research article

Racial Profiling Data Collection Policies: A Study in Depolicing

Bradley R Anders1* 

1Department of criminal justice, State Fair Community College, Sedalia, USA

*Corresponding author: Bradley R. Anders, State Fair Community College, Sedalia, USA, Tel: + 660-827-1812

 ; E-mail: banders0907@gmail.com 

Received date: 30, December, 2016; Accepted date: 03, March, 2017; Published date: 08, March, 2017

Citation: Anders BR (2017) Racial Profiling Data Collection Policies: A Study in Depolicing. Forensic Stud 2017: J104. DOI: 10.29011/FSTD-104. 100004

Claims of racial profiling by police services have prompted many states to collect demographic data on those with whom police have contact in both traffic stops and regular patrol. However, the possibility for police officers to disengage, or depolice, when faced with data collection policies that are viewed as lessening the officer’s discretion is real. Depolicing can include police officers refusing to patrol minority populated neighborhoods for fear of data reflecting over-representative minority contacts. As an unintended consequence, data tracking policies may negatively impact the very minorities they are designed to protect. This exploratory study analyzes the correlation found between police officers’ decisions to stop or not stop a speeding motorist identified as a racial or ethnic minority and four related factors; 1) Statutory racial data tracking, 2) frequency of contacts data reviews, 3) officer’s length of service and 4) any history of discipline for violating racial profiling policy. A sample of regular duty police officers from Midwest states were surveyed using their responses to a traffic stop scenario presented as a vignette. Results of a logistic regression model showed the only significant predictor of a police officer’s decision was the presence of a state statute requiring the collection of racial identity in contact data. Findings offer a potential explanation of individual officer minority contact ratios, and may prompt policy revision to effect equal treatment of all citizens regardless of race or ethnicity.

 

1. Introduction

In the mid-1980s, the United States Drug Enforcement Administration (DEA) engaged in a narcotics trafficking venture entitled “Operation Pipeline” with the goal of identifying drug couriers engaged in narcotics trafficking along major highways within U.S. borders [1]. The problem, according to [1] was the identification of such couriers involved the use of race and ethnicity. The training program implemented during Operation Pipeline specifically outlined certain indicators, such as race and gender, to identify would-be drug traffickers [1]. What followed were the filings of civil suits in which police were accused of using race in an inappropriate manner when deciding to conduct investigatory, as well as probable cause, stops [1]. Consequently, racial profiling, defined roughly as targeting minorities for disparate investigatory practices based on the belief that their race or ethnicity suggests a greater potential for criminality, was brought to the forefront of American legal proceedings [2].

Barnum and Perfetti [3] observed that the foundation of the racial profiling legal battle can be identified in two notable court cases: Wilkins vs. Maryland State Police (1993) and The State of New Jersey vs. Soto (1996). In each of these cases, the plaintiffs, minority citizens, alleged that police officers used their race as a primary motivating factor in the decision to conduct a traffic stop as opposed to any observed violation, traffic or criminal [3]. Between the years of 1991 and 2006, approximately 135 cases were heard on the federal level that directly addressed the issue of racial profiling. Furthermore, in regard to the Wilkins (1993) decision, the State of Maryland was required to start a data collection campaign to track demographic information of every traffic stop conducted in the state; this represented the beginning of data tracking campaigns across the country [2].

As Congress failed to pass any comprehensive racial profiling legislation, several individual states were successful in passing bills that addressed law enforcement’s use of race as an indicator of criminal activity, some calling for the mandatory collection of demographic data that characterized each police/citizen contact. However, as Laney [4] noted, with such an ambiguous definition of what constitutes racial profiling, any accurate measurement of the phenomenon could be difficult. Schafer, Carter, Katz-Bannister, and Wells [5] further added such data collection policies may show a problem when, in fact, a problem does not actually exist. The result of such a policy, as noted by Cooper [6] and Miller [7], could be a phenomenon known as depolicing, or the systematic refusal of police to engage in pro-active policing. It is this depolicing that could potentially keep police out of minority populated neighborhoods, which according to Capers [8], often have the highest crime rates and the greatest need for police presence.

The research of Ingram [9] and Novak and Chamlin [10], amongst others, addressed the numerous variables that play into a police officer’s decision to stop a motorist. What was missing from the existing literature is an analysis of how racial profiling policies impact proactive policing.The purpose of this exploratory, quantitative, cross-sectional study was to identify and analyze the possible relationship between racial profiling policy, state statutes, and a police officer’s decision to stop or not stop a motorist when that motorist is observed to be a racial or ethnic minority. Furthermore, this study utilized a vignette to address police officer behavior in a hypothetical situation. As noted by [11], vignettes can be used to identify behavioral patterns not identified through other data collection methods.

      1.1 Literature Review

Racial disparity in traffic stops is a troubling issue facing many law enforcement agencies. Researchers of racial profiling have addressed multiple variables that play a part in an officer’s decision to stop a motorist. Officer behavior is not an understudied topic by any means, but there is less research on policy influence as it relates to both officer decisions and racial profiling, which is a notable gap in current research.

