The Threatening Nature of " Rap " Music

APA NLM tapraid5/zdn-law/zdn-law/zdn00316/zdn2260d16z xppws S⫽1 7/12/16 7:01 Art: 2016-1370 Psychology, Public Policy, and Law 2016, Vol. 22, No. 3, 000 © 2016 American Psychological Association 1076-8971/16/$12.00 http://dx.doi.org/10.1037/law0000093 The Threatening Nature of “Rap” Music This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. AQ: au AQ: 1 Adam Dunbar, Charis E. Kubrin, and Nicholas Scurich University of California—Irvine Rap music has had a contentious relationship with the legal system, including censorship, regulation, and artists being arrested for lewd and profane performances. More recently, rap lyrics have been introduced by prosecutors to establish guilt in criminal trials. Some fear this form of artistic expression will be inappropriately interpreted as literal and threatening, perhaps because of stereotypes. Only a handful of studies have examined whether rap lyrics are evaluated using stereotypes, yet these studies were conducted in the 1990s—a period of heightened scrutiny for rap—and used nonoptimal methods. This study presents 3 experiments that examine the impact of genre-specific stereotypes on the evaluation of violent song lyrics by manipulating the musical genre (rap vs. country) while holding constant the actual lyrics. Study 1, a direct replication of previous research, found that participants deemed identical lyrics more literal, offensive, and in greater need of regulation when they were characterized as rap compared with country. Study 2 was a conceptual replication (i.e., same design but different stimuli), and again detected this effect. Study 3 used the same approach but experimentally manipulated the race of the author of the lyrics. A main effect was detected for the genre, with rap evaluated more negatively than country or a control condition with no label. However, no effects were found for the race of the lyrics’ author nor were interactions were detected. Collectively, these findings highlight the possibility that rap lyrics could inappropriately impact jurors when admitted as evidence to prove guilt. Keywords: stereotypes, decision making, evidence, criminal law, rap music misogynistic attitudes (Gan, Zillmann, & Mitrock, 1997), and antisocial behavior (Hansen & Hansen, 1990). Not surprisingly then, from its inception, rap music has been policed and monitored in exceptional ways. For example, early on, rap music’s distribution was limited by legal sanctions and police disruption, as law enforcement attempted to disrupt sales and get records by rap artists pulled from store shelves (Hirsch, 2014). Some of the music was also censored because it was believed that the lyrics were obscene (Crenshaw, 1991; Dixon & Linz, 1997). In the late 1980s and early 1990s, artists across the country such as LL Cool J, Too Short, and 2 Live Crew were arrested for performances that authorities regarded as lewd or profane (Blecha, 2004; Crenshaw, 1991; Dixon & Linz, 1997), while other artists were denied opportunities to perform in public venues, often because of police pressure (Rose, 1994).1 Rap music is rhymed storytelling that represents the political and social experiences common to inner-city communities throughout the United States (Perry, 2004; Rose, 1994). Scholars consider rap an outgrowth of the slave trade (Gilroy, 1995), originating from a tradition of African storytelling (Keyes, 2002), that is often used as a vehicle to express the economic and social frustration of the Black community (Rose, 1994). These characterizations, in one way or another, all emphasize that rap is a form of cultural expression that prioritizes Black voices. Perhaps more than any other music genre, rap has had a contentious relationship with the legal system. This is due, in large part, to the perception that rap music is threatening and dangerous. In an analysis of how the news media portray different music genres, Binder (1993) found that rap is presented through a “danger to society” frame, based on the idea that listeners of the genre are likely to become threats to society. Binder (1993) also found that the media is more likely to characterize listeners of rap as being prone to violence compared with listeners of heavy metal. In voicing their opposition to rap, opponents commonly reference studies that purport to show a link between listening to rap music and acceptance of violence (Johnson, Jackson, & Gatto, 1995), 1 AQ: 2 Fn1 As just one example, in 1990, a case was brought against 2 Live Crew based on the claim that their album, As Nasty As They Wanna Be, was obscene under state law. A judge ruled that the lyrics met the legal definition of obscenity because they: (a) had an excessive interest in sexual matters, (b) described sexual conduct as defined by state law, and (c) lacked serious literary, artistic, political, or scientific value (Skywalker Records, Inc. v. Navarro, 1990). By defining the lyrics as obscene, the AQ: 18 album was barred from being sold in record stores and the music was banned from being performed in concert. After ignoring the ruling, 2 Live Crew was arrested for performing obscene material at an adult-only nightclub. During the trial, Henry Louis Gates Jr., an historian, literary scholar, and cultural critic, testified that the songs purposefully used exaggeration to critique stereotypes about Black hyper-sexuality and were merely satire. While a jury acquitted 2 Live Crew, an appellate court maintained that the album was offensive and provided no artistic value. Critics of the decision believe that the obscenity label stemmed from stereotypes about rappers and rap music that were not applied to other sexually explicit or offensive material from White entertainers, such as Madonna or Andrew Dice Clay (Crenshaw, 1991; Dixon & Linz, 1997). Adam Dunbar and Charis E. Kubrin, Department of Criminology, Law and Society, University of California—Irvine; Nicholas Scurich, Departments of Criminology, Law and Society, and Psychology and Social Behavior, University of California—Irvine. Correspondence concerning this article should be addressed to Charis E. Kubrin, Department of Criminology, Law and Society, University of California—Irvine, Social Ecology II Room 3379, Irvine, CA 92697–7080. AQ: 17 E-mail: ckubrin@uci.edu 1 APA NLM tapraid5/zdn-law/zdn-law/zdn00316/zdn2260d16z xppws S⫽1 7/12/16 7:01 Art: 2016-1370 2 This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. Fn2 DUNBAR, KUBRIN, AND SCURICH Indeed, rap is unique in how it has been responded to, policed, and regulated by legal authorities.2 This has been especially true for the subgenre of gangsta rap, the most controversial type of rap music, which has gained global attention for “its vivid sexist, misogynistic, and homophobic lyrics, as well as its violent depiction of urban ghetto life in America” (Abrams, 2000, p. 198). Gangsta rappers chronicle life in poor and working class Black neighborhoods (Keyes, 2002; Kitwana, 1994; Kubrin, 2005a; Rose, 1994) and their lyrics reference poverty (Perry, 2004), crime and violence (Kubrin, 2005a), misogyny (Weitzer & Kubrin, 2009), the loss of loved ones (Kubrin, 2005b; Rose, 1994), and police surveillance (Rose, 1994). The regulation of rap continues today, but it has changed in form and function. As Kubrin and Nielson (2014) recently argued, we are currently witnessing a shift from censoring rap lyrics to using rap lyrics as evidence in criminal trials to secure convictions. In courtrooms across the country, rap lyrics, specifically gangsta rap lyrics, are being introduced by prosecutors to establish guilt. In so doing, the lyrics are treated like autobiographical confessions rather than art or entertainment. The vast majority of cases where this is happening involve aspiring rappers, nearly all of whom are young men of color from impoverished neighborhoods (Kubrin & Nielson, 2014). The practice of using rap lyrics as evidence in criminal trials raises questions about whether prosecutors, judges, and jurors may be relying on perceptions and stereotypes about rappers and rap music in their interpretation of the lyrics. Our study addresses the question of whether, in fact, violent lyrics are perceived as more threatening, dangerous, and literal when they are described as rap, compared with another music genre. Seeking to replicate and extend previous research on this issue (Fried, 1996, 1999), we conducted a series of experiments in which participants were presented with a set of violent music lyrics. Participants were randomly assigned to learn that the lyrics were from a rap song or country song. Regardless of assigned condition, all participants read the same set of violent lyrics. Afterward they were asked to make an evaluation of the lyrics, answering a series of questions related to the offensiveness of the lyrics, the threatening nature of the lyrics, the need for the lyrics to be regulated, and the autobiographical or literal nature of the lyrics. Our aim was to determine whether lyrics categorized as rap are perceived as more threatening, literal, and in need of regulation compared with identical lyrics categorized as country, replicating and extending a previous study (Fried, 1999). Additional experiments are reported that build on this direct-replication effort by testing whether the effect holds for a different set of lyrics (Experiment 2) and when the race of the artist is specified (Experiment 3). In the remainder of the article we first review the substantial literature on stereotyping, paying close attention to studies that explicitly examine how rap music is stereotyped. Next we describe the details of our experiments and report our findings. Finally, we discuss the implications of the findings for defendants who face criminal charges