Protecting Whiteness

Abstract

Focusing on intergroup anti-non-White bias in the criminal justice system, little attention is given to how Whites may additionally be protected from negative police treatment. This study examines intragroup bias via perceived suspect phenotypic racial stereotypicality (e.g., how strongly members possess physical features typical of their racial group) on severity of police use of force. It is hypothesized that the Whiter one appears, the more the suspect will be protected from police force. Internal use of force case files from a large police department were coded for severity of police force, and suspects’ booking photographs were scored for phenotypic racial stereotypicality. Regression analyses confirmed that police used less force with highly stereotypical Whites, and this protective effect was stronger than the effect for non-Whites. Results suggest that intragroup bias is a protective factor for Whites, but not for non-Whites, providing an additional route through which racial disparities in policing operate.

Keywords

race phenotypic stereotypicality intragroup bias police stereotype

The shooting deaths of unarmed Black men and women at the hands of law enforcement since the summer of 2014 continue to shock the nation’s conscience, both because of the frequency and the seeming disparity of how the state uses force against its citizens (Pew Research Center, 2014). From the Department of Justice’s recently launched National Initiative to Promote Community Trust and Justice (Department of Justice, 2014), to the President’s Task Force on 21st-Century Policing (President’s Task Force, 2015), and to seemingly countless local and national civil rights efforts around the United States (e.g., see Black Lives Matter movement), America’s political will seems turned toward protecting non-White citizens from police abuses.

Although a positive step, this framework privileges intergroup distinctions (over intragroup ones) and assumes that the bulk of the racial disparities observed in police/community interactions stem only from anti-Black or anti-non-White bias. However, there is robust evidence that intragroup distinctions (e.g., differences within a racial group) are often at least as large as intergroup ones (Eberhardt, Davies, Purdie-Vaughns, & Johnson, 2006) and that positive in-group biases are often stronger than out-group antipathy (Mummendey & Otten, 1998). What would it mean if that were true in the context of criminal justice? Might the protection that Whites receive compared to the negative treatment of non-Whites in criminal justice outcomes be an independent effect, effectively widening the racial gap in police treatment? The current research investigates those questions by examining positive intragroup bias for Whites via perceived phenotypic racial stereotypicality on the severity of police use of force used against arrestees held for comparable offenses.

Intergroup Bias in Policing Outcomes

Cultural discourse and empirical investigations of race and policing primarily focus on the intergroup level of analysis; that is, differences in negative treatment between racial groups, presumably reflecting out-group biases (particularly anti-Black biases). Racial disparity studies highlight the consistent disparate treatment that Blacks, Latinos, and other non-White subordinate group members receive within the criminal justice domain compared to Whites. Non-Whites are more likely to be stereotyped as criminals (Eberhardt, Goff, Purdie, & Davies, 2004; Niemann, Jennings, Rozelle, Baxter, & Sullivan, 1994), to be stopped by police (Ridgeway, 2006), are more likely to be searched (Gelman, Fagan, & Kiss, 2007), are exposed to more force in interactions (Walker, Spohn, & DeLone, 2007), are more likely to be arrested (Hartney & Vuong, 2009; Kochel, Wilson, & Mastrofski, 2011), and are more likely to report police mistreatment (Weitzer & Tuch, 1999, 2005) compared to Whites. These racial disparities are often viewed by the public as driven by police officers’ racially biased attitudes against non-Whites (Goff & Kahn, 2012). The Black Lives Matter movement echoes this refrain, emphasizing that police violence against Blacks denies basic humanness, rights, and ultimately life itself to Black citizens (Black Lives Matter, 2015).

Social psychological examinations have similarly centered on negative intergroup outcomes against non-Whites, focusing primarily on anti-non-White bias at both the implicit and explicit level. Studies of shooter bias, the phenomenon in which racial stereotypes impact decisions to shoot, find that, in video simulations, participants are more likely to shoot unarmed Blacks compared to unarmed Whites (Correll, Park, Judd, & Wittenbrink, 2002) as well as display bias against Latinos and other non-White subordinate group members (Sadler, Correll, Park, & Judd, 2012; Unkelbach, Forgas, & Denson, 2008). Police are also susceptible to this bias against non-White groups (Plant & Peruche, 2005, see also Correll et al., 2007; Sadler et al., 2012). In a nationally representative sample, the more individuals viewed Latinos as criminals, the harsher criminal justice crime control measures they supported (Welch, Payne, Chiricos, & Gertz, 2011). Implicitly, Blacks are more associated with weapons than Whites, such that individuals are more likely to misperceive a weapon when primed with Black faces (Payne, 2001). Studies have further examined the causal role of police officer racial attitudes on their field interactions with non-Whites compared to Whites. In one such study, police officers’ implicit dehumanization predicted the use of force against Black children (Goff, Jackson, Di Leone, Culotta, & DiTomasso, 2014). In trying to counteract these biases against non-Whites, police trainings to reduce implicit bias against non-Whites have been implemented across the country (Presidential Task Force on 21st-Century Policing, 2015), again targeting between group biases at the intergroup level.

