Concerns by the citizens of Missouri and the Missouri legislature regarding allegations of racial profiling by law enforcement prompted the passage of state law Section 590.650, RSMo (2000), which was enacted Aug. 28, 2000. Racial profiling has been defined as the inappropriate use of race by law enforcement when making a decision to stop, search or arrest a motorist.
Missouri’s state law requires that all peace officers in the state report specific information including a driver’s race for each traffic stop made in the state. Law enforcement agencies are required to turn in the data to the Attorney General, and the Attorney General is required to compile the data and report to the Governor no later than June 1 of each year. The law allows the Governor to withhold state funds for any agency which does not comply with the law. State law requires that all information be reported to the Attorney General’s Office by March 1.
The summary of statewide racial profiling data has been provided by Scott H. Decker, curator’s professor of criminology and criminal justice; Richard Rosenfeld, professor of criminology and criminal justice; and Jeffrey Rojek, doctoral candidate in criminology and criminal justice; all at the University of Missouri-St. Louis.
Table 1. 2004 statewide summary of results
|Statewide Metrics||Total||White||Black||Hispanic||Asian||Am. Indian||Other|
|Statewide population %||100%||85.24%||10.25%||1.85%||1.17%||.41%||1.08%|
|Disparity index||– – –||.97||1.34||1.07||.67||.20||.63|
|Contraband hit rate||20.61%||22.43%||15.33%||14.36%||14.84%||13.76%||15.80%|
|Notes: Population figures are from the 2000 census for persons 16 and older who designated a single race. Hispanics may be of any race. Other includes persons of mixed race and unknown race.
Disparity index = (proportion of stops / proportion of population). A value of 1 represents no disparity; values greater than 1 indicate over-representation, values less than 1 indicate under-representation.
Search rate = (searches / stops) X 100.
Contraband hit rate = (searches with contraband found / total searches) X 100.
Arrest rate = (arrests / stops) X 100.
II. STATEWIDE FINDINGS
The 2004 Annual Report on Missouri Traffic Stops summarizes the data from 603 law enforcement agencies in Missouri for 2004. This represents 83 percent of the 728 law enforcement agencies in the state.
The agencies filing reports recorded a total of 1,365,575 traffic stops, resulting in 106,430 searches and 74,484 arrests. Table 1 breaks out the stops, searches and arrests by race and ethnic group.
Four summary indicators are included in Table 1 that may be useful in initiating further assessments of racial profiling in traffic stops. The first, termed the “disparity index,” relates each group’s proportion of total traffic stops to its proportion of the driving-age population 16 and older.
A value of 1 on this index indicates that a group’s proportion of traffic stops equals its population proportion: it is neither “under-represented” nor “over-represented” in traffic stops. Values above 1 indicate over-representation, and those below 1 indicate under-representation. For example, the 1,129,748 whites who were stopped accounted for 82.7 percent of all traffic stops in 2004.
According to the 2000 census, whites comprise 85.2 percent of Missouri’s driving-age population. The value for whites on the disparity index is, therefore, .827/.852 or .97. Whites were stopped, in other words, at just slightly below the rate we would expect based on their fraction of the population.
The same is not the case for several of the other groups. African-Americans represent 10.3 percent of the population 16 and older but 13.7 percent of all traffic stops, for a value on the disparity index of 1.34.
African-Americans were stopped at a rate 34 percent greater than expected based solely on their proportion of the population 16 and older.
Hispanics were stopped at a rate slightly above their population proportion, and Asians, American Indians and persons of mixed or unknown race were stopped at a rate below their proportion of the driving-age population.
The values on the disparity index for the groups can be compared directly to one another. For example, the likelihood that an African-American motorist was stopped is 1.38 times that of a white motorist (1.34/.97). In other words, African-Americans were 38 percent more likely than whites to be stopped in 2004.
The disparity index is a gauge of the likelihood drivers of a given race or ethnic group are stopped based on their proportion of the residential population 16 and older, and not of the population of motorists on the state’s streets, roads, and highways.
A group’s share of the residential population 16 and older may or may not equal its proportion of drivers. Although in most instances the two proportions should be close, that may not always be the case.
