Perceived Stereotypicality of Black Defendants Predicts Capital-Sentencing Outcomes

Researchers previously have investigated the role of race in capital sentencing, and in particular, whether the race of the defendant or victim influences the likelihood of a death sentence. In the present study, we examined whether the likelihood of being sentenced to death is influenced by the degree to which a Black defendant is perceived to have a stereotypically Black appearance. Controlling for a wide array of factors, we found that in cases involving a White victim, the more stereotypically Black a defendant is perceived to be, the more likely that person is to be sentenced to death. Race matters in capital punishment. Even when statistically controlling for a wide variety of nonracial factors that may influence sentencing, numerous researchers have found that murderers of White victims are more likely than murderers of Black victims to be sentenced to death (Baldus, Pulaski, & Woodworth, 1983; Baldus, Woodworth, & Pulaski, 1985, 1990, 1994; Baldus,Woodworth, Zuckerman,Weiner, & Broffitt, 1998; Bowers, Pierce, & McDevitt, 1984; Gross & Mauro, 1989; Radelet, 1981; U.S. General Accounting Office, GAO, 1990). The U.S. GAO (1990) has described this race-of-victim effect as ‘‘remarkably consistent across data sets, states, data collection methods, and analytic techniques’’ (p. 5). In one of the most comprehensive studies to date, the race of the victim and the race of the defendant each were found to influence sentencing (Baldus et al., 1998). Not only did killing a White person rather than a Black person increase the likelihood of being sentenced to death, but also Black defendants were more likely than White defendants to be sentenced to death. In the current research, we used the data set from this study by Baldus and his colleagues (1998) to investigate whether the probability of receiving the death penalty is significantly influenced by the degree to which the defendant is perceived to have a stereotypically Black appearance (e.g., broad nose, thick lips, dark skin). In particular, we considered the effect of a Black defendant’s perceived stereotypicality for those cases in which race is most salient—when a Black defendant is charged with murdering a White victim. Although systematic studies of death sentencing have been conducted for decades, noprior studies have examined this potential influence of physical appearance on death-sentencing decisions. A growing body of research demonstrates that people more readily apply racial stereotypes to Blacks who are thought to look more stereotypically Black, compared with Blacks who are thought to look less stereotypically Black (Blair, Judd, & Fallman, 2004; Blair, Judd, Sadler, & Jenkins, 2002; Eberhardt, Goff, Purdie, & Davies, 2004; Maddox, 2004; Maddox & Gray, 2002, 2004). People associate Black physical traits with criminality in particular. The more stereotypically Black a person’s physical traits appear to be, the more criminal that person is perceived to be (Eberhardt et al., 2004). A recent study found that perceived stereotypicality correlated with the actual sentencing decisions of judges (Blair, Judd, & Chapleau, 2004). Even with differences in defendants’ criminal histories statistically controlled, those defendants who possessed the most stereotypically Black facial features served up to 8 months longer in prison for felonies than defendants who possessed the least stereotypically Black features. The present study examined the extent to which perceived stereotypicality of Black defendants influenced jurors’ death-sentencing decisions in cases with both White and Black victims. We argue that only in death-eligible cases involving White victims—cases in which race is most salient—will Black defendants’ physical traits function as a significant determinant of deathworthiness. PHASE I: BLACK DEFENDANT, WHITE VICTIM Method Weused an extensive database (compiled by Baldus et al., 1998) containing more than 600 death-eligible cases from PhiladelAddress correspondence to Jennifer L. Eberhardt, Department of Psychology, Stanford University, Jordan Hall, Building 420, Stanford, CA 94305-2130, e-mail: [email protected]. PSYCHOLOGICAL SCIENCE Volume 17—Number 5 383 Copyright r 2006 Association for Psychological Science at University of North Carolina at Chapel Hill on January 5, 2015 pss.sagepub.com Downloaded from phia, Pennsylvania, that advanced to penalty phase between 1979 and 1999. Forty-four of these cases involved Black male defendants who were convicted of murdering White victims. We obtained the photographs of these Black defendants and presented all 44 of them (in a slide-show format) to naive raters who did not know that the photographs depicted convicted murderers. Raters were asked to rate the stereotypicality of each Black defendant’s appearance and were told they could use any number of features (e.g., lips, nose, hair texture, skin tone) to arrive at their judgments (Fig. 1). Stanford undergraduates served as the raters. To control for potential order effects, we presented the photographs in a different random order in each of two sessions. Thirty-two raters (26 White, 4 Asian, and 2 of other ethnicities) participated in one session, and 19 raters (6 White, 11 Asian, and 2 of other ethnicities) participated in the second session. The raters were shown a black-and-white photograph of each defendant’s face. The photographs were edited such that the backgrounds and image sizes were standardized, and only the face and a portion