Racial Classification: Measuring a Multifaceted, Intersectional, and Geospatial Construct

The bifurcated emphases in race scholarship correspond to differences in how the concept of race is treated and measured. Those focusing on racial formation look at the social processes that define and transform racial categories themselves. This work examines racial boundary-making and asks when race, ethnicity, and/or nation become salient, to whom, and how so. It moves beyond using race as an explanatory variable for other outcomes and turns “race” into something to be analytically explained.Footnote ⁴ Studies in this vein explore how processes of institutional classification and individual identification construct the meaning of race. For instance, some look at how the evolving interests of the state construct racial and ethnic options on the census (Nobles Reference Nobles2000; Snipp Reference Snipp2003; Loveman and Muniz Reference Loveman and Muniz2007). Others examine how social movements and individuals engage in projects of shifting or crossing racial boundaries (Nagel Reference Nagel1995; Roth Reference Roth2005; Mora Reference Mora2014). Some interrogate the relationship between the categories used by institutional actors (including scholars) and the “groups” these categories are meant to represent (Loveman Reference Loveman1999; Brubaker Reference Brubaker2004). Together, this research on the social construction of racial categories positions the categories themselves as the outcome of interest and examines the processes through which they are formed and transformed.

By contrast, efforts to document racial inequality often rely on comparisons of some other outcome (e.g. incarceration rate, employment, educational attainment) across racial categories. This reverses the direction of causal inquiry, exploring the social consequences of existing racial categories (American Sociological Association 2003). Often, these categories derive from data sources like the census or survey instruments that capture an individual’s racial/ethnic self-identification, chosen from a limited set of options. Though comparisons of outcomes across racial categories usefully reveal disparities; the use of single measures of race reinforces the illusion that racial categories are static, ahistorical, and mutually exclusive (Martin and Yeung Reference Martin and Yeung2003). How, then, can scholars engage in the impactful work of explaining racial inequalities, without reifying racial categories?

Pulled Over offers a promising answer to this question—one that aligns with other efforts to take social constructionism seriously in studies of inequality. These efforts emphasize that race is multifaceted, perceptual, and context-dependent; and advocate for measures of race that reflect this (Greiner and Rubin Reference Greiner and Rubin2011; Sen and Wasow Reference Sen and Wasow2016; Roth Reference Roth2016; Saperstein et al. Reference Saperstein, Krizer and Penner2016). Studies of inequality that take this call seriously often incorporate multiple measures of race or measures that capture how race is salient in the specific interaction of interest (e.g. Pager et al. Reference Pager, Western and Bonikowski2009; Monk Reference Monk2015; Penner and Saperstein Reference Penner and Saperstein2015). Pulled Over is one such study. The book largely fits within the vein of research on racial inequalities in criminal justice practices as its primary purpose is to investigate differences in police treatment by race. The authors thus analyze race as an independent variable that affects investigatory stop practices. Yet, the study’s approach incorporates social constructionist insights into how the race variable is measured. The authors move beyond the typical nods to the socially constructed nature of the race variable, including several measures that reflect the multifaceted nature of race as it is experienced in everyday life. In addition, they analyze interaction effects that account for the intersectional and spatially embedded qualities of racialized identities and experiences.

Epp et al. draw on measures that capture both the asserted and ascribed components of “race.” While one measure—commonly used—asks respondents to self-identify, another asks, “if you met someone who did not know you, what race do you think they would think you are?” (171). This is the primary measure incorporated into their analysis, as it “best represents the perspective of the driver’s race as viewed by a police officer on the street” (ibid.).Footnote ⁵ They also include a measure of “street style,” asking respondents to identify the extent to which they look like a lawyer or doctor, construction worker, suburban parent, hip-hop star or deejay, etc. The authors hypothesize that these self-presentational attributes could be associated with the likelihood of being stopped and treatment during the stop. Using multiple measures of racial classification expands the definition of race beyond the common measure of self-identification. As a result, the authors address how race may become salient in the specific context of police stops by asking respondents questions that capture how an officer may have perceived them based on their physical features and self-presentation.Footnote ⁶

