Abstract
In adolescence, many students align their music participation choices with the expectations for their gender group. While cultural norms may explain gender disparities in music activities, there is no empirical evidence of how gender salience may influence students’ perceptions of music activities. The purpose of this study was to experimentally test the influence of gender salience on adolescents’ perceptions of music activities. Using a quasi-experimental design, middle school participants (N = 246) were assigned to either a primed condition where students were asked to consider their gender identity or a control condition; all participants took a survey on interest in music activities. Participants in the experimental condition rated music activities, including stereotypical masculine and feminine activities, significantly more positively than the control group. Music students rated music activities significantly higher than their non-music peers. Gender was only identified as having a significant relationship with feminine music activities, as ratings of stereotypical masculine activities were statistically similar between boys and girls. This experiment provides evidence that gender salience can impact students’ perceptions of music activities. With a more complete understanding of how gender operates within the environment, music educators can understand how gendered messages within the environment may impact perceptions of music activities.
Decisions to participate in music activities, especially school music electives, are complex and multifaceted (Demorest et al., 2017). Attitudes about music, family experiences with music, and peer influences can impact the decision to participate in school music. Within certain cultural contexts, such as Australia, China, and the United States, researchers in music education have also identified gender as being a possible contributing factor in school music participation decisions (Elpus, 2015; Ho, 2003; McPherson et al., 2015).
Gendered messages from within a culture may influence children to view certain music activities to be the domain of a particular sex. From a very early age, children assign masculine or feminine labels to certain musical instruments, with higher wind and string instruments being perceived as feminine and percussion and brass being perceived as masculine (Abeles, 2009; Abeles & Porter, 1978; Delzell & Leppla, 1992; for a review, see Eros, 2008). These gender norms of instruments are established by 5-year olds (Pickering & Repacholi, 2001) with evidence that even preschoolers are aware of such norms (Marshall & Shibazaki, 2013).
Within the United States, gender disparities in music participation have persisted in both instrumental and singing activities (Abeles, 2009; Elpus, 2015). In US school music programs, girls have been significantly overrepresented within high school music ensembles over the past 30 years, with approximately two-thirds of choirs and orchestras comprising girls (Elpus, 2015). Some researchers, such as those investigating the disparity of boys and girls in choral singing, believe a ‘hegemonic masculine’ culture, within some contexts, prevents boys from crossing gender norms and participating in music activities at all (Harrison, 2008). While cultural expectations for gender groups may help explain these disparities, it is still unclear within the research literature how gender operates in the environment and how it may impact perceptions of music activities.
Social identity theory
Social identity theory was used as a theoretical framework for this project. Tajfel and Turner (1986) developed SIT on the idea that individuals create social categorizations of groups based on membership. These social categories are created and maintained to classify, partition, and organize the social world; an individual’s membership or relation to a group can inform one’s behavior. Furthermore, social categorization provides individuals a tool for orientation and self-reference, as well as an ability to define themselves within society (Tajfel & Turner, 1986). Individuals maintain positive perceptions of the groups and collectives of which they are members (Tajfel & Turner, 1986).
Individuals can have many varied social identities that are meaningful to them, including political affiliation, ethnicity, gender, and allegiance to a particular sports team (Ethier & Deaux, 1994; Hogg, 1992; Hogg & Abrams, 1988). These social identities have a prototypic representation of an ideal member, based on the beliefs, attitudes, and behaviors of collective membership. Identifying with social identities that are meaningful to the individual has positive benefits such as heightened self-esteem (Luhtanen & Crocker, 1992). However, threats to social identity can lead to negative outcomes such as antisocial behavior, decreased self-esteem, depression, and stress (Aquino & Douglas, 2003). Moreover, environmental cues and social contexts can activate social identities within an individual and when threatened can influence behavior or performance (Murphy et al., 2007). When membership to a social category is made salient, distinctions between the in-group and out-group become more pronounced, and individuals within a group will minimize differences between themselves; it is from this intergroup position that stereotypes about individuals in the out-group are formed (Tajfel & Turner, 1986).