of the neck were visible. Raters were told that all the faces they would be viewing were of Black males. The defendants’ faces were projected one at a time onto a screen at the front of the room for 4 s each as participants recorded stereotypicality ratings using a scale from 1 (not at all stereotypical) to 11 (extremely stereotypical). In both sessions, raters were kept blind to the purpose of the study and the identity of the men in the photographs. The data were analyzed for effects of order and rater’s race, but none emerged. Results We computed an analysis of covariance (ANCOVA) using stereotypicality (low-high median split) as the independent variable, the percentage of death sentences imposed as the dependent variable, and six nonracial factors known to influence sentencing (Baldus et al., 1998; Landy & Aronson, 1969; Stewart, 1980) as covariates: (a) aggravating circumstances, (b) mitigating circumstances, (c) severity of the murder (as determined by blind ratings of the cases once purged of racial information), (d) the defendant’s socioeconomic status, (e) the victim’s socioeconomic status, and (f) the defendant’s attractiveness. As per Pennsylvania statute (Judiciary and Judicial Procedure, 2005), aggravating circumstances included factors such as the victim’s status as a police officer, prosecution witness, or drug-trafficking competitor; the defendant’s prior convictions for voluntary manslaughter or violent felonies; and characteristics of the crime, such as torture, kidnapping, or payment for the murder. Mitigating circumstances included factors such as the defendant’s youth or advanced age, extreme mental or emotional disturbance, lack of prior criminal convictions, minor or coerced role in the crime, and impaired ability to appreciate the criminality of his conduct. The Baldus database of death-eligible defendants is arguably one of the most comprehensive to date; using it allowed us to control for the key variables known to influence sentencing outcomes. The results confirmed that, above and beyond the effects of the covariates, defendants whose appearance was perceived asmore stereotypically Black were more likely to receive a death sentence than defendants whose appearance was perceived as less stereotypically Black, F(1, 36)5 4.11, p< .05,Z 2 p 1⁄4 :10 (Fig. 2a). In fact, 24.4% of those Black defendants who fell in the lower half of the stereotypicality distribution received a death sentence, whereas 57.5% of those Black defendants who fell in the upper half received a death sentence. PHASE II: BLACK DEFENDANT, BLACK VICTIM Method Using the same database and procedures described earlier, we examined whether this stereotypicality effect extended to cases in which the victims were Black. Of all cases that advanced to penalty phase, 308 involved Black male defendants who were convicted of murdering Black victims. The photographs for all of these defendants were obtained. The death-sentencing rate for these 308 defendants, however, was only 27% (as compared with 41% for the cases withWhite victims). Given both the low deathsentencing rate and the large number of cases involving Black defendants and Black victims, we selected 118 of these 308 cases randomly from the database with the stipulation that those defendants receiving the death sentence be oversampled. This oversampling yielded a subset of cases in which the deathsentencing rate (46%) was not significantly different from that for the cases withWhite victims (41%; F5 1). Using this subset provided a conservative test of our hypothesis. We then preFig. 1. Examples of variation in stereotypicality of Black faces. These images are the faces of people with no criminal history and are shown here for illustrative purposes only. The face on the right would be considered more stereotypically Black than the face on the left. With the exception of defendant’s attractiveness, all of the covariates employed here were included in the Baldus database and have been described in detail elsewhere (e.g., see Baldus et al., 1998). We added defendant’s attractiveness, basing this variable on 42 naive participants’ ratings of the defendants’ faces using a scale from 1 (not at all attractive) to 11 (extremely attractive). 384 Volume 17—Number 5 Looking Deathworthy at University of North Carolina at Chapel Hill on January 5, 2015 pss.sagepub.com Downloaded from sented this subset of Black defendants who murdered Black victims to 18 raters (12 White and 6 Asian), who rated the faces on stereotypicality. Results Employing the same analyses as we did for the cases with White victims, we found that the perceived stereotypicality of Black defendants convicted of murdering Black victims did not predict death sentencing, F(1, 110)< 1 (Fig. 2b). Black defendants who fell in the upper and lower halves of the stereotypicality distribution were sentenced to death at almost identical rates (45% vs. 46.6%, respectively). Thus, defendants who were perceived to be more stereotypically Black were more likely to be sentenced to death only when their victims were White. Although the two phases of this experiment were designed and conducted separately, readers may be interested in knowing whether combining the data from the two phases would produce a significant interactive effect of victims’ race and defendants’ stereotypicality on death-sentencing outcomes. Analysis confirmed that the interaction of victims’ race (Black vs. White) and defendants’ stereotypicality (low vs. high) was indeed significant, F(1, 158) 5 4.97, p < .05, Z 2 p 1⁄4 :03.