where rap lyrics are introduced, focusing specifically on the effectiveness of rap lyrics as evidence in the courtroom. We also discuss how the findings may contribute to theories about stereotyping. Stereotypes and Rap Music Relevant to the study of rap lyrics as evidence in criminal trials are stereotypes both about the genre as well as those who create the music—typically young men of color from the inner city (Dennis, 2007; Kubrin & Nielson, 2014). While lyrics from other genres are interpreted as satirical or artistic, rap music is considered offensive and threatening (Crenshaw, 1991). Police, judges, and jurors run the risk of using stereotypes about rap music when interpreting lyrics, particularly when prosecutors present rap lyrics as autobiographical confessions (Perry, 2004). In a 2013 report, the American Civil Liberties Union (ACLU) of New Jersey argued that songs from other genres that reference illicit acts, like Johnny Cash’s Cocaine Blues, are not interpreted with the same literality and offensiveness as rap music because other genres are stereotyped differently. Stereotypes about rap music, therefore, are central to an understanding of why the introduction of lyrics in criminal cases may be so effective. In 1954, Gordon Allport (Allport, 1954) defined stereotypes as categories of seemingly associated traits that are used to process information more quickly. This definition has been revised to include knowledge and expectations about a social group (Hamilton & Sherman, 1996). Bodenhausen and Wyer (1985) describe stereotypes as “judgmental heuristics” (p. 279) that help decision makers process information. In other words, decision makers rely on available information, even if incorrect, about groups or types of events to inform a specific judgment (Tversky & Kahneman, 1974). Individuals may be consciously aware of stereotypes, yet stereotyping can also occur at an unconscious level (Devine, 1989). According to research, stereotypes are most likely to be utilized when the decision maker has minimal information (Kunda & Spencer, 2003), needs to simplify a complex decision (Bodenhausen & Lichtenstein, 1987), and has to make an evaluative judgment (Kunda & Spencer, 2003). When interpreting ambiguous behaviors or events, stereotypes are often used to make inferences about a person or a situation (Darley & Gross, 1983; Devine, 1989; Hamilton, 1979; Kunda, 1999; Kunda & Thagard, 1996). Research finds that stereotypes play a role in evaluating threatening situations and individuals, especially under conditions of ambiguity. Duncan (1976) tested the effect of racial stereotypes on the interpretation of an ambiguous shove and found that when a shove was committed by a White confederate, it was viewed as more playful compared with when the shove was committed by a Black confederate, where it was viewed as more aggressive or violent. Duncan inferred that the stereotype of Black people as aggressive was used to interpret the ambiguous shove. In more recent research, Correll, Park, Judd, and Wittenbrink (2002) had 2 This is not to say that other genres have not had a contentious relationship with the legal system. Indeed, some heavy metal music has been censored because of the concern that it would incite listeners to become violent (Blecha, 2004). For example, artists like Ozzy Osbourne and Judas Priest were accused of inciting listeners to commit suicide because of their music (Martin, Clarke, & Pearce, 1993; Weinstein, 2000). In other cases, such as the West Memphis Three, prosecutors have introduced defendants’ preference for heavy metal music as evidence of a propensity toward violence (Leveritt, 2002). However, these practices differ from the treatment of rap music in the courts, which involves the introduction of defendant-authored lyrics as evidence of criminal involvement. AQ: 3 AQ: 4 APA NLM tapraid5/zdn-law/zdn-law/zdn00316/zdn2260d16z xppws S⫽1 7/12/16 7:01 Art: 2016-1370 This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. RAP MUSIC participants first determine if a target person in a video game simulation is armed or not armed, and then respond accordingly. If the target person was armed, the participant was supposed to “shoot” the person; if the target person was unarmed, the participant was not supposed to shoot. The researchers manipulated the race of the target person to test how race affected the speed and accuracy in identifying an armed person. Correll and colleagues (2002) discovered that participants were faster to accurately shoot an armed target when he was Black and faster to not shoot an unarmed target when he was White. They explained their findings by suggesting that stereotypes about Black people being dangerous were used to evaluate the target person in the simulation. Related research finds that individuals identify pictures of a weapon more quickly when they are primed with a Black face compared with a White face (Eberhardt, Goff, Purdie, & Davies, 2004; Payne, 2001). Stereotypes have also been shown to play a role in how individuals recall facts in criminal cases. Research on stereotyping and memory demonstrates that the activation of racial stereotypes results in individuals recalling more incriminating details about a criminal case (Banaji & Bhaskar, 2000). Bodenhausen and Wyer (1985) found that when a crime is stereotype congruent—that is, when a crime is commonly associated with a particular racial group—participants are more likely to remember incriminating facts from the case than when the crime is stereotype incongruent. Stereotyping is even found to impact evaluations of a defendant’s culpability. Graham and Lowery (2004) tested the effect of racial stereotypes on a series of decisions about juvenile defendants. They primed participants with either neutral (e.g., heaven, loneliness) or race-related (e.g., homeboy, basketball) words and tested the effect of the prime on participants’ perceptions of a youth’s culpability, risk of recidivism, and deserved punishment. Graham and Lowery (2004) found that the activation of racial stereotypes resulted in increased perceived culpability and a harsher punishment for the defendant. Racial stereotypes have real life consequences for defendants when it comes to sentencing decisions (Bridges & Steen, 1998; Eberhardt, Davies, Purdie-Vaughns, & Johnson, 2006; Steen, Engen, & Gainey, 2005). In one study on race and sentencing, for example, researchers found that defendants who were perceived as more stereotypically Black—in other words, who had more Afrocentric facial features—were more likely to be sentenced to death compared with defendants who were perceived as less stereotypically Black (Eberhardt et al., 2006). In another study that analyzed narrative reports about juvenile offenders written by probation officers, researchers discovered pronounced differences in probation officers’ attributions about the causes of delinquency by White versus minority youth, differences that translated into longer sentences for Black youth (Bridges & Steen, 1998). Fewer studies have explored stereotypes about music, particularly music genres that are considered violent or harmful to society. However, findings from this small body of literature are revealing. Research documents that a song is evaluated as more graphic when it is categorized as “banned” (Neguţ & Sârbescu, 2014) and more suicide-affirming when it is framed as “potentially harmful to the listener” (North & Hargreaves, 2005). Related research shows that stereotypes about violent music extend to the individuals who write the lyrics. As just one example, Fischoff (1999) conducted an experiment to determine 3 the impact gangsta rap lyrics might have on potential jurors. Participants were presented with basic biographical information about a hypothetical 18-year old Black man but only some were shown a set of violent, sexually explicit rap lyrics that he had written (the lyrics had been used as evidence in a 1995 murder trial). Participants were then asked about their perceptions regarding the young man’s personality (e.g., is the young man caring or uncaring; selfish or unselfish; likable or unlikeable; capable of murder or not capable of murder). Fischoff (1999) found the lyrics exerted a significant impact, generating more negative evaluations of the young man’s character on all dimensions. Most revealing, however, was the finding that those who read the lyrics were significantly more likely to think the man was capable of committing murder. Research also finds that stereotypes are genre specific. For example, country and pop are frequently stereotyped as less threatening than both rock and rap (Ballard, Dodson, & Bazzini, 1999; Rentfrow & Gosling, 2007). It turns out, however, that stereotypes associated with rock music are different from those associated with rap (Binder, 1993; Fried, 2003). Binder (1993) found that rap music is perceived as more likely to cause listeners to hurt others whereas rock music is perceived as more likely to cause listeners to hurt themselves. She posited that the difference in reactions to the genres is because rap is associated with Black audiences while rock genres, like heavy metal, are associated with White audiences. In related research, Fried (2003) compared stereotypes about rap music fans and heavy metal music fans. Participants were asked to describe either the prototypical rap or heavy metal fan. Fried (2003) found that fans of heavy metal are viewed as more self-destructive while fans of rap are seen as more threatening to society. Research also reveals that stereotypes related to different music genres extend to perceived differences among genre fans with respect to alcohol and drug preferences, personality types, and personally held values (Rentfrow & Gosling, 