to get you to use multiple models in your thinking, and the plethora of models provided will be an aid to any person out there.

That said, I found it rather difficult to get through. My attention wandered far more than I thought it would. I think that is just the nature of non-fiction sometimes. Even when interested in a subject, a book on it is not always enjoyable. Thus, my star rating ...more

flag 4 likes · Like · see review

Jan 19, 2019 Alexis Bauer Kolak rated it really liked it

The title of this book is a little misleading. While the author provides a clear, informative overview of the different types of models and the kinds of problems they are regularly used to solve, he never touches on examples of where you might want to apply them. I was expecting a little more guidance on what this information means to me.

In spite of that, I could see this being very useful, and it did make me want to read more on some of the topics covered here.

flag 3 likes · Like · see review

Jun 09, 2019 Ramanathan Palaniappan rated it really liked it

Charlie Munger (the billionaire investor/partner of Warren Buffett) said: "The first rule is that you’ve got to have multiple models—because if you just have one or two that you’re using, the nature of human psychology is such that you’ll torture reality so that it fits your models, or at least you’ll think it does. You become the equivalent of a chiropractor who, of course, is the great boob in medicine.

It’s like the old saying, “To the man with only a hammer, every problem looks like a nail.” Charlie Munger (the billionaire investor/partner of Warren Buffett) said: "The first rule is that you’ve got to have multiple models—because if you just have one or two that you’re using, the nature of human psychology is such that you’ll torture reality so that it fits your models, or at least you’ll think it does. You become the equivalent of a chiropractor who, of course, is the great boob in medicine.

It’s like the old saying, “To the man with only a hammer, every problem looks like a nail.” And of course, that’s the way the chiropractor goes about practicing medicine. But that’s a perfectly disastrous way to think and a perfectly disastrous way to operate in the world. So you’ve got to have multiple models."

We solve problems every day and our approach to problems is biased based on what we've learned
and experienced so far. The saying "To the man with only a hammer, every problem looks like a nail" is an apposite summarization of the above behavior. The problem with this approach is that it can be dangerous and result in completely opposite outcomes deviating from the expected. When workers join a new organization and the conventional statement in the meeting room is: "It's always been done this way." We have to give the problem its due respect and action in an appropriate way. This is where model thinking comes into play; Pague's book performs a comprehensive review of all the major models that can be used to explain how the world works - from policy decisions by governments/disease prevention organizations to equilibriums in game theory to almost even time distribution of parents' time with their kids. :)

In short, once we validate (with available historical data) that a model can be fit to a problem, the model will be able to predict the future state of the system. In the machine learning world, this can be used to predict what books you might like to read, people you might want to date, etc., In another world, models can predict the eventual state of a system that might appear random at first. For example: Pague introduces to the Balancing Process in the chapter on Path Dependence. Imagine a urn containing 1 white ball and 1 grey ball. You pick a ball at random: if the ball is white, you add the white ball back but adding 1 more gray ball and vice versa - this is the balancing process to ensure you will likely see a grey ball in the next turn with higher probability. Repeating this process indefinitely shows that the system will converge to an equilibrium state consisting of almost equal proportions of white balls and grey balls in the urn! The real life example of this is parenting with 2 kids. If you spend time with your elder one in the morning, you might most likely spend time with the younger one in the evening.

The book is an exhaustive summary of many models that could be applied at home, work and government organizations. It was intriguing to get a peek into how centers like CDC model disease outbreaks and control the spread (these come under network models which also covers social network phenomena like Facebook). The book is more textbook style narrative although Pague annotates every model with simple numerical examples (to illustrate the models) and relevant real world applications of the models. What surprised me was the plethora of scenarios where models could be applied and in some cases, either the same model could be applied to many domains or many models can be combined to explain one scenario (thus the many model approach referred to as model ensemble).

Pague's book is deep, dense, highly practical and very helpful to get clarity about how the world operates. Most of this is rooted in consumer psychology and rational behavior. Some of the models/chapters might need extra research to get a better understanding and that's where I felt Pague could have compressed too much into one book. IMO, the book could have been better off as a 2 part series but nevertheless, it's highly worth your time particularly if you are in the knowledge industry. Go for it! ...more

flag 2 likes · Like · see review

Apr 23, 2019 Héctor Iván Patricio Moreno rated it it was amazing · review of another edition

Shelves: everyone-must-read, programmers-must-read

This book is awesome. First, it heightened my appreciation for math. I've never had the opportunity to learn about models in such great detail and extensiveness. The book explains formally many models used to understand several problems and phenomenons that happen in the current world. The models help you understand in a mathematical and formal way some thoughts you may have intuitively and explain it. Other times, applying the same model to different problems makes you understand things you ...more

flag 1 like · Like · see review

Jul 21, 2019 Jasper Burns rated it it was amazing · review of another edition

Shelves: favorites, business, science, politics, life, psychology

I was first introduced to Prof Scott Page by listening to Farnam Street's podcast called The Knowledge Project. I've been increasingly curious about mental models, and this book was my first full book on the topic.

