ARE OUR COPS COLORBLIND?

EARLIER this year, San Jose was one of the first California communities to
commit to documenting the race and ethnicity of drivers that the police
pull over.

Last week, it became the first to release numbers. The results don't prove
racial profiling, but they won't end the debate over it either.

In the first three months they kept tabs, the police did pull over a
disproportionate number of Latino and African-American drivers. While wide
enough to feed suspicion, the disparity is also narrow enough to give
credence to Chief Bill Lansdowne's explanation for it.

If, indeed, there is racial profiling in San Jose -- hauling over drivers
based on their race and ethnicity alone -- it's nothing on the magnitude of
New Jersey. There, in a 20-month study, three-quarters of the drivers whose
cars were searched on the turnpike were black or Latino, yet those groups
make up about a fifth of the state's population.

In San Jose's case, Latinos, who are an estimated 31 percent of San Jose's
population, constituted 43 percent of 23,425 car stops between July 1 and
Sept. 30. African-Americans, with 4.5 percent of the population, made up 7
percent of the stops. Whites were underrepresented, with 43 percent of the
population but only 29 percent of the stops, as were Asian-Americans, who
make up 21 percent of the city but only 16 percent of stops. American
Indians and others were the remaining 5 percent.

To civil rights organizations, which have received complaints from minority
drivers who say they've been hassled by the police, the numbers speak for
themselves. ``I'm sad to say it is obvious by the way that minorities are
being stopped that they are being targeted,'' said Victor Garza, chairman
of La Raza Roundtable, a Latino organization.

But Lansdowne says that demographics other than race appear to account for
the variations. The department draws precinct boundaries and assigns
officers to them based on crime statistics and calls for assistance. So
higher crime sections -- parts of downtown and East San Jose -- have a
higher concentration of police officers in smaller areas, substantially
raising the odds that speeders and drivers wanted on a warrant will be
spotted and stopped. These are also the areas with the highest
concentrations of minorities and the poor, who are more likely to be
driving clunkers without tail lights.

Furthermore, if there were profiling, you might expect disproportionate
numbers of minorities to be stopped in areas where few live, such as
African-Americans in Almaden Valley. But nothing indicating this jumps out
of the precinct numbers.

Lansdowne's theory sounds plausible, but it's based only on a sense of
where people live. The department didn't do a breakdown of race and
ethnicity by precinct -- a picture that can be compiled with Census data.
If, for example, high concentrations of Asian-Americans also live in the
busier precincts, their numbers of stops also would be higher, instead of
lower. (Unfortunately, accurate information won't be available until April
2001, with the release of the 2000 Census.)

The report also doesn't answer another key question: whether minority
drivers and their cars are searched disproportionately. In New Jersey, it
wasn't the percentage of stops as much as the ratio of searches that was
blatantly discriminatory. Lansdowne says the problem is technical --
computer capacity -- but he's willing to consider solutions, such as
revising questions and eventually buying new computers.

Finally, the computerized reporting system is designed to track stops by
shift and by precinct, not to detect profiling by rogue cops. Earlier this
year, an African-American youth minister claimed he was stopped for no
reason in San Jose, treated like a criminal, then released with no
citation. That incident and national publicity on race profiling have led
some Latinos and African-Americans to mistrust the police.

Lansdowne kept his promise of releasing the first batch of data on car
stops and has offered to continue meeting with community groups. His
openness is encouraging.

But the department needs to collect more and better information. And it
should hire an independent statistician to do the next analysis. Perhaps
then Lansdowne can prove indisputably that profiling is a misperception, at
least in San Jose.