Critics of social identity theory have characterized the framework as too reductionist, diminishing the complex amalgam of the individual self to simplistic categories, as well as a lack of defining what psychological processes need to be in place for a person to feel connected or influenced by group membership (Rabbie et al., 1989; Rabbie & Horwitz, 1988). For these critics, social identity theory fails to explain why people of the same social categories behave differently within the same context. Tajfel, one of the co-creators of social identity theory, would contend that the goal of the theory is not to explain every action by any group member, but to explain why so many people, especially those within specific social categories, ‘behave in unison – as they often do’ (Tajfel, 1977, p. 66). For Tajfel, social behavior is rule-governed and SIT helps elucidate behavior decisions (Turner, 1996).
Gender identity
Gender is ‘one of the most important, salient, and pervasive social categories’ (Maass et al., 2003, p. 854). Gender is the socially or culturally constructed characteristics identified for each sex, whereas sex is the biological categorization within a male–female binary (World Health Organization [WHO], 2019). When a particular social identity (such as gender) is activated, membership within that social group becomes salient (e.g., gender salience) and a ‘vigilance process’ or a heightened sensitivity to the social cues within the environment is initiated, which can be associated with physiological responses and stress (Murphy et al., 2007). The heightened arousal state of vigilance can lead individuals to conclude that innocuous cues, such as an instructor’s sex or race, may be seen as a potential source for affirmation or devaluation (Wout et al., 2009). Any proficiency or interest in an activity that does not align with one’s gender could activate a gender identity threat (Rudman et al., 2007). In gender identity threat, an individual’s membership with their gender group is questioned, causing the individual to actively restore their membership by adhering to gender norms (Bosson & Vandello, 2011; Sinclair & Carlsson, 2013). For example, gender identity could be threatened when a boy is teased by his peers for being interested in singing (see Palkki & Caldwell, 2018).
Gender salience, or overtly considering one’s gender in a given situation, has been identified as affecting attitudes and preferences (Coyle & Liben, 2016; Hilliard & Liben, 2010). Sinclair and Carlsson (2013) reported that adolescents’ preferences for certain professions were significantly more stereotypical in a gender threat condition; gender identity threat was caused by having participants compare their experiences with those that were ‘typical’ for their gender group (p. 6); this wording was chosen to allow participants to make an inferential response to the prompt instead of relying on presupposed ideas about what behaviors belong to a specific gender group. The findings suggest that when gender identity is activated, their preferences are more aligned with socially constructed gender norms. In addition, Sinclair and Carlsson (2013) found that gender identity threat affected both adolescent boys and girls in a similar manner, with both preferring stereotypical careers more highly when under gender identity threat.
Examining gender and social identity through quasi-experimental designs is relatively new within the field of music education. Watson et al. (2017) examined whether performance and attitudes of choristers would be influenced by a situation where membership to their gender group is threatened. The researchers had choirs perform in a non-stereotype threat condition (performance for peers in choir) and a stereotype threat condition (performing for the school at large), where participants may be worried about judgment from their gender group peers. The researchers reported that the performance ratings of male choirs in the stereotype threat condition were significantly lower than those in the non-stereotype condition, although there was no significant difference in the performance of female choirs. The authors contend that the stereotype threat presented to male choirs negatively impacted their performance, or in other words, that participating in ‘gender transgressive’ acts can cause decrements in performance (p. 240). This study may provide evidence that motivation in music performance may be influenced, perhaps unconsciously, by environment and gender expectations.
The purpose of this study was to examine how perceptions of music activities may be influenced by making gender salient in a group of adolescents. By activating gender identity and gender group membership within a quasi-experimental setting, it can be observed that perceptions of music activities may differ by gender group. Participants, especially those who have been asked to consider their gender identity in relation to others, may have more positive perceptions of music activities that align with the values of their gender group as found in Sinclair and Carlsson’s (2013) study on careers. The literature on SIT suggests that participants will actively pursue activities that align with the values of the in-group and devalue activities that are associated with the out-group (Brewer, 1979; Hewstone et al., 2002). Moreover, when a group membership is made salient, conformity to group norms increases (Skinner & Stephenson, 1981). This project examined one research question:
To what degree are adolescent perceptions of various music activities influenced by gender salience or activating gender group identification within the environment?
