Abstract
The main objective of this research was examining the dynamics of stereotype activation in rock and hip-hop music, through the exposure of participants to priming (specific information of stereotypical nature, about the two music genres). The comparative experimental research was performed on a sample of 72 participants, aged between 19 and 21, using The Multidimensional Iowa Suggestibility Scale (short form) and The Lyrics Evaluation Scale. The research used a one-factorial experimental design with the two experimental conditions (absence or presence of priming) realized between groups. In the experiment, participants listened to eight songs belonging to the two music genres and evaluated the lyrics of the songs. The results have shown that stereotype activation occurs in the experimental group, through the more negative evaluation of the songs listened to, in comparison with the control group. The dynamics of stereotype activation is discussed in relation to theories postulated in the field. The implications of these results are discussed, pointing out the idea that the negative outcomes related to these music genres might owe more to the negative stereotypes associated with them than to the music itself.
Rock and hip-hop are two music genres which are said to have caused controversy among parents, public institutions, media and academics for over two decades (Brown & Campbell, 1986; Hansen & Hansen, 1990, 1991; Mulder, ter Bogt, Raaijmakers, & Vollenberg, 2007; Selfhout, Delsing, ter Bogt, & Meeus, 2008; Sherman & Dominick, 1986). As a consequence, for protecting the youth against the negative effects that rock and hip-hop have on behavior, censorship has become a common practice in countries like the United States, Mexico, France, Turkey, Lebanon, Palestine, Israel, South Africa, Zimbabwe, Afghanistan, North Korea and Romania (Binder, 1993; Nuzum, 2001; Korpe, 2004).
Rock and hip-hop music have been very often considered as problem music because they are believed to promote aggressive behavior, encourage drug and alcohol abuse, sexual promiscuity and satanic practices (Anderson, Carnagey, & Eubanks, 2003; Bashe, 1985; Gan, Zillmann, & Mitrook, 1997; Greenfield et al., 1987; Hansen, 1995; Hansen & Hansen, 1991; North & Hargreaves, 2006; Selfhout et al., 2008; Took & Weiss, 1994). A common concern among both the media and researchers is the possible link between rock and hip-hop music and suicidal thoughts or attempts (King, 1988; Litman & Farberow, 1994; Martin, Clarke, & Pearce, 1993; Stack, Gundlach, & Reeves, 1994). Although studies on the harmful effects are not conclusive (Ballard & Coates, 1995; Greenfield et al., 1987; North & Hargreaves, 2006; Rustad, Small, Jobes, Safer, & Peterson, 2003), a stigma is put upon these music genres, as they are considered deviant genres (Mulder et al., 2007), rebellious rhapsodies (Reddick & Beresin, 2002) and deviant music (Lynxwiler & Gay, 2000).
We could argue that nowadays neither genre can be labeled as ‘deviant’ anymore, because a lot of rock and hip-hop artists can be seen on popular music channels worldwide such as MTV or on international charts such as Billboard. However, these music genres have many subgenres, and while some of them can be considered mainstream, others remain underground. As an example, MTV Networks owns another two music channels playing rock and hip-hop underground music – MTV 2 and MTV Jams. Although a clear distinction between mainstream and underground subgenres/artists is beyond the goal of this paper, we must specify that despite their current popularity, rock and hip-hop still raise concern among media and academics (Mulder, ter Bogt, Raaijmakers, & Vollenberg, 2007; Reyna et al., 2009; Selfhout, Delsing, ter Bogt, & Meeus, 2008). For instance, in the case of rock, this situation is quite similar to the late ’80s when, although rock music was very popular among teens, it was still considered a deviant genre (Brown & Campbell, 1986; Hansen & Hansen, 1991; Sherman & Dominick, 1986).
