Driving While Black or Brown

An analysis of racial profiling in Arizona

The American Civil Liberties Union of Arizona is the states premier guardian of liberty, working daily in the courts, legislature and communities to defend and preserve individual rights and freedoms guaranteed to all by the Constitution and the laws of the United States and Arizona. The ACLU of Arizona is an affiliate of the ACLU, the largest civil liberties organization in the country, with more than 500,000 members. The ACLU of Arizona Campaign Against Racial Profiling seeks to end discriminatory police stops and searches through public education, legislative advocacy and litigation. This special report is designed to educate the public and enlist individuals in the fight to eliminate racial profiling in Arizona.

Table of Contents
I Introduction II Executive Summary of Findings III Overview of Data a Searches b Seizure of Contraband c Duration of Stops d Stops i Methodology Involved in the Collection of Stop Data 2 3 5 5 9 11 13 13 14 14 15 17

Acknowledgements The data analysis was completed by Dr. Frederic I. Solop, Director of the Social Research Laboratory at Northern Arizona University (NAU). The Social Research Laboratory (SRL) is a full service research unit within the College of Social and Behavioral Sciences at NAU. The SRL is best known for its work in the fields of criminal justice, municipal service satisfaction, health, environment, and electoral dynamics. For more information on the SRL, visit: www.socialresearchlab.com. Special thanks to ACLU-AZ cooperating attorneys Lee Phillips, Natalie Jacobs and Charles Babbitt for their unflagging commitment to racial justice, for standing up for the victims of racial profiling, and for going to court on their behalf. We also want to thank Meghan McDowell for her outstanding research on this project. Finally, we would like to acknowledge the work of ACLU attorneys Daniel J. Pochoda and Reginald T. Shuford for their vital roles in challenging civil rights violations in Arizona. ACLU of Arizona P.O. Box 17148 Phoenix, AZ 85011 602-650-1967 www.acluaz.org

ii Analysis of Highway Stops IV Conclusion V Background: Arnold v. Arizona Department of Public Safety VI Recommendations

Driving While Black or Brown

An analysis of racial profiling in Arizona
Published April 2008

I Introduction
Racial profiling occurs when police target people for humiliating and often frightening detentions, interrogations and searches based not on any evidence of criminal activity, but rather on their race, ethnicity, nationality or religion. Although normally associated with African Americans and Latinos, racial profiling and DWB or driving while black or brown have also become shorthand phrases for police stops of Native Americans and, increasingly after 9/11, of Arabs, Muslims and South Asians.

II Executive Summary of Findings

Arizona Department of Public Safety Officers made more than 500,000 stops between July 1, 2006 and June 30, 2007. Just under 200,000 stops were made on Arizonas interstate highways during this period. Of those 200,000 interstate highway stops, approximately 13,271 resulted in searches. African Americans and Hispanics stopped by DPS officers were more likely than whites to be searched on all major highways included in this analysis. Native Americans and persons of Middle Eastern descent also were more likely than whites to be searched on most highways. On average, Native Americans stopped by DPS officers were 3.25 times more likely to be searched than whites stopped by DPS officers. African Americans and Hispanics were each 2.5 times more likely than whites to be searched by DPS. Higher search rates for minorities were not justified by higher rates of transporting contraband. In fact, on average, whites were more likely to be carrying contraband than Native Americans, Middle Easterners, Hispanics and Asians on all major Arizona highways. African Americans were at least twice as likely as whites to be searched on all six interstate segments, despite the fact that the rate of contraband seizures for African Americans and whites was similar. Minorities, including African Americans, Hispanics and Middle Easterners, were consistently stopped for longer periods of time than whites traveling on all interstate highways in Arizona. In sum, this report concludes that DPS officers treated persons from different racial and ethnic groups unequally between July 2006 and June 2007. Minorities were more likely than whites to be searched and stopped for longer periods of time. This unequal treatment was not justified by higher contraband seizure rates from minority motorists.

Discriminatory police stops and searches have reached epidemic proportions in recent years fueled by the increased enforcement of immigration laws by local and state police and the misguided wars on drugs and terror. These policies have given police a pretext to target people simply because they look foreign, or fit a drug courier or terrorist profile.

For the past several years, the American Civil Liberties Union (ACLU) of Arizona has dedicated itself to fighting against the widespread, but unconstitutional, practice of racial profiling by the Arizona Department of Public Safety (DPS). In 2001, the ACLU, in conjunction with attorney Lee Phillips of Flagstaff, filed a class action lawsuit against DPS, charging that their officers engaged in a continuing pattern of race-based traffic stops, detentions and searches of African American and Latino motorists throughout Arizona. The case, Arnold v. Arizona Department of Public Safety, was resolved in 2006 when a federal district court in Phoenix approved a historic settlement agreement. That agreement called for substantial changes in DPS procedures. Most notably, the settlement required DPS to collect data on all traffic stops and modify search procedures to ensure that all officers obtain written permission from drivers before conducting so-called consent searches. This practice allows officers to conduct searches without evidence of criminal wrongdoing and even innocent people feel pressure to consent in this context. A nine-member Citizens Traffic Stop Advisory Board, which includes three ACLU representatives and other community members appointed by Governor Janet Napolitano, continues to monitor DPS compliance with the agreement.

