Search the Archive:

February 02, 2005

Back to the table of Contents Page


Palo Alto Online

Publication Date: Wednesday, February 02, 2005

Guest Opinion: 'Profiling' can occur in both directions, and is wrong either way Guest Opinion: 'Profiling' can occur in both directions, and is wrong either way (February 02, 2005)

by Lynne Johnson

Imagine for a moment that you're a Palo Alto police officer. You've been hired and trained to protect the safety of members of the community and their property. You take pride in the uniform and in the department.

One day, you're driving around on patrol and hear an alert on the police radio. A resident reports a possible burglary suspect leaving her next-door neighbor's house. She knows her neighbor is on vacation.

The witness describes the suspect simply as an African-American male, 18 to 25 years old, driving a white van. Your pulse quickens. This is what you've trained for. You drive toward the scene and enroute you observe an African-American male driving a white van traveling in the opposite direction.

Recognizing that the driver of this van matches the witness' description and may have just committed a burglary, you conduct a "felony car stop" -- the procedure is to draw your gun and order the person out of the car.

After checking his identification and other information, you determine that the man is not the burglary suspect. You explain why you stopped him, apologize for the situation and send him on his way. Still following procedure, you fill out a form that includes racial data on the stop.

The next day you get to work and you are informed that the man has filed a complaint against you for stopping him because he was African-American and accuses you of racial profiling.

In 2000, the Palo Alto Police Department voluntarily began collecting demographic data. We were one of the first agencies in the state to do so, and one of a handful nationally. Unlike other agencies -- such as San Jose and Sacramento police departments that collected data only on drivers of vehicles stopped by their officers -- we decided to capture significantly more data.

The effort was well intentioned. We wanted to ensure that we had the data to uncover any racial profiling that might be occurring and to reassure the public that we would not tolerate it.

In the process, in spite of our best intentions, we realized that the information collected could be highly misleading. Instead of collecting data that could be benchmarked against other cities to detect abnormal patterns, our data was essentially unusable for comparative statistical analysis.

Indeed, it was giving the wrong picture of our department's procedures because of how it was collected.

We decided to change our methods for two reasons: First, we wanted meaningful data that could be compared to other cities. Second, we needed to standardize and automate the process to keep our officers out on patrol instead of at their desks doing paperwork. We think the new process will better serve the community we are sworn to protect.

Here's an example of how previous data-collection efforts resulted in skewed data.

An officer was sent to Nordstrom's on the report of a shoplifter taken into custody by store security. Following procedure, the shoplifter's race was captured and reported in the demographic report. That situation clearly cannot be constructed as racial profiling, since the officer has no control over who was apprehended, yet the data showed that another person of color had been arrested.

Similarly, an officer working radar on Alma stops a car traveling at 55 miles an hour as registered on the radar gun. The officer has no knowledge of the sex, gender or race of the driver. All the officer knows is that the car is speeding. The officer stops the car, which turns out to be driven by a person of color. Two passengers also turn out to be people of color.

The data will show that the officer stopped three people of color -- and an erroneous perception of racial profiling could easily be developed, even though it was a legitimate traffic stop and the officer was just doing his or her job.

Then there is the question of accuracy. Officers sent from call to call frequently do not have time to complete the cards or are unable to provide all the information on the people they stopped. The cards have to be reviewed, compiled and analyzed manually, further calling into question the value of this process in terms of reliability of the data and the cost in terms of staffing priorities and efficiency.

Our new process will be automated. We will collect the same data that is routinely collected in other cities. It will be done on the spot by computer entry in the patrol car, and compilation and data analysis will be easy. We will have benchmarks for comparison, which was our original intent.

The job of a police officer is extremely difficult and rightly demands the highest accountability.

But that accountability should be based on factual information and fairness, just as everyone else wants to be judged. We must all work together to ensure that no pre-judgmental "profiling" occurs, whether it is focused on a person of color or a police officer.

Lynne Johnson was named chief of police in February 2003, becoming the first woman chief in the department's history, after rising from patrol officer to assistant chief.

E-mail a friend a link to this story.

Copyright © 2005 Embarcadero Publishing Company. All rights reserved.
Reproduction or online links to anything other than the home page
without permission is strictly prohibited.