Epp et al. also situate race as an element of intersectional identities. In addition to self-reported “street style,” the authors incorporate material indicators by including measures of vehicle model and condition. They argue that, on the street, the most visible symbol of class is a driver’s vehicle. Certain vehicle makes also carry racialized meanings. The authors find that drivers of domestic luxury cars like Cadillacs—which are considered a cultural marker of African Americans in the United States—are more likely to be subject to an investigatory stop than drivers of other car models. Their analysis of this point includes interactions among race, gender, and age. For example, they find that young African-AmericanFootnote ⁷ men driving an older domestic luxury car have a 44 percent likelihood of experiencing an investigatory stop over the course of a year; white men driving such vehicles have under a 20 percent likelihood (70). By including these interaction effects, Epp et al. account for the reality that race is experienced in intersectional ways—other demographic characteristics and the material and cultural signifiers of class or status intersect with race to form part of the imagery associated with threat and criminality.

Additionally, Epp et al. emphasize the importance of paying attention to race in relationship to specific spatial contexts. Place has played a central role as an “enabling technology” of racial domination in the United States (Delaney Reference Delaney2002, 7). The racial order is bolstered by “the placement of all people in specific, but highly visible, circumstances” (Kobayashi and Peake Reference Kobayashi and Peake2000, 395).Footnote ⁸ Cityscapes continue to reflect historic patterns of spatialized racial discrimination and exclusion, manifesting in persistent residential segregation. Pulled Over accounts for the powerful material and symbolic associations between race and place by analyzing variation in stop practices by location. Epp et al. compare the proportion of stops by race across three kinds of metropolitan areas: the urban core, the inner suburbs, and outlying areas.Footnote ⁹ They find that, while speeding traffic stops do not vary by race and location, investigatory stops are skewed toward African Americans in the inner suburbs (71). This reflects the ways in which the police—like many others—are affected by and implicated in maintaining the emplaced character of racial difference. While African Americans experience a higher proportion of investigatory stops than whites in all geographic areas, this difference grows more substantial in outlying areas and is most pronounced in the historically white inner suburbs (70).

Though they treat race as an independent variable that predicts the likelihood of experiencing an investigatory stop, Epp et al. include measures that capture several of the nuances of race as it operates in everyday life. They are particularly attentive to how race is likely to be salient in the specific context of the police stop. This multifaceted approach to measuring race and analyzing its effects accounts for the socially constructed qualities of race, while documenting racism and racial inequalities.

Racial Inequality: Documenting Racial Disparities and Their Institutional Origins

The criminal justice system is a key site of inquiry for those concerned with racial inequality. It is one of several subsystems that contributes to compounding racial disparities across various realms (Pager and Shepherd Reference Pager and Shepherd2008; Reskin Reference Reskin2012). Indeed, involvement in the criminal justice system adversely affects labor market prospects (Western Reference Western2002; Pager Reference Pager2003), political participation (Uggen and Manza Reference Uggen and Manza2002), health (Massoglia Reference Massoglia2008), and community stability (Rose and Clear Reference Rose and Clear1998), to name but a few consequences. Policing initiates entry into the criminal justice system and is therefore an important juncture where racial disparities emerge. Research on policing outcomes reveals persistent inequalities, particularly between African Americans and whites. Racial disparities are endemic to forms of involuntary police contact, such as traffic stops, subject stops, searches, and arrests (Langton and Durose Reference Langton and Durose2013; Fagan et al. Reference Fagan, Braga, Brunson and Pattavina2016). Research also documents salient racial disparities in patterns of police use of force (Eith and Durose Reference Eith and Durose2011; Fryer Reference Fryer2016; Morrow et al. Reference Morrow, White and Fradella2017; Scott et al. Reference Scott, Ma, Sadler and Correll2017). Taken together, these patterns indicate that policing is manifestly racially stratified.

Efforts to identify discrimination as a source of stratification often focus on racial profiling and prejudice, situating the problem at the level of the individual officer. However, the empirical challenges of capturing officers’ decision-making processes complicate the conclusive identification of racial bias (Engel, Calnon and Bernard Reference Engel, Calnon and Bernard2002; Atiba Goff and Barsamian Kahn Reference Atiba Goff and Kahn2012). In addition, though police officers could hold more racially prejudicial attitudes than the general population (Eberhardt et al. Reference Eberhardt, Purdie, Goff and Davies2004), the links between racial attitudes and actual behaviors are precarious. Profiling studies suggest that officers use race as a criterion in forming suspicion, but that that legal or behavioral characteristics largely account for the decision to take action (Alpert, Macdonald and Dunham Reference Alpert, Macdonald and Dunham2005; Correll et al. Reference Correll, Park, Judd, Wittenbrink, Sadler and Keesee2007). Though racial profiling is a pernicious phenomenon, it is not the only potential source of racial discrimination.

