Profiling by Police: Tales of Two Cities

By C.J. Oakes

The Issue: Racial profiling by police is an issue which must be avoided if police are to be effective in their mission, yet there is evidence that profiling does exist. What can be done?

Profiling, disparity, and discrimination; these concepts have become mainstays in criminal justice and law-related fields. In recent years, there has been an abundance of studies seeking to understand if and why discriminatory practices occur among police with no clearly acceptable conclusions drawn.

Some studies indicate that if racial profiling and discrimination is occurring, it is isolated to individual officers and particular regions whereas there are other studies to indicate the opposite; that profiling and discrimination are rampant among police departments nationwide. With such conflicts emerging from the same data sets, it can be difficult to determine what is true. Hence, we will examine two such studies with a view to better understanding this complex topic.

Two Racial Profiling Studies

The first profiling study we will examine was conducted by Miller who examines the issue from the context of warning and ticket stops. He sought to understand if there is a correlation between the issuance of a warning and a pretext to make a stop. Miller studied both local and state police and found differing results for the two groups (Miller, 2008).
In seeking the correlation, Miller used self-report data from North Carolinian licensed drivers; results indicate that whereas race does appear to play some role, it is not the dominant role. Instead, such factors as vehicle age, speeding, and driver age play strong roles in the decision to make a warning stop. Beyond this, the study did not attempt to probe but offered detailed suggestions for future research (Miller, 2008).

The second racial profiling study to be examined, by Vito and Walsh (2008) focused on suspicion. The researchers wanted to determine the role of suspicion when officers made a decision to stop. Data was collected using Scantron forms from officers who had made the stops examined. Of course because the stop information came after-the-fact, the information is subject to scrutiny. It is as the saying, hindsight is always 20/20.

None-the-less, Vito and Walsh (2008) found that police were four times more likely to view African American males with suspicion than their White counterparts.

Racial Profiling Research Conclusions

For two independent studies to come to opposing conclusions nullifies the results of both. However, although these two studies appear at first to be contradictory in fact they are complimentary.

These two studies highlight an important missive of criminal justice and demonstrates one of the key problems to expanding the reliable knowledge base of the field.

The mission of criminal justice is simple: To ensure that violators of law are dealt with justly and in accord with social custom. If any member of society is treated in a biased manner according to such differences as race, gender, or sexual orientation, justice has failed to be done and the opposite occurs; an injustice is done. One study suggests that African-American males are not a targeted minority whereas the other demonstrates a clear correlation between race and arrest during a traffic stop.

However, the first study only went to the point of the stop. The second study found that Whites were stopped more frequently, ticketed more often, and arrested far less. Therefore, the conclusion that an African-American male is four times more likely to be arrested during a traffic stop than a White male is both true and misleading. This happens when research is based on subjective data collection.

Abdulameer Yousef Habeeb, Iraqi refugee who br...
Abdulameer Yousef Habeeb, Iraqi refugee who brought suit, represented by the ACLU over racial profiling and related matters after being arrested by United States Customs and Border Patrol agents in Havre, Montana. (Photo credit: Wikipedia)

One of the major problems to attaining sound, provable theory in criminal justice is a seeming inability to test theories according to experimentation. Most studies appear to rely on extant data and various other data collection techniques. Few, if any rely on empirical evidence for conclusions; one study reviewed for this paper, though not used, admitted that their findings were “not empirical” (Warren; Tomaskovic-Devey; Smith; Zingraff; Mason, 2008, p. 710). Because data is always subject to the internal desires and beliefs of the researchers, it is always subject to subjection. In other words, data is open to interpretation and we tend to always find what we are looking for.

This is the complete opposite of proving theory by experimentation. For example, it is one thing to make a claim that a particular soda contains so much acid that it will eat up a rusty nail; it is an entirely different matter to watch a rusty nail be eaten by the soda. Until criminologists find a way to prove theory by experimentation, the field will continue to churn out studies proving both sides of many issues. The issue surrounding the case of potential racial bias in the system is both disturbing and enlightening. Looking further into these two studies, we can learn much.

Strong City, Weak City

Throughout history there have been strong cities. The issue of racial profiling is like an impenetrable strong city, but every city has a weakness. For every study proving that racial profiling occurs on a systematic level there is another proving it does not; that the racial profiling is isolated to certain individuals.

Just as a city has strengths and weaknesses, so too, any research project. Perhaps the greatest strength to the abundance of research done toward understanding racial profiling is the abundance of research. Circular reasoning to be sure. However, much or most of the research is based on subjective data. Needed at this point are empirical studies proving the various theories. Still, a look at the two current studies under review here provides lessons for future researchers.

The Miller study (2008) excelled at analyzing the known data but admits that the limitation of such data is that Blacks and Hispanics are known to return self report surveys in both larger numbers than Whites, but also because they more often believe they were wronged. Such problems make sound conclusions impossible.

