The Issue of Racial Profiling in Traffic Stop Citations

Abstract

This study addresses racial profiling when the traffic stop outcome is a citation. This study uses focal concerns theory as a theoretical explanation for police officer decision-making while using propensity score matching to provide similarly situated drivers based on race and/or gender. This study uses traffic stop data (N = 48,586) collected by the Louisville Police Department between January 1 and December 31, 2002. The statistical results show that focal concerns theory components matter the most for traffic stop data even though racial profiling is still an issue. Propensity score matching is a statistical technique that provides a better way to determine whether racial profiling was evident. Gender was not significant for female drivers. This study advances our understanding of race and traffic stop citations using a theoretical explanation.

Keywords

racial profiling citations focal concerns theory propensity score matching

Introduction

Racial profiling is not a new phenomenon in American policing. Recent events, however, have taken place across the United States in cities such as Ferguson, Missouri, and New York City along with the social justice movement of Black Lives Matter that has brought to light new racial issues involving the police. The issue that is present is questioning the discretionary behavior of law enforcement. These events show minorities harbor long-held attitudes or beliefs that law enforcement has been inherently biased against them.

Even with these recent tragic events, there have been an extensive number of studies conducted on racial profiling from various social science disciplines. Previous literature has brought up issues both with the majority of prior studies being atheoretical or limited from a methodological standpoint (Alpert, Smith, & Dunham, 2004; Engel, Calnon, & Bernard, 2002; Schafer, Carter, & Katz-Bannister, 2004; Walker, 2001). This article helps to expand on this literature by using focal concerns theory as a theoretical explanation for police officer decision-making during traffic stops that ended in a citation using propensity score matching as the analysis technique. This study uses traffic stop data collected from the Louisville Police Department from January 1, 2002, through December 31, 2002.

This article contributes modestly to the literature by addressing these four issues. To make this contribution, this article will summarize the studies conducted so far but will explain more in-depth the limited number of studies specifically examining citations. Second, this article will explain the importance of using theory along with how using propensity score matching as an advanced multivariate analysis would improve on what has been conducted previously. Third, by using propensity score matching allows the researchers to examine similarly situated drivers based on the driver’s race and then similarly situated drivers based on the driver’s race and gender. Fourth, this article will provide a brief summarization of what policy implications or recommendations could combat the issue or perception of racial profiling.

Using Traffic Stop Data to Examine Racial and Gender Bias

In racial profiling research, studies focus on how the race of the driver impacts what takes place. Previous research focused on the likelihood that police are stopping racial minorities more often than members of other racial groups. Racial profiling thus questions the legitimacy of law enforcement. If a police department is found to be using racial profiling, they lose trust with those certain racial groups (Barrick, 2014; Callanan & Rosenberg, 2011; Drakulich & Crunchfield, 2013; Engel & Calnon, 2004; Feinstein, 2015; Lever, 2007; Warren, 2011; Wu, 2014; Wu, Lake, & Cao, 2015) and this results in questioning all police–citizen interaction even when it is shown to be legitimate (Engel & Calnon, 2004; Lever, 2007; Warren, 2011).

While several studies have been conducted to date on racial profiling, the results have been mixed. A number of studies in both criminology and criminal justice found that racial minorities were more likely to be stopped compared with Caucasians (Alpert, Dunham, & Smith, 2007; Alpert, MacDonald, & Dunham, 2005; Carroll & Gonzalez, 2014; Eger, Fortner, & Slade, 2015; Engel & Calnon, 2004; Farrell, 2015; Hanink, 2013; Jacobs, 1979; Klahm & Tillyer, 2015; Lundman & Kaufman, 2003; Meehan & Ponder, 2002; Novak, 2004; Novak & Chamlin, 2012; Petrocelli, Piquero, & Smith, 2003; Regoeczi & Kent, 2014; Renauer, 2012; Rojek, Rosenfeld, & Decker, 2004; Ryan, 2016; Smith, Makarios, & Alpert, 2006; Stolzenberg, D’Alessio, & Eitle, 2004; Tillyer, 2014; Tillyer & Engel, 2013; Tillyer & Klahm, 2015; Warren, Tomaskovic-Devey, Smith, Zingraff, & Mason, 2006; Withrow, 2004a, 2004b, 2007).

An understudied issue, however, is how the gender of the driver (specifically male) impacts racial profiling. This issue is important because the gender of the driver could explain a police officer’s decision-making during a traffic stop. In racial profiling literature, studies concerned with the gender of the driver are focused on how the gender of the driver impacts the likelihood of racial profiling by the police. Researchers have found that police officers were more likely to stop, search, arrest, check records, and use force with male drivers (Barnum & Perfetti, 2010; Farrell, 2015; Higgins, Vito, Grossi, & Vito, 2012; Higgins, Vito, & Walsh, 2008; Lundman, 1979; Smith et al., 2006; Smith & Petrocelli, 2001). Among male drivers, African Americans, Hispanics, and Asians were more likely to be involved with police in stops, searches, arrests, records checks, and use of force (Cochran & Warren, 2012; Higgins et al., 2012; Higgins et al., 2008; Lundman, 1979; Lundman & Kowalski, 2009; Moon & Corley, 2007; Schafer, Carter, Katz-Bannister, & Wells, 2006; Terrill & Reisig, 2003; Tillyer & Engel, 2013). The following section will look at the previous literature on racial profiling using traffic stop data.

Racial Profiling Research After the Traffic Stop

Several methodological issues have been brought up from prior racial profiling studies. They range from the methods used to conduct the traffic stop to the outcome of the traffic stop. Using baselines or benchmarks was one way to combat certain methodological issues. The use of benchmarks came from studies conducted by economists and is a fraction where the number of recorded traffic stops is in the numerator of the fraction, and the number of eligible drivers of a specific race or a census count of racial minorities is in the denominator (Persico & Todd, 2004). However, researchers such as MacDonald (2001) raised concerns with the use of benchmarks because it cannot account for population variations on the roadways, the degrees of lawbreaking, or the police deployment/enforcement patterns used. Benchmarking was thought to be a way to examine racial bias because it could show differences in the stop and search rates of minority citizens and whether it was attributable to a few officers. Internal benchmarking is another type of benchmarking that allows a police department to see whether their police officers are under- or overenforcing the law upon minority citizens (Withrow, Dailey, & Jackson, 2008).

