Philip Ball, Nature News, February 2, 2009
Singling out individuals for surveillance, investigation or screening on the basis of their race or nationality is probably useless for catching those engaged in or planning criminal acts. That is the conclusion of a computer scientist who has exposed the basic mathematical flaw in this approach to crime prevention.
At first glance, the approach seems logical, despite many people’s moral objections. If all previous acts of politically motivated terrorism have been committed by a particular nationality, then doesn’t it make sense to focus searches on those groups?
Not necessarily, says William Press of the University of Texas at Austin. Do the maths and you discover that a simple-minded application of these actuarial methods is worthless: all you end up doing is repeatedly picking out the same innocent people [1].
{snip}
Worse than chance?
It is hard to know how often actuarial selection is really used. Airport security policy, for example, is a closely guarded secret. {snip}
Some think that ethnicity is so clearly a key factor in the process of identifying potential criminals that it has become satirized as the new ‘crime’ of FWM, or Flying While Muslim—the latest variant of the US driving ‘offence’ called DWB: Driving While Black.
{snip}
Press says that, if ‘prior probabilities’ of an individual being a potential offender are assigned accurately, and if screening always results in accurate detection (neither of which is guaranteed), then the chance of finding those with true criminal intent is best if the search is conducted from the top down: from the highest to the lowest probability.
{snip}
One ‘obvious’ method is to make the chance of screening proportional to the chance of being an offender. But Press shows that the number of times this approach repeatedly picks out innocent individuals means that offenders are found no more efficiently than in a random search.
Thus, if actuarial screening is being done, says Press, the chances are that it’s being done ineffectively.
Counterproductive
This doesn’t mean that the technique has no benefit. Press calculates that the best strategy picks out high-risk individuals with a bias proportional to the square root of their probability.
That might be feasible if reliable, quantitative actuarial data are gathered in advance and selection is made by computer. But Press feels that, “given how socially fraught the issue is already, and given how weakly profiling helps, we shouldn’t do it at all”.
These calculations add weight to other arguments against actuarial methods. Harcourt [Bernard Harcourt, a professor of law and political science the University of Chicago], for instance, says that these approaches may be counterproductive in crime prevention because people are likely to change their behaviour in response. {snip} In fact, says Harcourt, these techniques “may actually increase crime, depending on how different populations respond to increased policing”.
Given Press’s result, Harcourt adds, “there is absolutely no reason left to profile.”
References
{snip}
[Editor’s Note: The issue in which this article appears does not seem to have been posted yet. The website to check is here.
Original article
(Posted on February 3, 2009)
Comments
No logical argument is presented for ewhy profiling is ineffective - the reference to math seems to be an intimidation tactic.
If all crime were committed by bisexual Danes over 6 feet tall, then the police would need to focus on them and ignore everyone else.
There is no math needed, only logic.
This is nonsense.
If it looks like a Duck, walks like a Duck, sounds like a Duck…you need to be looking for Ducks.
“Do the maths and you discover that a simple-minded application of these actuarial methods is worthless: all you end up doing is repeatedly picking out the same innocent people [1].” (Why do the Brits put an “s” on math?)
Not true. One (or more) of those “innocent people” is the one(s) you are looking for.
If all the suicide bombers are arab muslims….key on anyone who might fit the bill….one of them is probably the bomber you seek.
If the rape victim says the attacker was a tall black male….there is no reason to consider any females, or short white guys.
You call it profiling. I call it effective police work.
God, don’t we NEED effective police work?
Is there actually any organization that ONLY uses racial profiling?
Doesn’t profiling involve a number of different aspects to narrow down suspects? Race or nationality alone may be almost the same as chance, but what about race, nationality, gender, height, weight, travel patterns, access to types of information, etc. etc.
It’s like saying “This one puzzle piece is worthless at telling me what the whole picture is, you might as well pick any puzzle piece at random.”
Guess I must be getting a little slow now that I’m an adult. RANDOM SEARCHES are more efficient than racial profiling? So if the perp is described as a 20ish Black punk with dreadlocks… it’s more efficient for cops to question long-haired Whites and/or Chinese with pigtails? I’d have sworn that if the perp was described as a Black, looking at suspicious Blacks is more logical. But, I guess I dozed during my p.c. indoctrination classes in school.
Finally, in the last section, the good Doctor admits profiling works. It just does not work well enough to couteract his “feeling” that it is “socially fraught”. Further, it might just make those in the targeted groups more likely to offend—i.e., don’t make Muslims, blacks, et. al., angry by attempting to enforce the law. Nothing new here, just the old sixties song.
Instead of fretting of racial profiling, perhaps Blacks, Latinos (and the liberals who love them) should be trying to fix the fact that their groups commit a massive and disproportionate number of crimes.
The author must belong to post-modernist movement that denies the existence of objective (for instance, mathematical) truth. Based on such denial, one can dismiss probablility theory and statistical reasoning based on known corelations as flawed or useless.
That is a typical proof by ignorance. I am surprised that they published it. Or the economy went so bad that they can’t afford a waste basket.
Poor litle Muslims. They are being oh so profiled. I sugest airport security focuses on some Japanese tourists instead.
I’m reminded of the old saying, “Figures don’t lie, lairs figures.”
