Project on Human Development in Chicago Neighborhoods

Part 1: Collective Efficacy

Jolene Hernon: This is a two-part podcast. Today, we are talking about collective efficacy, a very early and important finding from the “Project on Human Development in Chicago Neighborhoods.” In the companion interview, Dr. Liberman will talk about violence, race, ethnicity and exposure to firearms. Akiva, tell us about collective efficacy.

Dr. Akiva Liberman: Well, the question they were interested in was why some neighborhoods are able to better control deviant behavior and disorder than other neighborhoods. And their theory, collective efficacy theory if you will, is that the primary ingredients for that are a shared sense of values and trust of your neighbors and shared expectations about intervention in deviant behavior, to control disorder in deviant behavior.

Jolene Hernon: So that sounds a little abstract. Can you bring that into the real world for us?

Akiva Liberman: Sure, let me give you an anecdote. When I take my kids to the playground in my neighborhood, and there is a kid I don't know who's doing something I think is inappropriate, the question is, “Am I going to intervene?” If I think that my neighbors and I share standards of behavior, that they trust me and they expect me to intervene, then I'll intervene. But if I think that the parents of that kid have different standards of behavior then I do, or don't trust me and don't want to me intervene, then I'm likely to just let it go. That's the same concept here for collective efficacy, that you need to have shared standards of behavior and values and expectations that your neighbors are going to intervene on your behalf and vice versa.

Jolene Hernon: Okay, why is that such a big deal?

Akiva Liberman: It is a big deal for a couple of reasons. First, they show that the collective efficacy in a neighborhood was related to the amount of violence and victimization in that neighborhood. So that seemed to show that this is a process related to people being able to control disorder and deviant behavior in their neighborhoods. The other thing is that it's long been known that some neighborhoods have more violence than others and it's been shown that, for example, levels of poverty and disadvantage are related to level of violence in a neighborhood. But nobody really thinks that it's a direct connection. Being poor doesn't make you violent.

Jolene Hernon: Yes.

Akiva Liberman: So, why is that? And prior research didn't really satisfactorily get at the processes that might be related to the poverty on the one hand and violence and other negative outcomes on the other hand.

Jolene Hernon: The more collective efficacy a neighborhood has, the less violence. And the less collective efficacy, the more violence. Is that right?

Akiva Liberman: Yes, that is right.

Jolene Hernon: So what kinds of characteristics in a neighborhood cause a neighborhood to have more collective efficacy? Or maybe I shouldn't say cause, but what factors in a neighborhood contribute to collective efficacy?

Akiva Liberman: Neighborhoods have been found to have more collective efficacy when they have less poverty and disadvantage, more residential stability and also when they have more Latin American immigrants, who are most of the recent immigrants in Chicago.

Jolene Hernon: So it sounds like poverty is the big issue here, the big factor?

Akiva Liberman: Well, not precisely. Poverty, they would say, is a condition under which shared values, trust of your neighbors and shared expectations for intervention are undermined. Poverty and disadvantage.

Jolene Hernon: Okay.

Akiva Liberman:  But again, poverty and disadvantage don't seem to be . . . it seems hard to explain how poverty and disadvantage would directly explain levels of violence. The other thing to say is that when two neighborhoods have equivalent levels of poverty, but one has more collective efficacy, then the one with more collective efficacy has better outcomes.

Jolene Hernon: Okay.

Akiva Liberman: Another important thing about collective efficacy is that it's not a characteristic of individuals; no matter what my own resources are, how wealthy I am, what my own values are, I can't generate collective efficacy by myself. It's something about . . . I don't want to say relationship because I don't have to necessarily know my neighbor to share values with my neighbor, but it's a sense of the social fabric and the collective understanding about what we should do.

Jolene Hernon: Thank you very much, Akiva Liberman, social scientist analyst here at the National Institute of Justice.

Part 2: Race, Violence and Exposure to Guns

Jolene Hernon: I am here with Dr. Akiva Liberman, who's written a Research in Brief for NIJ on adolescents, neighborhoods and violence. Welcome Akiva.

Akiva Liberman: Thank you, Jolene.

Jolene Hernon: This is a two-part podcast and today we're talking about two studies out of the “Project on Human Development in Chicago Neighborhoods.” One study on race and violence and the other study on exposure to firearms violence. Akiva, tell us a little bit about the RIB first.

Akiva Liberman: This Research in Brief summarizes the findings from four published articles coming out of the “Project on Human Development in Chicago Neighborhoods.” Today, we are going to talk about two of them. One of them concerns race, ethnicity and violence and the other concerns the effects of the exposure to gun violence. I should say that I didn't conduct any of this research myself; I'm summarizing it. The paper on race, ethnicity and violence was written by Robert Sampson, Jeff Morenoff and Steve Raudenbush and the paper on the effects on the exposure to gun violence was written by Jeff Bingenheimer, Robert Brennan and Felton Earls. I've just summarized them for our audience.