Method
Participants
The participants of the study were recruited from 15 seventh- and eighth-grade classrooms at a middle school in the Pacific Northwest. After receiving proper authorization from the university’s Institutional Review Board and the middle school administration, the researcher presented the opportunity to the students in class to participate in the study approximately 1 week before the experiment. The school is highly diverse, with the student body being 39% Hispanic, 16% Asian, 6% African American, 5% Pacific Islander, 1% Native American, 26% White, and 7% two or more races (demographics taken from state school report card, 2013–2014). While the exact racial demographics of the participants are unclear as racial data were not self-reported by the participants, 11 of the classes approached for participation were math and science classes of various levels in an attempt to be as representative as possible. The remaining four classes were made up of music students; music students were defined as those enrolled in a band class at the middle school at the time of the study as no other music class was offered at this school.
A week prior to the administration of the instruments, the researcher invited students to participate in the study, obfuscating the purpose by explaining that they would be giving information about themselves and indicating what activities they would find pleasant among a list of various activities. Students were also given the necessary paperwork as approved by the Institutional Review Board.
Two hundred forty-six students participated in the study. Genders were equally represented, with 123 boys and 123 girls participating. One hundred eleven music students as well as 135 non-music students participated in the study. There were 134 students randomly assigned to the experimental condition and 112 students to the control condition.
Procedure
Each participant completed two instruments: the Gendered Music Activity Survey and the Collective Self-Esteem Scale: Gender Version. The Gendered Music Activity Scale was a researcher-created instrument based on the results of a pilot study. In the pilot study, a group of different participants (N = 59, 26 were female) were asked to rate various activities, including music activities, on a Likert-type scale that was anchored with ‘very masculine’ or ‘very feminine’; the 7-point rating scale also contained an anchor above the midpoint of the scale, labeled as ‘neutral’. ‘Masculine’ and ‘feminine’ were not defined for the pilot study participants in an effort to allow them to construct their own meaning of these constructs and evaluate accordingly.
In the pilot study results, participants rated six music activities as being significantly more masculine than neutral (one-sample t-test significantly different from ‘neutral’, p < .001); these activities included (1) singing in a garage band, (2) learning to play guitar, (3) performing in a drum circle, (4) working the soundboard at a theatrical show, (5) playing a music video game like Guitar Hero, and (6) creating mixes, loops, mash-ups, or playlists. Seven music activities were identified as significantly more ‘feminine’ than ‘neutral’. Six of these ‘feminine’ activities were selected to balance the ‘masculine’ activities; the ‘feminine’ activities included (1) singing in a choir, (2) singing karaoke, (3) performing in a musical, (4) learning to play the piano, (5) learning to play the flute, and (6) taking a dance class. The scale as well as each of the composites for the masculine or feminine activities had an acceptable reliability (Cronbach’s α greater than .80).
After the pilot study was conducted, the Gendered Music Activity Scale was created to gauge personal interest in these music activities. The Gendered Music Activity Scale prompted a different set of participants to rate ‘how pleasant do you find the following activities’ on a scale from 1 (‘very unpleasant’) to 7 (‘very pleasant’). This scale was directly modeled after the instrument created by Steele and Ambady (2006), using the language (e.g., ‘pleasant’) that they had used when examining attitudes toward math and arts after priming women to gender salience. On this created scale, a rating of 4 was identified as ‘neutral’. The scale included the six ‘feminine’ activities, the six ‘masculine’ activities, and five filler activities (i.e., learning to juggle, working out, and drawing) that were added to obscure the purpose of the study.
All participants also completed the Collective Self-Esteem Scale, an instrument taken from previous studies (Luhtanen & Crocker, 1992; Tobin et al., 2010). This instrument is a 16-question survey that asks the participant to evaluate her relationship to her gender group 1 by rating statements on a scale of 1 (‘strongly disagree’) to 7 (‘strongly agree’). Example statements on the scale include the following: ‘The gender group I belong to is an important reflection of who I am’ or ‘I often regret that I belong to my gender group’. The advantage of using this scale over other scales that intend to measure gender identity (i.e., Sex Role Inventory Scale by Bem, 1981) is that the participant infers what it means to be ‘male’ or ‘female’ rather than being supplied stereotypes by the instrument (Tobin et al., 2010). This approach is considered more valid in measuring gender identity, as an individual can appraise his or her gender holistically, rather than measuring one dimension of gender (e.g., aggression, sensitivity). It also does not pressure the participant to respond in a way that is stereotypical of their sex (Perry & Pauletti, 2011). The Collective Self-Esteem Scale has been used to prime participants in other experiments and has been validated to make gender identity salient within participants (Wang & Dovidio, 2017). Finally, in addition to the two scales that were administered, all participants answered two demographic questions. Participants were asked to identify their gender and their grade level.