The ACLU of Arizona commissioned the Social Research Laboratory (SRL) at Northern Arizona University to analyze the first year of DPS data collected under terms of the Arnold settlement. This report is the result of that effort. It is intended to provide a more comprehensive picture of the continuing problem of racial profiling in Arizona. The analysis examines data relating to highway stops and vehicle searches by DPS between July 1, 2006 and June 30, 2007. It focuses on who is being stopped, who is being searched, the frequency of contraband being found during searches, and the duration of highway stops. Within the pages of this report, you will learn that searches being conducted by the majority of DPS officers continue to target minorities, despite the fact that people of color are less likely than whites to be transporting drugs, weapons or other illegal contraband. The report also shows that minorities are detained for longer periods of time after being stopped by DPS officers. We ask that you as community members, law enforcement officers and elected officials use this information to increase dialogue in your respective communities and implement more effective and cost-efficient police practices that build stronger relationships of mutual confidence and trust between law enforcement and the community. Alessandra Soler Meetze Executive Director, ACLU of Arizona ameetze@acluaz.org

III Overview of Data

A. Searches
Arizona Department of Public Safety (DPS) policy states that every motor vehicle stop made by an officer must be documented with a ticket, warning or equipment repair order. Under the Arnold v. DPS settlement, these paper records are put into an electronic format and regularly made available to the ACLU for analyses. This procedure facilitates regular tracking of DPS officer activity. According to the electronic records provided by DPS, more than 500,000 stops were made by DPS officers throughout the state of Arizona between July 1, 2006 and June 30, 2007. The DPS dataset also includes information about stops resulting in searches of vehicles and/or people. Search information includes whether a search was conducted, the legal authority allowing for the search, and whether contraband (such as drugs or weapons) was seized during the process of completing the search. This section of the report examines search data resulting from approximately 200,000 stops made by DPS officers along four major interstate highways in Arizona: Interstates 8, 10, 17 and 40. The reason for focusing on interstate searches rather than all searches There should be little to made by DPS officers is because the origins of racial profiling of motorists begin with no variation in searches national Drug Enforcement Agency training programs, such as Operation Pipeline, across racial or ethnic that teach state officers how to engage in drug interdiction activities on the nations groups if everyone is being highways. The law enforcement operation instructs officers to identify profiles of people who engage in drug transportation. It is this so-called drug courier profile that treated equally by law encourages officers to employ racial stereotypes as they engage in drug interdiction enforcement. Yet, DPS activities. search data demonstrates Search data allows for meaningful analysis of whether minority drivers are disproportionately targeted by police. The proportion of people within each ethnic and that minorities were, racial group subjected to a search can be compared against the actual proportion of in fact, being treated people within each group that have been stopped by DPS officers. In this way, search data allows us to affirmatively say who is being searched, why they are being searched unequally during the period analyzed. and what is the outcome of the search. Generally speaking, Hispanics, African Americans, and Native Americans were searched at rates greater than the rate at which whites were searched during the study period. In other words, the likelihood of Hispanics, African Americans, and Native Americans being searched by a DPS officer was much higher than the likelihood of whites being searched during this period. A starting point for understanding the search data is to examine the universe of searches conducted on interstate highways and to ask what proportion of searches were conducted with African Americans, Asians, Hispanics, Middle Easterners, Native Americans, and whites. Table 1 (page 6) demonstrates the percentages by race of the total number of persons searched. The importance of information included in Table 1 is better understood when examined from a different angle. Table 2 (page 6) indicates the percentages of motorists within each ethnic or racial group who were subjected to a search by DPS officers. Looking at the column labeled African American, for example, we see that an average of 10% of African Americans stopped by DPS officers were searched between 2006 and 2007. Hispanics also were searched at an average rate of 10%. On average, Native Americans stopped by DPS officers were searched more frequently than Hispanics or African Americans stopped by DPS Officers. An average of 13% of Native American stops resulted in a search being conducted.

Table 1

Arizona Interstate Highway Searches by Ethnicity/Race as a Percentage of Total Searches

WHITE 39% I-10 99,625

Table 3
AFRICAN AMERICAN 10% ASIAN 1%

AFRICAN AMERICAN ASIAN HISPANIC MIDDLE EASTERN NATIVE AMERICAN

I-8 I-10 W I-10 E I-17 I-40 E I-40 W Average

NATIVE AMERICAN 1% MIDDLE EASTERN 2%

HISPANIC 47%

WHITE Average

I-8

I-10 W 27,877

I-10 W

I-10 E I-8 I-10 W 12% 1% 55% 1% 2% 29% 100% (1,942)

I-17 I-10 E 7% 1% 53% 1% 3% 35% 100% (5,036) I-17

I-40 E I-40 E 14% 1% 30% 2% 15% 39% 1% 102%* (1,796)

I-40 W I-40 W 11% 3% 26% 2% 5% 53% 1% 101%* (1,023)

African American Asian Hispanic Middle Eastern Native American White Unknown Total

10% 1% 47% 2% 1% 39% 100% (467)

8% 1% 40% 1% 7% 43% 1% 101%* (3,007)

*Total exceeds 100% due to rounding.