Mendias and Kehoe [12] observed that discretion employed by police officers must reflect the ideology, current social structure, and current paradigm espoused by the department with which they are employed. This suggests that police officers have not only drawn on departmental policy to guide their behavior but the police culture may have had an influence as well [13]. Reinforced the impact of police culture on individual officer discretion and stated that it is an organizational variable that should be taken into account when attempting to understand police officer behavior. However, police behavior and decision-making processes may not be an easily understood phenomenon [14]. Noted the different variables that impact an officer’s decision to stop or not stop a motorist. Phillips further identified issues such as vehicle characteristics as significant in influencing the decision to stop as opposed to driver characteristics, which were found to be not significant in the decision-making process [15]. Asserted that in making their decisions, police officers managed the information presented to them by using similar clues ascribed to similar people, stereotypes of sort.

Kennedy [16] Identified racial profiling as the conscious identification of race as an indication of potential criminality. As such, Kennedy delineated two primary groups, police and members of racial minorities, as key players in the practice of racial profiling. The issue arises as to whether or not racial profiling is an acceptable tactic used by police officers. Risse and Zeckhauser [17] noted that while racial profiling can have its uses from a utilitarian perspective, ignoring one’s individuality is a damaging practice. The damage, according to Kennedy, is monumental and has historically resulted in violent outbursts from racial minorities.

Identifying the damaging effects of using race as an indicator of criminality as [16] noted, many police departments created policies that ban the use of race as a proxy for criminality [7]. These policies, as noted by [4], often include the practice of tracking data to identify whether or not an officer is, in fact, engaging in the practice of racial profiling. Some police departments chose to implement racial profiling policies on their own while others were mandated by statutes adopted through legislation in their respective states [4]. Upon implementing such policies, however, police departments needed to ensure compliance, and as [18] noted, control over an employee comes by watching the employee and either rewarding desirable or punishing undesirable behavior. It is at this point that the decisions made by an employee are directly affected, according to Rowe et al. [19], by the control mechanisms chosen by an employee’s supervisor. Improper application of organizational control mechanisms result in negative behavior from

The employee [12] in the case of police officer behavior, this negative behavior may manifest itself in the form of depolicing [20].

Racial profiling policy, as noted by [6], may result in a police officer’s decision to under police neighborhoods populated predominately by minorities. According to Cooper, this practice of depolicing serves two purposes: (a) By under policing minority populated neighborhoods police avoid antagonizing any racial tensions and (b) Depolicing challenges police critics. In addition, Cooper noted that by engaging in depolicing, police officers get the chance to exert their autonomy and discretion in such a way that policy makers would have trouble controlling the action. The author further noted that the message conveyed with depolicing was, “Criticize our policing and you will get no policing” [6]. While Cooper addressed the urge to avoid antagonism, there was no mention of how an officer may fear disciplinary proceedings, or worse, receiving a label such as “racist.”

      1.2 Historical Analysis

Extant racial profiling literature has tended to focus on the social harms associated with using race as a proxy for criminal behavior. According to Tomaskovic-Devey and Warren (2009) [21], the Drug Enforcement Administration’s (DEA) Operation Pipeline prompted modern interest in racial profiling. The DEA trained officers to profile drug couriers, and this profile included race; specifically young males with dark skin [21]. From this point forward, police officers were believed to use the drug courier profile, which included race, as an indicator of criminal activity in the War on Drugs [2]. Research focused on the drug courier profile and its impact on the minority community while civil rights organizations condemned its use [21]. Meanwhile, law enforcement agencies continued to engage in the tactic with full support from the United States Department of Justice [21].

Research in racial profiling changed significantly after the terrorist attacks of September 11, 2001. According to [1], instead of concerns revolving around Black and Hispanic drug courier profiles, “new questions and concerns have been raised about racial profiling of Arab and Muslim Americans” (p.1197). Consequently, the topic of racial profiling jumped to the forefront of American homeland security as claims of racial profiling skyrocketed in both airport security checks and traffic stops [1]. Research into public approval of racial profiling as a police tactic also emerged, with results indicating a public propensity to approve of the tactic to prevent terrorism, but low approval ratings for crime prevention [22].

On the state level, according to [3], data collection policies began to emerge in the 1990s after the Wilkins (1993) and Soto (1996) decisions. However, Barnum and Perfeti, as well as [23], noted a recurring problem with racial profiling research founded in racial profiling data collection: the lack of a clear baseline for minority drivers in a given jurisdiction. The conundrum, according to [24], is that “the current literature suggests that police contact should be proportionate to population demographics and ignores all other intervening variables” (p.274). In addition, just because disproportionate stop ratios may be identified, that does not necessarily indicate disparate treatment at the hands of police [3]; there is just the assumption that the minority distribution identified in stops should be representative of the community [25].

Laney [4] noted the issue of accountability in racial profiling claims, stating that public sentiment varies concerning the proper response for officers found to have engaged in racial profiling. Some people feel that an officer found to be in violation of racial profiling policy should be subject to additional training, intense monitoring, or even removal from his or her position as an officer; others wanted the individual police officers subjected to civil litigation [4]. [26], in referring to public reaction to racially charged police-involved shootings in Cincinnati, called the reaction “a war against the defenders of law in Cincinnati, and in particular, against the defenders of law in the impoverished Cincinnati neighborhoods” (p.224). However, as a response to the public reaction, police administrators implemented control mechanisms in the form of policy to address the issue of racial profiling.