Overall, the main focus in the study of rock and hip-hop music and adolescent behavior regards the correlates of listening to these music styles and harmful behavioral outcomes. However, recent studies investigated the labeling effects on the negative consequences of listening to rock and hip-hop music. Results indicate that a priori labeling of a song by manipulations of its content and music genre influences the participants’ perceptions and evaluations of the song according to the initial frame (Ballard, Bazzini, & Dodson, 1999). For instance, if participants are exposed to a violent lyrical passage and are told that it represents a rap song, they evaluate the passage more critically compared to the situation in which they are exposed to the same passage which is presented as country or folk music (Fried, 1996, 1999). Furthermore, if a song is labeled as life-affirming or suicide-inducing, it could make people evaluate the song according to the label (North & Hargreaves, 2005).
Due to this current state of research in the field, we propose a similar approach in studying the strength of the labeling effects in the perception of rock and hip-hop music. This research question is of considerable interest because, in this case, the labeling of rock and hip-hop music as problem music makes these music genres problematic and prone to have negative effects on the listeners. These effects might owe more to the negative stereotypes associated with these music styles than to the music itself. To our knowledge, this is the first study which attempts to reveal by experimental design the activation of rock and hip-hop implicit stereotypes under priming effect, and the dynamics of this activation. Also, because of the reason stated earlier, our approach will be focused mainly on underground subgenres of rock and hip-hop music.
Priming and stereotypes
Stereotypes are categories that people use to organize their social world (Shih, Ambady, Richeson, Fujita, & Gray, 2002). Stereotypes are considered cognitive structures which contain knowledge, beliefs and expectations about a human group (Kim-Prieto, Goldstein, Okazaki, & Kirschner, 2010; Spencer-Rodgers, Hamilton, & Sherman, 2007). Research has recently shown that stereotypes activate with great ease, automatically and without conscious awareness, even with minimal exposure to the target. This process is known as implicit stereotyping (Bargh, 1999; Bartholow, Dickter, & Sestir, 2006; Blair, Ma, & Lenton, 2001; Devine, 2001; Kawakami, Dovidio, Moll, Hermsen, & Russin, 2000; Kunda, Davies, Adams, & Spencer, 2002). Stereotype activation is explained in terms of cognitive economy in order to preserve mental resources. According to this paradigm, stereotypes are viewed as cognitive shortcuts which help people to structure and simplify the world around them via the process of categorization (Fiske, 1998; Turner & Reynolds, 2002). Previous research has shown that people apply stereotypes when cognitive resources are insufficient. In a high cognitive load situation stereotypes are activated more easily compared to a low cognitive load condition when people are not cognitively busy (Biernat, Kobrynowicz, & Weber, 2003; Hansen & Hansen, 1991; Kawakami et al., 2000; Fiske, 1998).
In the case of rock music, words and lyrics are often difficult to understand due to loud music, instruments used and singing style which is sometimes screaming or raging (Hansen & Hansen, 1991; Kim & Baron, 1988). Also, in hip-hop music lyrics are frequently sung and pronounced fast so that they can be hardly understood by listeners. As a consequence, we propose that in the case of rock and hip-hop music cognitive resources are insufficient, and because cognitive capacity is diminished, stereotypes may activate easily and individuals are more prone to evaluate these music genres according to stereotype consistent information.
Priming is the process through which certain constructs stored in long-term memory are activated for short periods of time, under situational influences (Ford, 1997; Smeesters, Warlop, Avermaet, Corneille, & Yzerbyt, 2003; Werhun & Penner, 2010). The prime is any verbal or visual stimulus that makes a person think of something and can bias one’s thinking, behavior and evaluation. This influence is subtle and unconscious (Gan et al., 2007; Hull, Slone, Meteyer, & Matthews, 2002; Fiske, 1998; Meier, Robinson, & Wilkowski, 2007; Park, Yoom, Kim and Weyer Jr., 2001; Sparrow & Wegner, 2006; Smeesters et al., 2003). This activation is explained in terms of associative networks (Smith, 1998). The associative models of stereotype structure (Meier et al., 2007; Smith, 1998) propose that stereotypes are associative links between a node which represents a social group (in this case, rock and hip-hop music culture) and subsequent evaluations of these music cultures. In case of minimal exposure to the target (e.g. rock and hip-hop music), attitudes and evaluations associated with these music genres are activated and brought to mind. If individuals have stereotypes about rock and hip-hop music, they will evaluate the music genres according to these stereotypes, when exposed to information consistent with the stereotypes concerning this music, such as violent or obscene lyrics, censored by the National Broadcasting Council (NBC). This effect is known as the stereotype priming effect (Shih et al., 2002).
