Sociology papers online

April 14, 2009 6:38 AM

Harvard Sociologist Robert Samson, known for his work challenging the Broken Window hypothesis (previously on Metafilter), has a number of publications on neighborhoods, race and immigration, crime, and spatial dynamics posted publicly online. Here are just a few recent publications (all pdfs):

*Moving to Inequality: Neighborhood Effects and Experiences Meet Social Structure
*Durable effects of concentrated disadvantage on verbal ability of African American children
*Rethinking crime and immigration
*Neighborhood Selection and the Social Reproduction of Concentrated Racial Inequality
*"After School" Chicago: Space and the City

posted by lunit (22 comments total)

oops, that's Sampson.
posted by lunit at 6:42 AM on April 14, 2009

My biggest problem with "broken windows" and other crim theories that have emerged from the social disorganization/ecological perspective is that they only apply to American experiences. R.E. Park's (et al) works only applied to CHICAGO experiences. And it looks like Sampson's critiques also, in good American sociological fashion, also fail to recognize the existence of any city in any country anywhere in the world.

All that said, American experience is unique and important, and no crim theory has anything better than "fair" explanatory power, and that includes any and all social structural accounts, so Sampson does not in any way settle this debate. You only look at social structure (and that includes race-based perceptions), you ignore social learning, and you're left with a godawful lot of unanswered and very obvious questions- like "why don't ALL poor black men become criminals" and "why do SO FEW black WOMEN become criminals" and "why do ANY rich WHITE men become criminals."

Fun fact: the "broken window" in broken windows refers not to a building but to a "bait car" that had, in one case, windows intact, and in the other, a broken window. The one with the smashed-in window was further vandalized et voila.
posted by ethnomethodologist at 7:44 AM on April 14, 2009

But wouldn't a study that sampled the whole entire world and make a generalization about all the people in it be kinda weak?
posted by iamkimiam at 8:07 AM on April 14, 2009

But wouldn't a study that sampled the whole entire world and make a generalization about all the people in it be kinda weak?

What?
posted by delmoi at 8:25 AM on April 14, 2009

i would have to agree with iamkimiam...it would seem that unless otherwise specified, American sociologists would be exploring American social themes. if it was a English sociologist from Cambridge, i would expect his papers to be about British social themes.

re the fun fact from ethnomethodologist: thanks for letting me know about that. i like knowing weird details like that. i have no idea if academically/research-wise if that makes a difference, but it's still interesting to know.
posted by sio42 at 8:29 AM on April 14, 2009

This is great.
posted by fuq at 8:52 AM on April 14, 2009

There are two main causes for the decline in U.S. crime rates : First the legalization of abortion had an unprecedented impact across the whole country. Second high crime cities like NYC started making police chiefs report in about crime rates weekly, meaning they needed to actually know about the crimes going on.

I don't think we've much evidence that the broken windows methods have contributed much given the incredible significance of these two factors. Otoh, some broken windows supporters ala Giuliani deserves considerable credit for this idea that police chiefs can assign resources more wisely if they're forced to remain knowledgeable about the crimes themselves.
posted by jeffburdges at 9:01 AM on April 14, 2009

ethnomethodologist, when you say that "... they only apply to American experiences" is that because when the experiments were repeated in non-American locales they didn't produce significant results or because no repeated experiments were carried out at all?
posted by JeNeSaisQuoi at 9:02 AM on April 14, 2009

"Durable effects of concentrated disadvantage..." is a really interesting piece; thanks lunit.
posted by Mister_A at 9:03 AM on April 14, 2009

I can't wait to get home to read these. I am not being sarcastic.
posted by desjardins at 9:06 AM on April 14, 2009

What I thought was interesting about the Broken Windows piece seems acknowledged in the opening anecdote about community policing: that crime rates didn't actually go down when patrol officers were assigned, but that residents felt better. From there, it was kind of surprising that Bratton et al. would argue a causal relationship. (I'm also kinda fascinated by the referenced shift between police as order-keepers and police as crime-solvers).