The survey instruments were administered in classroom settings within a normally scheduled class period. Each participant was given a packet of surveys with a cover page. Each participant completed (1) the Collective Self-Esteem Scale: Gender Version, (2) the Gendered Music Activity Scale, and (3) Demographic Questions. Although all participants completed the same surveys, they were presented in different orders depending on the condition. For the experimental condition, the participants completed the Collective Self-Esteem Scale (used as an intervention), the Gendered Music Activity Scale (dependent variable), and then the demographic survey. For the control condition, the participants first completed the Gendered Music Activity Scale (dependent variable), followed by the Collective Self-Esteem Scale (used as a comparison for the experimental condition) and the demographic survey.
To randomize the conditions for participants, both conditions were administered simultaneously within classrooms as packets were distributed from a pre-sorted stack of randomized packets. The stack was randomized using a random number generator. After the completion of the quasi-experiment within each class, the researcher debriefed the participants and explained the purpose of the study.
Data analysis
To explore the research question, composites were tabulated from the Gendered Music Activity Scale to compare between conditions; the six items that rated pleasantness in ‘masculine’ music activities were added together to create a composite for ‘masculine’ music activities. Similarly, the six items that rated pleasantness in ‘feminine’ activities were combined to create a composite for ‘feminine’ music activities.
Results
Descriptive data for each musical activity presented are available in Table 1.
Participants’ Rating of Music Activities.
Examining ‘masculine’ music activities
A 2 (gender type; fixed, between-subjects) × 2 (music status; fixed, between-subjects) × 2 (condition type; fixed, between-subjects) factorial analysis of variance (ANOVA) was conducted on ratings of pleasantness of ‘masculine’ music activities. The results (displayed in Table 2) revealed two significant main effects: music status (p < .001) and experimental condition (p = .004). No significant main effect associated with gender was detected (p > .05), indicating that boys and girls have similar distributions in rating of ‘masculine’ music activities. Examination of the means across conditions shows that music students rated ‘masculine’ music activities higher than non-music students. Similarly, students who had been considered their gender (gender salience condition) rated ‘masculine’ music activities higher than students in the control condition.
Descriptives and Results for Ratings of Masculine Music Activities.
The significant main effects should be interpreted in light of a significant two-way interaction between music status and experimental condition (p = .029). To better understand the nature of the interaction, the group means of the main effects are plotted below, separated by gender (see Figure 1). The plots demonstrate that non-music students who were in the gender salience condition rated ‘masculine’ music activities more pleasant than their peers in the control groups; however, music students rated ‘masculine’ activities more favorably regardless of the condition. Finally, the plots illustrate that girls and boys rated ‘masculine’ music activities in a similar manner. The other two-way and three-way interactions in the model were not significant (p > .05).
Ratings of Masculine Music Activities.
Examining ‘feminine’ music activities
A 2 (gender type; fixed, between-subjects) × 2 (music status; fixed, between-subjects) × 2 (condition type; fixed, between-subjects) factorial ANOVA was conducted on pleasantness ratings of ‘feminine’ music activities. The results (see Table 3) indicated three main effects on ratings of ‘feminine’ music activities: gender type (p < .001), music status (p < .001), and experimental condition (p = .001). By examining the group means, it is evident that girls significantly rated these music activities higher than boys (see Figure 2). In addition, music students rated these activities higher than their non-music peers. Finally, students who had been asked to think about their gender rated ‘feminine’ music activities significantly higher than participants in the control group. The model did not detect any significant two-way or three-way interactions.
Descriptives and Results for Ratings of Feminine Music Activities.
Ratings of Feminine Music Activities.
Discussion
The purpose of this study was to examine whether adolescents’ views on various music activities were influenced by considering their gender. The results are clear, in that the intervention had a significant effect on the responses. Under the gender salience condition, both boys and girls rated stereotypical music activities for their gender significantly higher than their peers in the control group. Thus, it appears that the subtle intervention of gender salience may induce a response in adolescents, although that response was relatively weak in influence (ω2 = .03 for both boys and girls). These results are expected under the theoretical framework of social identity theory, in that gender group membership would be activated and in-group values might override personal interests.