Economists also proposed the outcomes test as a way to examine racial bias. Ayres (2002) provided two reasons supporting the outcome test: (a) In some contexts, evidence of racial disparities in the average outcome is strong evidence of disparities on the margin and (b) while problems may arise with inframarginality, it is not a problem when interpreting the outcomes analysis merely as a test of unjustified disparate impact (Persico & Todd, 2006). The issue presented from inframarginality is that the officers will never have all the information available to them at the time of the traffic stop that could influence their decision-making. This brings up an issue then in the prior literature that the officer does not have all the pertinent information (e.g., Is the driver under the influence of drugs and/or alcohol? Does the driver have any outstanding warrants? Does the driver have contraband?). These factors may be present to the officer while making observations during a traffic stop (e.g., expired tags, taillight out, speeding).

Criticisms of the use of a benchmark type study have been made by other researchers who show that the results of benchmark studies can only be described as tenuous. One of the key criticisms is that officers may believe racial minorities have a greater likelihood of committing traffic and legal code violations compared with Caucasians. Such a belief would lend credence to view that racial minorities then should be stopped for traffic violations (Becker, 2004; Borooah, 2001; Knowles, Persico, & Todd, 2001; Persico & Todd, 2004, 2006) and for minor traffic violations at a higher rate than Whites (Becker, 2004; Borooah, 2001; Knowles et al., 2001; Persico & Todd, 2004, 2006). The issue is that these stops then become “pretextual” (Engel et al., 2002). The pretextual stop allows the officers to stop a vehicle for a minor violation intending to find other or additional legal violations (Gizzi, 2011; Novak, 2004; Withrow, 2007). Pretextual stops are a consistent tactic used by the police, and they have been upheld in the courts (Whren et al. v. United States, 1996). Pretextual stops were one of the key police tactics used during the “war on drugs.” Pretextual stops also create an issue with researchers because it makes it harder to disentangle the causal role of race in traffic stops. It makes next to impossible to extract and examine racial motives.

This results in not allowing the researcher to reveal the impact that race has on traffic stops fully. However, researchers still could present results that show evidence of racial profiling from traffic stop data. Based on the difficulty that using traffic stop data presents, Engel and Johnson (2006) argued that the events that take place after the traffic stop could be better indicators that researchers need to consider when examining racial bias. The racial disparities that may be present could show evidence of racial profiling.

The result of a traffic stop outcome is important to analyze because it shows the official sanction that a police officer gives a citizen. This article examines the traffic stop outcome of citation. Citations indicate some level of involvement on behalf of the officer. The following section will examine racial profiling literature where traffic stop citation is the outcomes.

Citation as the Traffic Stop Outcome

A citation can be issued during a traffic stop for equipment or moving violations such as speeding. The decision made by a police officer in issuing a citation is a potentially important indicator of racial profiling. This article examines whether an officer is using race as a basis for decision-making with less serious offenses. To date, the research on citations has been mixed. Specifically, the research has shown that minority drivers were no more likely than White drivers to be given citations (Novak, 2004). In contrast, other studies have indicated that race, along with additional factors (i.e., legal and extralegal), makes it more likely that minorities are issued citations (Barnum & Perfetti, 2010; Ingram, 2007; Tillyer & Engel, 2013).

Novak (2004) examined whether racial profiling took place in traffic stops made by the Overland Park (KS) Police Department. The data were collected from July 1, 2000, to November 30, 2000, and comprised 10,473 traffic stops. Novak found that minorities were more likely to be the subject of pretextual stops. Minorities were less likely to be stopped during the day than at night. Drivers of all races stopped for unsafe driving, or moving violations were more likely to receive a citation instead of a warning.

Ingram (2007) examined what neighborhood factors may impact traffic citations made by the police. The data came from the 2000 U.S. Census and traffic stop data from a police department in a city in the southwestern United States were collected from January 1, 1999, to October 10, 1999. Several conclusions can be made from Ingram (2007). Areas with higher levels of disorganization had more traffic citations. Disadvantaged areas had higher traffic citation levels. In addition, the racial composition of the neighborhood impacted the likelihood of traffic citations issued (i.e., higher minority population areas had a higher number of citations issued).

Barnum and Perfetti (2010) examined what factors impact officer decision-making for stopping drivers based on race, citation, arrest, and search. The data, collected from June 1 to December 31, 2007, came from an unnamed police department and contained 5,417 traffic-related police–citizen contacts. The results showed evidence of racial profiling by police officers for making stops, citations, arrests, and searches. Male minority drivers were more likely to be stopped. Minority drivers were more likely to receive a citation and to be arrested. Officers were more likely to search a car if the driver was a minority.

Tillyer and Engel (2013) examined how a driver’s race, gender, and age impacted traffic stops. The data came from an unnamed police department and comprised 283,827 traffic stops from January 1, 2006, to December 21, 2006. The authors found that African American drivers were no more likely to receive a warning or citation than Caucasian drivers. However, drivers who were young and African American male drivers were more likely to receive a warning and also more likely to receive a citation than Caucasian drivers. Hispanic drivers were less likely to receive a warning and more likely to receive a citation than Caucasian drivers (Tillyer & Engel, 2013).

Farrell (2015) examined the differences in traffic stops between men and women. Farrell collected data from 149,883 traffic stops across 37 municipal jurisdictions in Rhode Island in 2005 and reached several conclusions. Drivers stopped in a disadvantaged area were more likely to be cited for both speeding and nonspeeding violations. Women received leniency from officers in both speeding and nonspeeding situations. The number of citations decreased as the police department hired more females. The pressure placed on an officer by the police department to give citations made it less likely that gender disparity existed. Black drivers were less likely to receive a citation for speeding or nonspeeding violation compared with White drivers.

The current literature on citations and traffic stops highlights that additional research is necessary for this area. The studies that analyzed citations in racial profiling have lacked a theoretical explanation for police officer decision-making. Only one study examined how the race and gender of the driver impacted citations and racial profiling. None of the studies used propensity score matching for statistical analysis. All three issues are important to the assessment of racial profiling and will be analyzed in this article.

Focal Concerns Theory

While theory has been used in some studies on racial profiling based on such facts as racial threat, concentrated disadvantage, social disorganization, and racial animus and organization profiles there has been no study to the researchers knowledge that that applied focal concerns theory to understand citations (Novak & Chamlin, 2012; Parker, MacDonald, Alpert, Smith, & Piquero, 2004; Petrocelli et al., 2003; Tomaskovic-Devey, Mason, & Zingraff, 2004; Warren, Tomaskovic-Devey, Smith, & Zingraff, 2006). The current article builds on Higgins, Vito, and Grossi (2012) by applying focal concerns theory as a theoretical explanation for police officer decision-making and using propensity score matching as the statistical analysis while focusing on the traffic stop outcome citation.