Media, rock and hip-hop music
Media’s influence on individual and social behavior has been widely debated in the scientific literature (Anderson et al., 2003; Kim-Prieto et al., 2010; Leaper, Breed, Hoffman, & Perlman, 2002; McArthur & Eisen, 1976; Myers, 2000). Media creates stereotypes through social communication (Brauer, Judd, & Jacquelin, 2001; Dates, 1980; Ford, 1997; Kim-Prieto et al., 2010; Yzerbyt & Schadron, 2002) and can function as a prime (Ford, 1997; Knobloch-Westerwick, Musto, & Shaw, 2008). Generally, media has focused on the antisocial and reckless themes of rock and hip-hop music (Ballard et al., 1999; Binder, 1993; Knobloch-Westerwick et al., 2008; Schneider, 2011). As a consequence, media may play a decisive role in the judgments people make about rock and hip-hop music. In some cases media focuses on the negative aspects of rock and hip-hop music providing negative information which, in turn, may create negative stereotypes associated with these music genres.
Furthermore, media shapes cognitions and stereotypes associated with these music genres. Rock and hip-hop stereotypes contain social category associations corresponding to these music genres, such as violence, antisocial behavior, drug and alcohol consumption, sexual promiscuity, satanic practices (Arnett, 1996; Ballard et al., 1999; Gore, 1987; Knobloch-Westerwick et al., 2008). The presence of sexuality, violence, death, suicidal tendencies, drugs and alcohol consumption is noticed, however, not to such a great extent as some claim. An analysis of rock and hip-hop songs showed that most songs contain rebellious messages, while the aforementioned themes are less frequently found (Knobloch-Westerwick et al., 2008).
Blaming music for crimes or suicides often masks the real problems or reasons for committing such acts. Some researchers consider that anti-hip-hop attitudes are in fact, anti-black attitudes (Fried, 1999; Noe, 1995). For instance, a study conducted by Reyna et al. (2009) showed that stereotypes of hip-hop music and culture influence anti-black attitudes and justify discrimination, because anti-black attitudes are associated with discrimination against blacks. Others state the hypothesis that for rock fans listening to rock, music may work as a catharsis (Arnett, 1991, 1993; Weinstein, 2000). Also, not all hip-hop and rock subgenres contain themes of sexual violence or antisocial behavior, which are present in subgenres such as gangsta rap and black or death metal (Perkins, 1996). In contrast, the positive aspects of these music genres, like involvement of different rock and hip-hop bands in anti-drug campaigns, literacy campaigns, anti-violence campaigns, humanitarian or donating money to charity are often overlooked or, at the most, briefly mentioned. Even more, the current use of rock (Tervo, 2001) and rap (Ciardiello, 2003; DeCarlo & Hockman, 2003) as an intervention method/therapy in dealing with adolescents with problem behavior is rarely mentioned by the media.
In Romania, rock and hip-hop music have developed a lot since the 1989 revolution. Several rock bands date back to the Ceausescu period, before the 1989 revolution, but in that period hip-hop bands were almost absent. Nowadays, both the rock and hip-hop music scenes are populated by a lot of bands, festivals and concerts with foreign bands internationally recognized. However, the perception of rock and hip-hop music in the media and society is similar to the one held in other European countries and North America: rock and hip-hop songs are considered deviant genres with negative implications for the listeners (Nedelea & Stoian, 2005), although no empirical study has investigated this topic in Romania.