Intragroup Bias and Policing

Because anti-non-White biases are often the focus of criminal justice research and intervention programs, intragroup biases that privilege lighter skinned non-Whites and disadvantage darker skinned ones are often ignored, despite a strong literature demonstrating their existence (Blair, Judd, & Chapleau, 2004; Eberhardt et al., 2006; Eberhardt et al., 2004; Livingston & Brewer, 2002; Maddox, 2004). Psychological research consistently has demonstrated that the most physically stereotypical racial group members—those deemed to be highly representative of their racial category—are subjected to greater levels of stereotyping and racial bias than lower phenotypically stereotypical racial group members (Eberhardt et al., 2006; Eberhardt et al., 2004). For example, highly phenotypically stereotypical Blacks (e.g., Blacks with darker skin tone, more stereotypic features) receive more negative implicit and explicit stereotyping, and more discrimination than low stereotypical Blacks (Kahn & Davies, 2011; Livingston & Brewer, 2002). Similar skin tone preferences favoring lighter skin tones over darker skin tones are also found for Hispanics (Uhlmann, Dasgupta, Elgueta, Greenwald, & Swanson, 2002).

Within the criminal justice domain, at the intragroup level, phenotypic racial stereotypicality negatively influences outcomes for non-Whites. Highly phenotypically stereotypical non-Whites are more likely to be perceived to be criminals by police and receive longer, and more deadly, criminal sentences than less stereotypical non-Whites (Blair et al., 2004; Eberhardt et al., 2006; Eberhardt et al., 2004; Viglione, Hannon, & DeFina, 2011). High stereotypical Blacks (e.g., darker skin tone, broader facial features) are also subjected to greater levels of “shooter bias” than low stereotypical Blacks, such that high stereotypical unarmed Black suspects are mistakenly “shot” in video game simulations more often than low stereotypical unarmed Black suspects (Kahn & Davies, 2011). The more the target is perceived to have racially stereotypical physical features, the more strongly the stereotype is applied, which leads to a greater number of racially biased shooting mistakes. Similar biases are evident in eyewitness identifications such that participants mistakenly remember a Black suspect as more phenotypically stereotypic of Blacks when committing a stereotypically Black crime (Osborne & Davies, 2013). Similarly, real-world political campaign ads that attempted to associate Barack Obama with crime showed him with darker skin tones (Messing, Jabon, & Plaut, 2015). Further, Whites that have more “Afrocentric features” typical of Blacks are more subject to Black racial stereotypes and receive more negative criminal justice outcomes, including more severe criminal sentences (Blair et al., 2004).

Pro-White Bias

With the emphasis on negative criminal justice outcomes for non-Whites, both within groups and between groups, less empirical attention has been given to the benefits that Whiteness in and of itself may play in police–suspect interactions. While non-Whiteness often involves punishment and negative treatment based on anti-non-White attitudes, pro-White bias describes the ways that, as the dominant group, Whiteness benefits, either overtly or indirectly, White individuals. According to Social Identity Theory (Tajfel & Turner, 1986), in-group members tend to favor in-group members often to a greater extent than derogating out-group members (Brewer, 1999; Mummendey & Otten, 1998; Mummendey, Otten, Berger, & Kessler, 2000). According to research on the positive–negative asymmetry, positive evaluations for the in-group can be more freely expressed because they are more socially acceptable than expressing negative attitudes toward an out-group (Mummendey & Otten, 1998). Whites as the dominant group may therefore favor their own group members in addition to negatively punishing the subordinate out-group.

Whites are also motivated to protect the in-group to maintain their privileged position in the social hierarchy (Sidanius & Pratto, 1999). This protection can take the form of denying the existence of racial prejudice (Branscombe, Schmitt, & Schiffhauer, 2007), downplaying their own social status (Kahn, Ho, Sidanius, & Pratto, 2009), and feeling threatened when privilege is brought up (Lowery, Knowles, & Unzueta, 2007). They may also draw sharper and stronger in-group distinctions regarding who is perceived to be an in-group or an out-group member (Castano, Yzerbyt, Bourguignon, & Seron, 2002), particularly when under threat (Miller, Maner, & Becker, 2010). America’s history with the one-drop rule can be viewed as the dominant group protecting Whiteness by excluding the privilege of the White race to anyone deemed to have even “one drop” of Black blood (Hickman, 1997). This legacy persists today and affects individuals’ perceptions and categorizations of mixed race individuals (Ho, Sidanius, Levin, & Banaji, 2011).

How might pro-White bias influence criminal justice outcomes? One additional way that Whiteness may be protected by those in power is through the application of police violence during a police–suspect interaction. While research has clearly shown that non-White individuals often receive harsher treatment involving the police, it is hypothesized that Whiteness, and the more White an individual appears, will bring with it corresponding protection during police interactions. This protective factor may be most evident at the intragroup, rather than intergroup, level, such that Whiteness will differentially protect Whites that appear most phenotypically representative of their racial category. That is, the inverse of the costs phenotypic racial stereotypicality has for Blacks and other non-Whites may also be true of Whites: the more stereotypically White they are, the less they experience harsh criminal justice outcomes.