The extremely low index values for American Indians, for example, could indicate that they are under-represented among the state’s motorists. In addition, motorists from other states are stopped on Missouri’s roadways. To the extent that out-of-state drivers do not reflect the race and ethnic composition of Missouri’s population, the disparity index will not accurately portray the probability that Missouri residents are stopped.
The second indicator that can be used to assess racial profiling is the “search rate,” or the number of searches divided by the number of stops (x 100). (Searches include searches of drivers or property in the vehicle.)
The search rate for all motorists who were stopped is 7.79 percent. Asians were searched at a rate well below the statewide average, and African-Americans, Hispanics and American Indians were searched at rates above the average for all motorists stopped.
The search rate for the groups also can be compared directly with one another. African-Americans were 1.71 times more likely to be searched than whites (12.02/7.03). Hispanics were 1.84 times more likely than whites to be searched (12.93/7.03).
The reasons for conducting a search and the outcome of the search (such as finding contraband) should be considered when making comparisons across groups.
Some searches are conducted with the consent of the driver, or because the officer observed suspected contraband in plain view, had reasonable suspicion that an individual may possess a weapon (Terry search), or other reasons. These searches may or may not result in an arrest.
Other searches are conducted incident to arrest, which means there is no other reason given for the search other than arrest. These searches are almost always performed when there is an outstanding warrant.
The third summary indicator, the “contraband hit rate,” reflects the percentage of searches in which contraband is found. Contraband was found in 20.6 percent of all searches conducted. However, there is considerable variation in the contraband hit rate across race and ethnic groups.
The contraband hit rate for whites was 22.4 percent, compared with 15.3 percent for African-Americans and 14.4 percent for Hispanics. This means that on average searches of African-Americans and Hispanics are less likely than searches of whites to produce contraband. This difference is most likely attributable to the higher arrest rates for African-Americans and Hispanics, circumstances that compel a search.
The “arrest rate” is the fourth summary indicator included in Table 1 that may be useful for assessing racial profiling. Just over 5 percent of all traffic stops resulted in an arrest (74,484/1,365,575). The probability of arrest varies across the race and ethnic groups.
Between 8 percent and 9 percent of the stops of African-Americans and Hispanics resulted in arrest, compared with 5 percent of the stops of whites. African-Americans and Hispanics are stopped and searched more often than whites, but they also are arrested more often.
This year’s report compares (PDF file 19,722K) the 2004 disparity index to the disparity indexes for 2000-2003.
For each agency, the disparity index for each race-ethnic group is presented for 2000-2003. For the state as a whole, the key indicators show very small changes between 2003 and 2004. The disparity index is highly stable across the two years. The search rate (percentage of stops in which a search is conducted) for American Indians drops to 9.7 percent in 2004 from just below 12 percent in 2002.
The arrest rate for American Indians also falls from 9.5 percent in 2003 to 6.9 percent in 2004. The contraband hit rate (percentage of searches in which contraband is found) for persons of other or mixed race fell from 22.7 percent to 15.8 percent.
A reasoned determination of the existence of racial profiling in a community requires a comprehensive evaluation of the full range of information compiled in the agency reports. This brief summary of selected indicators for the state as a whole is intended to stimulate those local evaluations and dialogue.