      1.3 Control

Discipline, as it relates to policy implementation, was the focus of Shane’s research, stating that “the intent conveyed by the organization when its disciplinary practices are perceived as unfair is that the employees are expendable and are not valued” (p.66) [12]. Noted, officer discretion must be employed in such a way that it agrees and meets organizational standards set forth in policy. Furthermore, controlling employee behavior must be done in such a way that meets organizational goals as well as promotes the proper responses in various situations calling for discretionary decisions [19].

Officer discretion is at the heart of the concept of depolicing [20]. Miller [7] noted Depolicing’s relation to policy implementation, stating that data collection policies may backfire, resulting in a police officer engaging in the practice of depolicing or the intentional misrepresentation of actual minority contacts. In addition, Cooper [6] suggested police officers may ultimately disengage from patrolling minority populated neighborhoods. This practice, as noted by Cooper, serves to both address critics of racial profiling practices and to send the message that police will “allow crime to go unchecked”.

      1.4 Decision Making       

As a behavior exhibited by police officers, depolicing might be viewed as an individual officer’s attempt to establish solidarity or exhibit his or her authority to employ discretion when he or she sees fit, as was the case when [27] referenced the practice. However, an analysis of police behavior revealed a multi-faceted approach to the decision-making process. Citing Wilson, [28] noted the differential policing styles of service, watchdog, and legalistic orientation, but suggested that officers differ in the way they approach problems and those behaviors cannot be attributed solely to the municipality’s political culture. Stroshine, Alpert, and Dunham [29] noted how individual interpretations of people and places have a direct influence on officer behavior and decision making processes. Notably, much of the existing literature has focused on a police officer’s decision-making process during a traffic stop.

Vito and Walsh [30] stated that the decision to make a traffic stop involved a conscious decision- making process on the officer’s part and understanding the thoughts and motives behind those decisions are of the utmost importance. In analyzing multiple variables associated with such a decision, [31] addressed the relationship of numerous variables such as sex, age, and race, age, and so on, on an officer’s decision to arrest or stop and question a person. Pollock et al. discovered that race did not have a significant relationship to an officers’ decision to stop and question or arrest a person. It was noted that their findings are consistent with much of the current body of knowledge addressing the insignificance of race and police contacts [31].

As politics may play a part in an officer’s decision making process [32], or the organizational leadership may influence police action [33], officers conduct themselves according to their descriptive perspectives of organizational justice [34]. According to Wolfe and Piquero [34], officers who view their departments as just in enforcing departmental guidelines are less likely to engage in undesirable behavior otherwise known as police misconduct. Shane echoed this assertion in noting the trend of increased desirable performance by employees when they felt connected or embraced by the organization. Conversely, Wolfe and Piquero cited research indicating that those who view their departments as treating their employees unjustly are more likely to engage in deviant behavior. Wolfe and Piquero utilized regression analysis in seeking their understanding perceptions of organizational justice as it effects officer attitude and beliefs in noble-cause or code-of-silence attitudes. What was learned was as officers felt their organizations were just, their rate of citizen complaints decreased. Wolfe and Piquero noted the importance of policy development that appears fair and just while explaining the importance and allowing for the officer to voice concerns about the policy.

In summary, researchers have shown the numerous other variables that may influence a police officer’s decision making process, and while race is typically the focus of traffic stop research, it is not always significant in the officer’s decision. Nonetheless, police administrators, often by the direction of new legislation, implement data tracking policies addressing the potential use of race as the single factor in decision making. These data tracking policies, as noted by Phillips [14], fail to identify the legal factors that play into an officer’s decision to stop a motorist. In addition, With row (as cited in Phillips [14]) suggested that data collection efforts fail to address those instances where officers choose to not stop a motorist, making comparisons between who was stopped and not stopped less valid. Furthermore, Phillips noted one of the problems associated with racial profiling data collection is the mere nature of self-reported data on a controversial topic, and according to Lundman [20], police officers have several reasons to inaccurately report data pertaining to racial profiling. Regardless of whether or not officers are accurately reporting their stops after it has already occurred, it is imperative that policy makers and administrators understand

What the officer is thinking prior to the stop. This study addressed the issue of race and not only the decision to stop, but the decision not to stop based on the officer’s observations.

2. Methodology

For this study a survey was conducted of a purposive sample of 412 sworn police officers in the Midwestern United States who were invited to respond to a descriptive traffic stop scenario presented as a vignette. Binary logistic regression was utilized to analyze the relationship between the dependent variable; an officer’s decision to stop or not stop a motorist for a minor traffic violation, when the race of that motorist is observed to be that of a racial or ethnic minority, and four predictor variables measured in the survey: 1) The presence of a statutorily implemented data tracking program, 2) years of service in policing, 3) previous discipline for violating the department’s racial profiling policy, and 4) the frequency of discussion relating to racial profiling data between an individual officer and his or her supervisor.

      2.1 Variables

Defining predictor variables for this study involved anecdotal understanding of police officer behavior combined with analysis of existing research resulting in the identification of four predictors. Although not a predictor variable used in this study, the perceived race of the driver must be addressed as it was analyzed as an influential variable in the decision to stop or not stop the motorist in the vignette. Visible racial or ethnic minority, as termed in this study, will include the remaining population that do not fall into the category of White.