Steffensmeier (1980) originally wrote focal concerns theory as a theoretical explanation for judges and other court actors decision-making for sentencing. In 1998, Steffensmeier, Ulmer, and Kramer (1998) emphasized that focal concerns theory is based on three components that impacted decision-making in sentencing: blameworthiness, protection of the community, and practical constraints and consequences. Blameworthiness is based on the culpability of the defendant and the desire to have the punishment fit the crime (Steffensmeier et al., 1998). Protection of the community is based on the goals of deterrence and incapacitation as indicated by the dangerousness of the offender and the probability of recidivism (Steffensmeier et al., 1998). Practical constraints and consequences are concerned with the organizational costs this has on the criminal justice system, including the monetary impact it has on the defendant’s family and other costs that may impact the public (Steffensmeier et al., 1998).

Focal concerns theory tries to take into account the ambiguity that can often occur in sentencing decisions (Albonetti, 1991). Judges are overloaded with information and develop a “perceptual shorthand” when certain behaviors or attributes are present in the individual case. This shorthand becomes reinforced creating a pattern where the judge may become resistant to change. Judges are then making a decision about defendants based on their character and future expected behavior. These behaviors are based on focal concerns theory.

In a society where everyone would be treated equally, such disparities would not exist. However, in the real world, the factors that go into focal concerns theory are often impacted by the person’s race or place in society (i.e., race or ethnicity). African American or Hispanic defendants are more likely to receive a harsher sentence compared with Caucasian defendants because they are viewed as more likely to recidivate, less likely to be deterred, or simply viewed as more dangerous (Steffensmeier, 1980, 1998). Ample evidence has shown support for this in the sentencing literature (Demuth & Steffensmeier, 2004; Johnson, Ulmer, & Kramer, 2008; Spohn & Beichner, 2000; Steffensmeier & Demuth, 2001, 2006). While focal concerns theory was applied to sentencing decision-making by judges or other court actors, this same theory could be applied to the decision-making of individual police officers.

Engel et al. (2002) wrote a criticism of the research conducted up to that point on racial profiling. One of the main criticisms that was brought forth was the lack of a theoretical foundation in the racial profiling literature. It was then that Tillyer and Hartley (2010) would make the argument that focal concerns theory could be applied to the racial profiling literature. However, Tillyer and Hartley did not apply it to an original study. Focal concerns theory will provide the researcher with data on when, where, and why racial disparities exist in traffic stops (Tillyer & Hartley, 2010). This study seeks to advance the literature on racial profiling by moving away from a singular focus on race.

During a traffic stop, a police officer does not have all the information available to them on the person who has been stopped. Because of this similar to judges, the officer also develops a shorthand to both simplify the situation and deal with the potential problem at hand. Smith and Alpert (2007) found that police officers develop profiles of citizens based on the exchange that takes place with a certain individual and the social identity of the person. The result then is a profile based not only on the person’s race but also their gender and age (Smith & Alpert, 2007).

How the media portrays a criminal has also been shown to impact police officer’s perceptions of certain individuals (Bobo, Kleugel, & Smith, 1997; Chiricos, Welch, & Gertz, 2004; Tillyer & Hartley, 2010; Weitzer & Tuch, 2004). What then takes place during a traffic stop could reinforce the profile of a certain individual from certain racial groups. The components of focal concerns theory allow the officer to make a decision on a person’s character and what type of behavior they expect the person to display. Tillyer and Hartley (2010) echoed this sentiment by stating that an officer’s previous experience could create an unconscious profile that could influence the officer’s decision-making when deciding whether to issue a citation.

Higgins et al. (2012) is the only study to date that has applied focal concerns theory as a theoretical explanation on the subject of racial profiling involving the decision-making of a police officer during a traffic stop. The study found that African American drivers were more likely to give consent for a search than Caucasian drivers. Police officers were more likely to search drivers when contraband was in plain view, which offers evidence of the focal concerns theory component blameworthiness. Caucasian drivers were viewed as more of a danger based on blameworthiness than African American drivers. This article builds upon this research by applying focal concerns theory to the traffic stop outcome citation with propensity score matching used for the statistical analysis. The following section provides an overview of the literature involving propensity score matching and how it can help a researcher turn cross-sectional data into a quasi-experimental design that would then allow the researcher to draw causal implications from the results.

Current Study

This article applies focal concerns theory as a way to explain police officer decision-making along with the use of propensity score matching. This method of analysis will build on Higgins et al. (2012) by applying propensity score matching with focal concerns theory for the traffic stop outcome citation among comparable racial groups and comparable male and female racial groups. We seek to answer three research questions:

Research Question 1: What is the relationship between focal concerns theory concepts and traffic stop outcome citation?

Research Question 1 is being addressed by using the three focal concerns theory components to explaining why a police officer would decide to give a citizen a citation.

Research Question 2: What is the relationship between race and traffic stop outcomes?

Propensity score matching will allow the researchers the possibility to examine whether racial differences exist for the traffic stop outcome situation among similarly situated Caucasian and African American drivers.

Research Question 3: What is the relationship between gender and race and traffic stop outcomes?

The gender of the driver is being considered in conjunction with the race of the driver by using propensity score matching to examine similarly situated female Caucasian and African American drivers and then similarly situated male Caucasian and African American drivers.

Data

This study examines traffic stop data collected by the Louisville Police Department between January 1 and December 31, 2002. The data were coded onto a two-sided Scantron form. Individual officers who made stops completed the forms, and their supervisors then reviewed these forms. After the district supervisors had completed their reviews, the forms went to staff services to examine the completeness and accuracy of each form. Any form containing errors or incomplete information was returned to the district to be corrected. The Scantron forms were then scanned directly into a database. The database was converted to the Statistical Package for Social Science (SPSS) 11.0. Incomplete forms were removed from the dataset.

The data form follows Fridell’s (2004) recommendations to police departments for collecting data on traffic stop searches. The form records the following: (a) whether a search was conducted, (b) whether contraband in plain view was a factor in the search, (c) whether a canine was used to detect drugs, (d) whether the driver consented to the search, (e) what items were found because of the search (e.g., money, weapons, drugs, stolen property, etc.), and (f) what was searched (e.g., vehicle, property, and/or passengers). Information that was also collected included the legal authority for the search (i.e., probable cause, consent, or warrant; Engel, Klahm, & Tillyer, 2010; Fridell, 2004).