|Table 2. Agencies that did not submit reports as required by state law|
|*Denotes agencies that submitted reports, but after the deadline. (Section 590.650, RSMo)|
|Altenburg Police Dept.||Anniston Police Dept.||Atlanta Police Dept.|
|Brunswick Police Dept.||Bull Creek Village Police Dept.||Burlington Junction Police Dept.|
|Butterfield Police Dept.||Callao Police Dept.||Camden Point Police Dept.|
|Canalou Police Dept.||*Chariton County Sheriff’s Dept.||*Conway Police Dept.|
|Dade County Sheriff’s Dept.||Delta Police Dept.||Diamond Police Dept.|
|East Lynne Police Dept.||*Edgerton Police Dept.||Elsberry Police Dept.|
|Eminence Police Dept.||Everton Police Dept.||Forest City Police Dept.|
|Gainesville Police Dept.||Glen Echo Park Police Dept.||Glenwood Police Dept.|
|Gower Police Dept.||Gravois Mills Police Dept.||Greendale Police Dept.|
|Hale Police Dept.||Hardin Police Dept.||Hawk Point Police Dept.|
|Higbee Police Dept.||Homestown Police Dept.||Hornersville Police Dept.|
|Howardville Police Dept.||*Huntsville Police Dept.||Jasper Police Dept.|
|Kidder Police Dept.||King City Police Dept.||Kinloch Police Dept.|
|Lilbourn Police Dept.||Lowry City Police Dept.||Maysville Police Dept.|
|Mayview Police Dept.||Mineral Point Police Dept.||Montrose Police Dept.|
|Neelyville Police Dept.||New Cambria Police Dept.||Newburg Police Dept.|
|*Novinger Police Dept.||Oran Police Dept.||Pleasant Hope Police Dept.|
|Portage Des Sioux Police Dept.||Rayville Police Dept.||Revere Police Dept.|
|Stanberry Police Dept.||Stewartsville Police Dept.||Stockton Police Dept.|
|Sunrise Beach Police Dept.||Tallapoosa Police Dept.||Taos Police Dept.|
|Trimble Police Dept.||Twin Bridges Police Dept.||Unity Village Police Dept.|
|Walker Police Dept.||Wardsville Police Dept.||Washburn Police Dept.|
|*Waverly Police Dept.||Wayland Police Dept.||Winston Police Dept.|
|Wright County Sheriff’s Dept.|
|Table 3. Agencies that submitted incomplete reports|
|Bell City Police Dept.|
|Table 4. Agencies that reported no stops (many contract out vehicle stops to other agencies)|
|Alexandria Police Dept.||Armstrong Police Dept.||Augusta Police Dept.|
|Avondale Police Dept.||Birch Tree Police Dept.||Black Jack Police Dept.|
|Breckenridge Police Dept.||Browning Police Dept.||Burlington Northern and Santa Fe Railway|
|Calhoun Police Dept.||Chamois Police Dept.||Clarksville Police Dept.|
|Clyde Police Dept.||Dardenne Prairie Police Dept.||Dearborn Police Dept.|
|Deepwater Police Dept.||DeWitt Police Dept.||Ellsinore Police Dept.|
|Essex Police Dept.||Fairview Police Dept.||Forest Park Police Dept.|
|Grant City Police Dept.||Hanley Hills Police Dept.||Hermitage Police Dept.|
|Holland Police Dept.||Homestead Village Police Dept.||Houston Lake Police Dept.|
|Hunnewell Police Dept.||Huntleigh Police Dept.||Irondale Police Dept.|
|Keytesville Police Dept.||Laddonia Police Dept.||Lake Annette Police Dept.|
|Lake Timberline Police Dept.||Meramec College Police Dept.||Mills Springs Police Dept.|
|Niangua Police Dept.||Oakland Police Dept.||Oregon Police Dept.|
|Oronogo Police Dept.||Pasadena Park Police Dept.||Prairie Home Police Dept.|
|Purdin Police Dept.||Qulin Police Dept.||Russellville Police Dept.|
|Shannon County Sheriff’s Dept.||Shelbyville Police Dept.||Sycamore Hills Police Dept.|
|Urich Police Dept.||Valley Park Police Dept.||Vinita Terrace Police Dept.|
|Westwood Police Dept.||Wheeling Police Dept.||Winchester Police Dept.|
|Wyatt Police Dept.|
III. ANALYSIS BY ATTORNEY GENERAL JAY NIXON
This report is the fifth annual analysis of traffic stops in Missouri issued by the Attorney General’s Office under Section 590.650 RSMo (2000). All previous reports, which date back to 2000, can be linked from this page to compare reports from year to year.
Viewing the statewide statistics for 2004 as a whole shows there was little change from previous years. The analysis of data from 2004 continues to show some degree of disparity in stop and search rates, particularly in the rates at which African-American drivers are stopped and at which African-American and Hispanic drivers who are stopped are then searched (although the rates in all three of these categories declined somewhat from 2003 to 2004). The continued disparity in stops and searches of these groups of drivers should be cause for ongoing review by law enforcement.