The first predictor variable was defined by the statutory requirement calling for the collection of data identifying the race of those with who police contact either via traffic stop or voluntary contact. According to Laney [4], several state governments passed legislation requiring police department’s to track demographic information as well as contact disposition to identify if officers are utilizing race as a primary factor in decision making or are engaging in disparate treatment of minorities. Some states, such as Missouri and Kansas, allow for police to be disciplined if found to be engaged in racial profiling or disparate treatment of minorities [35]. Other states, such as Iowa, have not passed any legislation forbidding the practice [36]. In the analysis this was coded as either Yes, for the existence of a statute, or No for no existing statute.

The second predictor variable in this study was that of time in policing. The number of years as a sworn officer can be an important variable in an officer’s behavior, and the use of race in discretionary decision-making may be no different. This study will analyze the correlation between how long an officer is employed in policing, by years, with his or her decision to stop or not stop a visible ethnic or racial minority. In the analysis this was coded in completed years of service and was the only continuous variable included in the study.

The third predictor variable involved any prior discipline or consultation, one or more, for violating bias-based policing policies within the individual officer’s police department. As noted by legislation in both Kansas and Missouri, officers can be subject to discipline if found to be engaging in racial profiling [35]. Consequently, as noted by [19] applications of behavioral control mechanisms such as disciplinary procedures may be an important variable. In the analysis this was coded as either yes for prior discipline or consultation or No for no prior discipline or consultation.

The fourth predictor variable identified for this study was the frequency of discussion relating to racial profiling statistics. In other words, any notification to the officer from their supervisor as to the current status of their racial profiling statistics was included as a predictor. In the analysis this was also coded as Yes for prior notification or No for any prior notification. These discussions may come in the form of formal or informal periodic evaluations or even as part of a disciplinary procedure.

Four research questions were created addressing the variables selected for this study, each one addressing a police officer’s decision making process as it relates to racial profiling policy. The study was geared to identify a correlation, if any existed, between; the presence of data collection policies, an officer’s years as a sworn police officer, any prior discipline for violating the department’s racial profiling policy, the frequency of supervisory

Discussion pertaining to the individual officer’s racial profiling statistics and that officer’s decision to stop or not stop a racial or ethnic minority when the race is observed prior to stop.

      2.2 Sampling

Data were collected from a purposive sample of 412 sworn police officers in the Midwestern United States. Police departments were chosen based on their similarities in size and their representativeness of varying levels of data collection policies. In addition, each department selected represented a varying level of racial profiling data collection, meaning one department was not required by statute to collect data, one was required to collect on traffic stops only, and one was required by law to collect data on traffic stops while encouraging data collection on voluntary pedestrian contacts.

This study employed binary logistic regression with demographic data collected from a self-administered survey that included as scenario presented as a vignette. Tabachnick and Fidell [37] noted that a simple rule in computing sample size is N ≥ 104 + m with m being the number of predictor variables in the model. This study employed a minimum sample size of N ≥ 104 + 4 or N ≥ 108. The alpha level for this study was set at (α) = .05, a power level of .80, and a medium observed effect size of .50.

      2.3 Participation and Data Collection

Prior to data collection, letters of cooperation were collected from each respective chief of police. An original survey was created that included questions about routine practices and a single vignette describing a hypothetical scenario commonly encountered by police officers on patrol. According to Jenkins et al.  [11], vignettes can be used to collect data that represent collective group behavior. This vignette presented the officer with a scenario in which he or she observed a vehicle traveling just above their individual allowance for speeding, but as they prepared to turn around and stop the motorist, they observed the race of the driver to be that of a racial or ethnic minority. This officer had also recently been informed that their minority contact ratios were close to the expected representative contact ratios for their jurisdictions.

The survey was composed of nine questions, two of which addressed a vignette describing a routine traffic stop scenario in which the officer observed the race of the driver prior to initiating a stop. These categorical questions were coded as a yes/no response with allowable open-ended responses to explain the respondents answer. The other seven questions were related to the police environment; policy, statutes, years of service, and demographic data. Each participant was sent an email with a survey link. The email included an informed consent letter that explained the study as well as the rights of each participant. A range of dates, exactly two weeks, was scheduled with each participating agency during which the officers could complete the survey at their own free will if they chose to do so. At the end of the time frame provided, data was collected and exported into SPSS. There was no debriefing of the participants nor was there any follow-up questionnaire or survey conducted. In addition, there was no payment for participation. Of the 412 potential respondents, 176 actually completed the survey (a response rate of 43%). During the two week data collection period, only one reminder was sent out to participants.

3. Results

Following are the results of this quantitative analysis. Frequencies and percentages are reported as well as χ2 statistics and correlation coefficients as appropriate. In addition, results of the binary logistic regression analysis are included.

      3.1 Descriptive Statistics

The demographic information collected pertained only to an officer’s years of sworn service. For the 176 officers who completed the survey, the range of experience was between 1 year and 34 years sworn service. Table 1 contains the mean and standard deviation for the officers’ years of experience.

