Tip:
Highlight text to annotate it
X
Sampson: One of the first concepts that we looked at was what you can think of as
collective efficacy theory, and this in a sense is charted out here in this particular
slide.
Collective efficacy really tries to unite two concepts: first of all, a sense of cohesion
or ties among residents.
Now, now we’re not talking about deep personal ties where you’re having dinner with your
neighbors every night, but a sense of working trust and acknowledgement of neighbors.
And secondly, a sense of the expectations about control — that is, what people will do
under certain conditions of challenge.
And shared expectations are crucial: When you’re in any environment, the shared
expectations for how you act is an important theoretical concept, and this goes back
actually in the history of social thought.
And we thought about this and carried it out with what we thought was a somewhat
innovative way, was to use vignettes where we asked residents, how likely is it, for
example, if kids were skipping school and hanging out on the corner, how likely is it
residents would do something about it?
And that could range.
It could range from intervening with the kids to calling the school to calling the
police.
So it’s informal social control, but it can involve formal institutions; that’s the key.
It’s that connection rather than doing nothing.
And cohesion and shared expectations for control are highly related to one another; they
vary tremendously across neighborhoods.
And it is related, as you might expect — collective efficacy, that is, the collectiveness
part, the efficacy, the intentions to do something, it’s undermined by concentrated
disadvantage, neighborhoods that are turning over rapidly.
It’s importantly related to network ties and organizations.
So we looked at, in a sense, the whole picture of what the causes of collective efficacy
are, but also, what the influence of collective efficacy is on crime, and that’s the — go
to the purple arrow down to the right — we found collective efficacy related to violence,
to other aspects of health.
So now, for a little bit of, just a little bit tougher work here, but I tried to make
these as straightforward as possible: This shows you the association of collective
efficacy and time with later homicide rates.
What we did is to look at, using our multiple surveys, we did it two different points in
time, 1995, 2002, and then we looked at the homicide rate after that, from ’96-2000, and
then after the 2002 survey, up to 2006, and this controls for a number of factors that are
important in the literature, such as poverty, friend/kinship ties, prior homicide, too,
so it’s almost change in homicide.
And the dark bars to the left are the lowest quartile in collective efficacy, lowest 25%
distribution of neighborhoods, and the right is the highest quartile.
And this difference, by the way, these are all homicide rates, about 16 to 4 to 10.4,
it’s about a 50% difference in homicide, so that’s a, it’s not only significant but it’s
substantively large.
Now, if you believe this model and you work out the math, it’s something like about 150
fewer killings over the period of about five years associated with that change in
collectively efficacy, net of the other characteristics.
What about change?
Neighborhoods differ; maybe our controls didn’t work so well.
Another way to look at this is the change within a neighborhood over time.
In this analysis, what I did was to take each year as a unit of analysis; that is, I
calculated the homicide rate in ’95, ’96, ’97, looked at it over time, then calculated the
rate of change, whether it accelerated or decelerated.
As we know, crime is going down.
So an interesting question one can ask is, is the neighborhood change, how the
neighborhood is changing, is it related to the rate of decline or the rate of
deceleration?
And it is.
If you look at this chart, I think the easiest thing is to compare the blue and the red
lines.
The blue line shows the homicide trajectory from ’96 to 2006, and those are basically
communities where there were decreases in collective efficacy and increases in poverty.
And basically, you see a pretty flat trajectory.
Again, there was a secular decline in crime, but it was not going down as fast in those
areas, whereas the red dashed line are areas with increases in collective efficacy and
decreases in poverty, and you see a distinct and significant difference as this controls
for other changes as well and I think provides a different way of looking at the matter.
I said a moment ago that it’s not just the internal characteristics; this is not a story
about, oh, there’s a magic bullet out there, collective efficacy.
This is just one factor.
And this chart shows you the importance of the surrounding context.
So what we can do is to say a neighborhood has a certain poverty rate, a certain rate of
collective efficacy, a certain rate of homicide, but what about the neighborhoods around
them?
Well, we can actually measure those and spatial statistics and spatial methodologies have
come a long way.
So what we were able to do is to take all the neighborhoods into Chicago and look at the
kind of the ripple effects of what does it mean for a community to be located in a
high-risk environment, or next to other high crime rate neighborhoods?
Does that have any independent effect?
And it does; it has a large effect.
This particular model, although it looks simple, there’s only four bars, actually
controls for a lot of the usual suspects out there in terms of racial composition and
poverty, density, and so forth.
But what you see when you compare the two bars on the left with the two bars on the right,
that’s the effect of collective efficacy, and this is a standardized homicide rate.
So what you can see basically is that these areas have higher homicide rates than these
areas, that’s the direct effect of collective efficacy, whereas this, even though a
community may internally have high rates of collective efficacy, that’s not enough, right?
That’s not necessarily a protective factor, because you can see that this area is
surrounded by high-poverty, high-crime areas, and that has direct influence net of the
internal characteristics.
Similarly, the rate of over here, for high collective efficacy areas, you can see that
when you’re surrounded, the rate matters.
So it’s saying that both characteristics matter.
That means then that in thinking about policy, in thinking about theory, we really have
to take the larger social structure into account.
It’s really a mediation of, well, individuals, neighborhoods, communities around the
city, and even macro-level city policies.
I’ll come back to that, too.