The Model Thinker reduces real-life interactions into mathematical models. If you have not had an undergraduate-level background in mathematics I'd imagine it difficult to follow along, but if you are familiar with math symbols and formula, the book is incredibly interesting and I was first introduced to Prof Scott Page by listening to Farnam Street's podcast called The Knowledge Project. I've been increasingly curious about mental models, and this book was my first full book on the topic.

The Model Thinker reduces real-life interactions into mathematical models. If you have not had an undergraduate-level background in mathematics I'd imagine it difficult to follow along, but if you are familiar with math symbols and formula, the book is incredibly interesting and useful.

Page's claim is that by understanding models of the world, even if imperfect, we can infer actual useful information and normative policies from them. This is, in fact, the entire premise underlying the field of economics. He covers game theory, network effects, path dependence, randomness, signaling, and many other complex attributes of the world that can be simplified to understandable systems.

The single most useful model he gave was his model-of-models, about why no model can be inerrant. He models mathematically why any rule which interacts with the complexity of the real world must have exceptions. This has so profoundly changed my thinking of the world that it earns the book five stars alone. The argument is as follows:

The realities of the world comprise near-limitless amounts of data (for example, individuals consumption of sugar and their weight). This data can be packaged into information (individuals in a sample who ate more sugar weighed more). Knowledge about the world is in understanding how information relates to other bits of information (eating sugar was the cause of people weighing more). Wisdom is the conditional understanding of how to utilize knowledge to some end (avoid sugar to weigh less). In the model, you see how wisdom is abstracted so far out from the data that it can only speak generally. There is some individual somewhere who will gain weight when they eat less sugar (maybe they over-substitute for fat or other carbs). While abstract things can still be absolute—this model does not invalidate math for example—it makes clear that any wisdom we have about the actual, tangible world must be conditional. It is as close to a formula for humility as you can come by.

This was one of the first, but most impactful of the many models on offer. Consumer behavior, opioid use, social networking, cooperation—all these things can be modeled and put to good use by businessmen, politicians, and also the standard layperson. Consulting the table of contents can give a good overview of what learning is on offer.

The book was also expertly written. It was some of the most concise, clear prose I've ever read. It felt like every word and every sentence was useful. Each paragraph fit together to make a point. It's refreshing to see the work of someone who can be as precise in their English as in their math.

Would not recommend this book if you are unfamiliar with math, but if you know basic notation and want to know how and why the world works in the way that it does, these models are an excellent start. Well worth reading.
...more

flag Like · see review

Nov 16, 2019 Jim Mehnet rated it it was amazing

Shelves: 2019-books-read

The case made by the author is that social systems are very complex and often have vast numbers of what could be causal forces at play. In physics, chemistry and biology, sciences from which social sciences emerge we have often had few enough variables that relatively simple models have been constructed since the enlightenment that are so surefooted at predicting fairly granular aspects of the future that we call them laws. The social sciences (economics, psychology, political science, The case made by the author is that social systems are very complex and often have vast numbers of what could be causal forces at play. In physics, chemistry and biology, sciences from which social sciences emerge we have often had few enough variables that relatively simple models have been constructed since the enlightenment that are so surefooted at predicting fairly granular aspects of the future that we call them laws. The social sciences (economics, psychology, political science, sociology, etc.) that are emergent from the physical sciences and have complexity that makes creating model that consistently predict future outcomes correctly enough to be called laws difficult at best. Such models would likely be too complex for our minds to use effectively.

So what to do? Should we examine with a single model that is more detailed and more complex. The author makes a strong case that often this leads to less understanding and is impossible to calibrate meaningfully. But there is a proven way to make more accurate predictions. If we simplify a problem in the social sciences we can model it and still gain valuable insight. And then we can simplify the the model in several other contexts and model it in those ways. The prime insight that I obtained from the book is that if we simplify a complex system problem in the context of different forces we gain a much better understanding of a system and can gain a better understanding of its future behavior and make better decisions.

The Model Thinker then goes on to explain and give examples of many types of model types. These include but are not limited to: linear models, power models, network models, broadcast models, diffusion models contagion models, path dependent models, local interaction models, Markov models, systems dynamics models, game theory models, cooperation models, collective action models, etc. Examples of problems are analyzed with every model introduced. Of course even though the math shows the power of this processes, most of us don’t have intuition related to more complex math. The author finishes the book in a way that addresses this. Two complex system problems, opioid addiction and wealth inequality are examined with multiple models to show the power of this approach.

...more

flag Like · see review

Sep 27, 2019 Ben Hughes rated it really liked it

This is a thorough coverage of various types of models that can help us apply to problems. A strong foundation in mathematics is helpful, as most of these models are spelled out formally via math, but not necessarily required. This book advocates for, importantly, a "many-models" approach to problem analysis: many people approach problems myopically, applying the models and intuitions that are most familiar to them to a problem, convincing themselves they have it all figured out - with ...more

flag Like · see review

Sep 18, 2019 Fadri Mokolintad rated it it was amazing

How to make sense the data? Use model!