Officers were asked whether or not they would stop an observed racial or ethnic minority for a minor traffic violation after being recently told by their supervisor during a routine evaluation review that their minority contact/stop ratio was slightly higher than acceptable by department standards. Further, officers were asked what influenced their decision if they chose to not stop the driver. Of the 176 respondents, 104 chose to go ahead and stop the vehicle (59%) and 72 chose to not stop the vehicle (41%). Sixty-seven of the officers who chose to not stop the vehicle (93%) reported either the observed race or a departmentally implemented policy prompted their decision. Frequencies and percentages of questions relating to the independent variables are included in Table 2. These frequencies and percentages account for the 176 officers surveyed (35%) of the total population for these jurisdictions.

Note: Not all percentages may equal 100% due to rounding.

4. Analysis

The results of the chi-square tests are presented in Table 3. To test the relationship between an officer’s years of service and their decision to stop or not stop a motorist, we ran a point biserial correlation, and those results are presented in Table 4.

Note: Parenthetical values represent expected counts.

Evaluation of the chi square statistics revealed only one significant relationship between the predictors and an officer’s decision to stop or not stop a vehicle, and that was the presence of a state law requiring the collection of racial profiling data, χ2(1) = 10.90, p < .001. An officer receiving prior discipline, χ2 (1) = 1.73, p = .188, and frequency of discussion, χ2 (1) = 1.72, p = .190, were not found to be statistically significant. Finally, an officer’s years of service were not found to be statistically significant with their decision to stop or not stop a motorist rpb(170) < .01, p = .969.

We addressed research questions 1, 2, 3, and 4 using binary logistic regression analysis. The presence of a state law mandating collection of racial profiling data, prior discipline for violating the department’s racial profiling policy, the frequency of discussion of racial profiling data with supervisors, and an officer’s years of service were used as predictors in the outcome officer’s decision to stop or not stop a motorist based on the given scenario. Of these four predictors used in the logistic regression model, only one, the presence of a state law mandating the collection of racial profiling dataχ2 (4) = 13.21, p = .010, was found to be statistically significant. Evaluation of the odds ratio shows that holding all other independent variables constant, respondents who are in a state that has a statutorily mandated racial profiling data collection policy are 4.30 times more likely to not stop a minority motorist if they are slightly exceeding their departments’ expected minority contacts. The results of the binary logistic regression are reported in Table 5.

We identified one constant variable between each department surveyed, and that is the presence of a departmental policy that addresses racial profiling or bias-based policing that cites discipline for violating the policy. In order to evaluate the influence of having such a policy on the decision to not stop a minority motorist, we asked respondents who did not stop the vehicle based on the scenario if their knowledge of the policy had any influence on their decision. Of the 72 officers who did not stop the vehicle based on the given scenario, 52 (72%) stated that their understanding of their department policy was influential in their decision making process to avoid the stop.

Finally, we allowed officers to fill in their own, open-ended response on what influenced their decision to not stop the motorist in the given scenario. Of the 72 officers who did not stop the vehicle based on the scenario, 67 (93%), cited either fear of violating the policy or race itself as the basis of the decision to avoid the stop.

      4.1 Discussion and Interpretation

The results of this study indicated that the presence of a state law requiring data collection policies implemented in every police department can significantly (Exp(B) = 4.303) impact police officer decision making when it comes to conducting traffic stops on racial or ethnic minorities, which is a new contribution to the existing literature. Those officers who chose to avoid stopping a racial or ethnic minority stated they were influenced by the very department policy that was mandated by their state’s legislators. Other potential predictors, such as years of service, prior discipline for violating the policy, and the frequency of statistic discussion were found to not be significant.

Respondents in this study were asked to respond to a scenario in which they were presented a hypothetical situation that would not be uncommon for many police officers and were asked if they would stop or not stop the vehicle based on the information given. Of the 176 sworn police officers responding to this survey, 72 stated they would let the driver go. Ninety-seven percent of these officers stated that would let the driver go because of skin color or because of the policy in place addressing minority contacts.

This current study was geared toward identifying factors that influence a police officer’s decision to stop, or not stop, a motorist when the race of that motorist was observed to be a racial or ethnic minority. The identification of a statute that requires data collection as a significant influence in a police officer’s discretionary decision making process raises concern. The potential for police officers to choose to avoid heavily minority populated neighborhoods that may, in reality, need police patrol is notable. In this study, 72 police officers out of the 176 surveyed stated they would not stop a visible racial or ethnic minority if their minority contacts were slightly above what was expected. If an officer wants to avoid any associated labels with having a disproportionate number of minority stops, then action must be taken to rectify the numbers. There really is no other choice in the matter as if the numbers do not equal out, discipline is looming on the horizon.

Officers who chose to not stop the vehicle in the given scenario were allowed to explain why it was they chose to avoid the stop. Officers explained they were afraid of being “terminated” from their employment, afraid it would “skew my numbers the wrong way,” or simply “the fact that the motorist is a minority.” In fact, one officer reported the following: “We are routinely told to look at the race of the driver, and if they are a minority, let them go and stop a white driver.” If there is a risk of discipline for violating an ambiguous policy then it might be best to avoid violating the policy by any means.

