Feminization of arts participation and extracurricular activities? Gender differences in cultural capital and bullying victimization

G Model POETIC 1262 No. of Pages 13

Poetics xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

Poetics journal homepage: www.elsevier.com/locate/poetic

Feminization of arts participation and extracurricular activities? Gender differences in cultural capital and bullying victimization Brett Lehmana,* , Susan A. Dumaisb a b

Auburn University at Montgomery, United States City University of New York, Lehman College, United States

A R T I C L E I N F O

Article history: Received 13 July 2015 Available online xxx Keywords: Cultural capital Bullying Gender Masculinity Victimization Extracurricular activities

A B S T R A C T

The study of gender differences in arts participation is extensive. Research on cultural capital often includes arts participation, but can also include participation in sports and extracurricular activities, particularly for youth in the United States. Far less research has examined the relationship between arts participation, other forms of cultural capital, and being the victim of bullying. Previous studies suggest that dominant definitions of masculinity help to explain the importance of gender in this relationship. We analyzed data from the 2002 Education Longitudinal Study to assess the extent to which male students were bullied in connection with participation in cultural activities often perceived to lack masculinity. We found that participation in clubs/academic extracurricular activities were associated with increased odds of bullying victimization for male students. On the other hand, participation in music courses was associated with increased odds of victimization for all students. Additionally, participation in varsity athletics was associated with decreased odds of victimization for all students. These findings suggest which cultural activities are stigmatized and for whom among a nationally representative sample of U.S. high school sophomores. © 2016 Elsevier B.V. All rights reserved.

1. Introduction In the United States, bullying in high schools has received a great deal of attention by researchers and the popular media. The U.S. Department of Health and Human Services sponsors a website called stopbullying.gov, indicating that bullying has become a national concern. The site defines bullying as “unwanted, aggressive behavior among school aged children that involves a real or perceived power imbalance. The behavior is repeated, or has the potential to be repeated, over time.” Male students tend to more likely be bullies and victims in physical incidents, while females participate more in relational, social, or cyber aggression (Carrera, DePalma, and Lameiras 2011; DeVoe & Bauer, 2011). Research has found an association between victimization and factors such as academic orientations at school, atypical gender behavior, race and ethnicity, and sexuality (Morris, 2012; Mueller, James, Abrutyn, and Levin 2015; Pascoe, 2007). Certain activities, such as interscholastic sports participation, have been associated with decreased risk of bullying victimization (Peguero, 2008).

* Corresponding author at: 331 Liberal Arts, P O Box 244023, Montgomery, AL 36124, United States. E-mail address: [email protected] (B. Lehman). http://dx.doi.org/10.1016/j.poetic.2016.12.003 0304-422X/© 2016 Elsevier B.V. All rights reserved.

Please cite this article in press as: B. Lehman, S.A. Dumais, Feminization of arts participation and extracurricular activities? Gender differences in cultural capital and bullying victimization, Poetics (2016), http://dx.doi.org/10.1016/j.poetic.2016.12.003