Jolene Hernon: Okay, the ethnicity and the race article that you summarized in this Research in Brief. Tell us about that one.

Akiva Liberman: Okay, that article was published in 2005 in the American Journal of Public Health. And it attempted to answer the question, “Why do adolescents of different, from different racial and ethnic groups, seem to commit violence at different rates?”

Jolene Hernon: Seem to or do?

Akiva Liberman: Well, they actually do. I mean, the first result of the study is that these participants who are asked about how much violence they themselves have committed differ in their reports as a function of race ethnicity. The black kids report committing more violence. The Mexican-American kids report committing the least violence, a bit less than the whites. And everyone else is somewhere in the middle.

And so, that is a long standing kind of finding, although sometimes it comes out of arrest reports. But here it is from self-reported violence. And they sought to see whether they could explain why that would be.

Jolene Hernon: And what did they find? Could they explain the differences?

Akiva Liberman: They found that they were able to explain about 60 percent of the black/white gap and all of the Mexican-American/white gap by three basic factors, those being whether the kids' parents are married, their immigrant generation and features of their neighborhoods.

Jolene Hernon: And was there one that was the largest influence?

Akiva Liberman: Of those three, the neighborhoods had the largest influence and that explained about a third of the black/white gap.

Jolene Hernon: Okay, so are we talking about collective efficacy in the neighborhood? Then, is that what they found?

Akiva Liberman: Actually, they didn't. In this study, again, this is from the longitudinal study, they found that the presence of lots of first-generation immigrants in the neighborhood and the presence of professionals living in the neighborhood are both protective factors. And neighborhood residents who are cynical about the law is a risk factor for those adolescents committing violence.

Jolene Hernon: Did I hear you correctly? That being an immigrant is a protective factor? How does that work?

Akiva Liberman: Well, first of all, it's worth pointing out that findings about immigration were consistent at both the individual level and at the neighborhood level, so individual kids who are first-generation immigrants commit less violence than second generation who commit less violence than the third generation. That is at the individual level. At the neighborhood level, kids living in neighborhoods with lots of first-generation immigrants commit less violence too.

Now it's important to just give this a little context. We are talking mostly about immigration from Latin America, primarily Mexico, in Chicago at the period being studied here. And contrary to some stereotypes, perhaps, yes, the first-generation immigrants have much better outcomes on a whole variety of things. Also, some health outcomes compared to whites of comparable means.

Jolene Hernon: But I thought immigrants tend to be poorer and more disadvantaged than whites?

Akiva Liberman: Right.

Jolene Hernon: So how does the poverty and disadvantage play into this?

Akiva Liberman: Well, we try to distinguish these things by comparing implicitly in statistical models by comparing immigrants and nonimmigrants at similar levels of poverty so we can disentangle those. And it turns out that immigrants do much better for the level of wealth or poverty that they have compared to whites and also compared to later generations of immigrants.

Jolene Hernon: Compared to later generations of immigrants? I think I have heard you talk a little bit about this, the “acculturation effect.” Can you explain to us what that is?

Akiva Liberman: Yea, well, there are two different hypotheses that haven't really been completely settled I think, but both might have an element of truth to them. One is that the first generation of immigrants who come over are especially law abiding, motivated, hard working. There's also health outcomes. Maybe they also are particularly healthy and hardy and so on because it is hard to get here.

Jolene Hernon: Yes.

Akiva Liberman: And then, their kids don't inherit that completely and then their grandkids, even less so. So they come with some kind of inborn advantage with regard to the kind of things that we are talking about that wears off with generations. That is one possibility.

The other possibility is that something negative is happening to their kids and grandkids as a function of being here. That's called the “acculturation paradox.”

We are talking here about immigrants from Latin America, primarily. And they come here and they are striving as best they can to fulfill kind of mainstream American values. They're striving to achieve mainstream American goals. Your kids and grandkids notice some rift between the Latino community and the mainstream white community and become somewhat alienated as a function of that, and, over time, they then start identifying with some kind of counter cultural tendencies. So, these are two possibilities, they both might be true to some degree.

Jolene Hernon: Akiva, can we move now to the exposure to gun violence findings?

Akiva Liberman: Yes, those came out also in 2005 in Science magazine and the first author is Jeffrey Bingenheimer. And that study explored, “What are the effects of being exposed to gun violence; either by witnessing it, being a victim yourself or having a family member who was victimized.”

