The Difference: How the Power of Diversity Creates Better Groups, Firms, Schools, and Societies - New Edition

Early in his career, political scientist and economist Scott Paige was messing around with a mathematical model of different groups trying to solve problems, when he noticed something: groups modeled to have smart members did only just as well - sometimes worse - as groups who had moderately smart but diverse members. What does that mean? Diversity might be equally important to group problem solving ability as the ability level of individual group members. So, Paige decided to go beyond the nice "Yay, diversity!" slogan, do some more math, and put this theory to the test.This book is the result. Section 1 explains the elements that make up diversity - that each person brings different vantage points, interpretations, heuristics (ways to solve problems), and predictive models to the table. Section 2 explores the mathematical - yes, this book proves a real challenge to people like me - proofs that diversity aids the predictive ability of groups. Section 3 discusses what happens when we switch from problem solving via diverse problem solvers to voting with diverse values and preferences (it doesn't go so well). Section 4 discusses the empirical literature showing that (or to what degree) Paige's mathematical models bear out in real life. Section 5 discusses implications for schools and firms (things like admission and hiring) as well as public policy.As others have noted, this book is heavy on the math. That's good because it puts some hard science to the intuition about diversity's benefit. But it also means that the book, at times, is a real challenge. Now, in some ways, it is obvious that the idea that diverse groups solve problems better on average than homogeneous groups; when problems are complex and have many facets, it is likely that groups will do better when different folks notice different things and approach aspects of the problem differently. But Paige puts some math formulae to this, like his own Diversity in Prediction Theorem (the squared collective error equals the squared individual error minus the diversity of the group). To put the DPT differently, diverse groups will do no worse in their averaged-together prediction as any individual in the group does on theirs. This is not a slogan, reminds Paige; it is a mathematical certainty.But not all diverse groups are working to predict something (the size of a heifer, the order of NFL draft picks, next month's computer sales). Some groups deliberate about what public (or company) policy should be, and their diversity is less in how they solve problems but in what they value (less in what next month's computer sales should be and more in how the company should try to expand its market share). In these cases, I"m afraid, Paige not only suggests that diversity of values brings costs that can easily outweigh the benefits, but has not many kind things to say about democracy. (Kenneth Arrow long ago proved that democracy often fails to aggregate preferences in a way that satisfies a majority. Others have shown that when given more than two options, people will often decide their 'votes' strategically rather than based on true preferences). Anyhow...The long and short is that Paige gives a lot of support to the idea that diversity is good, especially in helping us solve collective problems. But this means diversity of how we think, not NECESSARILY identity diversity. So, those with diverse sexual preferences, or different sexes and genders, or different ethnic backgrounds, will only make a good group to the degree that those differences actually map to differences in cognition (does being gay help this person in some ways think differently than straight people? The answer, says Paige, is "Sometimes.... only when this differences has led to different experiences that might have led different people to develop different cognitive toolboxes.) Also, while diversity is good when it means that people trying to solve the same problem come at it from different angles, it is less good - has costs that outweigh benefits - when people differ in fundamental values, such that they are not trying to solve the same problem, but disagree on what the problem is.This book is really insightful. It is very short on practical application for Paige's theoretically dense writing (section 5 is short compared to sections 2 and 3). But if you take the time and energy to get through the book attentively, it is guaranteed that you will think about the world and diversity a bit differently. Scott Paige takes a well-worn idea and slogan - diversity - and added some substantive grounding to our intuitions about it.

In support of his main claim that diversity trumps ability (page 148, and elsewhere), the author cites one real world example - Kasparov vs. 50,000 players - of a contest between ability and diversity. (page 138). But, Kasparov won.In a book with the subtitle "How diversity creates better groups, firms, schools and Societies, " one would expect at least a half dozen real world examples. Although he alludes to a few real world examples of group problem solving (Bletchley Park, DNA and Watson-Crick-Wilkins-Franklin) it's not clear that he claims the successful results are the result of diversity. With regard to the discovery of DNA, credit goes to four intelligent, well educated scientists - hardly a cross section of society. (I've just started Brenda Maddox's book on Franklin, and she is no ordinary person)The author does present some computer models, and some theoretical conjectures in support of his claim, but he does not support them with real world facts. He does begin to evidence some understanding that diversity has at least two dimensions - identity and cognitive, but I think he confuses informational ability with intellectual ability. Most chemists, physicists, engineers, biologists, and mathematicians share a high level of intellectual ability, but differ, because of their education and experience, in informational ability. It is no surprise, contrary to his statement on page 158, that a group of very able people with different informational ability are necessary to and often sufficient for the solution of difficult problems. IBM's development of the moving head hard disk drive in San Jose in 1956 was the result of a group of mechanical and electrical engineers, chemists, and physicists who were all extremely bright, but differed in their areas of knowledge. Similarly for the development of Telstar at Bell Telephone Laboratories in 1961. Not only is the need for diverse informational ability unsurprising in such endeavors, it is astounding that anyone would find it surprising.On page 159, the author says that if we have a difficult, unsolved math problem, we would want to ask a diverse collection of mathematicians. He has a point, in that providing a network to link the problem with potential solvers is a powerful idea, and somewhere in his book he makes reference to such a network. But it is individual ability that solves the problem, and tying in diversity is unnecessary. Is Shouryya Ray diverse?There are other problems where a high ability and a modicum of informational diversity is necessary and sufficient. Development of the theory of relativity, polio vaccine, an online bookstore, an online payments system. Most real world examples show one or two very capable individuals (two or three heads are better than one, if they are the right heads) creating breakthroughs of significance. Edison; Bell; Wright Brothers; Salk; Hewlett and Packard; Gates, Ballmer and Allen; Sergey Brin and Larry Page.Dr. Scott's book will provide comfort to the politically correct whose lives are ruled by emotion and belief, but will fail to pass muster with realists who ask for proof.I'll put my money with Damon Runyon, and bet on the wise and discerning of great ability.

A qualitative description of how and when diverse sets of people can be more effective than groups of people who are less diverse. This book easily presents conclusions based on solid logic and maths, but doesn't get bogged down in detail. It remains high-level and therefore very accessible. I really recommend it for the unique perspective it presents on diversity. If anyone's interested the author also published an awesome course on model-based thinking on Coursera ([...]) The course covers a toolkit of ideas and approaches that can be used to arrive at the startling conclusions in the book. Understanding model-based thinking will help to understand the ideas presented in the book, although the content of the course and the book aren't directly linked.

Scott E Page has made some real insightful discoveries about how important diversity really is for teams and how diversity is not caused by race, gender or culture but correlates with it.

レディガガは何で居るんでしょうか、変な奴は不要でしょうか、戦争って本当に罪でしょうか、少数意見は意味の無い戯言でしょうか、組織に沿わない人材は馘首されて当然でしょうか、混乱とは無意味でしょうか、気に入らない奴は自分にとって本当に不要でしょうか、流血とはムダでしょうか、政党や派閥はムダでしょうか？日本語でに翻訳されておりますんで、何か疑問を感じている方なら一読をオススメします。でも、翻訳された本は原書とは雰囲気が違って偉そうな装丁で嫌いです。以上。