Their simulations of more than 20 million virtual “neighborhoods” demonstrate a troubling paradox: that community and diversity may be fundamentally incompatible goals. As the authors explain, integration “provides opportunities for intergroup contact that are necessary to promote respect for diversity, but may prevent the formation of dense interpersonal networks that are necessary to promote sense of community.”

Urbanists and planners like to imagine and design for a world of diversity. Diversity, we like to think, is both a social good and, as I’ve argued, a spur to innovation and economic growth.

But to what degree is this goal of diverse, cohesive community attainable, even in theory?

That’s the key question behind an intriguing new study, “The (In)compatibility of Diversity and Sense of Community,” published in the November edition of the American Journal of Community Psychology. The study, by sociologist Zachary Neal and psychologist Jennifer Watling Neal, both of Michigan State University (full disclosure: I was an external member of the former’s dissertation committee), develops a nifty agent-based computer model to test this question.

Their simulations of more than 20 million virtual “neighborhoods” demonstrate a troubling paradox: that community and diversity may be fundamentally incompatible goals. As the authors explain, integration “provides opportunities for intergroup contact that are necessary to promote respect for diversity, but may prevent the formation of dense interpersonal networks that are necessary to promote sense of community.”

Their models focus on the emergence of the “community-diversity dialectic” based on two simple principles: homophily – the tendency of people to bond with others like themselves – and proximity – the tendency of people to bond with those nearby. Their models look at how the strength of these basic tendencies affect the evolution of neighborhoods comprised of two distinct populations (say by race, class, ethnicity and so on). In these simulated neighborhoods, the possible levels of integration ranged from 0 percent (totally segregated) to 50 percent (totally integrated).

In the images below, the authors show three sample neighborhoods with low, medium, and high levels of integration. Notice how much denser the resulting social networks are (in the bottom row) in the highly segregated neighborhood at far left. (You can also play around with an interactive version of the model on Neal’s website, adjusting the levels of homophily, proximity, and integration yourself).

Read More.

Read more on this study: Study Asks: Is A ‘Better World’ Possible?