Eli Reed / Magnum Photos

'Collective Action Can Make a Difference': A Conversation with Robert Sampson

Elizabeth Glazer

March 28, 2023

On disadvantage, race and neighborhood vitality

On disadvantage, race and neighborhood vitality

Vital City · Robert Sampson in His Own Words

To read Robert Sampson is to have the curtain pulled back on the gears that make our neighborhoods whir and go — and sometimes stick in tragic dimensions; on the lives of our neighbors and on the connections, thick and mainly thin, that bind us together. He is a sharp observer of the social currents of history that shape our streets and ourselves — and how all of these things come together and inhabit our everyday.

Sampson is a Harvard University sociologist who has spent the better part of his life on the great questions of urban life: violence, poverty, race, crime and equity. His answers have helped shape our understanding of how society works. Coining the term “collective efficacy,” Sampson brought rigor to Jane Jacobs’ insight about “eyes on the street,” showing how the willingness or unwillingness of neighbors to intervene on behalf of one another is a strong indicator of the level of violence in a community. He has revealed how “sticky” the character of a place is, enduring from generation to generation despite the flows of people, and indeed imprinting itself on its residents — the “neighborhood effect.”

In meticulous detail, he has demonstrated how disadvantage concentrates in a few places, with deep impacts especially in the lives of Black and Latino Americans, including literally entering their bodies: for example, the disproportionate levels of lead in children and the attendant effects on learning and achievement.

All of this combines to create the conditions in which violence can take root. The effects of violence, in turn, reverberate through generations. For example, high levels of violence have a direct effect on school achievement. 

This distress is not just stuck in place: a neighborhood’s well-being is not simply about its own conditions but also affected by the neighborhoods its residents visit and are visited by. This “triple disadvantage” is a highly distilled indicator of high homicide rates.

A theorist of cities, Sampson develops his ideas through a deep commitment to empiricism. If the ideas about how society works sound inchoate, he has developed a specific way to measure them: ecometrics. In a massive longitudinal study on Chicago neighborhoods, he interviewed 9,000 residents in 300 neighborhoods, videotaping the physical conditions of 22,000 blocks, and he has been following the progress of a cohort of people from those neighborhoods for 40 years. In his most recent work on “triply disadvantaged” neighborhoods, he analyzed data from nearly 32,000 neighborhoods and 9,700 homicides in the nation’s 37 largest cities.

A man of enormous personal modesty, the honors bestowed on him from around the world are anything but; they include the Stockholm Prize, considered the Nobel of criminology. He has written widely about his work, including his capstone book, “Great American City.”

He sat down with Vital City last month to talk about his work and the life of cities. The conversation has been edited for length and clarity.

Credit: Arthur Mount. Portrait of Robert Sampson.

Elizabeth Glazer: Your theoretical work is backed by such deep empirical work. The most famous is perhaps your Chicago study that started in the 1990s, at the peak of homicides, and is still going on.

Robert Sampson: That’s the Project on Human Development in Chicago Neighborhoods. Now, why neighborhoods? Spatial inequality has long been a key feature of cities, actually going back to ancient cities. It’s a fundamental feature of urban life. Inequality is persistent, and it varies across multiple dimensions. It’s not just poverty, which we might think of initially, or racial segregation, but also crime, violence, housing quality, environmental toxicity. Our goal was to measure these and other social characteristics of neighborhoods at the same time that we were studying the kids growing up in them. It was really this dual focus: individual development, while taking seriously the neighborhood context.

We tried to create a science of assessing community. Through surveys, we interviewed residents and asked them about their neighborhoods. We paired that with two other kinds of data collection. One was systematic social observation, where we literally videotaped every street. We wanted to “see” the city and measure it in a scientific sense so we could compare that with what people tell us about, say, disorder and garbage in the streets. The other was a key informant study, where we talked to community leaders. These three tools were the empirical backdrop of what we called community “ecometrics.”