Measures

This article has one dependent variable based on traffic stop outcome being a citation. The independent measures for this study are primarily based on the theoretical concepts of focal concerns theory (i.e., blameworthiness, protection of the community, and practical constraints and consequences) and control measures.

Dependent Measures

The only dependent measure for this study was whether a citation was given. Whether the individual was given a citation was coded as 0 for no and 1 for yes. The purpose of examining citations is that issuing a citation is an action that a police officer can take against a citizen.

Independent Measures

According to Steffensmeier et al. (1998), blameworthiness is the culpability of the individual and that the punishment should fit the crime. Culpability is not always a clear issue in the racial profiling literature. A police officer’s job is not to determine the guilt of a citizen, and guilt is the root of culpability. The closest an officer comes to determining guilt is establishing probable cause. When an officer is establishing probable cause, a sliding scale may be used based on the amount of evidence that is present. An individual is considered blameworthy once an officer has enough probable cause present. In this article, blameworthiness is operationalized using two items. During the traffic stop, each officer reported two different pieces of information: (a) whether contraband was in plain view and (b) whether the officer was able to smell the odor of drugs. Each item response was coded as 0 for no and 1 for yes. The minimum for this measure is 0, and the maximum is 2 (Higgins et al., 2012). On the scale, higher scores show a greater level of blameworthiness.

Protection of the community is based on the goals of incapacitation and general deterrence. Protection of the community is an assessment of an offender’s future behavior, such as if the person is a danger to society (Steffensmeier et al., 1998). To measure the dangerousness of an individual, the study examines whether a warrant check was performed. If a warrant check took place, this measure was coded as 0 for no and 1 for yes. An officer is protecting the community when performing a warrant check (Higgins et al., 2012). The performance of the warrant check is a proxy that the officer is concerned with the protection of the community. At the time of data collection, the chief had directed the police in Louisville to conduct a check to see whether any warrants (felony or misdemeanor) were outstanding for the driver of the vehicle. The warrant check consists of running the license plate; if the warrant check were positive, the officer would stop the vehicle (Grossi, Vito, & West, 2003). This action does not affect the probability that the citizen would be cited.

In sentencing research, practical constraints and consequences are the organizational costs incurred by the criminal justice system, such as the disruption of ties between children and family members, and the potential impact that offender recidivism has on public distress (Steffensmeier et al., 1998). In this article, two items are used to measure this concept. First, did the officer have preexisting knowledge of the individual? Second, was there a call for service? The practical constraints and consequences for an officer are the duty to answer the call for service and/or investigate the known individual who has been stopped. The two items are combined to create a scale that goes from 0 to 2 with higher scores showing the police officer is more likely to pursue an investigation (Higgins et al., 2012).

According to Tillyer and Hartley (2010), demographic measures help make important distinctions and identify potential interactions. This article has five proxy measures. First, gender is coded as 0 for female and 1 for male. Second, the race of the driver is coded as 0 for Caucasian and 1 for African American. Third, residency is coded as 0 for non–city resident and 1 for city resident. Fourth, the race of the officer is coded as 0 for Caucasian and 1 for African American. The final control measure age is left as an open-ended measure.

Analysis Plan

Propensity score matching is used in this article to analyze the differences between similarly situated Caucasian and African American drivers. Propensity score matching is a five-step process. Step 1 is calculating the descriptive statistics for all measures. The descriptive statistics will show the distribution of the data. The measures used with descriptive statistics are minimum, maximum, mean, standard deviation, variance, skewness, and kurtosis of the data. The mean value for each measure is interpreted when using the descriptive statistics for this study. Skewness and kurtosis are examined to determine whether the data are normally distributed. If skewness is measured at ≥3 or ≤10 and the kurtosis is ≤3 or ≤−10, the measures are normally distributed (Field, 2013).

Step 2 is matching individuals, and this study uses the nearest neighbor technique. That allows for 1-to-1 matching of individuals. A caliper (i.e., standard deviation) of 0.20 is used for this article. The caliper matches similarly Caucasian and African American drivers based on the dependent variable being a traffic stop outcome of a citation. Step 3 is to assess the quality of the matching and is done in two steps. First, the mean for the matched groups is identical or almost identical. Second is that this article uses Rosenbaum and Rubin’s (1985) approach to standardized bias so it must be ≤10 after matching for the propensity score matching to be acceptable.

Step 4 is a logistic regression that is conducted based on the weighted matches of the propensity score and produces an odds ratio that will indicate the propensity for the traffic stop outcome ending in a citation. The treatment used is Caucasian versus African American drivers for this article. Propensity score matching will balance the independent measures based on similarly situated Caucasian and African American drivers. The logistic regression will allow for an interpretation of the odds ratio based on the propensity score. The odds ratio accounts for the matching because the independent variable in the regression is the treatment assignment.

The four-step analysis is conducted on three different groups of individuals: (a) all similarly situated Caucasian and African American drivers, (b) all similarly situated female Caucasian and African American drivers, and (c) all similarly situated male Caucasian and African American drivers.

Propensity Score Matching Results

Table 1 presents the descriptive statistics for this article. Citations resulted in a traffic stop for 67% of drivers. The mean for blameworthiness is 0.05. The mean for practical constraints and consequences is 0.04. Officers conducted protection of the community for 78% of drivers. Of the drivers examined, 34% were African American, and 70% were male. The mean driver age was 33.38 years. Regarding residency, 63% of drivers were city residents. Only 22% of the officers were African American.

Table 1.

Descriptive Statistics Results.

Measure	Minimum	Maximum	Mean/percentage	SD	Skewness	Kurtosis
Citation	0 (16,241)	(32,345)	0.67	0.47	−0.70	−1.56
Blameworthiness	0	2	0.05	0.23	5.68	34.88
Practical constraints and consequences	0	2	0.04	0.19	5.21	26.93
Protection of the community	0 (10,664)	1 (37,394)	0.78	0.42	−1.34	−0.21
Race of driver	0 (30,481)	1 (15,792)	0.34	0.47	0.67	−1.55
Gender	0 (14,517)	1 (33,675)	0.70	0.46	−0.87	−1.25
Driver age	16	101	33.38	12.72	1.03	1.05
City resident	0 (17,351)	1 (30,088)	0.63	0.48	−0.56	−1.69
Officer race	0 (34,169)	1 (9,916)	0.22	0.42	1.32	−0.26

The percent bias was ≤10 after matching, suggesting that bias has not occurred and that the measures are properly balanced.¹ Caucasian and African American drivers were similarly situated regarding the traffic stop outcome citation.