Table 2

Search Rates by Ethnicity/Race

I-8 African American Asian Hispanic Middle Eastern Native American White Average 5% 1% 3% 6% 2% 2% 3% I-10 W 10% 3% 10% 5% 17% 4% 7% I-10 E 8% 4% 10% 4% 15% 4% 7% I-17 12% 5% 14% 7% 13% 5% 8% I-40 E 13% 3% 11% 7% 10% 5% 7% I-40 W 10% 3% 9% 3% 8% 4% 5% Average 10% 3% 10% 5% 13% 4% 6%

In contrast, white motorists were searched an average of four percent of the time after being stopped. Asians and Middle Easterners like whites were searched less frequently than Hispanics, African Americans, and Native Americans. Compared to average search rates in the double-digits, Asians and Middle Easterners stopped by DPS officers were searched three to five percent of the time, on average. Table 2 data can be translated into a ratio that reflects the likelihood of members of different racial or ethnic groups being searched relative to the rate at which whites were searched. This likelihood ratio speaks directly to the question of whether minorities stopped by DPS officers were treated similarly or differently than whites stopped by DPS officers. Ratios closer to 1:1 indicate relatively similar treatment. The greater the deviation from a ratio of 1:1, the greater the disparity of treatment. This information is presented in Table 3. Taking the average across all roadways examined in this analysis, African Americans stopped by DPS officers were twoand-a-half times more likely than whites to be searched. A similar portrait emerges for Hispanics, who also were consistently more likely than whites to be searched on interstate roadways between 2006 and 2007. On average, Hispanics were twoand-a-half times more likely than whites to be searched by DPS officers on interstates in Arizona. Search rates for Native Americans exceeded search rates for whites on every roadway except Interstate 8. In fact, comparative search rates for Native Americans on some roadways are the highest seen in this analysis. Native Americans driving along Interstate 10 West were more than four times more likely than white motorists to be searched by DPS officers. On average, Native Americans stopped by DPS officers were more than three times as likely as whites stopped by DPS officers to be searched between 2006 and 2007. While few Middle Easterners were stopped by DPS officers and subjected to searches, the likelihood of a stop of a Middle Eastern motorist resulting in a search was also higher than the likelihood of a stop with whites resulting in a search. This analysis of search rates demonstrates that people stopped by DPS officers on Arizona interstate highways were treated differently based on their race and ethnicity during the study period. However, some people could suggest that proof of differential search rates makes for an incomplete analysis. Differential search rates, some may argue, are justified if people from specific racial and ethnic groups transport contraband more frequently than others. The next section of the report examines the relationship between search rates and seizures of contraband within the DPS data for people from different racial and ethnic groups.

Likelihood of Being Searched Relative to White Search Rate

AVERAGE AFRICAN AMERICAN ASIAN HISPANIC MIDDLE EASTERN NATIVE AMERICAN WHITE Average I-8 AFRICAN AMERICAN ASIAN HISPANIC MIDDLE EASTERN NATIVE AMERICAN WHITE Average I-10 W AFRICAN AMERICAN ASIAN HISPANIC MIDDLE EASTERN NATIVE AMERICAN WHITE Average I-10 E AFRICAN AMERICAN ASIAN HISPANIC MIDDLE EASTERN NATIVE AMERICAN WHITE Average I-17 AFRICAN AMERICAN ASIAN HISPANIC MIDDLE-EASTERN NATIVE AMERICAN WHITE Average I-40 E AFRICAN AMERICAN ASIAN HISPANIC MIDDLE EASTERN NATIVE AMERICAN WHITE Average I-40 W AFRICAN AMERICAN ASIAN HISPANIC MIDDLE EASTERN NATIVE AMERICAN WHITE Average 1.0 1.25 0.75 2.0 0.75 2.2 2.5 1.0 1.4 1.4 2.0 0.6 2.2 2.6 1.0 1.6 1.4 2.6 1.0 2.8 2.4 1.0 1.75 1.0 3.75 1.0 2.5 2.0 1.0 1.75 1.25 4.25 0.75 2.5 2.5 1.0 1.0 1.5 0.5 1.5 3.0 2.5 1.0 1.5 1.25 3.25 0.75 2.5 2.5

7 66

B. Seizure of Contraband
Some analysts argue that disproportionate search rates are justified if there is an established pattern of people from one racial or ethnic group transporting contraband at disproportionately higher rates than people from another group. Within this scenario, higher search rates would be expected to yield a relatively larger proportion of seizures of contraband. For example, the data demonstrates that Hispanics stopped by DPS officers were, on average, two-and-a-half times more likely than whites to be searched. If Hispanics were shown to be transporting contraband two-and-a-half times as often as whites, some might suggest that higher The current study finds search rates are nothing more than good police that people of color were work. searched more frequently As seen in Table 4 (page 10), contraband was actually found during 34% of searches conducted than whites (at least twice with whites. Contraband seizure rates for whites as often, in most cases) ranged from 24% (Interstate 10 East and West) to on Arizona Interstate 50% (Interstate 40 West). Searches of Hispanics highways between 2006 resulted in the seizure of contraband less frequently than searches of whites on all interstate and 2007. However, white motorists were more likely segments. The average rate of contraband seizure for searches conducted with Hispanics during the to be carrying contraband last year was 22%, with seizure rates ranging from than Hispanics, Middle 15% (Interstate 17) to 33% (Interstate 40 West). African Americans were at least twice as likely Easterners and Native as whites to be searched on all six interstate Americans. segments, despite the fact that the rate of contraband seizures for African Americans and whites was somewhat similar (38% compared to 34%, respectively). On closer inspection, the higher overall rate of seizures for African Americans was largely driven by higher seizure rates on two specific interstate segments (Interstates 8 and 40 East). The so-called hit rates for African Americans were within one percentage point of hit rates for whites on three of six interstate segments and substantially lower than the white hit rate on one interstate segment. A similar portrait of contraband seizures from searches conducted with Middle Eastern motorists is evident in the data. While Middle Easterners were searched more frequently than whites on four of six interstate segments, contraband was found during searches of Middle Easterners less frequently on those four interstate segments than was found during searches of whites. The average hit rate for Middle Easterners was 24%, compared to 34% for whites. Native Americans were searched more frequently than whites on five of six interstate segments. In one case, Native Americans were searched more than four times as often as whites (Interstate 10 West).