No model is bad. One model is probably good. Many models are better.

Dunia ini kian hari semakin kompleks. Dan for most of the time since cognitive revolution, hanya manusia yang bisa "mengerti" kompleksitas dunia. Manusia melakukan itu dengan membangun "model" di dalam kepalanya. Model itu mungkin cuman intuisi/feeling yang tidak ada bentuk formal matematisnya, tapi sejauh ini terbukti cukup efektif untuk mengendalikan dunia.

Sekarang, dunia yang kompleks ...more

flag Like · see review

Nov 03, 2019 JQAdams rated it liked it

In this quasi-textbook, Page introduces a wide range of elementary social-science (mostly economics and political-science, with some sociology mixed in) models, arguing that having a wide variety of potential approaches to a problem allows for richer and more robust insights. In his culminating example, for instance, he talks about where economic inequality comes from, using the models he has shown to provide a range of potential insights and explanations about inequality.

It feels like Page ...more

flag Like · see review

Mar 17, 2019 Safiya rated it really liked it

Shelves: rmtd, vuca

Data -> Information -> Knowledge -> Wisdom

Basically sums the "Model Thinking" Course by the same professor.
The style in which the book was written, makes it a perfect handbook/manual for class...
And I guess that regardless of one's background, and with enough openness, you can discern magical phenomena (obviously it's just that we ignore how stuff work), while reading this book.
Now back to the oldest idea about models as mere reductions, and are not reliable or so: Yes they are ...more

flag Like · see review

Apr 24, 2019 Adrian rated it really liked it

Well crafted with clear examples, and no excessive formulae to intimidate non-experts (the detailed formulae are boxed away or footnoted rather than in the main body of text). The book is an overview only, not diving deep into any particular model, and achieves what it intends to: Suggest the many available models and how combining them can help people plan better, to achieve better results in a complex interactive world. Such an approach is extremely useful today, when you can easily look up ...more

flag Like · see review

Jun 23, 2019 Aaron Lee rated it really liked it

Overall, the book accomplishes its goal of introducing many different models. The writing is clear, and technical concepts are explained well without superfluous detail or jargon. That said, the book would benefit from better organization of the models — grouping chapters together, or contextualizing them so readers start each chapter with an idea of how and when they might apply the model. As is, each model is explained clearly, but it is on the reader to generalize and recognize when the ...more

flag Like · see review

Apr 25, 2019 Christopher Alleman rated it it was amazing

For the thinking minds that desire clarity with complex data or social issues. I had a really great time reading this book. Having recently graduated with a Mechanical Engineering degree, I found some chapters to be trivial and some to be concepts I had never thought of or ever actually understood. The book has been beneficial during the read and I hope to reference this book as I move forward in life. This by no means goes into detail of any one model but rather adequately explains the ...more

flag Like · see review

Nov 04, 2019 Spencer rated it it was ok

Read all but the 2nd-to-last and 3rd-to-last chapters, because I ran out of time before the book was due back at the library. I did read the final chapter. I found thee book to be very dense, but interesting at times. The first 100 pages were tedious and largely review for me. There were some interesting nuggets from game theory, signaling models, random walks, and spatial vs. hedonic models. All in all, like an encyclopedia that would be best digested in 6+weeks, each chapter lulling you to ...more

flag Like · see review

Jan 05, 2019 Mitch Malone rated it really liked it

There are some useful models in here that I've been able to apply directly to my work. The overall thesis—using many models to explain, predict, and act is more helpful than using one—is something that really came into focus.

It's a very dense book with "textbook" style writing. I wish it had more narrative prose to make the reading more enjoyable. There aren't as many real world examples in the book—it's mostly math and visuals of the models themselves.

flag Like · see review

Sep 14, 2019 Mikhail Filatov rated it liked it

Изначальный посыл автора-применение нескольких моделей для сравнения и «усреднение», как улучшение в соответствии с “jury theorem” интересен. Но практического применения он не приводит за исключением нескольких общих рассуждений. В результате книга похожа на справочник, причём с непонятной аудиторией-математики много для популярной литературы, но она не достаточно строга для учебника. А примеры слабо связаны с приведёнными моделями.

flag Like · see review

Feb 18, 2019 Ganapathy rated it really liked it · review of another edition

A good reference for knowing model names. Nothing more. It is not for people without basic statistics, computer science, and math knowledge. And It is not for people who were looking for advanced models. I was not really sure who the author's intended readers are. On the plus side, the book gives a very broad view of the models used in various fields--from economics to epidemiology.

flag Like · see review

Feb 16, 2019 Denis Romanovsky rated it it was amazing

Shelves: favorites

This is a great source of wisdom. 25 something general prediction models with simple math explanation, good examples and comprehensive insights. Many model approach reasonable as nothing else in this world. Invaluable food for thought!

flag Like · see review

Dec 05, 2019 Joel Temple rated it liked it

More of a reference manual than a book to be read straight through. Had some good tidbits here and there and the last chapter was great. But I mostly found the real world examples to be the only part of this book that were intelligible.

flag Like · see review