However, the intervention also influenced the ratings of non-stereotypical music activities for each gender group. This finding seems to conflict with social identity theory, where participants would align their personal interest with the in-group values as well as distance themselves from the out-group. This distancing from the other gender group has been documented in other studies (Sinclair & Carlsson, 2013; Steele & Ambady, 2006); thus, the gender salience condition in this study does not align with what would be expected according to social identity theory. While it is not feasible to ascertain the cause of the higher ratings of non-stereotypical music activities from this study alone, there are several possible explanations.
First, the higher ratings of the non-stereotypical activities (as well as the ratings of the stereotypical activities) could be a result of an order effect. Participants in the experimental group were first given the Collective Self-Esteem Scale: Gender Version, an instrument that asked participants to consider their relationship with their gender group. These participants then rated their perception of pleasantness in music activities on the Gendered Music Activity Scale, a decidedly less burdensome task on their cognition, while the control group started the experiment by rating the music activities. Thus, the order of the tasks may have influenced how participants rated music activities. While an order effect is a possible explanation, music students were affected differently by the experimental condition than non-music students. It is likely that if an order effect had influenced the dependent variables, it would have affected music and non-music students in a similar manner.
The unexpected results may have been produced from a lack of precision in the intervention. The intervention instrument (the Collective Self-Esteem Scale: Gender Version) was selected as a subtle means to have participants consider their gender identity; however, the open-ended nature of the instrument may have been too subtle for participants to consider their gender group membership’s values in a way that would impact their response. Although a manipulation check was employed to a limited sample to ensure that this intervention was similar to a priming agent that induced gender identity threat in Sinclair and Carlsson’s (2013) study, the small number of participants may have been susceptible to a Type II error, where there is a significant difference between groups, but a lack of power prevented it from being detected.
The intervention may have influenced the participants in other ways than intended. Rather than making participants consider their membership to their gender group, the intervention for these students may have prompted an openness toward gender issues. Within the sample, two participants reported being transgender. The presence of transgender students within this middle school may have prompted the school administration and teachers to discuss gender identity with their students. In discussion with the teachers at the study site, they stated that their students were very affirming of students with different gender identifications. Perhaps, for this sample, the intervention promoted a more global attitude of gender affirmation as opposed to activating a social category. Future study is needed to explore whether these results are idiosyncratic to this sample where gender identity may be more widely discussed as well as participants are more accepting of different gender expressions. Finally, perhaps the data of this study have stumbled onto a weakness of social identity theory, in that the design of this study overly simplifies the individualistic character and interests of the participants, in that, even when instructed to consider the values of their gender group, there are other influences and factors that guide their responses.
Regardless of why the effect occurred on non-stereotypical activities, its influence on non-stereotypical music activities bears some discussion. While there were positive gains in the ratings of non-stereotypical music activities by participants in the experimental condition, the gains have to be qualified. For boys, the experimental group rated ‘feminine’ music activities significantly higher than the control group. However, when looking at the mean values of ratings of the ‘feminine’ activities, boys in the experimental group rated every ‘feminine’ music activity as lower than 4 (neutral) despite the influence of the intervention. Thus, for boys, gender salience influenced them to view the ‘feminine’ activities to a lesser degree of unpleasantness rather than being pleasant.
For girls’ ratings of ‘masculine’ music activities, the influence of the intervention on the results is more nuanced. For some ‘masculine’ music activities (performing in a drum circle, singing in a garage band), gender salience influenced girls in the same manner as boys, where the mean scores remained below the neutral threshold and overall rated the activities as unpleasant. Several music activities (learning to play the guitar, creating playlists, mash-ups, mixes, and loops) were viewed more positively than neutral by girls in the control group; the intervention influenced girls in the experimental group to view them even more positively than their control group counterparts. The mean scores of two activities (playing a music video game like Guitar Hero, working the soundboard at a theatrical production) crossed from being viewed as unpleasant in the control group to being viewed as pleasant in the experimental group. These results may indicate that the intervention may have caused girls to view some music activities as pleasant that they would normally view as unpleasant.