The responses given by officers confirm that there is more involved in their decision making process than the mere observation of a violation. Officers are thinking about policy, what might be the repercussion of this stop, and whether or not they will be labeled erroneously based on their actions. All of these thoughts, in this study, impacted the decision to make a traffic stop on a visible racial or ethnic minority and, as such, were fresh on the minds of these officers.

The only significant variable identified in this study was that of a law being present that mandated racial profiling data collection. As officers worked in a state with a law mandating such practice, the odds of not stopping a visible or racial ethnic minority increased by 4.30 times holding all other independent variables constant. Stroshine et al. [29] noted how an officer’s individual interpretation of his or her surroundings can influence their decision making. Analysis of the data collected in this study reinforces this assertion. Even though the majority of officers did not specifically cite the state law as being significant in their decision making process, the fact that it was there proved to be statistically significant. One interpretation of this significance might be that an officer’s knowledge of this statutory requirement is influencing his or her discretion whether he or she consciously knows it or not. Another interpretation could be that the officers were simply not willing to share this information in their responses for any number of reasons.

It is apparent that the tone is set within each jurisdiction, perhaps each department, as to how a department will handle certain actions. Take for example, the handling of search incident to arrest after [38]. Some police officers began to tow every vehicle involved in a custodial arrest as it was a way around the warrantless search of the vehicle that had just been deemed in violation of the fourth amendment. When a vehicle is towed, an “inventory” must be completed of the vehicle to identify the suspect’s belongings in the vehicle. Police administrators either supported this decision, making it common practice, or they did not and issued unwritten directives informing their officers that they will not be towing every vehicle based simply on an arrest. The same interpretation falls from department to department when it comes to racial profiling policies. Some departments are going to take into account the demographic makeup of the officer’s district, or even surrounding districts, and hold him or her accountable accordingly, or they are going to take the total demographic makeup of the entire city and hold everyone to the same standard. Much like the search incident to arrest interpretation, neither one is technically wrong, but one is a perversion of the law’s intent and, with scrutiny, may even be deemed as violating someone’s rights.

The problem, as noted by [3], is that disproportionate minority contacts do not equate to racial profiling. The officers responding to this survey were keenly aware of their policies forbidding the practice of racial profiling, but the majority of those who chose to not stop the motorist believed that higher numbers did equate to a policy violation, at least in the eyes of their supervisors. The fear of discipline or termination was observed numerous times as a reason chose to disengage.

A large portion of the research surrounding racial profiling addressed how race influences the decision to stop [7, 24]. In fact, the focus tends to be on those variables that play into an officer’s decision when making stops; race is just one of the many variables. However, the influence of policy had not been included in any previous studies that we could find. Of the officers participating in this study who chose to not stop the vehicle, 97% stated that their department policy was influential in their decision. This variable was analyzed for frequencies and percentages only. Officers who responded to this question were only prompted to do so if they stated they were not going to stop the vehicle, and due to the follow-up nature of the question, the variable was not included in the logistic regression analysis. It is difficult to find a police agency of any size that does not have a policy banning the use of race as an indicator of criminal activity, and understandably so. This makes analysis of the policy’s influence somewhat problematic, but we cannot ignore the large number of officers who are citing its influence in this study. Again, the influence of policy in a police organization is proven to be quite strong and reflective of the organizational goals of the department.

      4.2 Limitations of the Study

As with any survey addressing sensitive topics, honesty of the respondents is a concern. While there were some officers who did not hold back, the data reveals some discrepancies in what the responses in the survey were and what the actual outcome of the analysis was. The presence of a state law mandating data collection policies was the only variable found to be statistically significant in the study. However, when asked directly about the influence of this state law, only 14% responded that they were influenced by the law. I am unsure as to why the discrepancy is identifiable. One interpretation could be that the officers were simply not willing to share this information. Alternatively, the officers may have not understood the question. At any rate, there is a concern with response bias due to the nature of the question. Anonymity was promised and explained in the informed consent document, but that promise comes with no concrete guarantees. The officers would have to take that promise for what it is worth in their own minds.

Generalizability is an issue. For the current study, the results should only be generalized to the departments from which data was drawn. Application of this data to outside agencies should be done with caution. These cities were predominantly urban with mixed races and cultures common in the Midwest. While I believe this data can be used to characterize the majority of police officers in the United States, there is no evidence to support such an assertion and the study was not constructed in such a way to be interpreted; as such, policies and laws were analyzed only to characterize the departments chosen.          

Conclusion

We found it quite troublesome that a respondent reported that his or her supervisor ordered them to seek out White drivers and stop them. A suggestion like this does not do little to promote proactive policing, nor does it bode well for the police department when this sort of order becomes public knowledge. Racial profiling laws are intended to eradicate the police use of race as the primary factor in stopping or investigating disproportionate members of any particular race, and that includes Whites whether they are at the heart of the policy or not. These policies are most certainly not intended to bring about what some call “reverse racial profiling,” which is nothing more than racial profiling. There must be an understanding on what constitutes racial profiling and we must move away from strict data analysis to identify the phenomenon.

