Cities less friendly? Phooey!
A study called “Social Interaction and Urban Sprawl” came out in December and caused me to raise an eyebrow. Print and web articles reported that University of California researchers had found suburbs to be friendlier places to live than cities. For every 10 percent decrease in density, the researchers said in a news release, people are 10 percent more likely to talk to their neighbors once a week and 15 percent more likely to belong to hobby-oriented clubs.
That finding contradicts widely cited research by Harvard sociologist Robert Putnam, author of Bowling Alone — thus it made a good news story. Because the research looked only at density — not urban design — I didn’t think it had much bearing on New Urbanism. But some writers, a few of them with an ax to grind, crowed that the study undercuts arguments in favor of New Urbanism and smart growth. So I read it, with some skepticism but also with a willingness to consider inconvenient facts.
It turns out that the original claim — widely reported in the media and, as of January 3, still being presented in a press release on the UC website — is inaccurate. The study really found that a 93 percent decline in density results in only a 5 percent higher chance of talking with neighbors regularly. Rather than a small decrease in density having a significant effect, it takes a 15-fold decline to boost social activity even a little, according to the study.
But even that small affect may be illusory. The survey responses actually show no significant difference in levels of social activity at various densities, with the exception that in higher-density areas people talk to their neighbors and hang out with friends more frequently. That’s right — more frequently, not less.
Manipulation for the sake of conjecture
The authors manipulate the findings based on the conjecture that people who want more social activity choose to live in more dense places. Although unproven, that may be partly true, but even so, these people are likely seeking a particular kind of social activity that they can’t find on a cul-de-sac. So to get rid of this hypothetical bias that the authors admit is “unobservable,” they add a subjective, probably unprovable factor to their method. That leads to an absurd conclusion: If a man who has chosen to live in a city were to be picked up and moved to a low-density cul-de-sac, his social interaction would increase — notwithstanding that he may have nothing in common with his new neighbors and the social life available in the new location may be of no interest to him. The authors go so far as to suggest that, consequently, public officials might want to encourage more sprawl. As Dave Barry says, I’m not making this up.
Readers of recent news articles should be forgiven if they came to the conclusion that people in the suburbs socialize more than city dwellers — a false conclusion. The way this study has been conducted and presented says more about the researchers’ willingness to manipulate data to tease out findings that will generate buzz than it says about social life and density.
People like to live in a range of urban environments, called the Transect. They choose which part of the Transect for many reasons — to be near a job, because they like the museums or the schools, because the houses suit their needs, because they like the nightlife, and perhaps because they feel they have more in common with their neighbors in a particular part of town. The latter factor is not necessarily related to density. Some probably feel they have more in common with suburbanites, and would be inclined to live in a low-density subdivision. Others may be drawn to hipsters in an edgy downtown area. Still others feel most comfortable in an urban single-family neighborhood.
Density is a crude measure and it is unlikely that, by itself, this attribute would have much impact on social interaction. New urbanists advocate higher densities, but not to increase socializing — rather because it brings more activities within walking distance. Furthermore, new urbanists believe good design and offering a wide range of densities are crucial to quality of life in urban environments. No one has studied whether — or which — design principles facilitate social interaction at various densities. Now that might be a useful piece of research.