The theoretical framing was what sociologists call social organization and, in particular, informal social controls. The notion is that peaceful communities are peaceful not because there are cops on every corner, but because of everyday interactions and relationships. We felt that we could measure them systematically. If you’re going to take part in a community activity — maybe a community cleanup or work at the local school, or maybe saying something to local kids who are causing trouble in the community — that’s usually something that takes place under conditions of mutual trust.

What we did was think about the coalescence of cohesion and control in communities of trust, though not necessarily a deep-rooted trust. It’s akin to the idea of weak ties as opposed to close friendships. It’s trust and cohesion combined with the willingness of residents to intervene in the community for the public good. That’s what we were getting at with the term “collective efficacy.”

We asked people, for example, what would your neighbors do if kids were skipping school and hanging out in the street and causing trouble? What if the local fire station were about to be closed down? Would your neighbors get together to do something about it? Do you trust your neighbors?

We created a scale of collective efficacy from such items, which was highly reliable and predicted lower rates of crime and violence even after controlling for poverty, prior crime and other commonly studied factors. We published a paper on that back in the ’90s, and there has been a lot of work since then testing the theory. Indeed, communities that are poor or burdened by, let’s say, high rates of past violence have lower rates of collective efficacy.

We still see that a lower rate of collective efficacy is a very strong predictor of crime, but it has also been shown to relate to other aspects of human well-being and health.

EG: So what builds collective efficacy? What manufactures it? It seems like a lot of times, it’s organic. Neighbors get together because there’s a shooting on the block. But is there something more durable that can be done to build those loose connections that make for collective efficacy?

RS: It’s a great question. And I think you put your finger on it two ways. One is the trigger of a larger challenge. It could be an earthquake, it could be a shooting, and sometimes people rise to the occasion to take action. Just like Black Lives Matter did. But collective capacity also comes from rather prosaic, everyday interactions and social ties.

Relationships are built out of everyday activities, not so much with the purpose of building something broader. Let’s say you have a bake sale for the PTA or the kids’ basketball team or for a church or whatnot. To the extent that people are engaged in shared tasks, a certain kind of cohesiveness is built.

I lived in New York for several years, and I’m sure you’ve had the experience where you go by the same person at a newsstand or coffee shop regularly, with nodding relationships. There’s a sense of who’s a regular and who’s not, even in a dense urban environment. That’s another kind of collectiveness. Eyes on the street, as Jane Jacobs would have put it.

What, then, would be a structural predictor of those kinds of activities and shared sense of community? Well, the longer you’ve lived in a neighborhood, the more likely you are to engage in local activities. Economic resources matter too, to the extent that people have time and the ability to engage in these events. Homeownership is another structural predictor.

I have tried to build a theory that says what you want to do is stabilize and invest in communities so that you provide the economic and housing and resource opportunities, but also so that you provide or incentivize shared tasks.

EG: Those connections potentially create a virtuous cycle, right? So neighbors can mobilize to get more resources from government deployed in their neighborhoods, and make them better.

RS: Absolutely. And that’s one of the reasons we asked about the fire station closing, or the school closing, or the park. The extent to which people can engage — that, I think, is really important. And that’s what we’ve tried to tap into.

EG: You’re the godfather of the analysis of neighborhood disadvantage and how it clusters. Your most recent work is around what you call triply disadvantaged neighborhoods. What are they, and what’s the connection to violence?

RS: I’ve challenged myself on the idea of neighborhood because there’s a tendency to view them as islands or isolated units. But we don’t live just in our neighborhoods. We go beyond to work, we visit friends, we go to sports events, we visit family, we go to restaurants or a museum. Neighborhoods have flows of people coming in and going out all the time. The idea was to be able to look at how they intersect to affect the character of the neighborhood.

For example, most of the work on racial segregation up to a few years ago was based on the isolation of the neighborhood based on its racial characteristics relative to the rest of the metropolitan area. What we did was measure the movement of people from a given neighborhood to all other neighborhoods in the metropolis and then visits into the neighborhood. From that, we created new measures of racially segregated mobility in the 50 largest American cities.

