Debunking Connections Between Urbanism and Alienation

In response to a recent essay about an apparent relationship between urbanism and social alienation, Robert Steuteville argues that the study in question -- and its press -- twists the facts.

 Robert SteutevilleThey say a lie can travel around the world in the time it takes for the truth to get its boots on. Richard Carson's "Bowling Alone in Urbanistaville" recently published on the website ArchNewsNow repeats false statements about a study that compares social activity in the suburbs and the city. "The statistical revelation behind all of these findings is that for every 10% increase in density, there is a 10% decrease in socialization. That's a simple, one-to-one inverse relationship that everyone can understand," he writes.

That's not even close to being true - as Carson would know if he read the study. "Social Interaction and Urban Sprawl," by Jan Brueckner and Ann Largey, finds that people talk to their neighbors and hang out with friends more in cities than in suburbs on a per capita basis. They found no statistically significant difference in other observed social activity, such as the number of friends and confidants, based on density.

To fully explore how that simple truth got twisted 180 degrees, we have to look more deeply at the study, its methods, its presentation, and how the authors publicized the findings.

The 'Unobservable' Tendency

Brueckner and Largey use an instrumental-variables (IV) estimation to put numbers on what they call an "unobservable propensity," the tendency of gregarious people to locate in cities. IV estimation is used to correct for what the authors assume is the self-selection bias of people who want to live close to their neighbors.

How do the authors know that something that they can't detect actually exists and can be measured? They don't. They assume that a normal person would seek to maximize lot size - that is, locate in a suburban place - all things being equal. Gregarious people, they say, like living closer to their neighbors and are willing to sacrifice lot size in proportion to their desire to be social.

In the authors' view, this would create higher social activity in cities, and so the authors sought to take out the bias - that is, weight the results according to the tendency of outgoing people to live in denser environments. The entire credibility of the study depends upon that assumption, which, as they say, is based on something that is "unobservable."

Without the IV estimation, the study results are, as I have stated earlier, slightly in favor of cities. With the IV estimation, Brueckner and Largey find an advantage for suburbs - but you have to go from very high density (11,591 people per square mile) to very low-density (779 people per square mile) to see any appreciable effect. This 93 percent drop in density increases the chance of talking with neighbors by 5 percent, hanging out with friends by 7 percent, and having friends over to your house by 13 percent. Why use such an extreme drop in density? That's because a less dramatic drop - say, 10 or 20 percent - yields numbers that are not statistically significant.

Mixed-Up Message

So much for the study and its methods, but its presentation and publicity is even stranger. Brueckner and Largey bury the straightforward, non-IV results of their study. They weren't in the press release or the introduction. Rather, you have to read deep into a paper full of barely readable jargon, comparing statements with tables not designed to enhance clarity. Here's a sample of the authors' prose: "As can be seen, the density coefficients are insignificant in the non-IV case in the NEISOC, CONFIDE, FRIENDS, FRNDHOM equations, becoming significantly negative when IV estimation is used. By contrast, the non-IV coefficients are significantly positive in the NEITALK and FRNDHNG equations, while IV estimation leads to coefficients that are negative and insignificant." These acronyms refer to social activity measures, and what the two sentences mean is that the assumption that gregarious people live in cities reverses the observed findings of the study. The authors then claim that the findings support the assumption - an example of circular logic if I ever saw one.

If the writing and logic were bad, the publicity was even worse. University of California Irvine, where Brueckner is an economics professor, reported in a press release the figures that Carson repeated - that a 10 percent decrease in density yields a 10 percent increase in social activity. Brueckner told me in an email dated January 1, 2007, more than a year and a half ago, that the error was a miscalculation made under deadline pressure, but that nothing could be done to correct it. The false press release figures remained on the UC Irvine website until July of this year, when I complained directly to the publicity department after seeing Carson's essay.

This all adds up to a pattern of mendacity - perhaps by a professor with a pro-sprawl agenda, perhaps by one seeking publicity, perhaps both - yet the twisted and exaggerated results of this dubious study have been reported by various news organizations that were more willing to pull material off of a press release than to dig deep into gobbledygook.

The Choices of the Gregarious

All of that only scratches the surface of what's wrong with this study. The idea that complex urban development patterns and human behavior can be meaningfully studied according to one primary criteria - density - is wrong from the start. There's good density and bad density from the standpoint of how humans act and interact. Consider a walkable, mixed-use, but primarily single-family urban neighborhood compared to an apartment complex surrounded by surface parking off of an eight-lane arterial. The latter is higher density but the former is more likely to attract people who want to get to know their neighbors. Urban public spaces can be designed to be inviting, comfortable, and even inspiring - an approach that helps to bring diverse people out of the private realm into contact with one another.

People choose different environments - some dense, some less dense - at various time in their lives as their needs change, for reasons too numerous to mention. It doesn't mean that they have become more or less gregarious.

Brueckner and Largey conclude that someone who is looking for a richer social life would do better to move to a cul-de-sac on a one- to two-acre lot. That cul-de-sac is populated by introverts, according to their own formula. That's a parody of social science at its most ridiculous.

But it works for Carson, whose favored term for urban planners, "urbanistas," is a takeoff on Sandinistas, a cheap attempt at red-baiting. A true-blue defender of sprawl, Carson is willing to take any nonsense and run with it.