Intragroup research finds that the more phenotypically representative of the racial group an individual is, the more he or she is the target of racial stereotypes associated with that racial group (Eberhardt et al., 2006; Eberhardt et al., 2004). Because Whites are stereotyped as safer in contrast to non-Whites (Devine & Elliot, 1995), the Whiter one phenotypically appears, the less threatening and less likely to be dangerous he or she may be perceived. This perception can influence how police behave in an interaction and make it less likely that an interaction escalates in severity of use of police force. If such an effect is found, then the function of race in policing not only is about negative treatment against non-Whites but also includes more positive treatment for Whites as the dominant group. It is only by examining both pieces—the discrimination enacted against non-Whites and the advantages that Whites receive—that the full scope of the issue can be observed.

Current Study

The current study naturalistically examines the effect of pro-White and anti-non-White intragroup bias on criminal justice outcomes, focusing on police officer use of force. With use of force case files and suspect booking photographs, we hypothesize that high phenotypically White suspects will receive less overall force in police interactions than less phenotypically White suspects. That is, the Whiter the suspect phenotypically appears, the more he or she will be shielded from police force. In line with prior literature, intragroup differences in phenotypic racial stereotypicality may exacerbate police use of force for non-White suspects, providing evidence of negative anti-non-White subordinate group bias. Due to racial stereotypes linking non-Whites with criminality (e.g., Devine & Elliot, 1995), highly phenotypically stereotypic non-Whites may receive more force than less stereotypic non-Whites, furthering the gap in overall police treatment between Whites and non-Whites. Importantly, pro-White intragroup bias may be as strong a protective effect as negative bias against non-Whites.

Method

Police Use of Force Case Files

In collaboration with the Center for Policing Equity, the research team requested a random sample of 250 police use of force cases collected from 2009 to 2010 in a large police department on the West Coast of the United States for the study. The police department represented a large urban city with a diverse population of over 1 million people. The police officer who accessed the data for the research team used a random number generator and randomly selected the cases. Of the 250 cases, 177 had complete data necessary for the study, including comprehensive case details, use of force ratings, and the minimum number of suspect trait ratings (8). The suspects in the cases consisted of 55% non-Whites (73 [44.5%] Latino, 24 [14.6%] African American, 18 [11.0%] Asian, 4 [.02%] Other) and 85.4% males, with an average age of 33.03 (SD = 11.14, median = 30.50). The case files contained incident information, which included case number and details about the arrest and suspect. The following variables were coded from the case files: (1) suspect gender (0 = female or 1 = male), (2) chemical influence: whether police indicated that there were signs of chemical influence involved (0 = no or 1 = yes), (3) mental health concerns: whether police indicated that there were signs of mental health issues (0 = no or 1 = yes), and (4) type of crime: nonviolent (0 = e.g., disturbance, pedestrian contact) or violent (1 = assault, domestic violence, and gang activity). These controls were selected a priori from the available case data, discussed with the collaborating police department, and were used in similar other police use of force studies (e.g., Goff et al., 2014).

Use of Force Severity Ratings

All cases in the sample involved at least one officer employing force, of varying severity, to a suspect in the arrest situation. Use of force ratings from the case files were recorded as the highest reported use of force from the officers involved in the suspect–police interaction. Ratings ranged from 1 (control hold, complaint, and injury), 2 (takedown), 3 (body weapons), 4 (OC/pepper spray), 5 (taser or impact weapon), 6 (canine apprehension), 7 (projective impact weapon) to 8 (deadly force or carotid). Higher numbers indicated more severe levels of use of force. The severity rating scale was established with assistance by the participating police department and reflects a traditional “use of force continuum” model employed by departments across the country (Garner, Schade, Hepburn, & Buchanan, 1995). It is also similar to the previously employed use of force severity scales (Goff et al., 2014).

Booking Photograph Demographics and Trait Ratings

Each use of force case was associated with a booking photograph of the arrested suspect taken as the suspect was being processed and booked after the incident. Each photograph contains a head shot from the shoulders up of the suspect. To obtain trait ratings of the photographs, a survey was posted on Amazon Mechanical Turk (MTurk) and participants were paid for their participation. This procedure has been done in prior research to obtain trait ratings with publically available suspect photographs in the criminal justice domain (Wilson & Rule, 2015) and in other real-world skin tone studies (Messing et al., 2015). Described as a study of person perception, participants were asked to rate one randomly selected photograph on a variety of demographics and trait dimensions. Each photograph received a minimum of 8 and a maximum of 20 ratings, with a total number of 2,891 ratings used for the study.

The sample of raters was 1,056 (36.53%) females and 1,810 (62.61%) males, with an average age of 35.38 years (SD = 12.6, median = 33.00). Forty-one (1.42%) of the raters had less than a high school degree, 441 (15.25%) completed high school degree, 1,057 (36.56%) had some college, 1,019 (35.25%) a college degree, and 308 (10.65%) an advanced degree. The sample leaned slightly liberal in political views, with 815 (28.18%) of the raters identified as conservative, 1,094 (37.85%) as liberal, and 957 (33.10%) as neither liberal nor conservative. There was also a range on reported self-reported socioeconomic status (SES), with 405 (14.01%) participants indicating high SES, 1,618 (55.97%) indicating average SES, and 843 (29.15%) indicating low SES. The diversity of the Mturk sample in the current study matches other reported MTurk sample distributions, which have been found to be as reliable as traditional laboratory samples (Buhrmester, Kwang, & Gosling, 2011). Photographs were rated on the following dimensions:

Perceived suspect race

The study is primarily interested in participants’ perceptions of the suspects’ race, as their perceptions activate associated group stereotypes to influence judgments and behaviors. Perceived suspect race was measured by asking the participant to indicate the race/ethnicity of the person in the presented photograph as White, African American/Black, Asian/Asian American, or Latino/Hispanic. Answers were coded as White or non-White (ratings of African American/Black, Asian/Asian American, Latino/Hispanic).