Citation and Race

Examining the traffic stop outcome of citation² with race addresses Research Questions 1 and 2. Table 2 provides the results of the weighted logistic regression for search and race. African American drivers were 42% less likely to be cited. As the level of blameworthiness increases by one unit, the likelihood of a citation for the driver decreases by 56%. As the level of practical constraints and consequences increases by one unit, the likelihood of a driver being cited decreases by 82%. Protection of the community shows that drivers are 18% less likely to be cited when a warrant check is conducted. Male drivers are 26% less likely to be cited than female drivers. The driver’s age was not statistically significant. City resident drivers are 45% percent less likely to be cited than nonresidents. African American officers are 2.32 times more likely to cite drivers in comparison with Caucasian officers.

Table 2.

Weighted Logistic Regression: Citation and Race.

Measure	b	SE	Exp (b)
Race of driver	−0.54**	0.02	0.58
Blameworthiness	−0.81**	0.02	0.44
Practical constraints and consequences	−1.73**	0.01	0.18
Protection of the community	−0.20**	0.03	0.82
Gender	−0.30**	0.02	0.74
Driver age	−0.0001	0.001	1.00
City resident	−0.61**	0.02	0.55
Officer race	0.84**	0.08	2.32
−2 log-likelihood = −16,484.67
Chi-square = 2,625.69**
Pseudo R² = .07

p < .05. **p < .01.

Citation, Female, and Race

To address Research Questions 2 and 3, the traffic stop outcome of citation in conjunction with the race of the driver and the driver being female was examined. The propensity score matching results on race for female drivers with the outcome being a citation shows that bias has not occurred because the mean for the matched groups between the treated and comparison groups is identical or almost identical, and the percent bias was ≤10 after matching.³ All similarly situated female Caucasian and African American drivers were properly balanced for the traffic stop outcome citation.

Table 3 presents the results of the weighted logistic regression for citation, female, and race. The race of the female driver was not a statistically significant variable. As the level of blameworthiness increases by one unit, the likelihood that a female driver receives a citation increases by 1,464.13 units. As the level of practical constraints and consequences increases by one unit, the likelihood that a female driver receives a citation increases by 8.32 units. Protection of the community makes it 15.70 times more likely that a female driver receives a citation. As the driver’s age increases by one unit, it makes it 1% less likely that a female driver will receive a citation. City resident, female drivers are 1.45 times more likely to be cited than nonresidents. African American officers are 61% less likely to cite female drivers than Caucasian officers.

Table 3.

Weighted Logistic Regression: Race on Female and Citation.

Measure	b	SE	Exp (b)
Race of driver	−0.05	0.09	0.95
Blameworthiness	7.29**	1,527.74	1,464.13
Practical constraints and consequences	2.12**	1.39	8.32
Protection of the community	2.75**	4.71	15.70
Driver age	−0.02**	0.001	0.99
City resident	0.37**	0.19	1.45
Officer race	−0.93**	0.06	0.39
−2 log-likelihood = −1,671.55
Chi-square = 981.88**
Pseudo R² = .23

p < .05. **p < .01.

Citation, Male, and Race

Examining the traffic stop outcome of citation in conjunction with the race of the driver and the driver being male addresses Research Questions 2 and 3. The propensity score matching results on race for male and citation show that the mean for the matched groups between the treated and comparison groups is identical or almost identical, and the percent bias is ≤10 after matching, so bias has not occurred.⁴ The measures are properly balanced for all similarly situated male Caucasian and African American drivers for the traffic stop outcome citation.

Table 4 presents the results of the weighted logistic regression for citation, male, and race. African American male drivers are 1.71 times more likely to receive a citation. As the level of blameworthiness increases by one unit, male drivers are 795.69 times more likely to get a citation. As the level of practical constraints and consequences increases by one unit, it is 5.82 times more likely that a male driver receives a citation. Protection of the community makes it 7.76 times more likely that a male driver is cited. As the male driver’s age increases by one unit, it makes it 2% less likely that the driver will receive a citation. If the male driver is a city resident, he is 1.64 times more likely to be cited. African American officer is 37% less likely to cite male drivers than a Caucasian officer.

Table 4.

Weighted Logistic Regression: Race on Male and Citation.

Measure	b	SE	Exp (b)
Race of driver	0.53**	0.07	1.71
Blameworthiness	6.68**	358.21	795.69
Practical constraints and consequences	1.76**	0.47	5.82
Protection of the community	2.05**	0.89	7.76
Driver age	−0.02**	0.002	0.98
City resident	0.50**	0.09	1.64
Office race	−0.46**	0.03	0.63
−2 log-likelihood = −8,333.81
Chi-square = 4,892.42**
Pseudo R² = .23

p < .05. **p < .01.

Conclusion

While this study provides a great insight of what could be used in future studies on racial profiling, it is not without its limitations. The data are cross-sectional and were only collected over a 1-year period January 1 and December 31, 2002. The data were self-report, so police officers may be reluctant to underreport information pertinent to the stop (Maxfield & Babbie, 2009). Police officers may have changed the race of the driver when completing the traffic stop form so it would not show that the officer was involved in racial profiling.

The purpose of this article was to examine the impact of focal concerns theory as a theoretical explanation for police officer decision-making when the traffic stop outcome is a citation. The use of focal concerns theory while also using propensity score matching for the statistical analysis allowed the researchers to examine similarly situated Caucasian and African American drivers. The overall findings of this study led to several conclusions.

First is that focal concerns theory can offer a theoretical explanation for police officer decision-making during traffic stops. All three focal concerns theory components had a greater impact on the police officer’s decision to give a citation than race alone. Yet, driver’s race was an important factor in certain situations. Among all similarly situated drivers, African Americans were less likely to be issued a citation. The reason for this could be that the police officer felt that African American drivers could be involved in more serious types of crime. When considering the race and gender of the driver, it was only significant that African American male drivers would be more likely to be issued a citation. This tends to support the criminological literature that African American males are more often viewed as the offender and serve more time.