It is vital to keep in mind that statistical disproportion does not prove that law enforcement officers are making traffic stops based on the perceived race or ethnicity of the driver. Racial profiling, whether as an agency policy or as a practice by individual officers, cannot be proved or disproved by statistics alone. That said, analysis of the data — particularly the data supplied by individual law enforcement agencies — has proven, for the past five years, to be a springboard for constructive dialogue between the agencies and the communities they serve.
Missouri was the first state in the nation to undertake such analysis on this scale, and our state continues to be a national leader in committing resources to assess traffic stops and eliminate racial disparities. State law also requires every agency to not only have a written policy regarding racial profiling, but also to provide training to officers, and to promote the use of effective, non-combative methods for carrying out their duties in a racially and culturally diverse environment.
Law enforcement agencies in Missouri have overwhelmingly complied with the law requiring reporting of traffic stop data, both by collecting the data and by working with the Attorney General’s Office to ensure that our analysis is as accurate as possible. I greatly appreciate the efforts of law enforcement, which have been carried out during a time when many agencies are understaffed and underfunded. In addition, it is important to note that fears that requiring collection of traffic stop information would result in “depolicing” — a willful refusal of law enforcement to carry out its duties — have proven to be unfounded, as the number of stops has remained largely stable.
I am concerned that the number of agencies that did not comply with the law by meeting the reporting deadlines increased from 56 in 2003 to 70 in 2004. A list of these agencies has been turned over to the Governor for imposition of the statutory sanctions. In 2004, the Missouri Department of Public Safety withheld more than $7,200 from 16 non-complying agencies that received state funding.
Section Il of this report, titled Statewide Findings, provides an analysis of the 2004 data, and compares that information to prior years. This analysis is done statewide and is not a reflection on any individual jurisdiction. In addition, as with any statistical analysis, there are limitations in the data.
For example, I remind readers of this report that the residential demographics for both the state and individual jurisdictions are taken from the 2000 census. The accuracy of this census in reflecting the true demographics of Missouri will lessen with each passing year. In particular, many areas of Missouri (and the United States) have experienced a marked rise in their Hispanic population. For the state as a whole, an examination of the 2004 population estimates indicate that the 2000 census figures remain a reliable basis for estimating the rate at which drivers of different race and ethnic groups are stopped by the police. However, the 2000 figures may be less accurate for particular localities.
Almost every jurisdiction will have factors that may necessitate increased law enforcement patrols in certain areas or impact the demographics of drivers using its streets and thoroughfares. Agencies are given an opportunity in this report to provide a synopsis of that information.
In conclusion, I must reiterate what I have said in my analysis of each of these reports: law-abiding drivers have the right to travel throughout Missouri without the fear that they will be stopped based solely on their race or ethnicity. I join with law enforcement officers from all parts of our state and with Missourians of all races in this commitment.
Footnote 1: The figures for whites, blacks, Asians, and American Indians in the table represent the number of non-Hispanic persons 16 and older who designated that race alone in the 2000 census. Hispanics may be of any race. About 1 percent of the population designated two or more races. These persons are included in the “other” category along with persons of unknown race.
Footnote 2: Investigative stops are included in the 2004 report for the first time as a separate category for the “reason” for a stop. However, since the majority of agencies reported no such stops for the four-month period during which these data were collected, they are not included in the statewide figures, but investigative stops are included in the tables for each jurisdiction. We anticipate that with a full year of collecting this information in 2005, it will be included in the statewide figures for that year.
Footnote 3: Geolytics Inc. provided estimated 2004 populations for whites, African-Americans and Hispanics. Estimated totals for other races were unavailable. No substantively meaningful differences in the statewide results were found when the estimated population figures were substituted for the 2000 census figures. Therefore, to maintain continuity with previous reports, the census-year population counts are used for the 2004 report. In future years, however, estimated population totals may be substituted.
Footnote 4: Caution should be used when comparing 2000 to subsequent years, especially for smaller agencies, because the 2000 figures are based on only four months of traffic data, while those for 2001-2004 are based on the full calendar year.