Perceived suspect phenotypic racial stereotypicality

Suspect phenotypic racial stereotypicality was measured using 1 item used in prior phenotypic stereotypicality research (Kahn, 2010, see also Eberhardt et al., 2006; Eberhardt et al., 2004). After first selecting a race for the suspect, participants were asked to rate the booking photograph on a scale of 1 (very low stereotypical) to 7 (very stereotypical) with the following instruction: “Not all people in the same racial group look the same physically. Some people look more stereotypical of their racial group than others, which is a dimension called stereotypicality. Please rate the following photographs on how stereotypical of their racial group they physically appear. You may use things like skin color, hairstyle, and other physical features to make your judgments.” Consistent with other stereotypicality research, this item focuses on individuals’ own definitions and perceptions of stereotypic racial group features, and thus lets their perceptions drive their phenotypic stereotypicality ratings.

Perceived suspect age

Participants indicated the perceived age of the individual in the presented photograph (open ended).

Perceived injury severity

In order to control for any visible injuries that the suspect obtained, participants rated “how physically beat up” the individual appeared to be on a scale of 1 (not at all) to 5 (very).

Results

Because the goal of the current study was to determine if suspects’ phenotypic racial stereotypicality related to the severity of force they received, participants’ ratings were aggregated into an overall mean value for each trait for the given suspect photograph. Means and standard deviations for the study variables are presented in Table 1. Participants’ ratings scaled reliability on perceived suspect demographic items (perceived age: α = .99; perceived injury severity: α = .93), and acceptably on the most subjective rating of perceived phenotypic racial stereotypicality (α = .63), and did not differ between Whites and non-Whites.

Table 1.
Mean and Standard Deviation of Study Variables.

Study variables White (n = 75) Non-White (n = 89) Overall (N = 164)

M (SD) M (SD) M (SD)

Highest use of force 3.12 (1.45) 2.89 (1.47) 2.99 (1.46)

Perceived racial phenotypic stereotypicality 4.55 (.48) 4.69 (.58) 4.63 (.54)

Suspect gender 0.87 (.34) 0.85 (.36) 0.86 (.35)

Perceived suspect age 35.17 (10.19) 30.93 (6.63) 32.87 (8.68)

Perceived injury severity 2.52 (.84) 2.12 (.76) 2.80 (.82)

Chemical influence 0.72 (.45) 0.72 (.45) 0.70 (.46)

Suspect mental health concerns 0.29 (.46) 0.29 (.46) 0.19 (.39)

Type of crime (violent) 0.21 (.41) 0.21 (.41) 0.18 (.39)

The perceived race for each suspect was coded by selecting the majority race indicated by participants for each photograph. The ratings from participants who matched the coded race were then used for the perceived phenotypic racial stereotypicality ratings (e.g., participants who perceived the photograph as White then answered how stereotypical of the White racial group the person was) and the subsequent suspect trait ratings. Cases in which there was no majority race rating (13 cases, 7.3%) were excluded from the analyses, as these cases most likely represent individuals on the borders of racial categories. Phenotypic racial stereotypicality analyses require that members consistently be perceived as a member of one category in order to examine variations within a given category, rather than between them (Kahn, 2010). This left a sample of 164 cases, with 75 cases perceived as White and 89 cases perceived as non-White (16 [9.8%] Asian, 23 [14.0%] African American, and 50 [30.5%] Latino). The units of analysis in the subsequent results are the suspect booking photograph, aggregated suspect ratings, and associated case details.

To test the effects of perceived suspect phenotypic racial stereotypicality and perceived suspect race on severity of police use of force, a regression-based bootstrapping method was used (Preacher & Hayes, 2004). With 5,000 samples, perceived suspect phenotypic racial stereotypicality (centered), perceived suspect race (White vs. non-White), and the interaction term were entered, with the addition of suspect gender, suspect mental health concerns, suspect chemical influence, type of crime, and perceived suspect age, and perceived injury severity entered as controls, in a regression predicting police use of force (see Table 2).^1
–3 Results confirmed a significant main effect of perceived racial phenotypic stereotypicality, b = −1.01, SE = .35, 95% confidence interval (CI) [−1.71, −.32], p = .004, which was qualified by a significant phenotypic stereotypicality by perceived suspect race interaction, b = 1.00, SE = .43, 95% CI [.15, 1.85], p = .02. The only significant covariate was suspect gender, b = 1.08, SE = .33, 95% CI [.42, 1.74], p = .002, such that more force was used against male compared to female suspects.

Table 2.
Results of Regression Analysis Predicting Use of Force.

