Statistical Profiling Is Not Racial

The new Arizona immigration law has caused quite a stir from Los Angeles to Phoenix to Washington, D.C., on the basis that it supposedly encourages racial profiling. But this is incorrect, since it is actually more restrictive on “racial profiling” than the federal law. It actually says in so many words that racial profiling is prohibited.

But the whole argument has reminded me that back in August 2005 I wrote a MidWeek column, “The Reality of Statistical Profiling,” which outlines how the policy of statistical profiling by the New Jersey state troopers on the Jersey Turnpike back in the mid-1990s, in order to more effectively interdict drug trafficking, got semantically sidetracked to racial profiling. And we’ve never been able to retorque the discussion back around to the fact that all so-called racial profiling is first and foremost based upon statistics and not race.

In view of the controversy over the new Arizona law, I felt it might be appropriate to reprise a few points from that column about racial profiling.

By keeping statistics over time, the New Jersey troopers began to form a profile of “high probability” automobiles most often found to be carrying drugs: make, type and condition of the vehicle, origin of license plates, time of day/night and erratic driving patterns. Once the vehicle was stopped, another set of criteria came into play: driver behavior, IDs of driver and passengers, driver/passenger relationships, inconsistent stories about destination or reason for trip, and scratch marks or loose panels around usual drug hiding places. The profile did not include race of the driver.

Although this profiling system was effective and efficient in netting more drug dealers and drug-carrying “mules,” it also incidentally netted a higher percentage of minority criminals than their percentage of turnpike drivers, therefore the troopers “must be” using race as a criterion for stopping drivers. Suddenly racial profiling was born, and became the latest politically incorrect bugaboo on everyone’s radar screen. And drug dealers effectively now had a green light on the Jersey Turnpike.

Rudy Giuliani used statistical profiling to stem rampant drug deals in Harlem: certain popular street corners, frequency of cars stopping at curbside, times of day/night, dress and demeanor of the suspect.

Using such criteria, a police crackdown almost eliminated the trafficking. But – you guessed it – all the Harlem drug dealers were the same minority race, so “racial profiling” was charged, the crackdown ceased and the dealing flourished.

In the past 30-plus years of worldwide terrorist attacks on U.S. citizens and facilities, aside from only one exception (Tim McVeigh’s Oklahoma City bombing), statistically, 100 percent have been perpetrated by fanatical male Islamist Middle Easterners with previous ties to a terrorist organization – a strong statistical profile. In effect, our “No Fly” lists are based upon statistical profiling.

So with such non-racial profiles available, why must everyone bow at the altar of P.C. when going through airport security?

Like New Jersey state troopers, Arizona Highway Patrol officers and police can easily statistically profile during traffic stops and routine questioning based upon a number of non-racial factors, and still easily identify an illegal alien.

Statistical profiling is fact-based, unemotional, impersonal and objective! We should call it what it is.