Being an old farm boy, if I wanted milk, I’d profile female cows of a certain age group. I wouldn’t walk around the barnyard randomly trying to milk the various animals till I got a pail full of milk.
http://www.nr.com/whp/
Another sad example of “science” being used(abused) to allegedly dismantle what even simple people can plainly see as logical and factual. Just as we are now told emphatically that science essentially “proves” that race is all basically in our minds, I suppose math can be used to “prove” that something so fundamentally obvious just MUST be wrong and unseccessful at fighting crime. The age we live in becomes more fantastically Orwellian by the hour…
Guess I must be getting a little slow now that I’m an adult. RANDOM SEARCHES are more efficient than racial profiling? So if the perp is described as a 20ish Black punk with dreadlocks… it’s more efficient for cops to question long-haired Whites and/or Chinese with pigtails? I’d have sworn that if the perp was described as a Black, looking at suspicious Blacks is more logical. But, I guess I dozed during my p.c. indoctrination classes in school.
Posted by Fed Up at 8:52 PM on February 3
Fed Up,
I can’t see any way that this is going to turn around other than through a violent upheaval. Our so called leaders are all asleep at the switch in a PC induced stupor as is so much of the law abiding citizenry.
This makes perfect SENSE to me, how about YOU?
Two black males driving a 5 year old Bue Escalade rob a store. Stopping every Ford Focus with two white males in it is without a doubt the quickest and smartest way to apprehend these culprits?
YES ___. “Because” stopping,blue Escaldes with two blacks in it is considered PROFILING and is not treating black folks fairly) ?
I think that part of the confusion over this is how they have removed all of the actual logic and analysis from the article.
The guy is not talking about finding someone who has committed a crime based on a description. He is talking about the case where you are screening for future criminals based on the race of past criminals. I.e., you screen Arabs at the airport because many past terrorists have been Arabs. You are not looking for someone for whom you already have a description; rather, you’re just checking the Arabs because previous terrorists were Arabs.
What he’s saying is that in this situation, you are going to end up screening almost as many innocent people as you would without racial profiling. I think this is probably true, since terrorists are needles in a haystack, even among Muslims or Arabs…
However, there is still an advantage, because you avoid wasting time on a huge number of non-muslims and non-Arabs, and this still says nothing about situations where a certain race has a higher innate propensity for crime, or situations where a perpetrator has already been identified…
If a perp is described as black, it is not profiling to look for a black. That is different from looking for a possible bomber at an airport where there is no perp and, in fact, there is not yet even a crime. Checking all Muslims in this case would be profiling. Stopping blacks to find out if maybe he has committed a crime is profiling.
Some of these posts have missed the point. If the perp was described as a black man the police will not be looking for a white female. They will be looking for a black man. That’s not profiling. That’s looking for the perp.
- Realist in Atlanta
Just as we are now told emphatically that science essentially “proves” that race is all basically in our minds,
Nothing furhter from truth. For some two decades now genetic research has validated many (most?) racial stereotypes. It’s just “Liberal” propaganda that claims that race is a purely “social construct”, but there is no credible evidence for that claim.
Similarly with math: probablility theory and mathematical statistics are sound and useful tools that allow for statistical reasoning about the facts (“profiling” belongs to tha category) with a small margin of error. Those, like the author of this article, are denying that fact are math-challenged ignorants or liars.
People are likely to change their behavior? Isn’t that the idea? We know that of the 19 bad guys on 9/11 all were Arabs, so now we should check just white old ladies at airports?
How many of the posters here have undergraduate, masters, or Phd degrees in statistical analysis or criminal justice? The majority of crimes do not have eyewitness reports, and eyewitness reports are often unreliable—this guy has hard data he’s crunched, and you have an anecdote that you heard from a “friend of a friend.” And you wonder why the movement is not taken seriously. Disregarding all data you disagree with as wrong, and treating all data you agree with as right without any real understanding of the issues just marginalizes this discussion further. People need to go here to get new information—instead they go here to get their message repeated back to them.
And YOUR point, Shadrach, is? …
Bottom line being we KNOW Blacks commit far more crimes per total numbers than Whites or any other race. So WHY would racial profiling be anything but logical? You blather about statistical analysis or degrees in criminal justice… yet can’t come to terms with the reality of so much crime by so many Blacks!
Granted, in rare cases, a dark complexioned person could be misidentified as a Black… particularly when the victim is mentally stressed and thus capable of error. But all you need to do is go to any criminal courthouse, watch the defendants herded in by the bailiffs. Review your local newspapers and watch the TV newscasts… doing an analysis of criminals and their respective race. Try having a meaningful dialogue with LEOs and get THEIR perspective as to what (racial identified) perps to be on the lookout for on their police radios and computers. Surely a lot more significant, more meaningful than your “you have an anecdote that you heard from a “friend of a friend.” In other words “HARD” facts you can compile and analyze yourself.
If police made it a practice of randomly stopping blacks only (unless a perp were actually known to be white) it would be make law enforcement far more effective, given that there’s an 800% greater chance that a randomly selected black African is a criminal than an Hispanic or white. (Jared Taylor’s “Color of Crime,” a must-read for anyone interested in the statistics of crime in America.
“Race” is more than a social “construct”. It is known that certain medications have different levels of effectiveness, dependent on the race of the individual being treated..