Predictor b (SE) Confidence Interval

Suspect race (0 = White, 1 = non-White) −.22 (.25) [−.71, .27]

Perceived racial phenotypic stereotypicality −1.01 (.35)** [−1.71, −.32]

Race × Stereotypicality interaction 1.00 (.43)* [.15, 1.85]

Suspect gender (0 = female, 1 = male) 1.08 (.33)** [.42, 1.74]

Perceived suspect age 0.00 (.01) [−.03, .03]

Perceived injury severity .14 (.15) [−.15, .44]

Chemical influence 0.18 (.26) [−.33, .69]

Suspect mental health concerns 0.14 (.33) [−.51, .78]

Type of crime (0 = nonviolent, 1 = violent) 0.06 (.30) [−.53, .66]

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

The significant interaction was decomposed using simple slopes analyses (see Figure 1). For White suspects, perceived phenotypic racial stereotypicality was associated with a decrease in use of force severity, b = −1.01, SE = .35, 95% CI [−1.71, −.32], p = .004. The more phenotypically White an individual was perceived to be, the less police force was used during the interaction, indicating that phenotypic racial stereotypicality is a protective factor against police use of force for Whites. For non-White suspects, however, phenotypic racial stereotypicality did not significantly affect police use of force, b = −.01, SE = .26, p = .96, 95% CI [−.54, .51].

Figure 1.
Perceived suspect phenotypic racial stereotypicality by perceived race interaction on use of force.

Next, we examined the strength of pro-White intragroup bias compared against anti-non-White bias. The simple slopes for White and non-White suspects were significantly different from each other, t(149) = 3.04, p < .01 (Robinson, Tomek, & Schumacker, 2013).⁴ This result indicates that perceived racial stereotypicality affected Whites to a stronger degree than non-Whites, shielding Whites from use of force more than for non-Whites.

Discussion

Within-group variation in facial appearance and skin tone imbues significant real-world life advantages for Whites. Using a naturalistic sample of police use of force case files from a large police department on the West Coast, intragroup variations in phenotypic racial stereotypicality differentially predicted the severity of police use of force against White and non-White suspects, protecting Whites to a greater degree than non-Whites. For White suspects, controlling for relevant case and suspect variables, phenotypic racial stereotypicality was associated with a decrease in the severity of police use of force. This was not the case for non-White individuals, as phenotypic racial stereotypicality did not significantly affect the amount of force received in the interaction.

That Whiteness provides an additional benefit for Whites broadens the overall influence of racial disparities within policing. Protecting Whiteness may be as, or potentially more, important than derogating non-White out-groups. Racial disparities in policing represent more than simple anti-Black or anti-non-White attitudes but also reflect a simultaneous pro-White bias that protects Whites from harm. These findings are also consistent with research showing that Whites with more Afrocentric features typical of Blacks receive longer sentences in the criminal justice system compared to those with less of such features (Blair et al., 2004).

These findings further highlight the importance of examining differences in experiences within racial groups at the intragroup level rather than only at the intergroup level. Intragroup differences have received less empirical attention compared to intergroup disparities, despite their significant influence. Focusing on differences only at the categorical between-group level obscures important within-group variation and undervalues important sources of disparate treatment. Whiteness carries benefits at the group level but also for those most representative of the category within it. Understanding the effect of race requires both inter- and intragroup examinations.

A strength of the current study is its focus on individuals’ perceptions of suspect race and phenotypic stereotypicality. Rather than objective measures, which may not accurately reflect how suspects are perceived in the field, the perceived ratings provide a closer examination of the racial categorization process and its effects on subsequent judgments and behaviors. Its focus on perceptions is also consistent with the past stereotypicality research procedures (e.g., Eberhardt et al., 2006; Eberhardt et al., 2004) and other examinations of the effect of perceived suspect individual traits (e.g., trustworthiness) on criminal sentencing (Wilson & Rule, 2015). It further allows individuals to define “Whiteness” based on their own conceptions about the racial category. While the current study did not use police officer ratings of stereotypicality, taking the average rating shows how the suspect is consistently perceived across individuals. Future research could extend this process by focusing on police ratings of phenotypic racial stereotypicality (e.g., see Eberhardt et al., 2004) to connect directly to police behavior in the field. Further, these data speak to the extent that force is used when it is employed and not whether force is used or not in a given interaction. It may be that phenotypic racial stereotypicality similarly predicts whether or not force is ever used in an interaction.

These results are consistent with and reflect the subtle nature of contemporary forms of bias. In an era where egalitarianism is a purported cultural norm, modern racial bias is often hidden or more subtle in nature (Dovidio, Glick, & Rudman, 2005). Implicit bias or less conscious forms of prejudice and stereotyping produce discriminatory outcomes that are unnoticed by the perpetrators. Whiteness can serve protective functions largely under the surface, while the explicit focus remains on overt negative discrimination and prejudice against non-Whites. Precisely because of its invisible nature, bias is able to persist and have its influence remain relatively unabated.

Inconsistent with the past literature (Eberhardt et al., 2006; Eberhardt et al., 2004), phenotypic racial stereotypicality was not associated with an increase in the severity use of force that non-White suspects experienced. This nonfinding, however, may be due to the study’s focus on non-Whites as a comparison group, due to its primary interest in examining pro-White bias. The current study centered on the effect of Whiteness on use of force, which, along with the small sample size and lack of power, necessitated non-Whites together (Blacks, Latinos, and Asians) as the comparison out-group. Most phenotypic racial stereotypicality research has involved Blacks as an independent group, finding the strongest negative intragroup effects. However, as described in the notes, limiting the comparison sample to both Blacks and Latinos, who have stronger criminality stereotypes associated with their racial groups (Devine & Elliot, 1995; Marín, 1984; Niemann et al., 1994), did not significantly change the results. Since the non-White grouping may have limited the power to find predicted negative effects, we would expect that anti-Black bias, in addition to pro-White bias, would be present in larger scale evaluations of police behavior.