Several other factors were significant based on the findings of this study. Older drivers were less likely to be issued a citation. Drivers who were city residents were more likely to be issued a citation. Yet, the most significant finding from the proxy measures is that African American officers overall were less likely to issue a citation. This supports the notion that African American officers may hold different perspectives than Caucasian officers that could influence their decision-making. This finding supports the notion for a more diverse workforce in policing.

The findings of this study could be used to help impact a police departments’ policy to combat racial profiling. A major goal of any police department is that police officer decision-making is free of racial bias. Police departments should make every effort to prevent racial profiling from taking place. Several recommendations are being made that police departments should consider if they’re dealing with the problem of racial profiling or the perception that minority citizens feel they are racially profiled.

First is implicit bias that even people who hold politically correct beliefs may have unconscious beliefs that implicitly associate certain people based on their race with crime (Fridell, 2008). One way to combat this is to build better community relations between the police and minority communities. Peruche and Plant (2006) found that officers who have more positive experiences in their personal lives with minority citizens are less likely to hold an implicit bias. Second is the use of an early warning system, which is a data-driven program, that identifies officers whose behavior may be problematic. Identifying these officers allows police departments to provide an intervention that could involve additional training or counseling to change the problematic behavior (Alpert, 2007; Alpert & Walker, 2000; U.S. Department of Justice, 2001; Walker, 2001; Walker, Alpert, & Kenney, 2000). Third is making sure that any police department has a clear policy on the issue of racial profiling (International Association of Chiefs of Police, 2006). The policy should contain a clear and concise definition of what is racial profiling and that the department will not tolerate it. Fridell et al. (2001) developed a racial profiling policy that police departments could use to develop their own policy. The final recommendation is that police departments continue to recruit and hire from minority communities. Racial profiling can be reduced by having a more diverse police department. This could be done by making an effort to hire people who will police in an unbiased manner or increase minority hires to reflect the racial demographics of the community they serve (Fridell et al., 2001).

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biographies

Anthony G. Vito is an assistant professor in the Department of Criminology at the University of West Georgia. He has articles published in the International Journal of Police Science and Management, Journal of Ethnicity in Criminal Justice, Deviant Behavior, and the American Journal of Criminal Justice. His research interests include policing, drug research, public policy and advanced statistics.

Elizabeth L. Grossi is an associate professor in the Department of Criminal Justice at the University of Louisville. She is actively involved in the Southern Criminal Justice Association (SCJA) where she served as president (2001), Region 2 Trustee (1996-1999), and executive director (1994-1999). Her research interests include domestic violence, racial profiling, and program evaluation.

George E. Higgins is a professor in the Department of Criminal Justice at the University of Louisville. He received his PhD in Criminology from Indiana University of Pennsylvania in 2001. His most recent publications appear in the American Journal of Criminal Justice, Violence and Victims, Substance Use and Misuse, and Computers in Human Behavior.

References

Albonetti

(1991). An integration of theories to explain judicial discretion. Social Problems, 38, 247-266.

Alpert

G. P.

(2007). Eliminate race as the only reason for police-citizen encounters. Criminology & Public Policy, 6, 671-678.

Alpert

G. P.

Dunham

R. G.

Smith

M. R.

(2007). Investigating racial profiling by the Miami-Dade police department: A multimethod approach. Criminology & Public Policy, 6, 25-53.

Alpert

G. P.

MacDonald

J. M.

Dunham

R. G.

(2005). Police suspicion and discretionary decision making during citizen stops. Criminology, 43, 407-434.

Alpert

G. P.

Smith

M.R.

Dunham

R.G.

(2004). Toward a better benchmark: Assessing the utility of not-at-fault traffic crash data in racial profiling research. Justice, Research, and Policy, 6(1), 43-69.

Alpert

G. P.

Walker

(2000). Police accountability and early warning systems: Developing policies and programs. Justice Research and Policy, 2(2), 59-72.

Ayres

(2002). Outcome tests of racial disparities in police practices. Justice Research and Policy, 4, 131-142.

Barnum

Perfetti

R. L.

(2010). Race-sensitive choices by police officers in traffic stop encounters. Police Quarterly, 13, 180-208.

Barrick

(2014). Latino confidence in the police: The role of immigrant enforcement, assimilation, and immigration status. Journal of Ethnicity and Criminal Justice, 12, 289-307.

10.

Becker

(2004). Assessing the use of profiling in searches by law enforcement personnel. Journal of Criminal Justice, 32, 183-193.

11.

Bobo

Kleugel

J. R.

Smith

R. A.

(1997). Laissez-faire racism: The crystallization of a kinder, gentler, antiblack ideology. In Tuch

S. A.

Martin

J. K.

Racial attitudes in the 1990s: Continuity and change (pp. 15-44). Westport, CT: Praeger.

12.

Borooah

V. K.

(2001). Racial bias in police stops and searches: An economic analysis. European Journal of Political Economy, 17(1), 17-37.

13.

Callanan

V. J.

Rosenberger

J. S.

(2011). Media and public perceptions of the police: Examining the impact of race and personal experience. Policing & Society, 21, 167-189.

14.

Carroll

Gonzalez

M. L.

(2014). Out of place: Racial stereotypes and the ecology of frisks and searches following traffic stops. Journal of Research in Crime and Delinquency, 51, 559-584.

15.

Chiricos

Welch

Gertz

(2004). Racial typification of crime and support for punitive measures. Criminology, 42, 358-390.

16.

Cochran

J. C.

Warren

P. Y.

(2012). Racial, ethnic, and gender differences in perceptions of the police: The salience of officer race within the context of racial profiling. Journal of Contemporary Criminal Justice, 28, 206-227.

17.

Demuth

Steffensmeier

D. J.

(2004). Ethnicity effects on sentence outcomes in large urban courts: Comparisons among White, Black and Hispanic defendants. Social Science Quarterly, 85, 994-1011.

18.

Drakulich

K. M.

Crunchfield

R. D.

(2013). The role of perceptions of police in informal social control: Implications for the racial stratification of crime and control. Social Problems, 60, 383-407.

19.

Eger

R. J.

III Fortner

C. K.

Slade

C. P.

(2015). The policy of enforcement: Red light cameras and racial profiling. Police Quarterly, 18, 397-413.

20.

Engel

R. S.

Calnon

J. M.

(2004). Examining the influence of drivers’ characteristics during traffic stops with police: Results from a national survey. Justice Quarterly, 21(1), 49-90.

21.