2007). Only a handful of studies have examined the direct impact of rap music stereotypes. In these studies, experimenters ask respondents to evaluate a set of lyrics, manipulating the genre label in an attempt to isolate the effects of the genre. Dixon and Linz (1997), for example, presented respondents with sexually explicit rap lyrics or sexually explicit non-rap lyrics, both of which were viewed as equally explicit in a pretest. They found that the sexually explicit music was considered more offensive and less artistic when it was labeled as rap compared with when it was labeled as non-rap, revealing that similar lyrics are evaluated differently depending on the genre. To more precisely isolate the genre effect, Fried (1999) conducted a study where participants read identical violent lyrics but were told they were from different music genres. In particular, she had participants read a set of lyrics from folk group Kingston Trio’s 1960 song, Bad Man’s Blunder, and told them that they were either from a rap or country music song. After reading the lyrics, participants evaluated them by responding to items, which conceptually measured the offensiveness of the song, the threatening nature of the song, the need for regulation of the song, and if the song would incite violence. Fried (1999) found that every item was evaluated more negatively when the lyrics were categorized as rap compared with country. APA NLM tapraid5/zdn-law/zdn-law/zdn00316/zdn2260d16z xppws S⫽1 7/12/16 7:01 Art: 2016-1370 DUNBAR, KUBRIN, AND SCURICH 4 This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. Current Study The current study builds on this small but important literature related to stereotypes and rap music. In particular, we aim to test whether the evaluation of violent lyrics is influenced by genrespecific stereotypes. Consistent with Fried (1999), we test whether lyrics labeled as rap are perceived as more threatening, dangerous, and in need of regulation compared with identical lyrics labeled as country. As such, we seek to directly replicate her early study. Given the current “replication crisis” in psychology (see Pashler & Wagenmakers, 2012; Open Science Collaboration, 2015), we believe it is imperative to examine whether her earlier findings are replicable. However, we also extend the work done by Fried (1999) in important ways, in particular by addressing several concerns. One concern is that Fried’s experiment was conducted during a time period of heightened scrutiny of rap (i.e., the 1990s), which could make stereotypes about the genre more salient. Indeed, Fried identifies this heightened scrutiny and references severe reactions to controversial songs such as Cop Killer as the impetus for her research. While rap music is still scrutinized today, the genre does not receive the same media and congressional attention as it did in the 1990s. The implications of this shifting context for the findings remain unknown. Also unknown is whether the historical context related to race relations may impact study findings. Given that Fried’s experiment was conducted nearly 20 years ago, we wanted to determine if the findings could be replicated in what many consider to be a postracialized context. A final concern is the representativeness of Fried’s (1999) sample. The location of the experiment, described only as a “mid-size southwestern city” (p. 709), reveals little about the area’s specific demographic composition. In addition, other than mentioning the recruitment of subjects at “public areas such as malls, coffee houses, and so forth” (p. 710), Fried (1999) provides little information about where participants were recruited and the demographics of the resulting sample. It is possible that her results may be unique to the community where she conducted her study. Two additional experiments are presented that seek to extend the original work by Fried. Study 2 is a conceptual replication (see Simons, 2014), in which the same study design is utilized but with different violent lyrics. Accordingly, this study tests whether the original effect detected by Fried (1999) is dependent on any idiosyncrasies of the specific lyrics used in the study. Replicating the effect with a different set of lyrics would enhance the generalizability of the findings. A third study is presented that experimentally manipulates the genre label of the lyrics as well as the race of the author of the lyrics. The previous studies did not specify the race of the lyrics’ author and, as such, participants may have made different assumptions about the author’s race. These assumptions could have potentially influenced the results. Thus, Study 3 principally seeks to unconfound the rap label effect. human intelligence tasks (HITs). HITs commonly involve surveys and questionnaires for academic or market research. Individuals can peruse the list of HITs and elect to participate in the task if they are eligible to participate and if they find the terms and conditions satisfactory. The participant pool, while skewing more liberal and educated, yields samples that are more representative of the U.S. population than are college samples or other online samples (Paolacci et al., 2010). Our HIT required workers to be U.S. citizens over the age of 18 who could read English. Only workers with an IP address from within the United States were able to participate. Workers were paid $0.70 for their participation, a highly competitive rate on MTurk (see Paolacci et al., 2010). There were 126 U.S. residents who participated in this experiment. The age of participants ranged from 18 – 66 with a mean age of 34.5 (SD ⫽ 9.4), median age of 33.5, and interquartile range (IQR) of 6. Table 1 displays the demographic composition of the sample, as well as their listening habits (i.e., number of hours spent listening to music per week) and music genre preferences. Procedure and Design After opting to participate in the study, participants were instructed that they would read some music lyrics and be asked to Table 1 Experiment 1 Participant (n ⫽ 126) Demographics Variable Gender Race Ethnicity Education level Number of children Preferred music Music listening time Study 1 Political ideology Participants Participants were recruited through Amazon’s Mechanical Turk (MTurk) website (see Paolacci, Chandler, & Ipeirotis, 2010 for review). MTurk is an online platform where individuals can post Political affiliation Description N Percentage of participants Female Male Black White Asian/Pacific Islander Other Non-Hispanic Hispanic High school Vocational school College courses Completed university Graduate school 0 1 2 3 4 Classical Heavy metal Country Rap Jazz Electronic Rock ⬍1 hr 1 hr 2–5 hr 6–10 hr 10⫹ hr Liberal Moderate Conservative Republican Democrat Other 53 73 8 109 5 4 116 10 14 4 60 44 3 86 15 14 9 2 4 5 11 13 3 16 74 7 14 41 31 33 71 22 33 25 56 45 42.1 57.9 6.3 86.5 4 3.2 92.1 7.9 11.1 3.2 47.6 34.9 3.2 68.3 11.9 11.1 7.1 1.6 3.2 4.0 8.7 10.3 2.4 12.7 58.7 5.6 11.1 32.5 24.6 26.2 56.4 17.5 26.2 19.8 44.4 35.8 T1, AQ:5 APA NLM tapraid5/zdn-law/zdn-law/zdn00316/zdn2260d16z xppws S⫽1 7/12/16 7:01 Art: 2016-1370 RAP MUSIC evaluate them along several dimensions using only the limited information provided. Participants were told that there is no right or wrong answer and that they should respond with their honest impression of the lyrics. Participants were then randomly assigned to one of two conditions, which experimentally manipulated the genre of the lyrics. The lyrics were either characterized as from a rap song or a country song. Participants then read an excerpt from the folk song Bad Man’s Blunder by group Kingston Trio: Well, early one evening I was rollin’ around I was feelin’ kind of mean, I shot a deputy down. Strollin’ on home, and I went to bed. This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. Well, I laid my pistol up under my head. Well, early in the morning ’bout the break of day, I figured it was time to make a getaway. Steppin’ right along but I was steppin’ too slow. Got surrounded by a sheriff down in Mexico. This is the same passage used in Fried’s (1999) original study. Because our primary goal in Study 1 is a direct replication, the actual lyrics remained the same. Measures After reading the lyrics, participants evaluated them by responding to 14 different items, presented in random order. For each item, the participant read a statement and rated their agreement with that statement on a 9-point Likert scale that ranges from 1 ⫽ strongly disagree to 9 ⫽ strongly agree, with 5 indicating a neutral posi- 5 tion. Eleven items conceptually measured offensiveness of the song, the threatening nature of the song, and the need for the song to be regulated. In light of the earlier discussion, the current study included additional items that were thought to measure the literality of the lyrics, or how true the lyrics were perceived to be by respondents. Participants thus indicated if they believed the lyrics were based on a real life experience, if the lyrics were written to brag about the experience, and if the lyrics were made-up. For these latter items, we wanted to test if the lyrics are more likely to be interpreted as literal speech when presented as rap compared with country (a complete list of the items is contained in Table 2). As previously indicated, the 14 items were intended to conceptually measure interpretations of and reactions to the lyrics. Based on Fried’s (1996, 1999) description of the measures, the items were aggregated into an “offensiveness” scale and a “regulation” scale. Similarly, the items related to the perceived autobiographical nature of the lyrics were aggregated into a “literality” scale. In addition to utilizing Fried’s (1996, 1999) theoretical framework to derive the