Still, Miller (2008) attempted to account for numerous variables when assessing the data. Such considerations as stops made as a result of a real or perceived threat, made as a result of a drug interdiction, or through a community policing measure are factored into the results. Never-the-less, the study continues to have the same limitation: All data is subjective and open to a multitude of interpretations.

The Vito and Walsh study (2008) perhaps provides greater evidence. As stated by the researchers, “Of course, these ‘self report’ data sources are subject to the same limitations that any study of this type faces” (Vito & Walsh, 2008, p. 93). This is to say, the study is not empirical. Hence, all conclusions are null and to be ignored until evidence is presented.
Still, the Vito and Walsh study of racial profiling provides many good concepts and ideas. For example, this study took the data collection beyond just the stop to the outcome of the stop: How many resulted in a warning, how many in a ticket, how many arrests made, was a warrant check made, was a search conducted, and was a search refused. Through careful consideration of these variables, Vito and Walsh could conclude that profiling does take place, but most notably on a local level. In addition, because of the careful attention to several measurable factors, this study could provide the framework for a future empirical study (Vito & Walsh, 2008).

How to Apply the Vito and Walsh Racial Profiling Study Empirically

Using the data collected in the Vito and Walsh study, it should be possible to construct simulations of events using computer driven software. In every simulation, variables are controlled and data collected, simultaneously compared, and the simulation adjusts accordingly. With each adjustment, data are collected as well as over regular intervals.

The data collection for the experiment would take into consideration a theory of human behavior that holds that all choices are driven according to internal needs/drives/motivations. This theory further holds that all choices regarding the fulfillment of needs are shaped according to values. Sometimes these choices conflict with the values of society; this most often results in the breaking of a law.

There is evidence that values play a major part in the decision to arrest or not (Crutchfield, 2010). Values are shaped by our environment; over time, we use combinations of values systems to develop our own unique set of values. These values play into every decision made, from the decision to conduct a stop to the decision to run a warrant check to the decision to warn, ticket, or arrest (Lewis, 2003).

Therefore, an empirical study into the legitimacy of stops, citations, and arrests must begin with a look into values. The software incorporated into the project would have parameters written into the program to account for certain basic or common values (beliefs). Numerical values and or codes would be assigned to each value and variable and accounted for.

Finally, volunteers from the street force, office staff, guards, supervisors, the public, and more would be asked to work through the scenarios as if playing a video game. Using artificial intelligence, the program will adjust and adapt as the values of the individual are noted and accounted for. The simulation will be enclosed and decorated so as to make it as real as possible. In addition, as the simulation progresses, the program will attempt to predict future choices based on data collected over time.

The study will require volunteers from a vast spectrum of society, as varied as possible. This will allow researchers to gather as much data relating to values and common values among groups as possible. This way, maximum data is collected, analyzed, and interpreted as soon as possible, perhaps even in real time.

From the beginning of empirical studies in the social sciences, Skinner Boxes sought to create a game-like feel to research projects. With the technology currently at our disposal, there is no reason for the absence of empirical studies into criminal justice and deviant behaviors. Through the use of simulations, the choices people make, whether biased or otherwise can be understood and even predicted.


Racial Profiling, like discrimination, is not likely to be entirely eliminated from the criminal justice system anytime soon. This is because most extant research and planned research is based on subjective, not empirical standards. Conclusions may be correct, but with the many conflicting voices, stronger proof is needed. There has been a need for sound theory and empirical studies to advance understanding within the criminal justice system; never has this been more evident and urgent than now.

The nation today stands more racially divided than ever, if the sheer volume of material attempting to understand the issue of profiling is any indication. What is needed is a sound theory coupled with empirical studies into values and needs.

Such a theory would explain why all choices are made and could reliably predict behavior; it would open a new era into criminal justice research. Such a theory does exist in raw form.



  • Crutchfield, R. D., Fernandez, A., & Martinez, J. (2010). Racial and Ethnic Disparity and Criminal Justice: How Much is Too Much? Journal of Criminal Law and Criminology, 100(3), 903-32.
  • Lewis, H. (2003). A Question of Values: Six Ways We Make the Personal Choices that Shape Our Lives (3rd ed.). Richmond, VA: Axios Press.
  • Miller, K. (2008). Police Stops, Pretext, and Racial Profiling: Explaining Warning and Ticket Stops Using Citizen Self-Reports. Journal of Ethnicity in Criminal Justice, 6(2), 123-49.
  • Vito, G. F., & Walsh, W. F. (Spring2008). Suspicion and traffic stops: Crime control or racial profiling. International Journal of Police Science and Management, 10(1), 89-100.

One thought on “Profiling by Police: Tales of Two Cities”

Leave a Reply