I was ordered out of the vehicle at gun point. I complied with all the commands that the DPS officers issued to me. I exited the vehicle with both of my hands in the air. Two DPS officers had their handguns pointed at me I was wearing blue jeans, a white t-shirt and an unbuttoned long sleeve shirt. The officers, with their weapons still drawn, had me turn around with my hands in the air and walk back to them. I complied with all of their orders. At the time of the incident, we were embarrassed about how the vehicle stop transpired and only wanted to leave the scene. I believe the actions of the officers were racially motivated.

Black female, 32, of Tempe. She is a police officer with the City of Tempe who was pulled over by several DPS officers in December 2004. She and her partner, both of whom were working undercover in Mesa, were stopped at gunpoint because the license plate on the unmarked police car was registered to a Dodge. They were driving a Chevy Impala. No citation or warning was issued.

Table 4

Rate of Contraband Seizures (Hit Rate) by Ethnicity /Race

AVERAGE 38% 26% 22% 24% 26% 34% 28% AFRICAN AMERICAN ASIAN HISPANIC MIDDLE EASTERN NATIVE AMERICAN WHITE Average I-8 55% 0.0% 16% 30% 50% 34% 27% AFRICAN AMERICAN ASIAN HISPANIC MIDDLE EASTERN NATIVE AMERICAN WHITE Average I-10 W 23% 19% 18% 0.0% 23% 24% 20% AFRICAN AMERICAN ASIAN HISPANIC MIDDLE EASTERN NATIVE AMERICAN WHITE Average I-10 E 25% 9.0% 19% 15% 16% 24% 21% AFRICAN AMERICAN ASIAN HISPANIC MIDDLE EASTERN NATIVE AMERICAN WHITE Average I-17 28% 30% 15% 17% 16% 27% 22% AFRICAN AMERICAN ASIAN HISPANIC MIDDLE EASTERN NATIVE AMERICAN WHITE Average I-40 E 52% 56% 28% 39% 15% 42% 36% AFRICAN AMERICAN ASIAN HISPANIC MIDDLE EASTERN NATIVE AMERICAN WHITE Average I-40 W 42% 43% 33% 45% 33% 50% 44% AFRICAN AMERICAN ASIAN HISPANIC MIDDLE EASTERN NATIVE AMERICAN WHITE Average

Table 5

Stop Length (Minutes:Seconds) Yet, on the same five segments, rates of contraband seizures from Native American searches were lower than hit rates for whites. Native American hit rates on these segments ranged from 15% (Interstate 40 East) to 33% (Interstate 40 West). Across all six interstate segments studied in this analysis, the average rate of contraband seizures for Native Americans was 26%, compared to 34% for whites. The results of the present Arizona study are consistent with the national literature that indicates Hispanics, African Americans and members of other racial and ethnic minorities transport drugs at lower rates than whites, yet are searched at higher rates. In a 2002 study conducted by the United States Department of Justice, African Americans and Hispanics were shown to be searched more frequently than whites across the nation.1 This study found that 8.1% of African American motorists stopped by police and 8.3% of Hispanics stopped were searched following the stop. At the same time, only 2.5% of whites stopped by police were searched. Contraband was discovered in 3.3% of African American searches and 13.0% of Hispanic searches, as compared to a higher 14.5% for whites. Seven studies completed around the country in recent years have reached similar conclusions: people of color are disproportionately targeted by police even though theyre less likely than whites to be involved in criminal activity.2 An analysis from 1999, conducted by the Attorney General of New Jersey, determined that discretionary consent searches of cars on the turnpike were even more racially disparate than the initial stops: 77.2% of all consent searches were of African Americans and other minorities.3 An extensive study of 175,000 pedestrian stops by the New York City Police Department found a highly disproportionate rate of minority stops. The Office of the Attorney General of New York State, which conducted the research, determined that: (1) African Americans were stopped six times more frequently than whites; (2) African Americans were stopped at a rate more than ten times their percentage of the population; and (3) stops of African Americans were less likely to result in arrests than stops of whites. Even when adjusting for crime rates by race, the differences in stops of minorities compared to stops of whites was statistically significant, with African Americans stopped twice as often as whites.4 One additional way of understanding how minorities are treated by DPS officers is to examine the average length of time members of different racial and ethnic groups are stopped by police. If people are being treated equally by police, then the average stop lengths should be roughly equal. Variations in the duration of stops taking place by DPS officers would indicate that racial and ethnic minorities are, in fact, being treated differently. An analysis of stop duration follows.
0