scales, responses to the 14 items were entered into a principal components analysis with a varimax rotation. It yielded a three factor solution with Eigenvalues of 8.71, 1.77, and 1.35, respectively (all other values less than 1), and the model explained 75% of the cumulative variance. This indicates that the items do indeed tap three distinct latent constructs, which parallel the items in the offensiveness, regulation, and literality scales, respectively. Table 2 displays which items loaded on which factor along with the inter-item correlations within each factor. We also created a unidimensional composite score by pooling the responses to all 14 Table 2 Scale Items Used to Evaluate the offensiveness, Regulation, and Literally of the lyrics Factor 1: Offensiveness ␣ ⫽ .914 Q1. I find the lyrics offensive. Q2: I object to the lyrics. Q3: The song is dangerous or harmful to society. Q4: The lyrics are threatening. Q5: The lyrics promote violence, riots, and civil unrest. Factor 2: Regulation ␣ ⫽ .933 Q6: Something should be done to warn consumers about (or otherwise regulate) this song. Q7: There should be mandatory warning labels for this song. Q8: They should ban such songs entirely. Q9: Regulations should be placed on these types of songs. Q10: These types of songs should not be played on the radio. Q11: I would be opposed to my younger sibling or young child listening to this song. Factor 3: Literality ␣ ⫽ .814 Q12: The lyrics are not based on a made-up story. Q13: The lyrics are based on the song writer’s real-life experience. Q14: The lyrics were written to brag about the song writer’s experience. Note. Country (mean, SD) Rap (mean, SD) 4.50 (2.47) 3.91 (2.49) 3.34 (2.21) 3.18 (2.45) 4.78 (2.44) 5.16 (2.54) 4.71 (2.71) 3.79 (2.35) 4.21 (2.51) 5.21 (2.74) Inter-item correlations Q1 1 X X X X Q2 .81 1 X X X Q3 .76 .76 1 X X Q4 .76 .75 .84 1 X Q5 .67 .65 .65 .70 1 Q6 Q7 Q8 Q9 Q10 Q11 3.50 (2.38) 4.37 (2.40) 1 .85 .55 .72 .69 .61 3.95 (2.54) 2.13 (1.76) 4.61 (2.43) 2.52 (2.09) X X 1 X .52 1 .75 .69 .67 .62 .57 .45 3.13 (2.37) 3.56 (2.23) X X X 1 .73 .49 3.59 (2.33) 4.44 (2.43) X X X X 1 .67 4.59 (2.78) 5.97 (2.48) X X X X X 1 2.52 (1.40) 2.45 (1.18) Q12 1.00 Q13 .56 Q14 .46 2.48 (1.51) 3.15 (1.63) X 1.00 .77 2.80 (1.78) 3.66 (1.92) X X 1.00 Note that groupings were validated by a principal components analysis using Varimax rotation. T2 APA NLM tapraid5/zdn-law/zdn-law/zdn00316/zdn2260d16z xppws S⫽1 7/12/16 7:01 Art: 2016-1370 DUNBAR, KUBRIN, AND SCURICH 6 items to be consistent with Fried (1999). We classify this variable as “total negative reaction score.” In addition to these scales, participants responded to a manipulation check. In particular, participants were asked if they knew the song. Participants who reported knowing the song (n ⫽ 1) were excluded from analysis. Finally, participants were asked a number of demographic questions, and then thanked for their participation. Before analyzing the effect of genre type on the offensiveness, regulation, and literality scales, Cronbach’s ␣s were conducted to determine each scale’s reliability; findings reveal a high degree of reliability for the offensiveness scale (␣ ⫽ .914), the regulation scale (␣ ⫽ .933), the literality scale (␣ ⫽ .814), and total negative reaction scale (␣ ⫽ .940). Independent sample t tests were then conducted with genre label (rap or country) as the independent variable and the total negative reaction scale as the dependent variable. Consistent with Fried, participants in the rap condition indicated a higher score on the unidimensional composite item, total negative reaction, compared with participants in the country condition, t126 ⫽ 2.06, p ⫽ .043, d ⫽ .37, confidence interval (CI) [⫺1.23, ⫺.02]. In other words, those who were told the lyrics were from a rap song perceived them to be more negative overall compared with those who were told the lyrics were from a country song. More specifically, there was also a statistically significant effect of genre label on the regulation scale, t126 ⫽ 2.18, p ⫽ .031, d ⫽ .39, CI [⫺8.73, ⫺.43] and literality scale, t126 ⫽ 2.0, p ⫽ .047, d ⫽ .36, CI [⫺2.90, ⫺.21], with participants in the rap condition evaluating the lyrics as in need of greater regulation and more literal than those in the country condition. The ratings of offensiveness were not significantly different between the two groups, t126 ⫽ 1.4, p ⫽ .16. Although no significant difference was found for offensiveness, the results are consistent with the other findings in this study: on average, participants in the rap condition rated the lyrics as more offensive than participants in the country condition. Although Fried (1999) split her sample into three categories based on participant’s age (i.e., “under 40,” “40 –52,” and “53 and over”), a median split was used in the current study to divide the Study 2 A second study was conducted to ensure that the detected effects were not simply because of idiosyncrasies of the lyrics. Thus, in a follow up experiment, we used the exact same study design but incorporated a different set of violent lyrics, which again were described as either rap or country. The lyrics in Study 2 were from the song, A Boy Named Sue, by Johnny Cash: Well, I hit him hard right between the eyes And he went down, but to my surprise, He come up with a knife and cut off a piece of my ear. But I busted a chair right across his teeth And we crashed through the wall and into the street Kicking and a’ gouging in the mud and the blood and the beer. A pretest was conducted using a separate sample of 35 Mturk workers to determine whether the lyrics would be primarily identified as from a rap or country song. This was done to ensure the content or style of the lyrics was not biased toward a particular genre of music. To determine whether the lyrics were perceived as belonging to either genre, participants in the pretest, who were not included in the following experiment, were asked to select the 5.5 Negative Reaction Score This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. Results sample into two groups based on age: “younger” (i.e., age 18 –33.5) and “older” (i.e., age 33.6 – 66) participants. A two-way analysis of variance (ANOVA) detected a significant interaction between age and genre label for the total negative reaction score (i.e., all 14 items combined) F(3, 126) ⫽ 4.57, p ⫽ .026, d ⫽ 1.34. As seen in Figure 1, participants in the younger category were insensitive to the genre label, while older participants evaluated rap significantly more negatively than those in the country condition. Several additional tests were conducted to examine whether other demographic variables moderate the evaluation of lyrics when characterized as rap versus country. These include participants’ gender, ethnicity, education level, number of children, music genre preference, music listening habits, political ideology, and political affiliation. No significant interactions were detected (all ps ⬎ .05). 5 Genre Label 4.5 Country 4 Rap 3.5 3 2.5 2 Younger Older Parcipant's Figure 1. Interactive effect of participant’s age (median split) and genre label ascribed to the lyrics on negative reactions to the lyrics. Note that error bars reflect 95% confidence intervals. F1 APA NLM tapraid5/zdn-law/zdn-law/zdn00316/zdn2260d16z xppws S⫽1 7/12/16 7:01 Art: 2016-1370 RAP MUSIC genre of the lyrics. Results confirm that the genre of the lyrics was ambiguous. Twenty percent of participants identified the lyrics as a rock song (n ⫽ 7), 20% identified the lyrics as a rap song (n ⫽ 7), 34% identified the lyrics as a country song (n ⫽ 12), and the remaining 26% of the sample identified the lyrics as some other genre (n ⫽ 9). This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. Participants T3 There were 244 U.S. residents who were recruited through Amazon’s Mechanical Turk to participate in Experiment 2. The age of participants ranged from 18 –73 with a mean age of 33.2 (SD ⫽ 9.89) and median age of 31. Thirteen out of the 244 participants were excluded for knowing the song. Table 3 displays the demographic composition of the sample, as well as their music listening habits (i.e., number of hours spent listening to music per week) and music genre preferences. Results For Study 2, Cronbach’s ␣s again reached acceptable reliability for the offensiveness scale (␣ ⫽ .891), the regulation scale, (␣ ⫽ Table 3 Experiment 2 Participant (n ⫽ 231) Demographics Variable Gender Race Ethnicity Education level Number of children Preferred music Listening time Political ideology Political affiliation Description N Percentage of participants Female Male Black White Native American Asian/Pacific Islander Other Non-Hispanic Hispanic High school Vocational school College courses Completed university Graduate School 0 1 2 3 4 5⫹ Classical Heavy Metal Country Rap Jazz Electronic Rock ⬍1 hr 1 hr 2–5 hr 6–10 hr 10⫹ hr Liberal Moderate Conservative Republican Democrat Other 100 131 13 197 2 14 5 220 11 32 3 89 70 37 154 33 28 11 2 3 7 13 20 32 10 32 104 10 16 88 45 72 99 50 55 42 101 88 43.3 56.7 5.6 85.3 .9 6.1 2.2 95.2 4.8 13 1.3 38.5 30.3 16 66.7 14.3 12.1 4.8 .9 1.3 3.0 5.6 8.7 13.9 4.3 13.9 45.0 4.3 6.9 38.1 19.5 31.2 54.5 21.6 23.8 18.2 43.7 38.1 7 .894), the literality scale (␣ ⫽ .766), and the total negative reaction scale (␣ ⫽ .919). Independent sample t tests were then conducted with genre label (rap or country) as the independent variable and the total negative reaction score as the dependent variable. Consistent with Fried (1999) and Study 1, participants in the rap condition once again indicated a higher score on the unidimensional composite item (M ⫽ 4.69, SD ⫽ 1.49), total negative reaction, than did participants in the country condition (M ⫽ 4.20, SD ⫽ 1.55), t231 ⫽ .2.45, p ⫽ .015, d ⫽ .32, CI [⫺.88, ⫺.01]. In other words, those who were told the lyrics were from a rap song perceived them to be more negative overall compared to those who were told the lyrics were from a country song. There was also a statistically significant effect of genre label on the regulation scale (M ⫽ 4.66 [SD ⫽ 1.78] vs. M ⫽ 4.04 [SD ⫽ 1.92] for rap and country, respectively), t231 ⫽ 2.54, p ⫽ .012, d ⫽ .33, CI [⫺1.10, ⫺.14] and the literality scale (M ⫽ 4.81 (SD ⫽ 1.50) vs. M ⫽ 4.31 (SD ⫽ 1.49) for rap and country, respectively), t231 ⫽ 2.7, p ⫽ .008, d ⫽ .35, CI [⫺.87, ⫺13], with participants in the rap condition evaluating the lyrics as more literal and in need of greater regulation than those in the country condition. The ratings of offensiveness were not significantly different between the two groups, t231 ⫽ 1.3, p ⫽ .20, also consistent with Study 1. Although the difference for offensiveness was not statistically significant, once again the pattern of results is consistent with Study 1 in that participants in the rap condition, on average, rated the lyrics as more offensive (M ⫽ 4.67, SD ⫽ 1.92) compared with participants in the country condition (M ⫽ 4.34, SD ⫽ 1.93). Consistent with Study 1, a median split was used to divide the sample into two groups based on age: younger (i.e., age 18 –31) and older (i.e., age 31.1–73) participants. A two-way ANOVA detected a significant main effect of genre label for the total negative reaction score (i.e., all 14 items combined), F(3, 231) ⫽ 6.34, p ⫽ .01, d ⫽ .33, consistent with the t test reported above. The main effect for age was not significant, F(3, 231) ⬍ 1, nor was the interaction F(3, 231) ⫽ 2.18, p ⫽ .141. Additional ANOVAs were conducted for participants’ gender, ethnicity, education level, number of children, music genre preference, music listening habits, political ideology, and political affiliation. No significant interactions were detected (all ps ⬎ .05). Study 3 The previous studies found the genre label effect detected by Fried (1996, 1999) is able to be replicated nearly 20 years later (Study 1) and with a different set of lyrics (Study 2). Of course, these studies have some limitations. One is that the race of the artist is not specified and thus participants might have made assumptions about the race of the songwriter when evaluating the lyrics. For instance, participants might have assumed that the author of the rap lyrics is Black and the author of the country lyrics is White; to the extent that this occurred, it is unclear whether the genre label or the assumed race of the author is driving the effect observed in the previous studies. A second issue concerns whether the previous results are indicative of positive stereotypes about country, negative stereotypes about rap, or both. In other words, a more informative comparison is a control condition in which no genre label is provided. Including this condition would allow us to directly test whether APA NLM tapraid5/zdn-law/zdn-law/zdn00316/zdn2260d16z xppws S⫽1 7/12/16 7:01 Art: 2016-1370 This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. 8 T4 F2 DUNBAR, KUBRIN, AND SCURICH the country label engenders positive or neutral evaluations and rap negative evaluations. These issues can be remedied by experimentally manipulating the race of the songwriter and by including a control condition where no genre is specified. Experiment 3 did just that. Participants were randomly assigned to one of six conditions, which experimentally manipulated the genre of the lyrics (no genre, country, or rap) and the race of the songwriter (Black or White). Thus, a 3 (genre label) ⫻ 2 (race of author) between-participants factorial design was utilized. After learning about the genre of the lyrics and the race of the songwriter, participants read lyrics from the folk song Bad Man’s Blunder (used in Study 1). After reading the lyrics, participants responded to the same 14 items evaluating the lyrics, a manipulation check, and a series of demographic questions that were used in Studies 1 and 2. Table 4 Experiment 3 Participant (n ⫽ 314) Demographics Participants Number of children Similar to Studies 1 and 2, participants were recruited through Amazon’s Mechanical Turk. There were 325 U.S. residents who participated in this experiment. The age of participants ranged from 18 –71 with a mean age of 33.7 (SD ⫽ 10.32) and median age of 31. Eleven out of the 325 participants were excluded for knowing the song. Table 4 displays the demographic composition of the sample as well as their music listening habits (i.e., number of hours spent listening to music per week) and music genre preferences. Variable Gender Race Ethnicity Education level Preferred music Listening time Results For Experiment 3, Cronbach’s ␣ once again reached acceptable reliability for the offensiveness scale (␣ ⫽ .939), the regulation scale, (␣ ⫽ .939), the literality scale (␣ ⫽ .784), and the total negative reaction scale (␣ ⫽ .951). A two-way ANOVA with genre and artist race as the independent variables and total negative reaction score as the dependent variable detected a main effect for genre F(2, 314) ⫽ 3.66, p ⫽ .03, d ⫽ .31. The main effect for artist race was not significant, F(1, 314) ⬍ 1, nor was the interaction F(5, 314) ⫽ 1.04, p ⫽ .354. Figure 2 displays this result. As seen below, Figure 2 shows that participants in the rap condition evaluated the lyrics more negatively (M ⫽ 4.72, SD ⫽ 1.81) than participants in the country (M ⫽ 4.10, SD ⫽ 2.15) and control conditions (M ⫽ 4.06, SD ⫽ 1.94), regardless of the songwriter’s race. A Fisher’s least significant difference (LSD) test confirms that the lyrics from the rap condition were evaluated significantly more negatively than the identical lyrics from the control (p ⫽ .02) and country conditions (p ⫽ .024), but that the country and control conditions were not significantly different from each other (p ⫽ .86). More important, the fact that no effect for the songwriter’s race was detected nor was there an interaction suggests that the previous studies, which did not specify the songwriter’s race, are robust and not an artifact of assumptions related to the songwriter’s race. Additionally, we ran a series of two-way ANOVAs on the subscales of the reaction scores (i.e., ratings of literality, offensiveness, and regulation). A two-way ANOVA with literality as the dependent variable (and race of artist and genre as the independent variables) failed to detect any main effects or an interaction (all ps ⬎ .05). A two-way ANOVA with offensiveness as the Political ideology Political affiliation Description N Percentage of participants Female Male Black White Native American Asian/Pacific Islander Other Non-Hispanic Hispanic High school Vocational school College courses Completed university Graduate school 0 1 2 3 4 5⫹ Classical Heavy metal Country Rap Jazz Electronic Rock ⬍1 hr 1 hr 2–5 hr 6–10 hr 10⫹ hr Liberal Moderate Conservative Republican Democrat Other 136 178 23 250 5 31 5 293 21 35 8 126 108 37 212 46 31 17 5 3 16 10 25 40 17 41 155 21 31 114 82 66 172 66 51 53 137 124 43.3 56.7 7.3 79.6 1.6 9.9 1.6 93.3 6.7 11.1 2.5 40.1 34.4 11.8 67.5 14.6 9.9 5.4 1.6 1.0 5.1 3.2 8.0 12.7 5.4 13.1 49.4 6.7 9.9 36.3 26.1 21.0 54.8 21.0 24.2 16.9 43.6 39.5 dependent variable detected only a main effect for genre F(2, 314) ⫽ 4.92, p ⫽ .008, d ⫽ .36, with rap being deemed more offensive (M ⫽ 5.51, SD ⫽ 2.25) than country (M ⫽ 4.62, SD ⫽ 2.54) or the control (M ⫽ 4.59, SD ⫽ 2.39). A Fisher’s LSD test indicated that responses from the rap condition were significantly different than responses from the control condition (p ⫽ .006) and the country condition (p ⫽ .008) but that the control condition and country condition were not significantly different (p ⫽ .89). Similarly, the two-way ANOVA with regulation as the dependent variable detected only a main effect for genre F(2, 314) ⫽ 3.43, p ⫽ .03, d ⫽ .30, with rap being deemed in greater need of regulation (M ⫽ 4.87, SD ⫽ 2.25) compared to country (M ⫽ 4.14, SD ⫽ 2.51) or the control (M ⫽ 4.13, SD ⫽ 2.56). A Fisher’s LSD test indicated that the rap condition was significantly different than the control condition (p ⫽ .02) and the country condition (p ⫽ .03) but that the country and control conditions were not significantly different from each other (p ⫽ .96). General Discussion There is a well-established literature showing that people use racial stereotypes to evaluate ambiguous information (e.g., Dun- APA NLM tapraid5/zdn-law/zdn-law/zdn00316/zdn2260d16z xppws S⫽1 7/12/16 7:01 Art: 2016-1370 RAP MUSIC 9 This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. Negave Reacon Score 5.5 Arst Race 5 White Black 4.5 4 3.5 3 No Label Country Rap Genre Label Figure 2. Interactive effect of genre label ascribed to the lyrics and race of songwriter on negative reactions to the lyrics. Note that error bars reflect 95% confidence intervals. can, 1976) and that the use of racial stereotypes translates into real world behavior with serious implications (e.g., Eberhardt et al., 2006). Far fewer studies have examined the role that stereotypes might play in the perception of rap music, a ubiquitous cultural expression today (Perry, 2004). A notable exception is a study conducted by Fried (1999), which found that merely manipulating the ostensible genre of a set of lyrics—while holding constant the actual words—induced more negative evaluations when the lyrics were described as rap compared with country. It is important to note that Fried’s study was conducted in the late 1990s, a period in which rap music received heightened scrutiny in the media. Moreover, there are methodological questions about the study’s sampling procedure and representativeness. However, the present research directly replicates Fried’s (1999) findings: participants deemed the exact same lyrics to be more offensive, in greater need of regulation, and more literal when characterized as rap compared with country. This replication suggests that the effect is robust, since our study contained several notable methodological differences, including the fact that we used a sample of participants from all over the United States as opposed to approaching individuals at a shopping mall in one location; we used factor analysis to empirically determine the offensive/regulation/literality dimensions as opposed to using a single Likert scale to measure each issue; we included a control condition to determine whether we were finding negative rap stereotypes rather than positive country stereotypes; and our data were collected in 2015, nearly two decades after the spate of media attention denouncing rap music. Additionally, the subsequent studies all replicated this effect. The studies all found a genre effect, irrespective of the actual lyrics (Study 2) or the race of the author (Study 3). The analyses of the subscales are also mostly consistent, though the genre effect on literality was not significant in Study 3 but it was in Study 1 and Study 2, and the genre effect on offensiveness was significant in Study 3 but not Study 1 or Study 2. Thus, we acknowledge some inconsistency in effects and effect sizes across the various experiments. It is not exactly clear why this occurred. Of course, Study 3 included information about the songwriter’s race, while Studies 1 and 2 did not. It is possible that the salience of race resulted in participants applying the stereotype to evaluations of the lyrics but not evaluations that more directly relate to the songwriter. That is, participants may have avoided making judgments about a person that could appear racially biased—a concern that may not have extended to judgments about the lyrics. Future research is necessary to test this possibility. We should note, however, that the main finding on the total negative reaction score was consistent across all three studies, suggesting a robust effect. It must be acknowledged that the samples used in these studies are not probability samples representative of the United States population. The findings should be qualified by this limitation. However, studies have shown that individuals recruited through Amazon Mechanical Turk behave in ways consistent with other commonly used subject pools and the general public (Bartneck, Duenser, Moltchanova, & Zawieska, 2015; Heen, Lieberman, & Miethe, 2014; Paolacci & Chandler, 2014; but see Chandler, Mueller, & Paolacci, 2014). Whether the findings would generalize to the U.S. population as a whole or to specific state populations is an unknown empirical question that must be addressed by future research. We have no reason to believe the results would systematically differ if the representativeness of the sample were increased. A more diverse sample would, however, enable one to conduct moderator analyses to determine whether the race of participants, and other individual characteristics, affects their evaluations of music genre label. Currently, it is unclear whether, for example, negative rap stereotypes are specific to the racial composition of our sample or whether the stereotype, like implicit racial stereotypes (Nosek, Banaji, & Greenwald, 2002), is generalizable across racial groups. Our findings have potential implications for the use of rap lyrics in criminal proceedings, where they are being increasingly admitted as evidence to prove guilt (Dennis, 2007; Hirsch, 2012; Kubrin & Nielson, 2014). A review of cases reveals that prosecutors use multiple strategies to introduce rap lyrics and secure convictions. In one strategy, prosecutors use lyrics as evidence to show that the defendant had intent, motive, or the necessary knowledge to commit a crime. For example, in Skinner v. State (2014), the prosecutor APA NLM tapraid5/zdn-law/zdn-law/zdn00316/zdn2260d16z xppws S⫽1 7/12/16 7:01 Art: 2016-1370 This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. 10 AQ: 6 DUNBAR, KUBRIN, AND SCURICH introduced 13 pages of rap lyrics written by aspiring rapper Vonte Skinner to argue that he was the perpetrator in an attempted murder case. During the investigation of a shooting, police found notebooks filled with violent rap lyrics in Skinner’s girlfriend’s car. Even though the lyrics did not reference the charged crime and it was not clear even when they were written, the prosecution argued that the lyrics demonstrated Skinner’s motive to shoot the victim. During the trial, Skinner’s attorney contested the admission of the lyrics, maintaining they were not relevant to the case yet the judge ultimately admitted them deciding they spoke to intent and motive. The jury convicted Skinner of attempted murder, aggravated assault, and assault with a deadly weapon. He received a 30-year prison sentence. Skinner appealed his conviction. The appellate court in Skinner’s case decided that, in fact, the lyrics should not have been admitted because their probative value was outweighed by their potential biasing effect. In particular, the appellate judges considered the violent, profane, and disturbing lyrics inflammatory and of little use to jurors. The court also pointed out that it is difficult to infer intent or motive from fictional work because fiction is not clearly tied to reality. While ruling the lyrics inadmissible in Skinner’s case specifically, the court left the door open for allowing rap lyrics in cases more generally, such as when there is a “strong nexus between specific details of the artistic composition and the circumstances of the offense” (Skinner v. State, p. 35). While the court signaled that lyrics may be admissible if they reference a specific crime, it did not describe how specific the references need to be or, more broadly, how fictional work should be interpreted. As another strategy, prosecutors argue that the rap lyrics constitute a true threat, which involves a statement that would be reasonably interpreted as imminent, causing fear, and threatening harm (see Rothman, 2001, for a detailed discussion of a true threat). For example, in People v. Oduwole (2013), police searched an abandoned car and found violent rap lyrics written by the defendant, Olutosin Oduwole, an aspiring rapper and college student who was being investigated for computer fraud and felony theft. The lyrics mentioned a “murderous rampage similar to the VT [Virginia Tech] shooting.” After obtaining a warrant, police searched Oduwole’s campus apartment and found guns along with notebooks filled with his violent and misogynistic rap lyrics. They arrested Oduwole and charged him with communicating a terrorist threat. The prosecutor presented the note in the car as indicative of a threat rather than as indicative of an early draft of rap lyrics, which is what Oduwole maintained. The jury apparently agreed. Oduwole was convicted and sentenced to 5 years in prison. Regardless of the type of case, the approach is generally the same. Prosecutors treat rap lyrics as literal, self-referential narratives that can be easily interpreted by the lay public (Dennis, 2007). Prosecutors reinforce the first-person narrative perspective by reading the lyrics at trial like a journal entry, without rhyme or music (Hirsch, 2014). As part of their argument, prosecutors claim that the lyrics are simply a reflection of the rapper’s lifestyle. In Dennis (2007), for example, the prosecutor argued that the lyrics should be admissible because they are “a reflection of the defendant’s soul” (as cited in Dennis, 2007, p. 7). The prosecutor also maintained that the lyrics are autobiographical because “the defendant is living his lyrics” (as cited in Dennis, 2007, p. 7). This strategy is evidenced in a prosecutorial handbook, Prosecuting Gang Cases: What Prosecutors Need to Know authored by former District Attorney Alan Jackson (Jackson, 2004). Jackson argues that prosecutors should aim to introduce the jury to the “real defendant” (p. 15) because at trial he will have “taken on the aura of an altar boy,” (p. 15) and that through “photographs, letters, notes, and even music lyrics, [the prosecutor] can invade and exploit the defendant’s true personality” (p. 16). The findings in the present study suggest that rap lyrics might influence jurors’ decisions independent of their actual content. That is, the mere label of rap is sufficient to induce negative evaluations, even when holding constant the actual lyrics. This has direct implications for judges who must consider and weigh potential prejudicial impact against probative value when deciding whether to admit rap lyrics as evidence. In particular, the findings suggest that judges might underappreciate the extent to which the label of lyrics—and not the substantive lyrics themselves—impact jurors’ decisions. In addition to other concerns expressed by scholars regarding judges’ problematic assumptions about rap music (see Dennis, 2007), the present findings suggest that judges should consider limiting the introduction of rap lyrics to instances in which the lyrics are highly probative of some relevant legal issue, and judges should realize that jurors might make inferences based merely on the genre of the lyrics and the stereotypes that they evoke. However, it must also be acknowledged that we did not test the effect of rap lyrics in an adjudicative context; rather, we examined perceptions of rap lyrics in a general context and in isolation. Important differences between this general context and an adjudicative context exist. For example, in a criminal trial such evidence would be presented as part of a narrative in conjunction with other evidence. It remains to be seen, therefore, how