While the results suggest that phenotypic racial stereotypicality is associated with decreased police use of force for Whites, causal conclusions cannot be drawn from the current data. It is possible that other co-occurring factors may be driving the relationship, including differential suspect behavior or resistance that vary by phenotypic stereotypicality. However, experimental findings demonstrate that, even when suspect behavior is identical, racial stereotypes bias shooting behavior and decision-making (e.g., Correll et al., 2002). It is therefore unlikely that these effects are solely caused by differential suspect behaviors.

Most importantly, however, the results of the current research suggest that any policy to reduce racial disparities in police use of force should not focus on anti-non-White bias to the exclusion of intragroup and pro-White biases. The research literature on intragroup and positive White bias is robust and translates to the policing context in at least this field test. What would interventions, trainings, and policies look like if they took earnestly the prevalence of pro-White in-group bias and the importance of intragroup distinctions? The focus would go beyond negative bias against subordinate out-groups and also highlight the benefits that Whites are receiving within the criminal justice domain and beyond. Any training program to counteract racial disparities within policing must consider both aspects of bias to reduce observed racial gaps in outcomes. These findings suggest that, in response to community calls that “Black Lives Matter,” society should at least also consider how White lives also disproportionately benefit and how skin color shapes the law enforcement context for both groups.

Study variables	White (n = 75)	Non-White (n = 89)	Overall (N = 164)
Highest use of force	3.12 (1.45)	2.89 (1.47)	2.99 (1.46)
Perceived racial phenotypic stereotypicality	4.55 (.48)	4.69 (.58)	4.63 (.54)
Suspect gender	0.87 (.34)	0.85 (.36)	0.86 (.35)
Perceived suspect age	35.17 (10.19)	30.93 (6.63)	32.87 (8.68)
Perceived injury severity	2.52 (.84)	2.12 (.76)	2.80 (.82)
Chemical influence	0.72 (.45)	0.72 (.45)	0.70 (.46)
Suspect mental health concerns	0.29 (.46)	0.29 (.46)	0.19 (.39)
Type of crime (violent)	0.21 (.41)	0.21 (.41)	0.18 (.39)

Predictor	b (SE)	Confidence Interval
Suspect race (0 = White, 1 = non-White)	−.22 (.25)	[−.71, .27]
Perceived racial phenotypic stereotypicality	−1.01 (.35)**	[−1.71, −.32]
Race × Stereotypicality interaction	1.00 (.43)*	[.15, 1.85]
Suspect gender (0 = female, 1 = male)	1.08 (.33)**	[.42, 1.74]
Perceived suspect age	0.00 (.01)	[−.03, .03]
Perceived injury severity	.14 (.15)	[−.15, .44]
Chemical influence	0.18 (.26)	[−.33, .69]
Suspect mental health concerns	0.14 (.33)	[−.51, .78]
Type of crime (0 = nonviolent, 1 = violent)	0.06 (.30)	[−.53, .66]

Footnotes

Declaration of Conflicting Interests

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

Funding

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

Notes

References

Black Lives Matter. (2015). We affirm that All Black Lives matter. Black Lives Matter. Retrieved from http://blacklivesmatter.com/guiding-principles/

Blair

I. V.

Judd

C. M.

Chapleau

K. M.

(2004). The influence of Afrocentric facial features in criminal sentencing. Psychological Science, 15, 674–679. doi:10.1111/j.0956-7976.2004.00739.x

Branscombe

N. R.

Schmitt

M. T.

Schiffhauer

(2007). Racial attitudes in response to thoughts of White privilege. European Journal of Social Psychology, 37, 203–215. doi:10.1002/ejsp.348

Brewer

M. B.

(1999). The psychology of prejudice: Ingroup love or outgroup hate? Journal of Social Issues, 55, 429–444.

Buhrmester

Kwang

Gosling

S. D.

(2011). Amazon’s Mechanical Turk a new source of inexpensive, yet high-quality, data? Perspectives on Psychological Science, 6, 3–5. doi:10.1177/1745691610393980

Castano

Yzerbyt

Bourguignon

Seron

(2002). Who may enter? The impact of in-group identification on in-group/out-group categorization. Journal of Experimental Social Psychology, 38, 315–322.

Correll

Park

Judd

C. M.

Wittenbrink

(2002). The police officer’s dilemma: Using ethnicity to disambiguate potentially threatening individuals. Journal of Personality & Social Psychology, 83, 1314–1329. doi:10.1037//0022-3514.83.6.1314

Correll

Park

Judd

C. M.

Wittenbrink

Sadler

M. S.

Keesee

(2007). Across the thin blue line: Police officers and racial bias in the decision to shoot. Journal of Personality and Social Psychology, 92, 1006–1023. doi:https://dx-doi-org.web.bisu.edu.cn/10.1037/0022-3514.92.6.1006

Department of Justice. (2014, 9 18). Justice Department announces national effort to build trust between law enforcement and the communities they serve. Retrieved from http://www.justice.gov/opa/pr/justice-department-announces-national-effort-build-trust-between-law-enforcement-and

10.