Jolene Hernon: What were the findings?

Akiva Liberman: The findings were that the kids exposed to gun violence were twice as likely to commit violence themselves as kids who were not exposed to gun violence.

Jolene Hernon: But maybe the kids who committed violence had a propensity for being violent anyway. I believe in your RIB you mentioned many of those kids were more impulsive and aggressive. Their parents used corporal punishment more. And so, how did the researchers tease out the differences between kids who were more likely to be violent from the kids who were exposed to violence?

Akiva Liberman: This is very important issue you are raising. It's the issue of trying to reach causal conclusions from non-experimental data.

Jolene Hernon: Yes.

Akiva Liberman: And there are a couple of things to say about it. First of all, they took advantage of the longitudinal design to at least know what came first and what came second. So, in the second project interview, you will remember there were three interviews with each kid, in the second interview kids were asked about whether or not they had witnessed or been victims of gun violence. At the third project interview, which is about two and a half years later, they were asked about any violence they had committed over the last 12 months.

So the first thing we know that it is not a chicken/egg problem. But, even so, as you mentioned, there are lots of differences between kids exposed to gun violence and kids who aren't. And the main challenge for this paper was to disentangle that. And the goal is basically to be able to compare kids who are similar on all those things but some were exposed to gun violence and some aren't. So, instead of just comparing the group of all kids who were exposed to gun violence simply to all kids who weren't, what you try and do is get groups who are very similar and do comparisons within those groups.

In this case, they used a method called “propensity score stratification.” They divided the kids up into 12 strata, 12 groups. And the difference between them is how much they were at risk of exposure to gun violence. And that takes into account all those other things that you were talking about; neighborhood characteristics, their individual psychological background, stuff about their family, peers, all that, is taken into account.

And then, kids who are at high-risk of exposure to gun violence because of all that stuff are looked at. And those, among that group with high-risk, those who were actually exposed are compared in outcomes to those who were not exposed. And similarly, kids who are at low-risk of exposure to gun violence but some of them were exposed are compared to similarly low-risk kids who were not exposed. And this way, most of that possible bias is eliminated.

Jolene Hernon: So now, let's ask the question that all the chiefs of police want to know, about how they can make their neighborhoods safer. Based on what the research shows, what are the policy implications?

Akiva Liberman: Well, let's work backwards. Starting with the paper on exposure to gun violence, this paper, I think, is the strongest methodological study to date about what the effects of witnessing gun violence are. You can't do an experiment on this question. And so any results . . .

Jolene Hernon: You can't do experiment on this question because?

Akiva Liberman: I can't experimentally decide I am going to have some kids get shot at or witness somebody get shot and see what the effect is.

Jolene Hernon: Right, Okay.

Akiva Liberman: So we are left to resort to what are called observational or correlation studies. And it's very difficult to be confident that you have gotten cause and effect relationships. This study is probably as strong, I would guess, as you are going to get at addressing this and it finds a pretty dramatic effect of exposure to gun violence that seems to confirm the pretty common sense notion that there is a kind of a cycle of violence. People who witness and are exposed to violence are more likely to themselves commit more. And then, of course, people who witness them committing it are more likely to commit more themselves. So this suggests you want to interrupt that cycle of violence and that it will have long-lasting effects. That is the first kind of policy implication. And particulars about how you control gun violence and how you limit access to guns and so on I will leave to others to discuss.

Jolene Hernon: Okay, okay. How about this question then: “How do you make your neighborhoods have more collective efficacy?” Because the more collective efficacy in my neighborhood, the safer my neighborhood is.

Akiva Liberman: Right, that's a great question and that is what everybody asks the researchers, and they always defer the question. And the reason is that they didn't do a study trying to change the level of collective efficacy. They just did a study finding that the level of collective efficacy is related to all kinds of good and bad things that happen. To find out what changes collective efficacy requires a whole new research project, intervention studies, which are difficult to do. And I would say that they've identified the target for the intervention. Now, we have to do the hard work to develop interventions that might actually effect collective efficacy.

Jolene Hernon: What about race? What implications are there for what we know about race and violence?

Akiva Liberman: Well, the most obvious implication is that it's not race per say that's creating differences in violence. And that might lead one to be somewhat more optimistic about the future. Let me conclude with Sampson, Morenoff and Raudenbush's own words. They conclude by writing, "We conclude that the large racial and ethnic disparities in violence found in American cities are not immutable. Indeed, they are largely social in nature and therefore amenable to change."

Jolene Hernon: Okay, that's a good positive note to end on. Thank you very much. Akiva Liberman, social science analyst here at the National Institute of Justice.

Akiva Liberman: My pleasure to be here.