Newsflash: Social Science Lacks Predictive Power

Experts have lots of theories about the dramatic drop in crime in the last decade, and at this year's meeting of the American Association for the Advancement of Science, they tried to predict whether a recent slight increase in the incidence of homicides and robberies was cause for panic. The NYT's John Tierney reports:

The experts don’t think crime rates will go up. Or at least not very much. Or at least they can’t see any terribly ominous trends. But then, these social scientists don’t claim to have great crystal balls.

“Our leading indicators stink,” said Franklin Zimring of the University of California at Berkeley. “Remember that nobody predicted the decline of the ’90s.” The experts have had a hard enough time trying to explain it in retrospect.

One interesting theory floated at the AAAS convention for the drop in crime in the 1990s: More immigrants. "Studies have shown that crime in a neighborhood drops as the concentration of immigrants increases."

Via TierneyLab

Editor's Note: We invite comments and request that they be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of Reason.com or Reason Foundation. We reserve the right to delete any comment for any reason at any time. Report abuses.

  • ||

    "Studies have shown that crime in a neighborhood drops as the concentration of immigrants increases."

    Make up your minds. Just a while back, many boards were linking to studies that showed that increases in diversity led to increases in crime.

  • Bagel||

    Is that based on reported crime or actual crime? I'm not familiar with the methodology but I would think that immigrants, many of them with tenuous legal standing, would probably avoid reporting crime to the police or other "authorities".

  • ||

    The inverse relationship between immigrants and crime makes a certain logical sense to the extent that folks "new in the area" would lack the social networks that make organized crime possible. Of course, if an immigrant population comes to predominate in a neighborhood, and perhaps imports criminal networks from the "old country", all bets are off.

  • Guy Montag||

    Um, what happened to that deal about males aged 13 - 25 (somewhere around that range) being the "best" predictor of violent crime rates?

    It was pointed out many places that the male target range had declined in the places that were experiencing lower crime rates, blah blah blah.

    Perhaps the war against "testosterone poisoning" is over and 'science' needs a new villain?

  • The Wine Commonsewer||

    Studies have shown that crime in a neighborhood drops as the concentration of immigrants increases.

    I don't think so. The famous gun study that compared Seattle to Vancouver and proved that gun control resulted in lower crime rate was blown out of the water after it was determined that when you correct for the fact that Vancouver has no immigrants and Seattle does that the crime rates were identical. Bottom line: the immigrants in Seattle were responsible for the lion's share of the reported crime.

  • ||

    Bagel,

    What's the difference between "actual crime" statistics and "reported crime" statistics?

  • ||

    "Studies have shown that..."

    Who HASN'T got a study to support what they want to believe?

  • ||

    I think a statement like that would have to be massively qualified. Which immigrants, from where, in what part of which city? Legal or illegal? How well are they doing economically?

  • ed||

    Or, in other words, we don't know shit.

  • ||

    Who HASN'T got a study to support what they want to believe?

    Theists.

  • The Wine Commonsewer||

    Bergamot, I'm pretty sure that the refutation of the Seattle/Vancouver gun study came from Reason, although it could have come from one of those other gun guys. The thing that was apparent is that Vancouver has a stable, homogeneous population base and Seattle doesn't. After a quick scan though I may have to retract. It may not be immigrants per se but minorities instead. See what you think.

    This is something Jacob wrote four years ago that has some useful links.

    http://www.reason.com/news/show/32165.html

  • ||

    I thought the crime rate went down because we aborted all the criminals.

  • ||

    Vancouver doesn't have immigrants?

  • ||

    TWC said:

    I don't think so. The famous gun study that compared Seattle to Vancouver and proved that gun control resulted in lower crime rate was blown out of the water after it was determined that when you correct for the fact that Vancouver has no immigrants and Seattle does that the crime rates were identical. Bottom line: the immigrants in Seattle were responsible for the lion's share of the reported crime.

    I don't know where folks are getting their data, but Vancouver has had an enormous Chinese immigration since the merger of Hong Kong with the PRC. Vancouver is not in the least a homogeneous city. Like Toronto it's got lots of immigrants, many of them recent.

  • MikeT||

    Its lack of predictive power is why "social science" is largely a useless field of study. Political Science and Sociology are only good for studying a minority of social phenomenon. The rest of it is just speculation. The amount of crap that passes for "science" in these fields render them a joke to any thinking human being.

  • ||

    FYI, you can listen to some of these social scientists, including Zimring, talk about what they don't know about the decrease/increase in crime on last Friday's Talk of the Nation:

    http://www.npr.org/templates/story/story.php?storyId=7453416

  • ||

    ...great crystal balls...

    how are their brass ones? great I hope...

  • ||

    MP

    They call them "The Bible" or "The Koran", etc.

  • ||

    to Greg Nathan:

    Would that account for the increase in drive by chop stick assaults?