Robert Steuteville is editor, publisher, and founder of New Urban News, the newsletter covering the New Urbanism and smart growth. He is president of New Urban News Publications - providing news, ideas, and connections for New Urbanism - which has played a leading role in this trend for more than a decade.



Josh Stephens's picture

Right on

"Urban public spaces can be designed to be inviting, comfortable, and even inspiring — an approach that helps to bring diverse people out of the private realm into contact with one another."

This is so undeniably true. And i love seeing marginal scholarship debunked. Or, even if it's not marginal, it's certainly exaggerated.

The study seems so unrealistic, so mired in academic jargon, and so reliant on misguided quantitative analysis that there is no way it could ever account for such an obvious, elegant notion of urbanism. While the authors were out crunching numbers, people in well designed places were out having a grand old time. In short, the fact that anyone would feel the need to study this sort of thing -- to debunk a bona fide truism -- almost proves its absurdity prima facie.

Having said that, I firmly believe that the suburbs can be social (and that the city can be anti-social), in part because people go out of their way to have Tupperware parties and such. But there's an insidious element of both exclusivity and conformity in the suburbs that does not exist in the city. Cities are where opportunities -- social, economic, culture, etc. -- reside. Suburbs are where you go if you just want to be content with having what you have and knowing whom you know.

Urbanism And Alienation

Thanks for clarifying this. I didn't realize that this study systematically distorted its findings by assuming that there is an "unobservable propensity" of gregarious people to locate in cities.

It would make just as much sense to assume the opposite. We could assume that people to the suburbs to find closer communities, correct for this "unobservable propensity," and conclude that lower densities prevent people from finding social contacts.

Charles Siegel

Only twisted if you don't understand

I only just found this essay but want to state for the record that it is not only weirdly ill-spirited, it dramatically mischaracterizes both the methods and results of the Brueckner/Largey study.

Steuteville concludes that the IV technique is a mendacious effort to bias the results in the wrong direction for some dark (i.e., pro-sprawl) purpose. Actually, it is the standard statistical strategy for testing for a lack of randomness in the underlying data, without which the results lack any credibility. Steuteville has it exactly backwards in the sense that the technique actually tests for and removes bias.

The clearest related example of this in this literature may be the association between weight and suburbanization. Heavier people seem to live in the suburbs. What we don't know is whether the suburbs make people heavy or whether heavier people prefer the suburbs, or some of both or neither. The standard statistical technique for sorting this out with cross-sectional data is IV estimation, which explicitly tests for these possibilities, or what is known as sampling bias. It then eliminates that bias. IV estimation does not assume that more gregarious people prefer higher densities -- Steuteville's main misstatement -- it merely tests for that bias, and controls for it if found.

He may like the non-IV results better, but they are wrong. They neither test for nor eliminate any possible sample bias, which is to say they are not credible to any established researcher in this field. They were reported in the paper, per standard practice, only to show what difference such tests make. If the IV and non-IV results were the same, this would indicate there was no self-selection bias in the sample, another point that Steuteville misunderstands.

The paper clearly is trying to do nothing more than get a leg up on this question, using the best data and methods available. It does not suggest it is the last word on the matter, or strangely question the motives of anyone reaching different conclusions, and invites anyone to do better with other data or better methods.

The result that higher densities are associated with less social interaction by some measures was surprising and thus reported as such, even if some of the details of the calculations were confusing in the presentation. The key result is the direction of the association. We could reasonably ask if there are problems with the data, or with the methods, or the choice of statistical instruments, or if these results make any sense on their face, but this essay is so uninformed as to be seriously unhelpful and deceptive.

To repeat, the study in question is neither misleading nor incorrect. It does not, for example, assume that more gregarious people live in cities and that suburbs are populated by introverts. It is a straightforward work using straightforward, if possibly unclear to some, methods that invites further examination and debate.

Randall Crane, UCLA

Statistical analysis is the wrong approach

Whether the researchers used proper statistical techniques or not, it seems like the only real conclusion of the study is that you can't run a regression to figure out people interact in space. When it comes to a subject as complicated as urbanism, the only meaningful studies come from direct observation.

Stats are wrong approach only if using the wrong approach.

But you have to have enough direct observations to determine whether they apply widely. And you then analyze these observations to determine that. Statistics is a tool to analyze the observations. Basic stuff.



Bowling Alone

The truth is that I quit writing commentary -- especially for Planetizen. There is a major bias against any planner who is a contrarian. And I am a contrarian. I think there must be something more productive for me to do in life than point out the obvious.

Prepare for the AICP* Exam

Join the thousands of students who have utilized the Planetizen AICP* Exam Preparation Class to prepare for the American Planning Association's AICP* exam.
Starting at $245

Essential Readings in Urban Planning

Planning on taking the AICP* Exam? Register for Planetizen's AICP * Exam Preparation Course to save $25.
Planetizen Courses image ad

Looking for the perfect last-minute gift?

A one-year subscription to Planetizen Courses is the gift that will continue giving all year long.
Rome grey gold tie

Tie one on to celebrate your city!

Choose from over 20 styles imprinted with detailed city or transit maps.