Present research
The major objective of this research is to investigate whether priming implicitly activates rock and hip-hop music stereotypes, even under low cognitive load condition. To achieve this goal we used a simple experimental design with two experimental conditions: the absence or presence of priming. Both participant groups listened to a total of eight songs (three rock music tracks, four hip-hop music tracks and one pop track, as a control song), after which they evaluated the lyrics in terms of different dimensions: violence, misogyny, alcohol or drugs consumption, sex and satanism.
One variable that might make people more prone to priming effect is suggestibility, as it seems to influence compliance and conformity to group norms (Brehm & Kassin, 1996). Therefore, we have decided to measure this variable, in order to statistically control its effect on the evaluations that participants made.
Information about the Romanian songs used is presented in Table 1. The use of a control song is necessary because if a song belonging to other music genres than rock or hip-hop is more negatively evaluated by the experimental group, then this could raise serious doubts whether the observed differences support stereotype activation. Also, consistent with our prior statements, all the seven songs belong to underground subgenres of rock and hip-hop. The only song which could be considered more mainstream is the first one, mainly because it is the only one that has a video and was actually broadcast on national music channels (with several restrictions from the NBC).
Information about the songs.
Note. * represents the total number of censored words in the song, including the repeated ones.
The present study contributes with new points to this field of research. Firstly, it attempts to show the strength of the prime to bias judgments, by using a wide range of songs, some of which contain lyrics with an explicit content related to targeted dimensions. Secondly, participants are exposed to the experimental conditions and assess the songs under a low cognitive load condition (because they are provided with written lyrics while they are listening to the songs), with sufficient cognitive resources in order to reduce stereotype activation. Thirdly, using a wide range of rock and hip-hop songs, which contain some of the main ‘problem’ themes present in these cultures, offers a better reflection of reality and provides participants with a wider perspective for their evaluation.
To our knowledge, this is the first experimental study using a wide range of the subspecies of these genres, which investigates whether rock and hip-hop music stereotypes activate even in the condition of low cognitive load. Thus, we hypothesize that participants exposed to priming will evaluate more negatively the content of the songs, in comparison with those not exposed to priming. We also anticipate that no differences will emerge in the evaluation of the control song.
Methodology
Participants
The sample used consisted of 72 undergraduates, out of which 68 (94.44%) were females. The participants were randomly divided into two groups: control group (37 participants) and experimental group (35 participants). The age of the respondents ranged from 19 to 21 years (M = 19.54, SD = .67). Although the sample is somewhat small, the a-priori statistical power analysis using PowerStaTim (Sava & Maricuţoiu, 2008) revealed an optimal statistical power (1−β = .88) for the moderate effect size expected. All the participants received extra credit in an introductive statistical course for participating in this experiment.
Instruments
The Multidimensional Iowa Suggestibility Scale (MISS, Kotov, Bellman, & Watson, 2004), short form, contains 21 items scored on a five-point Likert scale and measures suggestibility. Internal consistency was satisfactory (Cronbach’s α = .78).
The Lyrics Evaluation Scale, for measuring the participants’ lyrics appreciation (see Appendix). The scale was developed for this experiment, and assesses ten items: obscenity, violence, misogyny, sexual content, alcohol incitement, incitement to drug use, incitement to rebellion against the state, references to satanic practices, references to demonic creatures, and encouraging the development of antisocial behavior. Participants had to rate every item on a ten-point scale, 1 representing a minimum load of one song in that item, and 10 representing a full load. By summing up all items, a total score is obtained (participants completed eight such assessment scales, one for each song listened to). The internal alpha consistencies ranged from α =.75 (for song no. 2) to α = .90 (for song no. 7).
Procedure
The experiment used a one-factorial experimental design with the two experimental conditions (absence or presence of priming) realized between groups. In order to avoid experimenter effects, neither the participants nor the research assistant knew what the real purpose of the experiment was. They were told that the research is related to recognition memory performance. The priming phrase (by which the experimental group was exposed to the specific information related to the stereotype) is: ‘Before you begin, we should mention that all the songs you will listen to contain, to a greater or lesser extent, obscene or violent lyrics. Also, some of these songs have been banned or censored by the NBC’.
