The Difference -- a book by Scott Page
The Difference by Scott Page is one of the best management books I’ve read.
Diverse perspectives: people have different ways of representing situations and problems; they who see or envision the set of possibilities confronting them differently.
Diverse interpretations: people put things into different categories and classifications. To some people, I might be someone from the World Bank. To others, I might be an organizational storyteller. To others, I might be an author. All are true. They are different interpretations of the world.
Diverse heuristics: People have different ways of generating solutions to problems. Some people like to talk through their thinking about problems; others prefer to write out his solutions first and then talk
Diverse Predictive Models: Some people analyze the situation. Others may look for the story)
Does diversity lead to better outcomes? In this wonderful book, Scott Page: The Difference: How the Power of Diversity Creates Better Groups, Firms, Schools and Societies, (Princeton, 2007) Scott Page provides a definitive answer. He shows in great detail and considerable intellectual rigor when diversity does lead to better outcomes and how and why, as well as when it doesn’t. His short answer is that in some circumstances diversity doesn’t lead to better outcomes:
“… if a loved one requires open-heart surgery, we do not want a collection of butchers, bakers and candlestick makers carving open the chest cavity. We’d much prefer a trained heart surgeon, and for good reason.”
But in other circumstances, particularly difficult problems, such as constructing a welfare policy, cracking a secret code or evaluating post-heart attack treatment, diversity not only merits equal standing with ability.
Page goes beyond the conventional wisdom that other things being equal, diversity trumps like-mindedness. Page makes the startling claim that diversity often trumps ability. In some situations, a group of ordinary people who are diverse can defeat a group of like-minded experts. Page backs up his claim with detailed arguments and evidence.
The prediction markets also show the power of diversity in their ability to make better predictions about the outcome of presidential elections than the experts.
When are two heads better than one? When do too many cooks spoil the broth? Clearly all great restaurants have many cooks. So having lots of cooks won’t spoil the broth, if they are all following the same recipes. The chaos comes if they start to follow different recipes at the same time. In fact, having lots of cooks is essential to running a great restaurant. And when it comes to solving a difficult problem, like coming up with a better way to make coq au vin, having cooks with different points of view will usually help.
One of the useful things Page does in The Difference is to unpack the notion of diversity. He focuses on cognitive differences between people, not identity differences like race, gender, ethnicity or religion. He suggests that cognitive diversity has four dimensions: perspectives, interpretations, heuristics, predictive models.
This enables Page to explore exactly how these kinds of diversity might help to solve difficult problems or make better predictions.
A second thing that struck me about the book is the simile that he uses in comparing the solving of problems with climbing rugged landscapes. If our object is to climb as high as possible, our chances of accomplishing that depend on which mountain we decide to climb. If we climb a local hill, we might consider yourself doing well, because we have never the Rockies, let alone knowing about Mount Everest. And while we’re climbing up one mountain, often we can’t really see how high it is, or how it compares with other neighboring mountains, until we’ve already climbed it. Climbing the highest mountain may entail descending the mountain we are on, and moving to a completely different mountain range.
I find that this image of what it’s like to solve a difficult problem is illuminating in showing how and why having people with different perspectives might enable a group of diverse people to do better than a group of like-minded experts who think they know they are on climbing the highest mountain.
A third virtue of the book is his summary of the evidence as to whether diversity leads to benefits, including comparisons of cognitive diversity and identity diversity. Cognitive diversity doesn’t improve performance when it comes to routine tasks, like flipping burgers. But when we are dealing with complex tasks like engineering problems, or tasks requiring creativity and innovation, or managerial issues, cognitive diversity is a key explanatory variable in levels of performance.