  • ||

    Would that account for the increase in drive by chop stick assaults?

    I don't know about that, but it might explain the rapid growth in black-market rickshaws.

  • ||

    I don't trust crime statistics.

    Here's a "statistic": When I was mugged at gunpoint, I didn't report it.

  • ||

    The guy who wrote 'Freakonomics' established a conection between states that legalized abortion pre Roe v Wade and decreased crime statistics a generation later.

  • ||

    "The guy who wrote 'Freakonomics' established a conection between states that legalized abortion pre Roe v Wade and decreased crime statistics a generation later."

    On that Talk of the Nation show they talk about a possible connection between Roe v Wade in 1973 and the nationwide decrease in crime in the 1990's, and they largely debunk it. At least that was their take - of course they didn't go into a ton of detail on controlling for all of the various factors involved....

  • ||

    Economics is a 'social science'.... And thinking people practice it. 'Thinking people' shouldn't disparage any field of inquiry offhand. As with everything, there are bad and good examples

  • Edward||

    I think we're all missing the entrepreneurial spirit informs so much crime. Crime should be higher in a society that encourages initiative and independence from the nanny state. The sort of in-your-face rejection of the state's monopoly on violence is so libertarian. Tony Soprano is a libertarian hero.

  • The Wine Commonsewer||

    Geoff, the study was done in 1988. See the link I posted for more info.

  • stephen the goldberger||

    I read freakanomics, he doesn't show you any of the numbers and just essentially says "trust me the numbers work, you would find them boring." I think it logically makes sense that legalized abortions could play a role but his argument would be a lot stronger if we didn't have to take his word for it.

    I think the economic boom of the 90's has a lot to do with the crime rate drop as well. The fact of the matter is this world consists of so many different variables and attributes is hard to form a substantial cause and effect relationship between a limited number of factors in "reality." Thus why these predictions are always terrible.

  • ||

    bah. don't even start with the gun control crap here. Stats seem to show that the places with the most strict gun control in the US also have the highest crime. (DC being one, Chicago being another.) Also keep in mind the race of those behind most of the gun crimes: blacks. Canada barely has any. Low crime rate. DC has a lot. high crime. coincidence? Compare it to LA, Oakland, San Francisco, Chicago, Detroit, Atlanta, Miami, etc...

  • ||

    It's the economy stupid.

  • ||

    "It's the economy stupid."

    Than why was the crime rate low during the great depression?

  • ||

    "...Stats seem to show that the places with the most strict gun control in the US also have the highest crime. (DC being one, Chicago being another.) Also keep in mind the race of those behind most of the gun crimes: blacks."

    There is so much wrong with the causal reasoning here, where would you start?

    Social science studies more complex phenomena than "hard science." As a result, it is very important to think clearly about the limits of a particular study to answer a particular question.

    Social science can make predictions at certain analytical levels, and not at others. Same goes for physics. Knowing when you are going beyond the limit of your measures is a large part of the art of science.

  • ||

    Than why was the crime rate low during the great depression?

    Nobody had any money to steal?

  • ||

    "Than why was the crime rate low during the great depression?"

    Because Prohibition ended.

  • ||

    aresen,

    Good point. Both scriptures and social science research can serve the same purpose to the unscrupulous. It's an authoritative source that some people use to dictate other people's actions. The defense for both is the same. Believers who become good people by following the writings are doing right. People who go arround telling everyone else to follow the writings should be avoided.

  • ||

    Studies have shown that illegal Canadians are coming and can't find jobs because of racial discrimination. They then will turn to a life of crime. They will first infiltrate the Bloods, Crips and Hispanic gangs.Then they'll make a move on the Asian, Somilian and Nigerian gangs in the big cities. Statistics have borne this out repeatedly.

  • ||

    Economics is a 'social science'.... And thinking people practice it. 'Thinking people' shouldn't disparage any field of inquiry offhand. As with everything, there are bad and good examples

    Notice that they call themselves economists, not economic scientists. Also, name one good thing that has come from sociology. Some good things have come out of social science and many from economics, but sociology, I don't know. Please enlighten me.

  • ||

    More likely due to the neighborhood becoming more homogeneous. Even if they are "immigrants" they are likely to be from the SAME immigrant group since they tend to migrate to areas of like language culture etc.

    Hence the fallacy of the diversity crap shoot and the multi-culti hookah.

    Neighborhoods tend to be better behaved if they are alike and "understand the show" so to speak.

    That is what "assimilation" is all about. No matter what the assimilation is...its about fitting in.

  • ||

    On the immigrants and crime thing, I wouldn't be at all surprised if the legal status of the immigrants was the most important variable. Crime rates are of course based on reported crime, meaning numbers of calls to the cops. Illegal immigrants, who will likely get deported if they attract much attention from law enforcement, or who probably believe they'll be deported, won't call the cops. Ergo, number of reported incidents goes down, even though actual incidents may be quite common. (Also, I know that many jurisdictions don't check citizenship of crime victims, but I'm pretty sure the average undocumented worker doesn't know that, and also, most come from places where I wouldn't call the cops even if I were a citizen.)