We also conceptualized neighborhood disadvantage in the following way: We measured the poverty of a neighborhood, and then we took all the visits into that neighborhood from other neighborhoods around the city, and the poverty rates of those neighborhoods, so we could get at what network theorists call the in-degree, or inflow, of disadvantage. And then we measured where residents were going and the poverty rates of those visited neighborhoods. That’s the out-degree of disadvantage. 

The combination of all three we called “triple disadvantage” — it’s a neighborhood that’s poor, its residents visit other poor neighborhoods and the residents flowing into the neighborhood are from other poor neighborhoods. What we found is that triple disadvantage was an important predictor of homicide.

The flip side allowed us to see how wealth and advantage are also highly clustered in their mobility-based contacts. There is a certain kind of isolation in triply advantaged areas that’s not just geographic.

open in new tab
Download: Data, Chart Image
open in new tab
Download: Data, Chart Image

EG: And what makes that correlation between triply disadvantaged neighborhoods and homicides so strong?

RS: Take a neighborhood that’s poor but where you also have people coming in from other poverty-stricken areas, many of whom are strangers. It’s increasing the risk of friction, especially since interpersonal interactions among strangers in public spaces in poor neighborhoods can be fraught with danger. Especially if there’s a high rate of gun ownership or trafficking based on visits, or there’s a higher fear that gun violence will take place, or public drug transactions — then people may be more willing to take preemptive actions that can lead to violence.

For these reasons, neighborhoods that have the highest triple disadvantage tend to have more high-risk situations and illicit markets that tend to trigger incidents of violence and ultimately homicides. The city might have a better chance of reducing crime by putting more resources in those neighborhoods. It’s a different kind of metric that’s super important, because if we can’t measure something, for the most part, it’s hard to really understand it or demonstrate effectiveness of any interventions.

We tend to underestimate just how persistent and sticky some of these historical legacies of inequality are.

That’s why I pushed for many years on the idea of ecometrics and the social ecology of measuring disorder, measuring collective efficacy and, in this case, measuring things like mobility-based segregation and neighborhood triple disadvantage. And what we’ve tried to do is to put the metrics out there, put the data out there for the public to use.

EG: I once heard you say something like: Trying to measure the effect of concentrated disadvantage among whites is tantamount to estimating a phantom reality.

RS: Yes. We already know that African Americans tend to live in neighborhoods that are much poorer on average than white Americans — something like 10 times more likely for percent living in poverty. But when we looked at mobility-based poverty, it was something like 30 times more likely. In other words, triple neighborhood disadvantage is exacerbating racial inequality.

I think we tend to underestimate just how persistent and sticky some of these historical legacies of inequality are. That’s in part why we see neighborhoods that decade after decade reside in basically the same position in the overall hierarchy of resources and rates of crime and violence. Our work has tried to always be cognizant of that in terms of trying to look at both the historical and long-term effects while focusing on present conditions.

Inequalities are socially constructed, which means that they can be deconstructed.

EG: Reading your work, I can come away a little depressed because of the longstanding nature, the structural nature, the “where do we even start?” aspect. Yet you actually seem to be an optimist. Would you have advice for policymakers on how to square what seems to be the immutable characteristic of neighborhoods through decades with how to improve conditions for people?

RS: It’s a paradox. Because you’re right, there’s a lot of persistent inequality. What’s important to understand both theoretically and for policymakers, however, is that it’s not inherent that we have these kinds of inequalities. They are socially constructed, which means that they can be deconstructed. So yes, we have a history and legacies of inequality, but that doesn’t mean nothing can be done.

That’s why I always try to work back and forth between structural and policy types of efforts that have to do with housing, the environment, economic resources and law on the one hand, and citizen-based collective efficacy on the other.

Let’s take crime and violence. Crime did come down a lot from its peak in the 1990s. Evidence suggests that community-based organizations were responsible for a good chunk of the decline in crime rates.