Devine

P. G.

Elliot

A. J.

(1995). Are racial stereotypes really fading? The Princeton trilogy revisited. Personality and Social Psychology Bulletin, 21, 1139–1150. doi:10.1177/01461672952111002

11.

Dovidio

J. F.

Glick

P. E.

Rudman

L. A.

(2005). On the nature of prejudice: Fifty years after Allport. Malden, MA: Blackwell.

12.

Eberhardt

J. L.

Davies

P. G.

Purdie-Vaughns

V. J.

Johnson

S. L.

(2006). Looking deathworthy: Perceived stereotypicality of Black defendants predicts capital-sentencing outcomes. Psychological Science, 17, 383–386. doi:10.1111/j.1467-9280.2006.01716.x

13.

Eberhardt

J. L.

Goff

P. A.

Purdie

V. J.

Davies

P. G.

(2004). Seeing Black: Race, representation, and visual processing. Journal of Personality and Social Psychology, 87, 876–893. doi:10.1037/0022-3514.87.6.876

14.

Gelman

Fagan

Kiss

(2007). An analysis of the New York City police department’s “stop-and-frisk” policy in the context of claims of racial bias. Journal of the American Statistical Association, 102, 813–823. doi:10.1198/016214506000001040

15.

Garner

J. H.

Schade

Hepburn

Buchanan

(1995). Measuring the continuum of force used by and against the police. Criminal Justice Review, 20, 146–168.

16.

Goff

P. A.

Jackson

M. C.

Di Leone

B. A. L.

Culotta

C. M.

DiTomasso

N. A.

(2014). The essence of innocence: Consequences of dehumanizing Black children. Journal of Personality and Social Psychology, 106, 526–545. doi:10.1037/a0035663

17.

Goff

P. A.

Kahn

K. B.

(2012). Racial bias in policing: Why we know less than we should. Social Issues and Policy Review, 6, 177–210. doi:10.1111/j.1751-2409.2011.01039.x

18.

Hartney

Vuong

(2009). Created equal: Racial and ethnic disparities in the U.S. criminal justice system. Washington, DC: National Council on Crime and Delinquency.

19.

Hickman

C. B.

(1997). The devil and the one drop rule: Racial categories, African Americans, and the U.S. Census. Michigan Law Review, 95, 1161–1265.

20.

A. K.

Sidanius

Levin

D. T.

Banaji

M. R.

(2011). Evidence for hypodescent and racial hierarchy in the categorization and perception of biracial individuals. Journal of Personality and Social Psychology, 100, 492–506. doi:10.1037/a0021562

21.

Kahn

K. B.

(2010). The impact of Black stereotypicality on the experiences of stereotype threat. Los Angeles: University of California.

22.

Kahn

K. B.

Davies

P. G.

(2011). Differentially dangerous? Phenotypic racial stereotypicality increases implicit bias among ingroup and outgroup members. Group Processes and Intergroup Relations, 14, 569–580. doi:10.1177/1368430210374609

23.

Kahn

A. K.

Sidanius

Pratto

(2009). The space between us and them: Perceptions of status differences. Group Processes & Intergroup Relations, 12, 591–604. doi:10.1177/1368430209338716

24.

Kochel

T. R.

Wilson

D. B.

Mastrofski

S. D.

(2011). Effect of suspect race on officers’ arrest decisions. Criminology, 49, 473–512. doi:10.1111/j.1745-9125.2011.00230.x

25.

Livingston

R. W.

Brewer

M. B.

(2002). What are we really priming? Cue-based versus category-based processing of facial stimuli. Journal of Personality and Social Psychology, 82, 5–18. doi:10.1037/0022-3514.82.1.5

26.

Lowery

B. S.

Knowles

E. D.

Unzueta

M. M.

(2007). Framing inequity safely: Whites’ motivated perceptions of racial privilege. Personality and Social Psychology Bulletin, 33, 1237–1250. doi:10.1177/0146167207303016

27.

Maddox

K. B.

(2004). Perspectives of racial phenotypicality bias. Personality and Social Psychology Review, 8, 383–401. doi:10.1207/s15327957pspr0804_4

28.

Marín

(1984). Stereotyping hispanics: The differential effect of research method, label, and degree of contact. International Journal of Intercultural Relations, 8, 17–27. doi:10.1016/0147-1767(84)90005 -1

29.

Messing

Jabon

Plaut

(2015). Bias in the flesh: Skin complexion and stereotype consistency in political campaigns. Public Opinion Quarterly. doi: 10.1093/poq/nfv046

30.

Miller

S. L.

Maner

J. K.

Becker

D. V.

(2010). Self-protective biases in group categorization: Threat cues shape the psychological boundary between “us” and “them.” Journal of Personality and Social Psychology, 99, 62–77. doi:10.1037/a0018086

31.

Mummendey

Otten

(1998). Positive–negative asymmetry in social discrimination. European Review of Social Psychology, 9, 107–143. doi:10.1080/14792779843000063

32.

Mummendey

Otten

Berger

Kessler

(2000). Positive-negative asymmetry in social discrimination: Valence of evaluation and salience of categorization. Personality and Social Psychology Bulletin, 26, 1258–1270. doi:10.1177/0146167200262007

33.