Engel

R. S.

Calnon

J. M.

Bernard

T. J.

(2002). Theory and racial-profiling: Shortcomings and future directions in research. Justice Quarterly, 19, 201-225.

22.

Engel

R. S.

Johnson

(2006). Toward a better understanding of racial and ethnic disparities in search and seizure rates. Journal of Criminal Justice, 34, 605-617.

23.

Engel

R. S.

Klahm

Tillyer

(2010). Citizen demeanor, race, and traffic stops: A review of the evidence. In White

Rice

(Eds.), Race, ethnicity and policing: The issues, methods, research, and future (pp. 287-308). New York, NY: New York University Press.

24.

Farrell

(2015). Explaining leniency: Organizational predictors of the differential treatment of men and women in traffic stops. Crime & Delinquency, 61, 509-537.

25.

Feinstein

(2015). A qualitative analysis of police interactions and disproportionate minority contact. Journal of Ethnicity in Criminal Justice, 13, 159-178.

26.

Field

(2013). Discovering statistics using IBM SPSS statistics and sex and drugs and rock “n” roll (4th ed.). Los Angeles, CA: Sage.

27.

Fridell

L. A.

(2004). By the numbers: A guide for analyzing race data from vehicle stops. Washington, DC: Police Executive Research Forum.

28.

Fridell

L. A.

(2008). Racially biased policing: The law enforcement response to the implicit black-crime association. In Lynch

Patterson

E. B.

Childs

K. K.

(Eds.), Racial divide: Racial and ethnic bias in the criminal justice system (pp. 39-59). Monsey, NY: Criminal Justice Press.

29.

Fridell

L. A.

Lunney

Diamond

Kubu

Scott

Laing

(2001). Racially biased policing: A principled response. Washington, DC: Police Executive Research Forum.

30.

Gizzi

M. C.

(2011). Pretextual stops, vehicle searches, and crime control: An examination of strategies used on the frontline of the war on drugs. Criminal Justice Studies, 24, 139-152.

31.

Grossi

E.L.

Vito

G.F.

West

A.D.

(2003). Traffic stop practices of the Louisville Police Department: January 1 – December 31, 2002. Louisville, KY: University of Louisville.

32.

Hanink

(2013). Don’t trust the police: Stop question frisk, COMPSTAT, and the high cost of statistical over-reliance in the NYPD. Journal of the Institute of Justice & International Studies, 13, 99-116.

33.

Higgins

G. E.

Vito

G. F.

Grossi

E. I.

Vito

A. G.

(2012). Searches and traffic stops: Racial profiling and capriciousness. Journal of Ethnicity in Criminal Justice, 10, 163-179.

34.

Higgins

G. E.

Vito

G. F.

Grossi

E. L.

(2012). The impact of race on the police decision to search during a traffic stop: A focal concerns theory perspective. Journal of Contemporary Criminal Justice, 28, 166-183.

35.

Higgins

G. E.

Vito

G. F.

Walsh

W. F.

(2008). Searches: An understudied area of racial profiling. Journal of Ethnicity in Criminal Justice, 6(1), 23-39.

36.

Ingram

J. R.

(2007). The effect of neighborhood characteristics on traffic citation practices of the police. Police Quarterly, 10, 371-393.

37.

International Association of Chiefs of Police. (2006). Addressing racial profiling: Creating a comprehensive commitment to bias-free policing. Washington, DC: Community Oriented Policing Services (COPS).

38.

Jacobs

(1979). Inequality and police strength: Conflict theory and coercive control in metropolitan areas. American Sociological Review, 44, 913-925.

39.

Johnson

Ulmer

Kramer

J. H.

(2008). The social context of guideline circumvention: The case of Federal district courts. Criminology, 46, 711-783.

40.

Klahm

C. F.

Tillyer

(2015). Rethinking the measurement of officer experience and its role in traffic stop searches. Police Quarterly, 18, 343-367.

41.

Knowles

Persico

Todd

(2001). Racial bias in motor vehicle searches: Theory and evidence. Journal of Political Economy, 109, 203-229.

42.

Lever

(2007). What’s wrong with racial profiling? Another look at the problem. Criminal Justice Ethics, 26, 20-28.

43.

Lundman

R. J.

(1979). Organizational norms and police discretion: An observational study of police work with traffic law violators. Criminology, 17, 159-171.

44.

Lundman

R. J.

Kaufman

R. L.

(2003). Driving while Black: Effects of race, ethnicity, and gender on citizen self-reports of traffic stops and police actions. Criminology, 41, 195-220.

45.

Lundman

R. J.

Kowalski

B. R.

(2009). Speeding while Black? Assessing the generalizability of Lange et al.’s (2001, 2005) New Jersey Turnpike speeding survey findings. Justice Quarterly, 26, 504-527.

46.

MacDonald

(2001). The myth of racial profiling. City Journal, 11, 1-13.

47.

Maxfield

M. G.

Babbie

(2009). Basics of research methods for criminal justice and criminology (2nd ed.). Belmont, CA: Wadsworth Cengage Learning.

48.

Meehan

A. J.

Ponder

(2002). How roadway composition matters in analyzing police data on racial profiling. Police Quarterly, 5, 306-333.

49.

Moon

Corley

C. J.

(2007). Driving across campus: Assessing the impact of driver’s race and gender on police traffic enforcement actions. Journal of Criminal Justice, 35(1), 29-37.

50.

Novak

K. J.

(2004). Disparity and racial profiling in traffic enforcement. Police Quarterly, 7(1), 65-96.

51.

Novak

K. J.

Chamlin

M. B.

(2012). Racial threat, suspicion, and police behavior: The impact of race and place in traffic enforcement. Crime & Delinquency, 58, 275-300.

52.

Parker

K. F.

MacDonald

Alpert

G. P.

Smith

M. R.

Piquero

(2004). A contextual study of racial profiling: Assessing the theoretical rationale for the study of racial profiling at the local level. American Behavioral Scientist, 47, 943-962.

53.

Persico

Todd

(2004). Using hit rate tests to test for racial bias in law enforcement: Vehicle searches in Wichita (NBER Working Paper Series, pp. 1-31). Available at: https://ssrn.com/abstract=651862 or https://dx-doi-org.web.bisu.edu.cn/10.2139/ssrn.651862

54.

Persico

Todd

(2006). Generalising the hit rates test for racial bias in law enforcement, with an application to vehicle searches in Wichita. The Economic Journal, 116, 351-367.

55.