perceptions of rap lyrics might change as a function of the narrative or how the lyrics might interact with other evidence. Additionally, jurors are bound by legal rules when evaluating evidence and are instructed on the permissible inferences that may be drawn from such evidence. Again, it remains to be seen how such rules and instruction might affect the perception of rap lyrics. Building on the current study, further research should examine perceptions of rap lyrics in adjudicative and other contexts. It should also be acknowledged that these results cannot speak to the accuracy of stereotypes associated with rappers and rap music, or any other genre for that matter. One question is whether rappers are more likely to engage in crime and write violent lyrics that reflect that crime than are artists from other genres. Interestingly, research indirectly addressing this question yields mixed results. For example, Tapper, Thorson, and Black (1994) found that rap music videos, compared with videos from genres like heavy metal, country, and classic rock, do include more lyrical and visual depictions of violence. In contrast, Armstrong (1993) analyzed lyrics from country and rap songs and found that depictions of violence and masculinity were consistent themes found in both of the genres. Of course, these studies do not discern whether rappers are, in fact, engaging in more violence than are artists from other genres, and we know of no study that addresses this specific question. Still, it remains unknown as to whether rap lyrics have more diagnostic value as evidence than lyrics from other genres. Regardless, a key concern is that any value rap lyrics may have as evidence is likely to be artificially inflated by stereotypes associated with the genre. APA NLM tapraid5/zdn-law/zdn-law/zdn00316/zdn2260d16z xppws S⫽1 7/12/16 7:01 Art: 2016-1370 This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. RAP MUSIC AQ: 7 Kubrin and Nielson (2014) issued a call to scholars to “critically examine the growing movement to turn rap lyrics against their authors” (p. 19). Relatedly, Kang and colleagues (2012) have issued a broader call to further explore the impact of stereotyping in the courtroom. This series of experiments constitutes an effort to address these calls, and as is evident, the results speak to the continued need for additional research on stereotyping in police and juror evaluations of evidence. In particular, social science research can provide insight into additional ways that including rap lyrics as evidence might result in biased judgments. For example, according to research, rap lyrics also activate stereotypes related to race more broadly, that is, beyond rap music fans and listeners. Exposure to rap music has been shown to increase the ease of associating Black people with negative traits like hostility, being violent, and being sexist (Rudman & Lee, 2002) as well as making less empathetic judgments toward Black victims (Johnson, Bushman, & Dovidio, 2008). Johnson, Trawalter, and Dovidio (2000) found that participants who listened to violent rap music (compared with nonviolent rap and no music) were likely to evaluate the target male in an unrelated task as more inherently violent and less intelligent. Exploring the broader implications of this line of research can potentially provide greater insight into how stereotypes related to rap music can influence the perceived value of violent lyrics and other, related judgments. Given our current findings and previous scholarship (Dennis, 2007; Kubrin & Nielson, 2014; Wilson, 2005), we suggest that researchers also explore the effectiveness of potential policy recommendations. For example, presenting instructions or expert testimony that explain the genre conventions of rap music to jurors may reduce the chance of biased judgments (Dennis, 2007). Additionally, having jurors take the perspective of the songwriter rather than the listener may alter the interpretation of the lyrics. Although it is not likely that rap lyrics will be deemed inadmissible evidence by the courts (Dennis, 2007), it is important to develop solutions to mitigate any bias associated with the evidence and to determine the effectiveness of such policy recommendations. References Abrams, N. (2000). Gangsta rap. In T. Pendergast & S. Pendergast (Eds.), St. James encyclopedia of popular culture (p. 198). Farmington Hills, MI: Thomson-Gale. Allport, G. W. (1954). The nature of prejudice. Cambridge, MA: Perseus Books. Armstrong, E. G. (1993). The rhetoric of violence in rap and country music. Sociological Inquiry, 63, 64 –78. http://dx.doi.org/10.1111/j .1475-682X.1993.tb00202.x Ballard, M. E., Dodson, A. R., & Bazzini, D. G. (1999). Genre of music and lyrical content: Expectation effects. The Journal of Genetic Psychology, 160, 476 – 487. http://dx.doi.org/10.1080/00221329909595560 Banaji, M. R., & Bhaskar, R. (2000). Implicit stereotypes and memory: The bounded rationality of social beliefs. In D. L. Schacter & E. Scarry (eds.), Memory, brain, and belief (pp. 139 –175). Cambridge, MA: Harvard University Press. Bartneck, C., Duenser, A., Moltchanova, E., & Zawieska, K. (2015). Comparing the similarity of responses received from studies in Amazon’s Mechanical Turk to studies conducted online and with direct recruitment. PLoS ONE, 10, e0121595. http://dx.doi.org/10.1371/journal .pone.0121595 11 Binder, A. (1993). Constructing racial rhetoric: Media depiction of harm in heavy metal and rap music. American Sociological Review, 58, 753–767. http://dx.doi.org/10.2307/2095949 Blecha, P. (2004). Taboo tunes: A history of banned bands and censored songs. San Francisco, CA: Backbeat Books. Bodenhausen, G. V., & Lichtenstein, M. (1987). Social stereotypes and information-processing strategies: The impact of task complexity. Journal of Personality and Social Psychology, 52, 871– 880. http://dx.doi .org/10.1037/0022-3514.52.5.871 Bodenhausen, G. V., & Wyer, R. S., Jr. (1985). Effects of stereotypes on decision making and information-processing strategies. Journal of Personality and Social Psychology, 48, 267–282. http://dx.doi.org/10.1037/ 0022-3514.48.2.267 Bridges, G. S., & Steen, S. (1998). Racial disparities in official assessments of juvenile offenders: Attributional stereotypes as mediating mechanisms. American Sociological Review, 63, 554 –570. http://dx.doi.org/ 10.2307/2657267 Chandler, J., Mueller, P., & Paolacci, G. (2014). Nonnaïveté among Amazon Mechanical Turk workers: Consequences and solutions for behavioral researchers. Behavior Research Methods, 46, 112–130. http://dx .doi.org/10.3758/s13428-013-0365-7 Correll, J., Park, B., Judd, C. M., & Wittenbrink, B. (2002). The police officer’s dilemma: Using ethnicity to disambiguate potentially threatening individuals. Journal of Personality and Social Psychology, 83, 1314 –1329. http://dx.doi.org/10.1037/0022-3514.83.6.1314 Crenshaw, K. (1991). Mapping the margins: Intersectionality, identity politics, and violence against women of color. Stanford Law Review, 43, 1241–1299. http://dx.doi.org/10.2307/1229039 Darley, J. M., & Gross, P. H. (1983). A hypothesis-confirming bias in labeling effects. Journal of Personality and Social Psychology, 44, 20 –33. http://dx.doi.org/10.1037/0022-3514.44.1.20 Dennis, A. L. (2007). Poetic (In)justice? Rap music lyrics as art, life, and criminal evidence. The Columbia Journal of Law & the Arts, 31, 1– 41. Retrieved from http://ssrn.com/abstract⫽1104756 Devine, P. G. (1989). Stereotypes and prejudice: Their automatic and controlled components. Journal of Personality and Social Psychology, 56, 5–18. http://dx.doi.org/10.1037/0022-3514.56.1.5 Dixon, T. L., & Linz, D. G. (1997). Obscenity law and sexually explicit rap music: Understanding the effects of sex, attitudes, and beliefs. Journal of Applied Communication Research, 25, 217–241. http://dx.doi.org/10 .1080/00909889709365477 Duncan, B. L. (1976). Differential social perception and attribution of intergroup violence: Testing the lower limits of sterotyping of blacks. Journal of Personality and Social Psychology, 34, 590 –598. http://dx .doi.org/10.1037/0022-3514.34.4.590 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. http://dx.doi.org/10.1111/j.1467-9280.2006.01716.x Eberhardt, J. L., Goff, P. A., Purdie, V. J., & Davies, P. G. (2004). Seeing black: Race, crime, and visual processing. Journal of Personality and Social Psychology, 87, 876 – 893. http://dx.doi.org/10.1037/0022-3514 .87.6.876 Fischoff, S. P. (1999). Gangsta’ rap and a murder in Bakersfield. Journal of Applied Social Psychology, 29, 795– 805. http://dx.doi.org/10.1111/j .1559-1816.1999.tb02025.x Fried, C. B. (1996). Bad rap for rap: Bias in reactions to music lyrics. Journal of Applied Social Psychology, 26, 2135–2146. http://dx.doi.org/ 10.1111/j.1559-1816.1996.tb01791.x Fried, C. B. (1999). Who’s afraid of rap? Differential reactions to music lyrics. Journal of Applied Social Psychology, 29, 705–721. http://dx.doi .org/10.1111/j.1559-1816.1999.tb02020.x APA NLM tapraid5/zdn-law/zdn-law/zdn00316/zdn2260d16z xppws S⫽1 7/12/16 7:01 Art: 2016-1370 12 DUNBAR, KUBRIN, AND SCURICH This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. Fried, C. B. (2003). Stereotypes of music fans: Are rap and heavy metal fans a danger to themselves or others? Journal of Media Psychology, 8, 2–27. Gan, S., Zillmann, D., & Mitrook, M. (1997). Stereotyping effect of Black women’s sexual rap on White audiences. Basic and Applied Social Psychology, 19, 381–399. http://dx.doi.org/10.1207/s15324834 basp1903_7 Gilroy, P. (1995). The black Atlantic: Modernity and double consciousness. Cambridge, MA: Harvard University Press. Graham, S., & Lowery, B. S. (2004). Priming unconscious racial stereotypes about adolescent offenders. Law and Human Behavior, 28, 483– 504. http://dx.doi.org/10.1023/B:LAHU.0000046430.65485.1f Hamilton, D. L. (1979). A cognitive-attributional analysis of stereotyping. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 12, pp. 53– 84). New York, NY: Academic Press. Hamilton, D. L., & Sherman, S. J. (1996). Perceiving persons and groups. Psychological Review, 103, 336 –355. http://dx.doi.org/10.1037/0033295X.103.2.336 Hansen, C. H., & Hansen, R. D. (1990). Rock music videos and antisocial behavior. Basic and Applied Social Psychology, 11, 357–369. http://dx .doi.org/10.1207/s15324834basp1104_1 Heen, M. S., Lieberman, J. D., & Miethe, T. D. (2014). A comparison of different online sampling approaches for generating national samples. Retrieved from http://www.unlv.edu/sites/default/files/page_files/27/ ComparisonDifferentOnlineSampling.pdf Hirsch, L. E. (2012). Music in American crime prevention and punishment. Ann Arbor, MI: University of Michigan Press. http://dx.doi.org/10.3998/ mpub.4365408 Hirsch, L. E. (2014). Rap as threat? The violent translation of music in American law. Law, Culture and the Humanities, 1–19. http://dx.doi .org/10.1177/1743872114556858 Inc. v. Navarro, 739 F. Supp. 578 (S.D. Fla. 1990). Jackson, A. (2004). Prosecuting local gang cases: What prosecutors need to know. Alexandria, VA: American Prosecutors Research Institute. Retrieved from http://www.ndaa.org/pdf/gang_cases.pdf Johnson, J. D., Bushman, B. J., & Dovidio, J. F. (2008). Support for harmful treatment and reduction of empathy toward blacks: “Remnants” of stereotype activation involving Hurricane Katrina and “Lil’Kim”. Journal of Experimental Social Psychology, 44, 1506 –1513. http://dx .doi.org/10.1016/j.jesp.2008.07.002 Johnson, J. D., Jackson, L. A., & Gatto, L. (1995). Violent attitudes and deferred academic aspiration: Deleterious effects of exposure to rap music. Basic and Applied Social Psychology, 16, 27– 41. http://dx.doi .org/10.1207/s15324834basp1601&2_2 Johnson, J. D., Trawalter, S., & Dovidio, J. F. (2000). Converging interracial consequences of exposure to violent rap music on stereotypical attributions of blacks. Journal of Experimental Social Psychology, 36, 233–251. http://dx.doi.org/10.1006/jesp.1999.1404 Kang, J., Bennett, M. W., Carbado, D. W., Casey, P., Dasgupta, N., Faigman, D. L., . . . Mnookin, J. (2012). Implicit bias in the courtroom. UCLA Law Review, 59, 1124 –1186. Keyes, C. (2002). Rap music and street consciousness. Urbana, IL: University of Illinois Press. Kitwana, B. (1994). The rap on gangsta rap. Chicago, IL: Third World Press. Kubrin, C. E. (2005a). Gangstas, thugs, and hustlas: Identity and the code of the street in rap music. Social Problems, 52, 360 –378. http://dx.doi .org/10.1525/sp.2005.52.3.360 Kubrin, C. E. (2005b). “I see death around the corner”: Nihilism in rap music. Sociological Perspectives, 48, 433– 459. http://dx.doi.org/10 .1525/sop.2005.48.4.433 AQ: 12 Kubrin, C. E., & Nielson, E. (2014). Rap on trial. Race and Justice, 4, 185–211. http://dx.doi.org/10.1177/2153368714525411 Kunda, Z. (1999). Social cognition: Making sense of people. Cambridge, MA: MIT Press. Kunda, Z., & Spencer, S. J. (2003). When do stereotypes come to mind and when do they color judgment? A goal-based theoretical framework for stereotype activation and application. Psychological Bulletin, 129, 522– 544. http://dx.doi.org/10.1037/0033-2909.129.4.522 Kunda, Z., & Thagard, P. (1996). Forming impressions from stereotypes, traits, and behaviors: A parallel constraint-satisfaction theory. Psychological Review, 103, 284 –308. http://dx.doi.org/10.1037/0033-295X .103.2.284 Leveritt, M. (2002). Devil’s knot: The true story of the West Memphis Three. New York, NY: Atria Books. Martin, G., Clarke, M., & Pearce, C. (1993). Adolescent suicide: Music preference as an indicator of vulnerability. Journal of the American Academy of Child & Adolescent Psychiatry, 32, 530 –535. http://dx.doi .org/10.1097/00004583-199305000-00007 Neguţ, A., & Sârbescu, P. (2014). Problem music or problem stereotypes? The dynamics of stereotype activation in rock and hip-hop music. Musicae Scientiae, 18, 3–16. http://dx.doi.org/10.1177/1029 864913499180 North, A. C., & Hargreaves, D. J. (2005). Brief report: Labelling effects on the perceived deleterious consequences of pop music listening. Journal of Adolescence, 28, 433– 440. http://dx.doi.org/10.1016/j.adolescence .2004.09.003 Nosek, B. A., Banaji, M., & Greenwald, A. G. (2002). Harvesting implicit group attitudes and beliefs from a demonstration web site. Group Dynamics: Theory, Research, and Practice, 6, 101–115. http://dx.doi.org/ 10.1037/1089-2699.6.1.101 Open Science Collaboration. (2015). Estimating the reproducibility of psychological science. Science, 349, aac4716. http://dx.doi.org/10.1126/ science.aac4716 Paolacci, G., & Chandler, J. (2014). Inside the Turk: Understanding Mechanical Turk as a participant pool. Current Directions in Psychological Science, 23, 184 –188. http://dx.doi.org/10.1177/0963721414531598 Paolacci, G., Chandler, J., & Ipeirotis, P. G. (2010). Running experiments on amazon mechanical turk. Judgment and Decision Making, 5, 411– 419. Retrieved from http://ssrn.com/abstract⫽1626226 Pashler, H., & Wagenmakers, E. J. (2012). Eds.’ introduction to the special section on replicability in psychological science: A crisis of confidence? Perspectives on Psychological Science, 7, 528 –530. http://dx.doi.org/10 .1177/1745691612465253 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. http://dx.doi.org/10.1037/00223514.81.2.181 People v. Oduwole, No. 5–12-0039 (App. Ct. of Illinois, 5th Circuit 2013). Perry, I. (2004). Prophets of the hood: Politics and poetics in hip hop. Durham, Ireland: Duke University Press. http://dx.doi.org/10.1215/ 9780822386155 Rentfrow, P. J., & Gosling, S. D. (2007). The content and validity of music-genre stereotypes among college students. Psychology of Music, 35, 306 –326. http://dx.doi.org/10.1177/0305735607070382 Rose, T. (1994). Black noise. Hanover, Germany: Wesleyan University Press. Rothman, J. E. (2001). Freedom of speech and true threats. Harvard Journal of Law & Public Policy, 25, 283–367. Rudman, L. A., & Lee, M. R. (2002). Implicit and explicit consequences of exposure to violent and misogynous rap music. Group Processes & Intergroup Relations, 5, 133–150. http://dx.doi.org/10.1177/ 1368430202005002541 Simons, D. J. (2014). The value of direct replication. Perspectives on Psychological Science, 9, 76 – 80. http://dx.doi.org/10.1177/1745 691613514755 Skinner v. State, A-2201-08T2 (N. J. Supr. Ct. 2013). APA NLM tapraid5/zdn-law/zdn-law/zdn00316/zdn2260d16z xppws S⫽1 7/12/16 7:01 Art: 2016-1370 RAP MUSIC This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. Steen, S., Engen, R. L., & Gainey, R. R. (2005). Images of danger and culpability: Racial stereotyping, case processing, and criminal sentencing. Criminology, 43, 435– 468. http://dx.doi.org/10.1111/j.0011-1348 .2005.00013.x Tapper, J., Thorson, E., & Black, D. (1994). Profile: Variations in music videos as a function of their musical genre. Journal of Broadcasting & Electronic Media, 38, 103–113. http://dx.doi.org/10.1080/0883 8159409364249 Tversky, A., & Kahneman, D. (1974). Judgment under uncertainty: Heuristics and biases. Science, 185, 1124 –1131. http://dx.doi.org/10.1126/ science.185.4157.1124 Weinstein, D. (2000). Heavy metal: The music and its culture. Cambridge, MA: Da Capo Press. 13 Weitzer, R., & Kubrin, C. E. (2009). Misogyny in rap music: A content analysis of prevalence and meanings. Men and Masculinities, 12, 3–29. http://dx.doi.org/10.1177/1097184X08327696 Wilson, S. P. (2005). Rap Sheets: The constitutional and societal complications arising from the use of rap lyrics as evidence at criminal trials. UCLA Entertainment Law Review, 12, 345–376. Received April 1, 2016 Revision received May 22, 2016 Accepted May 26, 2016 䡲 JOBNAME: AUTHOR QUERIES PAGE: 1 SESS: 2 OUTPUT: Tue Jul 12 07:01:19 2016 /tapraid5/zdn-law/zdn-law/zdn00316/zdn2260d16z AUTHOR QUERIES AUTHOR PLEASE ANSWER ALL QUERIES 1 AQau—Please confirm the given-names and surnames are identified properly by the colors. ⫽ Given-Name, ⫽ Surname The colors are for proofing purposes only. The colors will not appear online or in print. AQ1—Author: Please be sure to provide the name of the department(s) with which you and your coauthors are affiliated at your respective institutes if you have not already done so. If you are affiliated with a governmental department, business, hospital, clinic, VA center, or other nonuniversity-based institute, please provide the city and U.S. state (or the city, province, and country) in which the institute is based. Departments should be listed in the author footnote only, not the byline. If you or your coauthors have changed affiliations since the article was written, please include a separate note indicating the new department/affiliation: [author’s name] is now at [affiliation]. AQ2—Author: Please add Blecha (2004) to the reference list or remove it from the text. AQ3—Author: Please add Tversky & Kahneman (1974) to the reference list or remove it from the text. AQ4 —Author: Please add Devine et al. (1999) to the reference list or remove it from the text. AQ5—Author: Please verify if “% of Ps” is meant to be “Percentage of participants” in Tables 1 and 3. AQ6 —Author: Please add Neblett v. Commonwealth (2006) to the reference list or remove it from the text. AQ7—Author: Please add Wilson (2005) to the reference list or remove it from the text. AQ12—Author: Please cite Kubrin (2005) in the text or remove it from the reference list. AQ17—Author: Please provide the department affiliation for the author footnotes. AQ18 —Author: Please add Skywalker Records, Inc. v. Navarro (1990) to the reference list or remove it from the text.