Finally, spacial dynamics research seems incredibly interesting to me. Anyone know what kind of math this requires and what grad schools are good at SDR?
posted by klangklangston at 9:33 AM on April 14, 2009

Finally, spacial [sic] dynamics research seems incredibly interesting to me. Anyone know what kind of math this requires and what grad schools are good at SDR?

The field of Geographic Information Systems (GIS) deals with this. I am not a math nerd, but ArcGIS will do a lot of the calculations for you, if you know which ones to use. Here's a book on the topic. UW-Milwaukee is a good school for GIS, not least because Milwaukee was one of the first cities to adopt a GIS (in 1976!!) and the guy who implemented it is the department chair. I do wish I'd been encouraged to take more programming classes, though.
posted by desjardins at 10:29 AM on April 14, 2009 [2 favorites]

There are a bunch of AskMe posts about GIS, also. Here's a good one.
posted by lunit at 10:37 AM on April 14, 2009 [2 favorites]

Locking everyone up for real and imagined offenses probably helps reduce crime. It's more expensive than crime, but it reduces crime.

The United States has more people incarcerated, per capita, than any other country in the world.

Bah, don't get me started.
posted by Xoebe at 10:47 AM on April 14, 2009

I promise that if I study spatial dynamics, I will first learn to spell spatial dynamics.
posted by klangklangston at 11:19 AM on April 14, 2009

About 4 articles in... lots of great stuff. good post and more reading to do...
posted by Lacking Subtlety at 11:19 AM on April 14, 2009

There are two main causes for the decline in U.S. crime rates : First the legalization of abortion had an unprecedented impact across the whole country. Second high crime cities like NYC started making police chiefs report in about crime rates weekly, meaning they needed to actually know about the crimes going on.

Actually, there's still some debate about the centrality of these causes for the 1990s crime drop. If you look at abortion legalization in a worldwide context, France is a counterexample that undermines the hypothesis that abortion reduces crime. France legalized abortion in 1975, but unlike the United States, France experienced an increase in youth crime during the 1990s. In addition, the change in policing practices is debatable as an explanation for the crime drop, because crime dropped all over the U.S.A. in cities with many different policing practices.

Other explanations for the crime drop include low youth unemployment during the Clinton years, the stabilization of drug markets (i.e., people stopped shooting each other for territorial rights over drug turf), and the "younger brother effect" (i.e., my older sibling is dead, in jail, or a crackhead, so I decide from a very young age to stay away from crime and drugs). Trust me. The criminologists are going to be busy for a long time debating why crime rates dropped in the 1990s.
posted by jonp72 at 11:30 AM on April 14, 2009 [1 favorite]

Interesting stuff, thanks
posted by Smedleyman at 12:28 PM on April 14, 2009

This is a terrific post lunit, nicely done.
posted by nola at 2:09 PM on April 14, 2009

There are other possibilities for the decrease as well: such as lead abatement programs.
posted by drezdn at 3:34 PM on April 14, 2009

The theories I've seen about the drop in Canada have centered on demographics: young people commit more crime. Less young people -> less crime.
posted by Monday, stony Monday at 5:26 PM on April 14, 2009

France is a counterexample that undermines the hypothesis that abortion reduces crime. France legalized abortion in 1975, but unlike the United States, France experienced an increase in youth crime during the 1990s.

True, but how does France's immigration profile modulate the crime rate? Crime isn't a single-variable causation.

I'm not saying that abortion reduces crime; most of these arguments are stupid because they don't take into account multiple downstream effects.

My hypothesis is that the greater the wealth gap, the greater the crime (modulated by the amount of resource spent on draconian policing). Exactly how to reduce the wealth gap (whether to bring down the rich, or raise the poor - and how to do this), I'll leave to others to speculate.
posted by porpoise at 10:49 PM on April 14, 2009