In the first phase of the experiment, the participants completed the demographic data and the MISS. The second phase of the experiment consisted in listening to the eight songs (while reading their lyrics) and in evaluating them. At the beginning of this phase, the research assistant explained the task to the participants. At this point, the experimental group was exposed to priming. After the experiment was over, the debriefing section began, and the real purpose of the experiment was revealed.
Results
A preliminary Exploratory Factor Analysis was used for each song, in order to verify whether summing up the items into a total score was appropriate. The analysis was accomplished using Principal Axis Factoring followed by Direct Oblimin Rotation. Both parallel analysis and scree plot suggested the extraction of only one factor (for each song), which explained 32.12% to 41.34% of the total variance. All items had their primary loading in the first factor. This analysis suggests that summing up the items into a total score for each song is an appropriate procedure. Table 2 presents the factor loadings for all eight songs (on the one extracted factor) and the internal alpha consistencies for each song.
Results of the Exploratory Factor Analysis and Cronbach’s alpha for all songs (on the one extracted factor).
None of the distributions for the total score (obtained by summing up all the items) of every song deviated substantially from normality (skewness and kurtosis < 2). Table 3 presents the t-test results for the total score of every song. Significant results were obtained for songs 2 (t = 2.17, p < .05), 5 (t = 2.62, p < .01) and 6 (t = 3.19, p < .01), with moderate to strong effect sizes (Cohen’s d from 0.52 to 0.76). As expected, no differences emerged in the evaluation of the control song.
t-test results for the total score – all songs (one-tailed).
Note. * p < .05, ** p < .01.
Because suggestibility positively correlated with a more negative lyrics evaluation for all songs (.23 to .31), we conducted an analysis of covariance (ANCOVA) for songs 2, 5 and 6, including suggestibility as a covariate. The results of the ANCOVA (Table 4) show that significant differences were identified for all three songs, with moderate effect sizes (partial η2 from .06 to .12). Suggestibility had a significant effect on all the songs, with moderate effect sizes as well (partial η2 from .05 to .09).
ANCOVA results for the total score (songs 2, 5 and 6).
Note. * p < .05, ** p < .01, EM = estimated mean.
Discussion
This research advances the following idea: exposing participants to a priming stimulus that contains information consistent with rock and hip-hop music stereotype (i.e. violent lyrics prohibited by the NBC) will bias participants’ judgments and evaluations in the experimental group, in a stereotype consistent manner. Specifically, participants exposed to priming will appreciate the lyrics’ content in a more negative way, compared with those not exposed to priming. Although some studies argue otherwise (Hansen & Hansen, 1991; Kawakami et al., 2000; Fiske, 1998), we anticipated that stereotypes will activate in a low cognitive load condition, under sufficient cognitive resources which allow an accurate and non-stereotypical assessment. Overall, our assumptions gained reasonable support.
The results showed that stereotypes were activated in three of the seven songs evaluated (excluding the control song) because participants in the experimental group evaluated more negatively the second, fifth and sixth song compared to the control group. This indicates that the prime has triggered rock and hip-hop music stereotypes in the experimental group.
The presence of the control song allows us to rule out the possibility that participants in the experimental condition would have been primed by a general idea of social disapproval, rather than any specific musical stereotype. As expected, there were no differences between the experimental and control group concerning the evaluation of the control song. This shows that, in the case of three of the other seven songs, participants have been primed with a specific music stereotype which, in turn, activated rock and hip-hop music stereotypes. We remind you that priming activates stereotypes only if the information is suitable to the target. In the case of the control song, stereotypes were not activated because the information from the song was not stereotype consistent. Three of the other seven songs contained stereotype consistent information, which in turn led to stereotype activation.
Overall, the main result of this research is that when stereotypes are activated by the instructions, participants evaluated the songs more negatively than the ones in the control condition did. Of particular importance is that stereotypes activated even in a low cognitive load condition, under sufficient cognitive resources. These results support the idea of the prime’s strength to bias judgments and assessments in case of explicit lyrics content, as well as the ease by which stereotype activates in three of the seven songs, even when cognitive resources are sufficient.