Niemann

Y. F.

Jennings

Rozelle

R. M.

Baxter

J. C.

Sullivan

(1994). Use of free responses and cluster analysis to determine stereotypes of eight groups. Personality and Social Psychology Bulletin, 20, 379–390. doi:10.1177/0146167294204005

34.

Osborne

Davies

P. G.

(2013). Eyewitness identifications are affected by stereotypes about a suspect’s level of perceived stereotypicality. Group Processes & Intergroup Relations, 16, 488–504. doi:10.1177/1368430212454927

35.

Payne

B. K.

(2001). Prejudice and perception: The role of automatic and controlled processes in misperceiving a weapon. Journal of Personality and Social Psychology, 81, 181–192. doi:10.1037/0022-3514.81.2.181

36.

Pew Research Center. (2014, 8 18). Stark racial divisions in reactions to Ferguson police shooting. Retrieved from http://www.people-press.org/2014/08/18/stark-racial-divisions-in-reactions-to-ferguson-police-shooting/

37.

Plant

E. A.

Peruche

B. M.

(2005). The consequences of race for police officers’ responses to criminal suspects. Psychological Science, 16, 180–183. doi:10.1111/j.0956-7976.2005.00800.x

38.

Preacher

K. J.

Hayes

A. F.

(2004). SPSS and SAS procedures for estimating indirect effects in simple mediation models. Behavior Research Methods, Instruments, & Computers, 36, 717–731. doi:10.3758/BF03206553

39.

President’s Task Force on 21st Century Policing. (2015). Final report of the president’s task force on 21st century policing. Washington, DC: Office of Community Oriented Policing Services.

40.

Ridgeway

(2006). Assessing the effect of race bias in post-traffic stop outcomes using propensity scores. Journal of Quantitative Criminology, 22, 1–29. doi:10.1007/s10940-005-9000-9

41.

Robinson

C. D.

Tomek

Schumaker

R. E.

(2013). Tests of moderation effects: Difference in simple slopes versus the interaction term. Multiple Linear Regression Viewpoints, 39, 16–24.

42.

Sadler

M. S.

Correll

Park

Judd

C. M.

(2012). The world is not Black and White: Racial bias in the decision to shoot in a multiethnic context. Journal of Social Issues, 68, 286–313. doi:10.1111/j.1540-4560.2012.01749.x

43.

Sidanius

Pratto

(1999). Social dominance: An intergroup theory of social hierarchy and oppression. New York, NY: Cambridge University Press.

44.

Tajfel

Turner

J. C.

(1986). The social identity theory of intergroup behavior. In Worchel

Austin

(Eds.), Psychology of intergroup relations (pp. 7–24). Chicago, IL: Nelson-Hall.

45.

Uhlmann

Dasgupta

Elgueta

Greenwald

A. G.

Swanson

(2002). Subgroup prejudice based on skin color among Hispanics in the United States and Latin America. Social Cognition, 20, 198–226.

46.

Unkelbach

Forgas

J. P.

Denson

T. F.

(2008). The turban effect: The influence of Muslim headgear and induced affect on aggressive responses in the shooter bias paradigm. Journal of Experimental Social Psychology, 44, 1409–1413.

47.

Viglione

Hannon

DeFina

(2011). The impact of light skin on prison time for Black female offenders. The Social Science Journal, 48, 250–258. doi:10.1016/l.soscjj.2010.08.003

48.

Walker

Spohn

DeLone

(2007). The color of justice: Race, ethnicity, and crime in America (4th ed.). Belmont, CA: Cengage.

49.

Weitzer

Tuch

S. A.

(1999). Race, class, and perceptions of discrimination by the police. Crime & Delinquency, 45, 494–507.

50.

Weitzer

Tuch

S. A.

(2005). Racially biased policing: Determinates of citizens perceptions. Social Forces, 83, 1009–1030. doi:https://dx-doi-org.web.bisu.edu.cn/10.1353/sof.2005.0050

51.

Welch

Payne

A. A.

Chiricos

Gertz

(2011). The typification of hispanics as criminals and support for punitive crime control policies. Social Science Research, 40, 822–840. doi:10.1016/j.ssresearch.2010.09.012

52.

Wilson

J. P.

Rule

N. O.

(2015). Facial trustworthiness predicts extreme criminal-sentencing outcomes. Psychological Science, 26, 1325–1331. doi:10.1177/0956797615590992

Study variables	White (n = 75)	Non-White (n = 89)	Overall (N = 164)
Study variables	M (SD)	M (SD)	M (SD)
Highest use of force	3.12 (1.45)	2.89 (1.47)	2.99 (1.46)
Perceived racial phenotypic stereotypicality	4.55 (.48)	4.69 (.58)	4.63 (.54)
Suspect gender	0.87 (.34)	0.85 (.36)	0.86 (.35)
Perceived suspect age	35.17 (10.19)	30.93 (6.63)	32.87 (8.68)
Perceived injury severity	2.52 (.84)	2.12 (.76)	2.80 (.82)
Chemical influence	0.72 (.45)	0.72 (.45)	0.70 (.46)
Suspect mental health concerns	0.29 (.46)	0.29 (.46)	0.19 (.39)
Type of crime (violent)	0.21 (.41)	0.21 (.41)	0.18 (.39)