C. Duration of Stops
Like the analysis of search rates and contraband seizures, a comparative analysis of stop duration allows us to understand if people of color were treated differently by DPS officers. This section of the report looks at how long motorists were detained after they were stopped by DPS officers. It is reasonable to assume that stop duration should be roughly equivalent regardless of ones racial or ethnic identity. There should be little Previous studies conducted to no variation in stop duration across population subgroups if everyone is in Arizona and other parts being treated equally. of the country have shown Overall, minorities were detained for longer periods than white drivers (Table that law enforcement 5). For all six interstate segments studied actions are regularly here, African Americans and Hispanics directed at minorities in a were stopped for longer periods of time than whites. Middle Easterners, on disproportionate manner, average, were stopped for dramatically despite the fact that traffic longer periods of time than whites. violations committed by DPS records stop duration in ten minorities are no greater minute increments (i.e., 0-10 minutes, 11-20 minutes, 21-30 minutes, 31-50 than those committed by minutes, and 51+ minutes). The analysis whites. that follows assumes the midpoint for each ten minute increment. In other words, if a stop is recorded as taking between 11-20 minutes, this analysis assumes the stop was 15 minutes in length. The previous analysis in Table 2 shows that DPS officers are searching African Americans, Hispanics, Middle Easterners, and Native Americans at a disproportionate rate compared to whites. It is fair to assume that stops involving searches are longer in length than stops that do not involve a search. Including stops with searches in this duration analysis would exaggerate the findings and obscure the central question: Are people of color being treated differently by DPS officers? To avoid double-counting a bias already known to exist, the stop duration analysis provided here only involves stops that did not result in a search. This analysis of stop duration demonstrates that minorities, other than Native Americans, were consistently stopped and detained by DPS officers for longer periods of time than whites during the time frame of this study. This finding supports the overall conclusion of this analysis that racial and ethnic minorities were treated differently on Arizona interstate highways during the study period.

NOT INCLUDING STOPS IN WHICH SEARCHES WERE CONDUCTED AVERAGE AFRICAN AMERICAN ASIAN HISPANIC MIDDLE EASTERN NATIVE AMERICAN WHITE Average I-8 AFRICAN AMERICAN ASIAN HISPANIC MIDDLE EASTERN NATIVE AMERICAN WHITE Average I-10 W AFRICAN AMERICAN ASIAN HISPANIC MIDDLE EASTERN NATIVE AMERICAN WHITE Average I-10 E AFRICAN AMERICAN ASIAN HISPANIC MIDDLE EASTERN NATIVE AMERICAN WHITE Average I-17 AFRICAN AMERICAN ASIAN HISPANIC MIDDLE EASTERN NATIVE AMERICAN WHITE Average I-40 E AFRICAN AMERICAN ASIAN HISPANIC MIDDLE EASTERN NATIVE AMERICAN WHITE Average I-40 W AFRICAN AMERICAN ASIAN HISPANIC MIDDLE EASTERN NATIVE AMERICAN WHITE Average -1:42 -0:42 17:12 +2:24 +2:00 +1:06 +3:30 +0:12 -0:18 15:42 +0:24 +8:12 +0:30 +2:12 -2:00 -0:48 17:00 +2:18 +2:54 +3:06 +4:24 -0:30 -1:54 17:12 +2:06 +6:06 +3:06 +3:54 -0:54 -0:48 18:36 +0:42 -0:42 +1:48 +0:30 +0:12 -1:06 15:54 0:00 +2:24 +6:06 +3:12 -0:42 -0:54 16:54 +2:02 +1:43 +1:49 +4:48

10

11

D. Stops
i. Methodology involved in the collection of stop data This section of the report examines traffic stops made by DPS officers on interstate highways in Arizona. There is some debate as I-8 I-10 W I-10 E I-17 to whether stop data alone indicate Table 6 biased decision-making by law DPS Stops by Interstate enforcement. One reason is that there are no comparative figures on the I-8 16,265 actual number of people using highways I-40 42,991 nor is there data on the number of I-8 16,265 traffic violators by race. However, I-40 42,991 both national and local studies have found that law enforcement actions are regularly directed at minorities in a I-17 disproportionate manner, despite the 35,372 I-10 fact that traffic violations committed 99,625 by minorities are no greater than those I-17 35,372 committed by whites. For example, I-10 99,625 an extensive study conducted in New Table 7 Jersey found that although 98% of Stops drivers traveling along the Turnpike were, DPSI-40 W I-8 18,370 by Interstate 16,265 Segment in fact, speeding, black drivers were I-8 several times more likely to be stopped I-10 W I-8 I-40 W I-40 E 16,265 27,877 18,370 than white drivers.5 24,595 I-10 W I-8 Although this report does not I-10 W I-40 E 27,877 construct a comparative benchmark, 24,595 I-10 W Arizona researchers have previously I-17 measured populations of traffic 35,372 violators. For example, two studies I-17 conducted in 2004 of traffic violators I-10 E 35,372 71,558 I-10 E on 1-40 in Coconino County and I-17 in Yavapai County found that the stop rate I-10 E 71,558 I-10 E for African Americans was on average 4.8%, yet African Americans made up only 2.0% of the violator population. Hispanics were targeted at considerably higher rates. They were stopped an average of 17% of the time in both studies, despite the fact that they committed traffic violations only 8% of the time.6 Since then, no new similar studies have been conducted in Arizona. This section of the report focuses solely on who was stopped along interstate highways. As shown in Table 6, slightly more than 194,000 stops were made on all four major interstate highways in Arizona between July 1, 2006 and June 30, 2007. To hone this analysis further, we break out the data by specific interstate segments in Table 7.
I-8 I-10 W I-10 E I-17 I-40 E I-40 W