Gender salience also influenced students differently depending on their status as a music student, as identified by the interaction effects between the experimental condition and the music status. For music students, the intervention had little effect on the ratings of music activities. These findings could suggest that music students are already at a heightened state of gender salience – although why they are is still unclear. Another explanation is that the survey concerning music activities activated components of their identity as a music student, and the result was similar to that experienced by non-music students in the experimental condition. Clearly, these explanations are only speculative and would need additional study to understand more fully.
Additional main effects other than experimental condition were detected in the results. Status as music students was identified as a main effect on ratings of music activities within the models. Music students viewed both ‘masculine’ and ‘feminine’ music activities as significantly more pleasant than their non-music peers. This finding seems logical as students who elect to take a music class during the school day may be more open to music activities than non-music students. The effect sizes of the main effects of music status on ratings of ‘masculine’ music activities (ω2 = .07) and on ratings of ‘feminine’ music activities (ω2 = .07) indicate a moderate association between music status and the dependent variable. It should be noted that music students in this study only refers to those students who were enrolled in curricular music classes; it is unknown what the musical lives of these participants were outside of school.
In the results of the study, gender was identified as a factor that produced a significant main effect for ‘feminine’ music activities. The results indicate a large disparity in ratings of ‘feminine’ music activities between boys and girls, with girls viewing ‘feminine’ music activities much more pleasantly. These results seem to typify the discussion of hegemonic masculinity as a cultural phenomenon (Freer, 2012; Harrison, 2008), where boys are not attracted to ‘feminine’ activities. The effect size for this main effect (ω2 = .20) was the largest one detected in the study, suggesting a very strong association between gender and perceived pleasantness of ‘feminine’ music activities.
While gender had a strong association with ratings of ‘feminine’ music activities, there was no significant difference in the ratings of ‘masculine’ music activities between boys and girls. This finding is consistent with some other studies inside and outside of music that reported girls are more willing to consider ‘masculine’ activities than boys are to consider ‘feminine’ activities (Conway, 2000; Green, 1993; McBride & Palkki, 2020). Girls may be more willing to cross gender norms to seek perceived benefits or be encouraged to challenge gender roles that they find restrictive.
While boys may be more attracted to ‘masculine’ music activities than ‘feminine’ music activities (as the data here suggest), boys may be guarded about such activities because they view the entire domain of music as a ‘feminine’ enterprise. Even though girls are equally attracted to ‘masculine’ music activities as boys, they are less drawn to ‘masculine’ activities than ‘feminine’ activities. However, girls are ostensibly more attracted to ‘masculine’ music activities than boys are attracted to ‘feminine’ music activities. Again, this finding could be a result of hegemonic masculine culture where girls encounter fewer social liabilities to cross gender barriers than boys. These findings are similar to those of Pickering and Repacholi (2001) who found that girls were more willing to consider counter-stereotyping of instruments than boys. This study cannot illuminate why there are not more pronounced differences between genders on ‘masculine’ music activities, and future research on this topic is warranted.
In conclusion, participants in this study were influenced by the intervention and rated stereotypical music activities for their gender group significantly more pleasant than their peers in the control group. The gender salience condition also influenced participants in the experimental condition to have more positive (or less negative) evaluations of non-stereotypical music activities for their gender group.
Limitations
This quasi-experimental study has several limitations that must be considered in interpreting the findings. First, the findings should be interpreted and generalized beyond this sample with caution. While the study site was chosen to be representative of a diverse population (including healthy proportions of various ethnicities and socio-economic levels), participants were not chosen at random and the findings may have resulted from environmental factors that were not examined. For instance, the study site was located in a part of the country that is socially progressive and in a suburb of a large metropolitan city. The findings here may be substantially different from other areas of the country, such as schools that are in more socially conservative areas. As mentioned earlier, the presence of transgender students within the student body may influence the students who participated in this study to view gender identity differently than school-aged students who do not know transgender peers, although such situations are becoming more common. Future replications of this study should consider using different types of school sites to assess the generalizability of these findings.