With the lack of common agreement in defining racial profiling, policies implemented to combat its existence tend to be ambiguous. We use the term ambiguous because there is no agreement on what constitutes racial profiling amongst scholars [4], yet there are laws passed that are left open to interpretation and that interpretation typically falls into the category of numbers and only numbers. It typically makes no difference whether or not there was an actual violation that prompted the stop. All that matters is the race of the driver. [39-43] Noted that at least half of the time, race is not even noticeable to the officer due to the veil of darkness. Time of day, actual violation observed, and whether or not the race of the driver was even noticeable prior to the stop are amongst the many variables that should be taken into account when labeling behavior as problematic [44,45].

So much attention has been given to studying what factors go in to making a traffic stop that we have ignored the numerous reasons why an officer chooses to not make a stop. It is not feasible to stop every violation. There are many factors influencing an officer’s decision to stop, or not stop, a vehicle, and research must include the possibility that the officer is allowing stops to pass as opposed to vilifying their actions as based on observed race or ethnicity.

Acknowledgements

We would like to thank those members of the Lee’s Summit Police Department’s command staff who took time to provide input on the survey. We would also like to thank the three anonymous departments for their participation in this study.

Funding

The author did not receive any funding for this study. This is derived from the author’s doctoral dissertation.

 


 

Survey Question

 

M

 

SD

How long have you been a police officer?

13.73

7.28

 
Table 1: Mean and Standard Deviation for Years of Service
 

 

Survey Question

 

n

 

%

Based on the scenario above, would you stop this vehicle?

 

 

Yes

104

59

No

72

41

Was your decision to not stop this vehicle influenced by your understanding of any state law addressing racial profiling?

 

 

Yes

23

14

No

53

32

Does not apply

88

54

Was your decision to not stop this vehicle influenced by your department’s racial profiling policy?

 

 

Yes

67

93

No

5

7

Have you received discipline for violating your department’s racial profiling policy?

 

 

Yes

14

8

No

159

90

Does not apply

2

1

Are your personal racial profiling stats discussed with you?

 

 

Discussed

104

60

Not discussed

70

40

 

Table 2: Frequencies and Percentages for Participant Survey Responses

 

 

Would you stop the vehicle?

 

Predictor

 

Yes

 

No

 

χ2(1)

 

p

Presence of Racial profiling law in state

 

10.9

< .001

No law

28 [20]

5 [13]

 

 

Law present

76 [84]

66 [58]

 

 

Prior discipline

 

1.73

0.188

Yes

6 [8]

8 [6]

 

 

No

98 [96]

63 [65]

 

 

Racial statistics discussion

 

1.72

0.19

Does not discuss

46 [42]

24 [28]

 

 

Does discuss

58 [62]

46 [42]

 

 

 

Table 3: Chi square analyses of categorical predictors and decision to stop or not stop

 

 

Predictor

 

Decision to stop or not stop

Years of service

-0.003

 
Table 4: Point Biserial Correlation between Years of Service and Decision to Stop or Not Stop

 

 

 

Predictor

 

Log Reg Coefficient

 

Wald Statistic

 

P

 

Exp(B)1

Years of Service

-0.01

0.202

0.653

0.99

Prior Discipline

0.735

1.434

0.231

2.086

Mandated Policy

1.459

7.474

0.006

4.303

Stat Discussion

0.087

0.061

0.804

1.091

(Constant)

-0.381

0.392

0.531

0.683

Table 5: Logistic Regression: Predicting Stop or Not Stop

 

 