I-40 E

I-40 W

I-40 W

In my mind those sent to protect me came instead to persecute me. I was visually raped, unconstitutionally violated, and racially profiled. Robbed of my innocence and branded with GUILT without validation or cause for reason. Looking down after my release from jail at a ticket that read IMPROPER RIGHT TURN. Huh? The whole ordeal has left me feeling UNAMERICAN.

I-40 E

I-40 W

I-40 E I-17

I-17

Black female, 27, of Glendale. She is a social worker who was pulled over by DPS officers while driving westbound on Peoria Avenue near I-17. She was stopped on January 4, 2007, for making an improper right turn and then strip-searched in front of male and female officers who were convinced she was concealing drugs. She spent a night in jail for obstruction of justice and possession of drug paraphernalia. Those drugrelated criminal charges were then dismissed, and she was issued a $115 fine for making an improper turn.

12

13

ii. Analysis of highway stops DPS records indicate that people of different racial and ethnic groups were stopped in different proportions on each stretch of highway examined in this analysis (Table 8). For example, there was large variation in the proportion of Hispanics stopped within the six interstate segments. Fifteen percent of stops on I-40 West and 19% of stops on I-40 East involved Hispanics. On the other side of the spectrum, 37% of stops on I-10 West, 36% of stops on I-8, and 35% of stops on I-10 East involved Hispanics. Hispanics were stopped at disproportionately higher rates on southern interstate segments than on northern interstate segments.
Table 8

V Background: Arnold v. Arizona DPS

In 2001, the ACLU of Arizona, in conjunction with Flagstaff attorney Lee Phillips, filed a class action lawsuit on behalf of African American and Hispanic motorists who were stopped and searched by DPS officers simply because of the color of their skin. The lawsuit challenged the discriminatory practice of racial profiling along Arizonas streets and highways and called for the implementation of a statewide system of record keeping to monitor all traffic stops and vehicle searches made by DPS officers. The ACLU maintained that data collection is necessary to document the problem of racial profiling because it provides police departments and community members with information about the types of stops being made by officers and the results of such stops, and identifies potential police misconduct. Throughout the course of the litigation, ACLU lawyers on the case were able to obtain more than 300,000 DPS documents, including traffic citations, traffic warnings and repair orders all of which contained information about stops, searches, detentions and arrests. At the same time, the Social Research Laboratory at Northern Arizona University was able to conduct a study of motor vehicle law violators. Dr. Frederic Solop then conducted a racial profiling analysis comparing the race and ethnicity of people stopped by DPS on Interstate 40 in Coconino County with characteristics of the population violating motor vehicle laws (the violator population) and eligible to be stopped. This report, published in 2004 and presented as part of the class action lawsuit, concluded that: Hispanics and African Americans are consistently being stopped by DPS officers at rates disproportionately greater than their representation within the violator population; and, white, non-Hispanics are consistently being stopped at rates disproportionately less than their representation within the violator population. These differences are statistically significant and fit the Supreme Courts definition of the presence of racial and ethnic discrimination. After years of litigation, the case came before the United States Court of Appeals for the Ninth Circuit. At that point, DPS and the State of Arizona opted to move forward with mediation, rather than pursue additional litigation. A federal magistrate approved the settlement agreement on Friday, July 28, 2006, ending a five-year legal battle between the ACLU and DPS. The so-called Arnold Settlement (referring to the last name of the lead plaintiff) has had serious implications for the State of Arizona. It requires DPS to make numerous internal and external changes all of which were intended to eliminate racial profiling by DPS officers by specifically addressing three broad areas: (1) policies and procedures; (2) data collection and analysis; and (3) the formation of a citizens advisory board. As a result of the settlement, DPS is now required to collect and store data relevant to the nature, duration, and grounds for all police-citizen contacts. Furthermore, the agreement stipulates that consent searches must be authorized with a written consent form that is available in English and Spanish and mandates that all officers attend a racial profiling training session. One of the more important policy changes that resulted from the settlement was that DPS agreed to install video systems in all DPS patrol cars and record all traffic stops conducted on the highways.7 To protect motorists from racially-biased policing, the agreement also compels DPS to perform semi-annual spot checking procedures to ensure the data is being reliably and appropriately recorded and scanned. In addition, DPS has hired an outside consultant to review and analyze the statistical data generated from the traffic-stop forms to monitor incidents of racial profiling. If the statistical data suggests that a particular officer is engaged in racial profiling, DPS must take corrective and/or disciplinary measures to end the behavior.

Arizona Interstate Highway Stops by Ethnicity/Race as a Percentage of Total Stops

I-8 African American Asian Hispanic Middle Eastern Native American White Unknown Total 5% 2% 36% 1% 1% 55% 100% I-10 W 8% 3% 37% 1% 1% 50% 1% 101%* I-10 E 6% 2% 35% 1% 1% 54% 1% 100% I-17 6% 2% 23% 1% 4% 64% 1% 101%* I-40 E 7% 3% 19% 2% 10% 59% 1% 101%* I-40 W 6% 5% 15% 4% 3% 67% 1% 101%*

*Total exceeds 100% due to rounding.