By comparison, the impact of identity diversity is mixed. One part of this is due to the fact that routine tasks are better done by individuals. A second part of it is due to the fact that identity diversity doesn’t necessarily lead to cognitive diversity. The whole idea of medical training, for instance, is to get medical students thinking alike, i.e. like doctors. It shouldn’t be surprising, then, that doctors who are diverse in identity terms are cognitively alike, and hence may do no better than doctors who are not diverse in identity terms. A third part is due to the fact that getting the benefits of diversity depends people being able to work together. We would expect that some people who are diverse in identity terms find it difficult to work together effectively.
The fourth good quality of the book is when Page goes on the offensive and addresses the question of: so what? Given what we have learned, what should we do differently? Page points out that diversity offers not merely the advantage of a diverse stock portfolio where different stocks do better in different conditions, adding up to an overall average that does reasonably in all conditions. Diversity in teams offers what he calls superadditivity. When a collection of people work together, and one person makes an improvement, the others can often improve on this new solution even further: improvements build on improvements. Diverse perspectives and iverse hueruistics apply sequentially: one gets applied after the other and in combination. As a result, one plus one often exceeds two.
What does this imply? Page has several suggestions that bear on the issue of creating high-performance teams:
Bring in outsiders with different, relevant perspectives. But be careful! Outsiders don’t stay outsiders for long. If outsiders become insiders, they will cease to think differently. And be careful of brining in “highly paid consultants in fancy suits to add credibility to decisions that directors have already made—‘Look, McKinsey agrees with me!’” And the diversity must be relevant to the task at hand: you don’t ask villagers from Papua New Guinea to advise on the implementation of Sarbanes-Oxley.
Encourage inter-disciplinary efforts: When faced with difficult problems, requiring innovation and creativity, the advantages of having cognitively diverse people working on them are overwhelming.
Diverse preferences can be beneficial: If we agree on the goal then disagreements about different ways to reach the goal can be helpful in expanding the array of solutions, if we we agree on the ultimate goal. But diversity in terms of fundamental preferences can also help. Although solving problems of fundamental differences will often require compromise, diversity in terms of fundamental differences may lead to improvements: Gwen and Tess may disagree on goals, but if Gwen and Tess are cognitively diverse, Gwen may find a solution that Tess improves on, which they both like better.
Diversity needs to be a factor in recruiting: If the work is mainly done by individuals or is routine, cognitive diversity is unlikely to lead to improved performance, although it might be pursued for other reasons. But where people have to work together on difficult problems, cognitive diversity should be very important in hiring. Page praises Google for trying to hire people with diverse interests and skills while also requiring that the recruits have basics skills in fields relevant to Google, i.e. computer science and mathematics.
Recruiters should assess the cognitive aspects of diversity: Identity diversity correlates to a certain degree with cognitive diversity. Since it is easier to assess identity diversity, that may be a first rough approximation of cognitive diversity. But it is also possible to test for cognitive diversity directly, and Page encourages firms to do so.
Overall, this is a terrific book—one of the best management books I’ve ever read. It takes a complex subject, moves beyond metaphor and mysticism and politics and places the claims of diversity’s benefits on a solid intellectual foundation. Using precise definitions, rigorous analysis and clear conclusions, Page tells you everything you need to know about this subject. His book is well-written and has many interesting apercus and examples, although, given the effort to be rigorously, it’s not always an easy read. Yet it’s a book that tries hard to make us think clearly and what more can we ask than that?
It’s interesting to compare The Difference to a long-running best-selling business book like
From Good to Great by Jim Collins. From Good to Great has unclear definitions, suspect methodology, and dubious conclusions.1/ It is not particularly well-written. Yet its catchy title enables it to keep on selling tens of thousands of copies, year after year. Even the obvious fall from “greatness” of half of its supposedly great companies has no impact on the book’s popularity: it just goes on selling. Apparently, its mythology of greatness is a siren call that hard-pressed managers are unable to resist. The Difference is an infinitely better book, but most people have never heard of it.
1./ The Halo Effect, and the Eight Other Business Delusions That Deceive Managers
by Phil Rosenzweig (Free Press, 2007)