Peruche

B. M.

Plant

E. A.

(2006). The correlates of law enforcement officers’ automatic and controlled race-based responses to criminal suspects. Basic and Applied Social Psychology, 28, 193-209.

56.

Petrocelli

Piquero

A. R.

Smith

M. R.

(2003). Conflict theory and racial profiling: An empirical analysis of police traffic stop data. Journal of Criminal Justice, 31, 1-11.

57.

Regoeczi

W. C.

Kent

(2014). Race, poverty, and the traffic ticket cycle: Exploring the situational context of the application of police discretion. Policing: An International Journal of Police Strategies & Management, 37, 190-205.

58.

Renauer

B. C.

(2012). Neighborhood variation in police stops and searches: A test of consensus and conflict perspectives. Police Quarterly, 15, 219-240.

59.

Rojek

Rosenfeld

Decker

(2004). The influence of driver’s race on traffic stops in Missouri. Police Quarterly, 7, 126-147.

60.

Rosenbaum

Rubin

(1985). Constructing a control group using multivariate matched sampling methods that incorporate the propensity score. The American Statistician, 39(1), 33-38.

61.

Ryan

M. E.

(2016). Frisky business: Race, gender, and police activity during traffic stops. European Journal of Law and Economics, 41(1), 65-83.

62.

Schafer

Carter

Katz-Bannister

(2004). Studying traffic stop encounters. Journal of Criminal Justice, 32, 159-170.

63.

Schafer

J.A.

Carter

D.L.

Katz-Bannister

A.J.

Wells

W.M.

(2006). Decision making in traffic stop encounters: A multivariate analysis of police behavior. Police Quaterly, 9(2), 184-209.

64.

Smith

M. R.

Alpert

G. P.

(2007). Explaining police bias: A theory of social conditioning and illusory correlation. Criminal Justice and Behavior, 34, 1262-1283.

65.

Smith

M. R.

Makarios

Alpert

G. P.

(2006). Differential suspicion: Theory specification and gender effects in the traffic stop context. Justice Quarterly, 23, 271-295.

66.

Smith

M. R.

Petrocelli

(2001). Racial profiling? A multivariate analysis of police traffic stop data. Police Quarterly, 4(1), 4-27.

67.

Spohn

Beichner

(2000). Is preferential treatment of female offenders a thing of the past? A multi-site study of gender, race, and imprisonment. Criminal Justice Policy Review, 11, 149-184.

68.

Steffensmeier

D. J.

(1980). Assessing the impact of the women’s movement on sex-based differences in the handling of adult criminal defendants. Crime & Delinquency, 26, 344-357.

69.

Steffensmeier

D. J.

Demuth

(2001). Ethnicity and judge’s sentencing decisions: Hispanic-Black-White comparisons. Criminology, 39, 145-177.

70.

Steffensmeier

D. J.

Demuth

(2006). Does gender modify the effects of race-ethnicity on criminal sanctioning: Sentence for male and female White, Black, and Hispanic defendants. Journal of Quantitative Criminology, 22, 241-261.

71.

Steffensmeier

D. J.

Ulmer

Kramer

(1998). The interaction of race, gender, and age in criminal sentencing: The punishment cost of being young, black, and male. Criminology, 36, 763-798.

72.

Stolzenberg

D’Alessio

S. J.

Eitle

(2004). A multilevel test of racial threat theory. Criminology, 42, 673-698.

73.

Terrill

Reisig

M. D.

(2003). Neighborhood context and police use of force. Journal of Research in Crime and Delinquency, 40(3), 291-321.

74.

Tillyer

(2014). Opening the black box of officer decision-making: An examination of race, criminal history, and discretionary searches. Justice Quarterly, 31, 961-985.

75.

Tillyer

Engel

R. S.

(2013). The impact of drivers’ race, gender, and age during traffic stops: Assessing interaction terms and the social conditioning model. Crime & Delinquency, 59(3), 369-395.

76.

Tillyer

Hartley

R. D.

(2010). Driving racial profiling research forward: Learning lessons from sentencing research. Journal of Criminal Justice, 38, 657-665.

77.

Tillyer

Klahm

I. V. C. F

. (2015). Discretionary searches, the impact of passengers, and the implications for police-minority encounters. Criminal Justice Review, 40, 378-396.

78.

Tomaskovic-Devey

Mason

Zingraff

(2004). Looking for the driving while black phenomena: Conceptualizing racial bias processes and their associated distributions. Police Quarterly, 7(1), 3-29.

79.

U.S. Department of Justice. (2001). Principles for promoting police integrity: Examples of promising police practices and policies. Washington, DC: U.S. Department of Justice.

80.

Walker

(2001). Searching for the denominator: Problems with police traffic stop data and an early warning system solution. Justice Research and Policy, 3(1), 63-95.

81.

Walker

Alpert

G. P.

Kenney

D. J.

(2000). Early warning systems for police: Concept, history, and issues. Police Quarterly, 3, 132-152.

82.

Warren

P. Y.

(2011). Perceptions of police disrespect during vehicle stops: A race-based analysis. Crime and Delinquency, 57, 356-376.

83.

Warren

P. Y

Tomaskovic-Devey

Smith

Zingraff

Mason

(2006). Driving while Black: Bias processes and racial disparity in police stops. Criminology, 44, 709-738.

84.

Weitzer

Tuch

S. A.

(2004). Race and perceptions of police misconduct. Social Problems, 51, 305-325.

85.

Whren et al., v. United States, 517 U.S. 806 (The Supreme Court 1996).

86.

Withrow

B. L.

(2004a). Driving while different: A potential theoretical explanation for race-based policing. Criminal Justice Policy Review, 15, 344-364.

87.

Withrow

B. L.

(2004b). Race-based policing: A descriptive analysis of the Wichita stop study. Police Practice and Research, 5, 223-240.

88.

Withrow

B. L.

(2007). When Whren won’t work: The effects of a diminished capacity to initiate a pretextual stop on police officer behavior. Police Quarterly, 10, 351-370.

89.

Withrow

B. L.

Dailey

J. D.

Jackson

(2008). The utility of an internal benchmarking strategy in racial profiling surveillance. Justice Research and Policy, 10, 19-47.

90.

(2014). Race/ethnicity and perceptions of the police: A comparison of white, black, Asian, and Hispanic Americans. Policing & Society, 24, 135-157.

91.

Lake

Cao

(2015). Race, social bonds, and juvenile attitudes toward the police. Justice Quarterly, 32, 445-470.