Pulled Over shifts attention from individual officers to institutional practices. Epp et al. thoroughly document qualitative and quantitative differences in police stop experiences by race, and they account for the ongoing production of these disparities by embedding stop practices within specific historical and institutional contexts, thus drawing attention to meso-level determinants of racial inequality. While attention often focuses on individual officers—centering their racial attitudes and implicit biases as the mechanisms driving racial profiling—Pulled Over embeds officers’ actions within the set of racialized institutional practices surrounding the investigatory stop. Though the practice of encouraging officers to seek out “suspicious” drivers to investigate potential criminal misconduct is widely legitimated, it can trigger implicit racial stereotypes that result in the disproportionate targeting of African Americans. The authors identify this kind of institutional practice as one source of the gap between whites and African Americans in trust of the police. Like racial disparities, perceptions of fairness have deeper origins than the actions of individual officers. As Epp et al. explain, “if people who are subjected to an ongoing, discriminatory pattern learn to recognize it as such, they will come to conclude that the process is deeply unfair even if the officials carrying it out are unfailingly respectful and polite” (6, emphasis in original). In its focus on the institutional level, Pulled Over offers a much-needed analysis of racial disparities in policing outcomes and their implications.

Epp et al. begin by situating current investigatory stop practices within legal and historical frameworks, describing the development of “Broken Windows” theory,Footnote ¹⁰ significant Supreme Court decisions like Whren v. United States (1996),Footnote ¹¹ the evolution of programs like the Drug Enforcement Administration’s Operation Pipeline,Footnote ¹² and commonplace police training materials like Tactics for Criminal Patrol (Remsberg Reference Remsberg1995).Footnote ¹³ These guidelines and strategies laid the foundation for widespread reliance by law enforcement agencies on investigatory stops. Investigatory stops continue to be a sanctioned and legitimated policing tactic, thought to be effective and drawing professional rewards from both local departments and policing associations. They are thus an institutionalized practice: a “common way of doing things that, while not required by any specific official policy, are supported and legitimated by rules, training, and law, and that spread widely to become a commonly accepted activity” (11).

Yet, through common practice, investigatory stops have a distinctly racial character. Epp et al. cite concepts of framing and implicit bias to account for how stop practices produce racial outcomes. They explain that the police—like others—frame their experiences in a social and historical context that includes “the enduring stereotype that blacks are more likely to be violent, aggressive, and engaged in crime” (42). These stereotypes can operate on both conscious and nonconscious levels. Though officers may not express overt prejudice, they may be subject to the unconscious activation of implicit racial stereotypes. The authors argue that, in requiring officers to identify those who look “suspicious,” investigatory stops inevitably trigger stereotypes that frame black drivers as potential criminals. Investigatory stops are, hence, a racially framed institutionalized practice—one that “grows from and reproduces negative racial stereotypes” (12).

After establishing this framework, Epp et al. systematically document racial disparities in police stops, revealing the discriminatory nature of investigatory stops, in particular. The authors use survey and interview data to describe racial disparities in how police stops are conducted and experienced. They find that 25 percent of black drivers are stopped in a year, compared to 12 percent of white drivers, even though black drivers do not typically violate traffic laws more than white drivers and, on average, black drivers speed less than white drivers (57). White drivers report being stopped for traffic-safety violations while black drivers are often pulled over for ambiguous or unstated reasons consistent with the tactics associated with investigatory stops. Indeed, black drivers are 2.7 times more likely to be stopped in an investigatory stop than white drivers. The authors find that race interacts with gender, age, and class indicators; black men under the age of twenty-five have a 28 percent likelihood of being subject to an investigatory stop over the course of a year, compared to white women, who experience a 7 percent likelihood (67). The police are 70 percent more likely to stop drivers in the lowest quartile of vehicle value, though drivers of those domestic luxury cars that Epp et al. identify as a “cultural marker of minority race” are more likely to be stopped than those with another vehicle make (69). Notably, it is only in investigatory stops that race and its interactions with other status characteristics have this influence. In traffic-safety stops, the most important determinant of the likelihood of being stopped is driving behavior; race has no substantive or statistically significant impact in this study. Racial disproportionality in police stops is driven solely by investigatory stops.