Lead exposure is another pathway that has contributed to not just racial inequality but all kinds of inequalities among children, such as cognitive deficits and behavioral problems. When I first saw the map of the distribution of lead poisoning among kids in Chicago, I almost fell out of my chair because it was so stark, so concentrated, so racially segregated — more so than poverty.

But the positive story is that the public health department in Chicago was incredibly aggressive about forcing landlords to remove lead paint in housing. And the lead levels just plummeted over time. It’s still true that poor neighborhoods and African American neighborhoods still have higher levels, but levels are so much lower for everybody, that has to be taken as good news. It’s not a perfect world, but it has improved in this way.

EG: Can we talk about COVID-19 for a minute, because so much of your work takes into account these kinds of enormous social changes. Do you think it frayed ties, or maybe bound us together more?

RS: COVID-19 unleashed something we don’t fully understand yet. A lot of things went awry. At first, there was a sense of pulling together. But over time, we saw social isolation: reduced contact with others, kids not going to school, people not interacting with one another, all leading to a destabilization of society.

I don’t think there was necessarily a direct line to crime. There’s some good evidence that there were mental health declines, and kids weren’t doing as well. Gun purchases were going up. Then there was economic hardship in the sense of the shutting down of society.

Car accidents were coming down at a rapid rate, and then all of a sudden started shooting up. There were more aggressive incidents on streets, more hit-and-runs. It’s almost as if self-interest took the front seat at a certain point in the epidemic. Even though, initially, it was the opposite. I hope we get out of it soon.

We also know it’s not just COVID-related. It goes back pre-COVID-19, so-called deaths of despair in terms of not just mental health, but drug and alcohol addiction associated with increased mortality.

That’s why I think it’s so important to keep our attention on crime, violence and safety because they have such profound effects on people’s well-being, on their fear, on their learning, on their cognitive abilities and outlooks on life. And those kinds of changes and the intersection with when kids are coming of age has longstanding consequences.

EG: That is a theory of everything.

RS: Yeah. I know I tend to get trapped by these sorts of big questions.

EG: It seems like there are a lot of stressors right now. The polar caps are melting. There’s a possible world war …

RS: Yes. All true. But again, we have agency. I think that one of the things that Black Lives Matter taught us is that pushing back and collective action can make a difference, which we also saw in the civil rights movement. There are lots of examples of successful action. That’s why it’s important not to be too pessimistic. But at the same time, it’s important to get the facts straight and not shy away from pointing out deep-rooted inequalities and persistent inequalities.

Sometimes the weight of those trends leads to a stasis, when really, we should always be pushing forward.

The city needs to publish not just crime statistics, but statistics on positive characteristics of community and social interactions.

EG: We’ve started at Vital City a project called Vital Signs. The initial iteration is just to show on a very regular and very accessible basis what the criminal justice system is doing. But the project is not just about the criminal justice system; it’s about what makes the city tick. I was wondering if you had any measures that you wish were collected?

RS: I love the idea of Vital Signs. It goes to the notion that people care deeply about crime, but that’s not all they care about. And cities tend to emphasize danger all the time. Of course, that’s important, but the unintended consequences are often that people become fearful. One of my recommendations would be that the city needs to publish not just crime statistics, but statistics on positive characteristics of community and community organizations and social interactions.

I like the idea of an ecometrics for the city that takes into account a broad range of social and organizational characteristics. Like how many times people have participated, let’s say, in the local school or in a community-based organization. How many times they visited the park or the zoo. How many times they have gone out to eat in another neighborhood. How many times they visited friends in another borough. How many times they have used public transportation. What’s the level of trust that people have in their neighborhoods? And by the way, it’s higher than we think.

I like asking about what people feel is going right in the community or in the city, along with ways to improve. Increasing people’s stake in the community is really about the participation and inclusiveness of citizens and increasing the sense of ownership, which enhances the collective willingness of people to intervene.

Publishing these kinds of statistics is not a magic wand, but nonetheless, it’s certainly better in my view than publishing only one kind of negative statistic.

The more vital signs, the better. The beating heart of New York City is not just cops on the street, let’s put it that way. It’s the people.