While interesting, it is difficult to draw definitive conclusions from this stop data alone. Scholars agree that in the absence of reliable information about the racial and ethnic composition of people who violate traffic laws on the highway, stop data itself does not indicate whether one group is being stopped at disproportionately large or small rates. Other proposed sources of benchmark information such as census data or accident reports do not have proven reliability for application in the context of understanding stops on rural interstate highways.

IV Conclusion
The Arizona Department of Public Safety has been recording information about all stops and searches conducted by DPS officers in compliance with terms from the settlement of Arnold v. Arizona DPS. The analysis presented here is centrally concerned with understanding whether various racial and ethnic subgroups were treated differently by DPS officers during the first year of settlement compliance. More specifically, this study examines information from the almost 200,000 stops made by DPS officers on Arizona interstate highways. This analysis demonstrates that African Americans, Hispanics and Native Americans were searched more frequently than whites and that these search rates are not justified by rates of contraband seizures. Also, African Americans, Hispanics and Middle Easterners were detained for longer periods of time when stopped by DPS officers. After analyzing search rates, rates of contraband seizures, and duration of stops, one can only reasonably conclude that minority motorists were treated differently by DPS officers during the study period in a manner consistent with allegations of racial profiling. 14

15

Lastly, the Citizens Traffic Stop Advisory Board has been created to review DPS practices, policies, and procedures related to racial profiling and the settlement agreement. The advisory board consists of nine members appointed by the Governor, including three nominated by the ACLU. The board has been granted full access to any DPS records and/or documents they deem necessary to make informed assessments and recommendations.

VI Recommendations from the ACLU of Arizona

Despite the historic settlement agreement in Arnold v. Arizona DPS, we are nowhere near declaring mission accomplished. The fact is: racial profiling in Arizona as this analysis clearly shows continues in Arizona. In the first place, the state refuses to admit there is a problem. A 233-page report released by DPS in November 2007 concurred for the most part with the ACLUs own analysis presented here.8 DPS found that highway patrol officers operating throughout Arizona were more than twice as likely to search vehicles driven by Hispanics and African Americans than those operated by whites between 2006 and 2007, and that minority motorists were far more likely to be arrested and to be hit with multiple traffic citations. However, the states own researcher Dr. Robin S. Engel said the ethnic-enforcement disparities may be explained by legitimate factors unmeasured by these data. She cited factors such as the severity of the traffic offense, motorist attitudes and socioeconomic status, adding that: Until I can get into the mind of an officer, I cannot determine whether he or she is making stops based on race. This has become a standard response from state-supported researchers, and reflects an environment where racial profiling has become part of the culture of law enforcement and society as a whole. Studies confirm what we all know: Most people speed or commit other traffic infractions while driving on our highways. So why is it acceptable for law enforcement to continue targeting minorities for race-based stops? The ACLU believes, quite simply, that the practice of treating motorists differently based on their race or ethnicity is unconstitutional. Law enforcement should only use race in limited circumstances when there is a specific description identifying a particular suspect by race. In this report, Dr. Frederic Solop has documented the ongoing problem of racial profiling in Arizona and deepened the publics understanding of the issue. But, where do we go from here? Law enforcement officials, citizens groups, legislators and public officials should consider the following recommendations for eliminating racial profiling in Arizona.

Clearly it was racial profiling because they were targeting corn vendors who fit a certain profile, and not the hot dog vendors, for example. Corn vendors are from South America. Theyre mostly Mexican; some are from Honduras or El Salvador. The officers would just surround them in their police cars, start questioning them and then just round them up in handcuffs for operating without a license. Most of the people who were rounded up ended up being deported, but there were a few who were U.S. citizens, who were also arrested.

Latino union organizer, 38, of Phoenix, describing a joint operation between Maricopa County Sheriffs Office and the Department of Public Safety to round up and deport corn vendors in Phoenix, despite repeated assertions from Governor Janet Napolitano that DPS is not going to be engaged in roundups. Maricopa County Sheriffs officers, along with Department of Public Safety officers, have been trained by Immigration and Customs Enforcement to enforce immigration laws.