Epp et al. then document differences in driver experiences during police stops, focusing on the demeanor of the officer and on intrusions and sanctions reported by drivers. For traffic-safety stops, they find small differences in intrusions and demeanor based on the race of the driver, but those differences are not statistically significant. However, for investigatory stops, they find large statistically significant differences. A much higher proportion of black drivers report experiencing intrusions and sanctions, including threats of arrest, vehicle searches, driver searches, and handcuffing. For instance, black drivers are five times more likely than white drivers to have their car searched (105). Black drivers also report more surly, hostile, and insulting behavior from the officer (82).Footnote ¹⁴ Epp et al. challenge the hypothesis that the more aggressive behavior from officers occurs in response to the driver’s demeanor. They find that black drivers are not significantly more disrespectful than white drivers during investigatory stops.Footnote ¹⁵ They subsequently developed a statistical model that considered a range of influences on intrusions and the officer’s demeanor. The model incorporates the drivers’ status characteristics like race, gender, age, and education, in addition to other hypothesized influences like vehicle value and appearance, driving behavior, the number of vehicle passengers present, the number of police officers present, and drivers’ level of distrust of authority. Epp et al. find that race is a statistically significant predictor of intrusions and officer demeanor in investigatory stops (103). They conclude: “African Americans are subjected to deeper investigatory intrusions not because they are poorer, less educated, more disrespectful, or more distrustful of the police than are whites. It is a racial disparity, pure and simple” (113). The scrutiny black drivers face stems from the scripts associated with the investigatory stop as a practice; officers are trained to conduct such stops through intrusive tactics and vague behaviors easily interpreted as disrespectful.

Epp et al. then draw out the implications of these disparities through an analysis of police stop experiences. They find a wide racial gap in how police stops are evaluated. African Americans, on average, evaluate their most recent police stop as less legitimate than whites across three dimensions: the extent the officer provided a legitimate reason to make the stop, the extent the officer behaved properly during the stop, and the extent the outcome was more severe than deserved. For instance, on a negative-to-positive scale ranging from one to five, African American men rate the legitimacy of the decision to stop an average of approximately 2.6, while white men gave an average rating of approximately 4 (128). Epp et al. situate racial differences in evaluations of stops in the differences in types of stops themselves. In a statistical model comparing black and white drivers’ evaluations of the legitimacy of police stops, they find that black drivers view investigatory stops as significantly less legitimate than traffic-safety stops,Footnote ¹⁶ while there is no such effect for white drivers.Footnote ¹⁷ This model controls for experiences during the stop, the drivers’ reported violation of traffic laws, and background predispositions like the driver’s political ideology, distrust in the government, and education (130). The authors argue that the practice of investigatory stops undermines perceptions of legitimacy, even if the stops are carried out in a procedurally fair manner. Drivers evaluate the fairness of the reason for the stop, rather than its outcome. In investigatory stops, the relationship between the stop and an actual traffic violation is unclear. Investigatory stops are therefore perceived as less legitimate and this perception comes to be collectively shared by African Americans based on both personal and vicarious experiences of such stops.

Through their analysis, Epp et al. isolate investigatory stops as a racially framed and racially biased institutional practice. In doing so, they illustrate the important connections between racial classification, institutional practices, and racial inequality. Racial classification is the determinative factor in the likelihood of experiencing an investigatory stop. Once stopped, the scripts and structures of the practice itself drive racial disproportionality in the experiences and evaluations of stops. Significantly, Epp et al. find that the widely employed and endorsed practice of the investigatory stop—in precipitating racial differences in experiences of law enforcement—ultimately undermines police legitimacy.