  • ||

    It is so simple. Global cooling slows crime. Global warming increases crime. Thus crime is increasing due to global warming.

    If you disagree then you must love Hitler.

  • ||

    Very old joke:

    The economics department of a university used the same tests year after year. "Don't you know the students pass the questions down every year?" "Of course. We just change the answers."

  • ||

    Opps, but I swear it was a joke.

    Cathy Young, a contributing editor at Reason magazine is quoted:
    "No more Nazi or Hitler analogies to describe policies or politicians you dislike. Unless, of course, those policies include actual mass murder and torture, or those politicians who engage in such acts. "

  • dhex||

    "Also, name one good thing that has come from sociology."

    but i realize this is one of those "but left wingers are associated with it, it must be bad!" but that's a hellaciously stupid thing to say.

    http://www.amazon.com/exec/obidos/ASIN/0674023552/reasonmagazineA/

    studying marginalized populations is important for a number of reasons, and you don't have to be a leftist to realize this.

  • dhex||

    then again, i'm an ethnography buff, so of course i would say that.

  • ||

    these social scientists don't claim to have great crystal balls

    How on earth did that get past the vaunted NYT editorial staff?

  • ||

    these social scientists don't claim to have great crystal balls.

    How on earth did that get past the vaunted NYT editorial staff?


    They claimed to have great chocolate balls. While the editorial staff licked their lips, drooled and mumbled "mmmmm chocolate", they hurried past.

  • ||

    Well that's enough to scald the balls of a brass monkey.

    Sorry, just another old joke.

  • ||

    "Studies have shown that crime in a neighborhood drops as the concentration of immigrants increases."

    [begin rant, balls scalding away]

    Which is "interesting" because it "fits" the slant around here (and other places).

    I'm not anti-immigrant on net balance. But I'd love to see somebody take a real, genuinely balanced perspective on this issue. The most dangerous neighborhoods I've ever been in were immigrant ghettos.

    It may be true that immigrants are a net plus in the overall balance. But flatly denying the very clear and obvious down sides to high immigration levels -- or at least failing to ever give it any serious attention and air time -- does not make the case for immigration clear, nor does it persuade anyone.

    My reason for being pro- has a lot more to do with humanitarian grounds than it does economics. I've yet to be fully convinced of the economic case, because the people who try to make either deny or don't address the down sides to what really happens.

    [end rant, balls melted away]

  • ||

    That study comparing Vancouver to Seattle is interesting. I wonder if they considered how to sort out the fact that THEY ARE IN DIFFERENT COUNTRIES.

  • JD||

    The most dangerous neighborhoods I've ever been in were immigrant ghettos.



    Correlation, causation, let's call the whole thing off... My feeling would be that the immigrants probably have to live in the dangerous neighborhoods because they can't afford anywhere else. That's not to say that being poor and marginalized helps the crime rate - there have been a bunch of stories about gangs which draw their membership largely from immigrant populations, like MS-13.

  • dhex||

    "Correlation, causation, let's call the whole thing off..."

    thanks for my first coffee spill of the day!

  • ||

    My feeling would be that the immigrants probably have to live in the dangerous neighborhoods because they can't afford anywhere else.

    Maybe so, in some cases. Not so in others.

    There are poor Mexican immigrant neighborhoods around here (AZ) where the immigrants can hardly go to the laundry mat and back without getting in a gun fight along the way.

    Around here we've had immigrants -- yes I said IMMIGRANTs -- driving up and down I-10 shooting at each other. No worries about whether they hit innocent bystanders.

    They somehow managed to shoot each other, both cars exited the highway at the hospital, and they were still shooting at each other as they walked across the parking lot into the emergency room.

    This would be an absolutely hilarious joke if it wasn't a true story. I lived a couple of blocks from the hospital at the time.

    I've never heard of other people (natives or immigrants) doing the kinds of things that regularly occur in the Mexican immigrant neighborhoods around here. The Mexican ghettos are a world unto themselves, and they are what they -- because the people who move into them, are who they are.


    Like I said, on net balance I'm pro-immigration. But I'm really, really tired of all the pro-crowd who run around and try to pretend that immigrants "are no problem".

    That just ain't the way it is. Nobody is going to sway the anti-crowd until they admit this little part of reality.

GET REASON MAGAZINE

Get Reason's print or digital edition before it’s posted online

  • Video Game Nation: How gaming is making America freer – and more fun.
  • Matt Welch: How the left turned against free speech.
  • Nothing Left to Cut? Congress can’t live within their means.
  • And much more.

SUBSCRIBE

advertisement