16

17

1. Ban consent searches. DPS should implement a policy that greatly restricts or prohibits the use of consent searches the practice of allowing police to use discretion to search completely innocent people, even when there is no evidence of criminal activity, as long as people are persuaded to give their consent. This practice is bad policy because it means police are wasting precious time and resources going after people based solely on race rather than evidence of real criminal activity. Further, it has long been demonstrated that consent in the context of police presence is illusory at best. 2. Establish meaningful internal and external accountability. Police should be held accountable for their actions. To ensure effective internal accountability procedures, DPS should conduct regular reviews of stop, search and citation data, identify problem officers through early intervention systems, and discipline them when required. External accountability procedures also are important in serving community needs. Victims of racial profiling rarely file complaints with their local or state police departments because of lack of trust and fear of retaliation. And when they do, most of those complaints are routinely dismissed as being unfounded. For example, all of the 19 racial profiling complaints filed with DPS in Arizona since the July 2006 settlement agreement have been dismissed because the agency found that there was no wrongdoing on the part of the officer. The Citizens Traffic Stop Advisory Board is authorized to hear from individuals who have experienced racially motivated harassment at the hands of state police. Although board members are limited in their ability to take action on specific complaints, they should regularly follow up with motorists to verify the accuracy of the DPS data collection process. A well-funded, well-staffed police oversight board, located within state government, but designed as an independent body, is also needed. It should review the actions of local departments, as well as the Department of Public Safety, and have subpoena power and the ability to ensure confidentiality. 3. Secure additional funding for audio-visual equipment for all DPS vehicles. When the settlement agreement was approved, DPS agreed to seek funding to install audio-visual (A/V) equipment in their vehicles. The use of A/V equipment and the retention of tapes would document searches and seizures, allow supervisors to monitor the conduct of officers and help explain disparities, and boost the publics confidence in law enforcement. It has proven to be a critical deterrent throughout the country. However, DPS has been unable to obtain the necessary funding to equip all of their patrol cars with video cameras. They should apply for federal grants to purchase cameras, and make the installation of video equipment in patrol cars a high priority. All funding for A/V equipment should be linked to a proportionate increase in money to improve the data collection process and accountability measures. 4. Enact state legislation on racial profiling for all state and local law enforcement agencies. The Arizona Legislature should pass, and the Governor should sign, a racial profiling bill that requires mandatory data collection regarding: (a) the date, time and location of the stop; (b) make and model of the vehicle and whether the motorist was local or from out-of-state; (c) the race and ethnicity of the motorist; (d) the reason for the stop; (e) the result of the stop i.e. whether a ticket was issued or an arrest was made, or even whether the driver was stopped and let go with a warning; (f) whether a search was conducted; (g) the type of search i.e. probable cause, consent, or inventory search after an arrest is made; (h) what, if anything, was found in the course of the search; (i) officer badge number or individual identifier; and (j) passenger activity if any. These ten categories of data are essential to determine the extent of racial profiling in any community. In addition, any bill should create a uniform, standardized reporting format (like a template or chart) to assist agencies in their data collection and mandate the creation of a statewide agency that will serve as a repository for racial profiling reports and oversee compliance with the law. Democratic and Republican governors have signed mandatory data collection bills in numerous other states, including Texas, Kansas, Minnesota, Missouri, North Carolina, Maryland, Rhode Island, Tennessee and Washington.

Endnotes
1 2 U.S. Department of Justice, Contacts Between Police and the Public: Findings from the 2002 National Survey, Bureau of Justice Statistics. Found on the Internet at www.ojp.usdoj.gov/bjs/pub/pdf/cpp02.pdf. Minnesota Department of Public Safety (MDPS) 2003 Minnesota Racial Profiling Report (found that blacks and Hispanics are more likely to be searched than whites during a traffic stop, but searches of whites are more likely to produce contraband). McCorkle, R.C. 2003 A.B. 500: Traffic Stop Data Collection Study. Carson City, NV: Office of the Attorney General (found that blacks searched at more than twice the rate of white drivers but the hit rate for blacks and Hispanics was lower than for whites and Asians). Zingraff, M.T. Mason, H.M. Smith, W.R., Tomastovic-Devey, D. (found that blacks are more likely to be searched than whites but contraband is less likely to be found in searches of vehicles operated by black drivers). Lamberth, J. 2003, Racial Profiling Data Analysis, final report for the San Antonio Police Department. Chadds Ford, PA Lamberth Consulting (found that black and Hispanic drivers are more likely to be searched than white or Asian drivers yet contraband is consistently found at lower rates for black and Hispanic drivers). Washington State Police, (WSP), 2001. Report to the Legislature on Routine Traffic Stop Data. Olympia: Washington State Police (found that nonwhite minorities are searched at a disproportionately higher rate than whites). Peter Verniero, Attorney General of New Jersey, Interim Clerk of the State Review Team Regarding Allegations of Racial Profiling 26-28 (April 20, 1999), available at http://www.state.nj.us//lps.intm419.pdf. Office of the Attorney General of New York State, Report on the New York City Police Departments Stop and Frisk Practices (1999) available at http://www.oag.state.ny.us//press/reports/stop_frisk/stop_frisk.html. Revised Statistical Analysis of the Incidence of Police Stops and Arrests of Black Drivers/Travelers on the New Jersey Turnpike Between Exits or Interchanges 1 and 3 From the Years 1991 Through 1998 http://www.lamberthconsulting. com/downloads/new_jersey_study_report.pdf. Solop, Frederic I. 2004. Statistical Analysis of I-40 Stop Data and I-40 Violator Data: Coconino County, Arizona. Both studies are available online at: http://www.acluaz.org/News/PressReleases/07_28_06.html. These are just a handful of the mandatory changes DPS must make under the settlement. For a full list please see the settlement agreement in full at: http://www.aclu.org/racialjustice/racialprofiling/16036lgl20050202.html. That report, conducted by the University of Cincinnati Policing Institute, is available online at: http://www.azdps.gov/ agreement/pdf/ADPS_Traffic_Stop_Data_Analysis_Study_Year_1_Final_2007_11.07.07.pdf.

3 4 5

6 7 8

18

19

20

13

ACLU of Arizona P.O. Box 17148 Phoenix, AZ 85011 602-650-1967 www.acluaz.org