The data obtained provides a complex perspective on the dynamics of stereotype activation. At the beginning of the trial, a slight inhibition of the prime’s influence may be observed. This fact is consistent with the idea that in complex tasks an initial filtering of information leads to a partial activation of associated concepts stored in long-term memory (Smith, Besner, & Miyoshi, 1994). After this inhibition, which probably occurs along the first three songs and takes about 15 to 20 minutes, follows a surprising stereotype boost on the fourth and fifth song. This can be interpreted as the peak of priming effect which leads to stereotype activation and application. Then on the last two songs there is a decrease of this effect. This aspect may be interpreted as follows: the more priming effect increases, the more stereotypes activate and apply. Thus, in the case of a reduced priming influence, stereotype activation is minimal. This explanation needs further evidence, but can stand for future research, given the fact that in priming research on stereotype activation and application it is considered that a mere exposure to a disadvantaged group can automatically activate implicit stereotypes on that group (Kunda et al., 2002).
Also, an alternative explication has to be considered for the results pattern in the beginning of the trial. One may have noticed that the first and third song (containing 11 and 23 censored words, respectively) achieved very high ratings in both participant groups, suggesting that both groups reacted strongly to the language used in these songs. This ‘language effect’ might have been stronger than the priming effect, at least in the beginning of the experiment.
In addition to the theoretical and empirical implications discussed above, the current study links to one of the potential mechanisms by which the media creates stereotypes regarding rock and hip-hop music. Media primes individuals’ attitudes and behaviors (Ford, 1997; Kim-Prieto et al., 2010; Knobloch-Westerwick et al., 2008; Leaper et al., 2002; McArthur & Eisen, 1976), and often communicates negative information about rock and hip-hop music (Binder, 1993; Schneider, 2011; Selfhout et al., 2008). The current research is based on the foundations of priming theory and claims a major issue: it suggests that priming effect induces negative judgments about rock and hip-hop music. Similar results are supported by the labeling effect (Fried, 1996; 1999; North & Hargreaves, 2005). This indicates that the pattern of labeling rock and hip-hop music as problem music may, in fact, create stereotypes of those music genres, making at least the lyrics of the music look more problematic and deviant than they actually are.
The present research has three notable limitations that should be taken into account when interpreting its results. Firstly, the overwhelming majority of female participants could raise some questions about the external validity of the results, although several studies have not identified differences among men and female regarding stereotype activation (Banaji & Hardin, 1996; Fiske, 1998). Nevertheless, other studies have identified that male participants are more involved in rock and hip-hop subculture and have a greater taste for this music (Arnett, 1991, 1993). Therefore, women might evaluate rock and hip-hop songs more critically than men. Secondly, because the song selection included an unequal number of rock and hip-hop songs, a more complex MANCOVA analysis could not be accomplished. Thirdly, in this study we did not use a counterbalanced presentation of the songs, in order to minimize carry-over effects. Because of that we cannot rule out the possibility that the results of the seven experimental songs might have additionally been caused by song differences.
Also, this study points out several future research directions related to this topic. One could focus on further verifying the stereotype priming effect in both high and low cognitive load conditions (i.e. listening to music with and without providing participants with written lyrics’ support). Another research direction could focus on discriminating between the negative effects caused by the instrumentals and the lyrics of these genres. This would be of great importance because, for example, although rock lyrics are quite often not as offensive as hip-hop ones, the instruments used may alter the song’s message, and make it seem harsher than it really is.
To sum up, the current research results are convergent with those that highlight the stereotype priming effect. Thus, we have focused on rock and hip-hop music stereotypes and have provided an insight into the dynamics of these stereotypes under priming influences. Even more, results highlight the presence of rock and hip-hop music stereotypes among participants and offer support for the labeling effect. Therefore, we could conclude that many of the negative outcomes related to rock and hip-hop might owe more to the stereotypes associated with these music styles than to the music itself.
Footnotes
Appendix
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