  1. Ramirez DH, Hoopes J, Quinla TL (2003) Defining racial profiling in a post September 11 world. American Criminal Law Review 3; 1195.
  2. Gabbidon SL, Marzette LN, Peterson SA (2007) Racial profiling and the courts: An empirical analysis of federal litigation, 1991 to 2006. Journal of Contemporary Criminal Justice, 23(3), 226-238. doi: 10.1177/1043986207306864.
  3. Barnum C, Perfetti RL (2010) Race-sensitive choices by police officers in traffic stop encounters. Police Quarterly 2: 180-208.
  4. Laney GP (2004) Racial profiling: Issues and federal legislative proposals and options. CRS Report for Congress.
  5. Schafer JA, Carter DL, Katz-Bannister AJ, Wells WM (2006) Decision making in traffic stop encounters: A multivariate analysis of police behavior. Police Quarterly 2: 184-209.
  6.  Cooper FR (2003) Understanding depolicing: Symbiosis theory and critical cultural theory. Villanova University Legal Working Paper Series. Villanova University School of Law Working Paper Series.
  7. Miller K (2007) racial profiling and postmodern society: Police responsiveness, image maintenance, and the left flank of police legitimacy. Journal of Contemporary Criminal Justice 3: 248-262.
  8. Capers B (2009) Policing race and place. Harvard Civil Rights-Civil Liberties Law Review 44: 43-78.
  9. Ingram JR (2007) the effect of neighborhood characteristics on traffic citation practices of the police. Police Quarterly 4: 371-393.
  10. Novak KJ, Chamlin MB (2012) racial threat, suspicion, and police behavior: The impact of race and place in traffic enforcement. Crime and Delinquency 2: 275-300.
  11. Jenkins N, Bloor M, Fischer J, Berney L, Neale J (2010) Putting it in context: The use of vignettes in qualitative interviewing. Qualitative Research 2: 175-198.
  12. Mendias C, Kehoe EJ (2006) Engagement of policing ideals and their relationship to the exercise of discretionary powers. Criminal Justice and Behavior 1: 70-92.
  13. Matrofski SD (2004) Controlling street-level police discretion. The ANNALS of the American Academy of Political and Social Science 593: 100-118.
  14. Phillips SW (2009) Using a vignette research design to examine traffic stop decision making of police officers: A research note. Criminal Justice Policy Review 4: 495-506.
  15. Higgins GE, Vito GF, Grossi EL (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 2: 166-183.
  16. Kennedy R (1997) Race, crime, and the law. New York, NY: Vintage Books.
  17. Risse M, Zeckhauser R (2004) racial profiling. Philosophy & Public Affairs 2: 131-170.
  18. Ouchi WG (1977) the relationship between organizational structure and organizational control. Administrative Science Quarterly 22: 95-113.
  19. Rowe WG, O’Brien J, Rouse MJ, Nixon RD (2012) Navy stories: Behavior versus professional control. Journal of Management Inquiry 1: 61-77.
  20. Lundman RJ (2012) Race and ethnicity missingness in the traffic stop data reported by 308 Massachusetts police agencies. Race and Justice 1: 42-65.
  21. Tomaskovic-Devey D, Warren P (2009) Explaining and eliminating racial profiling. Contexts 2: 34-39.
  22. Johnson D, Brazier D, Forrest K, Ketelhut C, Mason D, et al. (2011) Attitudes toward the use of racial/ethnic profiling to prevent crime and terrorism. Criminal Justice Policy Review 4: 422-447.
  23. Higgins GE, Gabbidon SL, Jordan KL (2008) Examining the generality of citizens’views on racial profiling in diverse situational contexts. Criminal Justice and Behavior 12: 1527-1541.
  24. Ioimo R, Tears RS, Meadows LA, Becton JB, Charles MT (2007) the police view of bias-based policing. Police Quarterly 3: 270-287.
  25. Reitzel J, Piquero AR (2006) Does it exist? Studying citizens’ attitudes of racial profiling. Police Quarterly 2:  161-183.
  26. Horowitz D, Levin M (2001) the legal, political, and public war on racial profiling and its unlikely victims. Texas Review of Law and Politics 1: 223-235.
  27. Boivin R, Cordeau G (2011) Measuring the impact of police discretion on official crime statistics: A research note. Police Quarterly 2: 186-203.
  28. Liederbach J, Travis III LF (2008) Wilson redux: Another look at varieties of police behavior. Police Quarterly 4: 447-467.
  29. Stroshine M, Alpert G, Dunham R (2008) the influence of “working rules” on police suspicion and discretionary decision making. Police Quarterly 3: 315-337.
  30. Vito GF, Walsh WF (2008) Suspicion and traffic stops: crime control or racial profiling. International Journal of Police Science and Management 1: 89-100.
  31. Pollock W, Oliver W, Menard S (2012) Measuring the problem: A national examination of disproportionate police contact. Criminal Justice Review 2: 153-173.
  32. Warren PY Farrell A (2009) the environmental context of racial profiling. The ANNALS of the American Academy of Political and Social Science 1: 52-63.
  33. Chappell AT, MacDonald JM, Manz PW (2006) the organizational determinants of police arrest decisions. Crime & Delinquency 2: 287-306.
  34. Wolfe SE, Piquero AR (2011) Organizational justice and police misconduct. Criminal Justice and Behavior 4: 332-353.
  35. Kansas Code of Criminal Procedure, Kan. Stat. Ann. 22-4601-4615 (2013).
  36. Iowa Code (2013).
  37. Tabachnick BG ,Fidell LS (2001) Using multivariate statistics (4th ed.). Needham Heights, MA: Pearson.
  38. Arizona v. Gant (2009) 507-542.
  39. Worden RE, McLean SJ, Wheeler AP (2012) Testing for racial profiling with the veil-of-darkness method. Police Quarterly 1: 92-111.
  40. Federal Bureau of Investigation (2011). Crime in the United States (Table 43).
  41. RSMO ch. 590 § 590.650. Racial profiling – Minority group defined – Reporting requirements – Annual report – Review of findings – Failure to comply – Fund for audio-visual equipment – Sobriety check points exempt. Retrieved from http://www.moga.mo.gov/statutes.
  42. Shane JM (2012) Police employee disciplinary matrix: An emerging concept. Police Quarterly 1: 62-91.
  43. Shultz M, Withrow BL (2004) racial profiling and organizational change. Criminal Justice Policy Review 4: 462-485.
  44. U.S. Census Bureau. (2011). The Hispanic Population. Retrieved from http://www.census.gov/prod/cen2010/briefs/c2010br-04.pdf.
  45. Wilson JQ (1968) Varieties of Police Behavior: The Management of Law and Order in Eight Communities. Cambridge, MA: Harvard University Press.

© by the Authors & Gavin Publishers. This is an Open Access Journal Article Published Under Attribution-Share Alike CC BY-SA: Creative Commons Attribution-Share Alike 4.0 International License. With this license, readers can share, distribute, download, even commercially, as long as the original source is properly cited. Read More About Open Access Policy.

Forensic Studies