Another possible limitation of the study is the language used within the researcher-created Gendered Music Activity Scale. While the focus of this project was to understand interest in music activities, the prompt was modeled after a previous study (Steele & Ambady, 2006) and asked participants how ‘pleasant’ they found various activities. The semantic difference may have affected results or may have influenced some participants to only consider activities that they had participated in as being ‘pleasant’ or ‘unpleasant’, and relegated all unexperienced activities as ‘neutral’. The results may have been quite different if the prompt have suggested, ‘how interested are you in the following activities?’ Also, on this scale, there were incongruous items for consideration (i.e., ‘learning to play piano’ vs. ‘singing in a garage band’); the addition of ‘learning to’ to the item may have had unintended influence on the participants’ responses. Future replications should revisit and refine this instrument for clarity and precision.
Implications and future research
The findings of this study offer important implications for the field of music education. First, the results confirm and extend understanding of how gender influences certain music activities. Consistent with many studies that have examined gender differences in music education (for a review, see Abeles, 2009; Warzecha, 2013), boys perceive music activities that are considered feminine (e.g., singing in a choir or performing in a musical) significantly less pleasantly than girls. Furthermore, no significant difference in the level of pleasantness in masculine music activities by gender was detected. These findings empirically corroborate Harrison’s (2008) and Freer’s (2012) contentions about the role of hegemonic masculinity in the participation choices of students. Boys’ perceptions of music activities seem to be constrained by gender expectations. Girls, on the contrary, have more freedom to express interest in music activities that do not align with their gender group; girls may also be more drawn to cross-gendered music activities than boys due to the higher status of masculinity (or masculine things) within US culture.
For music educators, these findings can inform curricular decisions of class offerings. For instance, if a school is looking to enlist more music students, schools should consider providing electives in which both genders have interest (e.g., a guitar class, a class using technology to compose/arrange music). Music teachers should recognize the cultural influences that may affect their students’ participation choices and acknowledge the social liability of students who cross gender norms.
The present experiment demonstrated that adolescent attitudes toward music activities can be manipulated by subtle stimuli within the environment. Participants who had considered their gender identity rated all music activities significantly more pleasant than the control group. Based on these findings, music teachers may be tempted to use gender salience as a recruitment tool. However, while the gender salience intervention influenced ratings of pleasantness in music activities positively, it is unknown whether this influence would affect participation decisions if students had to make a forced choice. For instance, if gender salience was activated in boys, they may choose to participate in more stereotypical activities for boys in general (e.g., participating in sports) than any music activity. Thus, while some music educators may use campaigns like ‘Real Men Sing’ or ‘Girls Play Jazz’ to fight the cultural stereotypes of music participation, students may experience a gender identity threat by these cues and be pushed toward more gender-aligned activities, possibly outside of music altogether. Future replications of this experiment using a forced choice as a dependent variable or the use of a stronger, threatening intervention could help validate these findings.
This experiment provides evidence that thinking about one’s gender identity for a few moments can influence his or her perceptions of various music activities. Future research in this line of inquiry should examine what other stimuli may have similar effects. Perhaps, physical objects that are displayed within typical music classrooms (e.g., choir robes, pianos, sound equipment, trophies) send gendered messages to students and inform them whether they are welcome within the space. This phenomenon, called ambient belonging (Cheryan et al., 2009), has been shown to influence women’s interest in computer science. This line of study has shown that very subtle stimuli within the environment, possibly subliminally, can influence an individual’s interest and motivation.
The intent of this study was to empirically explore the effect of gender salience on perceptions of music activities. While this study did illustrate that male and female participants within the gender salience condition led to viewing music activities that were stereotypical for their gender as more pleasant, this intervention also influenced participants to view cross-gendered activities more positively. Music educators and researchers in music education need to continue to explore the motivations for participating in various music activities and how cultural expectations may influence participation decisions. Future studies by researchers in music education should consider the use of intervention designs to explore how specific cultural norms may influence students’ views on music participation. Harrison and O’Neill’s (2000) study, in conjunction with this study, illustrates that very simple interventions (i.e., priming agent of gender salience, concert presentation that challenges cultural norms) may affect interest in various music activities. Studies that incorporate thoughtful or prolonged interventions (e.g., units within general music classes that have all students study ‘masculine’ or ‘feminine’ instruments concurrently, frank discussions with students about cultural gender norms of music activities) may provide evidence of ways to combat cultural norms that may push students away from a musical endeavor in which they are interested. With a more complete understanding of how gender affects music participation, music educators can address and possibly minimize the influences that inhibit music participation, supporting all students to pursue any music activity they wish.
Footnotes
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