Racial Formation: Theorizing the Reification of Racial Hierarchies, Differences, and Identities

Though we have discussed race scholarship as bifurcated—with emphases on the social construction of race, on the one hand, and racial inequality, on the other—these two social processes are intimately linked. Racial categories and racial inequalities are mutually constitutive; they are co-produced through various institutions, one of which is the law. Critical Race Theory (CRT) and empirical Critical Race Theory (eCRT)Footnote ¹⁸ pay particular attention to the ways the law constructs race and is therefore part of the process of racial formation (Crenshaw Reference Crenshaw1995, xxv; Obasogie Reference Obasogie2013, Reference Obasogie2014; Gómez Reference Gómez2012; Paul-Emile Reference Paul-Emile2015). Analyzing the “role of law in the construction of race becomes, then, an examination of the possible ways in which law creates differences in physical appearance, of the extent to which law ascribes racialized meanings to physical features and ancestry, and of the ways in which law translates ideas about race into the material societal conditions that confirm and entrench those ideas” (Haney López Reference Haney López2006, 10). In addition to documenting racial inequality in law enforcement, Pulled Over attends to how law enforcement, police policy, and practices re-inscribe racial meanings and experiences in ways that solidify the racial order.

Epp et al. tie their empirical findings to a theorization of racial formation by cogently arguing that police stop experiences shape the contemporary social meaning of race. This happens through several mechanisms. First, the visibility of stop practices reflects and reproduces a racialized hierarchy of citizenship and standing. For passersby, the images of black drivers—stopped, detained, handcuffed, or searched on the side of the road—are easy to interpret in ways consistent with broader social meanings. These images reinforce the idea that African Americans are criminal, on the one hand, or subject of discriminatory targeting by law enforcement, on the other. Both interpretations correspond to an understanding of the relational places that groups in society occupy. These images and experiences reinforce the precarious status of black Americans, while “affirming whites’ sense of their equality in a community ruled by law” (138). Investigatory stops communicate messages that “some citizens are not free to move about as equal members of society” (136).

Second, stop practices contribute to the differences in experience and perceptions that characterize the contemporary racial divide. A substantially higher proportion of African Americans report distrust in the police compared to whites, corresponding to their experiences of investigatory stops. Black drivers are more likely to agree with the statement, “the police are out to get people like me” (approximately 25 percent of black men between 30–39 years old, compared to under 5 percent of white men) and approximately one in five black men in their 40s report that they would be uncomfortable calling the police if they needed help compared to approximately one in twenty white men of the same age (141). Epp et al. find large, statistically significant differences between black and white respondents across various dimensions of distrust. These findings suggest that racialized citizens experience qualitatively different relationships to the police, and, hence, to the state. Some members of society expect service and protection, while others anticipate surveillance and social control. Over 60 percent of African American drivers do not trust the police to do the right thing, compared to under 25 percent of white drivers. These racially disparate stop practices and experiences with the police contribute to defining and maintaining racial segregation boundaries. Black drivers report avoiding certain areas for fear of how they might be treated by the police; 40 percent of black respondents, compared to 12 percent of white respondents reported such behaviors (146). Investigatory stops communicate messages about who belongs and who is “out of place.” Through this practice, the police shape and maintain the spatial and experiential character of racial difference.

Finally, investigatory police stops shape racial identity. They foster the “groupness” of individuals perceived to be Black by developing a sense of shared history and fate. While African Americans already report more racial identificationFootnote ¹⁹ than whites, the experiences of an investigatory stop make race even more salient to the identity. Epp et al. describe this dynamic by including measures of racial identity in their survey. While 38 percent of African Americans, compared to 11 percent of whites, reported the highest level of racial identification, the experience of an investigatory stop impacts these identifications. On average, the experience of an investigatory stop corresponds to a 1.4 point increase, on a scale of seven, in racial identification for African Americans, compared to a 1.1 point decrease in racial identification for whites. Both effects are statistically significant. The authors conclude: “enduring an investigatory stop significantly increases racial identification among African Americans but significantly decreases racial identification among whites,” controlling for other correlates of racial identity including education, income, political ideology, gender, age, etc. (149). Bearing the inconvenience and intrusion of a stop reminds black drivers of how vulnerable they are to this form of racial subordination. The authors speculate that the experience of an investigatory stop disrupts white drivers’ taken-for-granted assumptions of racial privilege, decreasing their identification with a racial group.Footnote ²⁰ By drawing out the implications of racial disparities in stop practices on racial hierarchies, differences, and identities, Pulled Over shows that race is not merely an input that determines the likelihood of experiencing a stop, it is an outcome of the stop as well; police stops are part of the process of populating and transforming racial categories.