G Model POETIC 1262 No. of Pages 13

2

B. Lehman, S.A. Dumais / Poetics xxx (2016) xxx–xxx

Inequalities such as the significance conferred on masculinity relative to femininity can be reflected in the ways and reasons why students bully each other in schools. In systems of education at large, inequalities can be driven by cultural capital: the resources, tastes, and behaviors that are rewarded in schools and that are largely held by students in privileged class positions. In this manuscript, we study the potentially problematic relationship between cultural capital and bullying, asking whether participating in cultural activities that can theoretically help one's social mobility actually results in negative sanctions from other students via the enforcement of cultural codes of masculinity. Therefore, we examine the extent to which the relationship between cultural capital and bullying victimization varies by gender. Boys who participate in dance, choir, or other arts activities may be seen as less “masculine” than other male students, and research has shown that appearing less masculine can lead to negative social outcomes at school (Pascoe, 2007), including teasing and bullying victimization. In this study we examine cultural capital in two forms – arts courses taken in school, and extracurricular (EC) activities in which students participate, both school- and non-school-affiliated – and consider how cultural capital is related to bullying victimization. The focus on gender differences in the relationship between cultural capital and bullying victimization contributes to sociological knowledge about both cultural capital and bullying victimization. Focusing on bullying extends cultural capital research by considering possible negative outcomes (such as bullying victimization) related to the embodiment of cultural capital and which forms of cultural capital are associated with victimization. This will shed light on the manifestation of gender inequalities within adolescent social life in U.S. schools. 2. Background 2.1. Cultural capital and gender Cultural capital has been a popular topic in the sociology of education since it was first introduced by Bourdieu (1973). Cultural capital provides an explanation for why social class inequality tends to reproduce from generation to generation. Defined by Lamont and Lareau (1988:156) as “institutionalized, i.e., widely shared, high status cultural signals (attitudes, preferences, formal knowledge, behaviors, goods and credentials) used for social and cultural exclusion,” cultural capital serves to maintain the privileged position of the middle and upper classes while limiting the social mobility of the working class and poor. Children in middle- and upper-class families are socialized into the tastes, knowledge, and language styles (all components of what Bourdieu (1986) calls “embodied” cultural capital) beginning at a very young age and continuing throughout their childhood. The culture into which these children have been socialized is the same one that is used in and favored by schools, resulting in a seamless transition as these students enter the school system. Teachers tend to be high in cultural capital (DiMaggio [123_TD$IF]and Useem, 1978) and may consciously or unconsciously favor students who possess cultural capital, giving them better grades and/or encouraging them to pursue higher education. Although Bourdieu’s argument was that cultural capital benefits the middle and upper classes, others have shown that cultural capital may enable individuals from the lower classes to achieve upward mobility if they can gain access to cultural capital. DiMaggio (1982) found that the grades of males whose fathers were not college educated benefitted more from cultural capital than the grades of males whose fathers had higher levels of education. He concluded that “teachers may have rewarded students from lower status backgrounds who exhibited interests and behavior expected from higher status students” (1982:198). More recently, Andersen and Jaeger (2015) used data from the Programme for International Student Assessment to show that returns to cultural capital tended to be higher in low-achieving school environments, again providing support for cultural mobility, rather than cultural reproduction. In quantitative empirical research, cultural capital has most often been operationalized as arts participation, including taking arts lessons, participating in arts activities, or attending arts events. Many of these studies have found a positive association between arts participation and academic outcomes (see Jaeger, 2011 for a table summarizing results of past quantitative studies). For example, Kaufman and Gabler (2004) found that some types of arts participation, particularly music and dance training, were associated with college attendance. Empirical studies have also found non-academic benefits to cultural participation. Tramonte and Willms (2010) found that across 28 countries, cultural capital (both attendance at cultural events and conversations between parents and children about cultural matters) was significantly associated not only with reading test scores, but also a sense of belonging at school and occupational aspirations. In their study of high school students, Fredericks and Eccles (2005) found that students who participated in school band, drama, and/or art were less likely to use alcohol than nonparticipants. In her study of eighth grade students, Dumais (2002) found that females had higher rates of cultural participation and received better returns to cultural capital than males; she postulated that gender socialization may encourage boys to downplay their cultural capital, especially during a time when students are developing their sexual and gender identities. Boys sometimes feel the need to protect their identities from peers’ derogatory accusations that they are not masculine (Morris, 2008; Pascoe, 2007). This is similar to what scholars have called “laddish culture” where boys purposefully display “an interest in pastimes and subjects constructed as masculine” (Francis, 1999: 357) while avoiding anything that could associate them with femininity (see also Warrington, Younger, and Williams 2000). Within a school-based adolescent society, the importance of avoiding being stigmatized as feminine is far more immediate compared to the social class benefits that cultural capital can transfer to an individual. Thus, we limit our focus to gender inequality in the relationship between cultural capital and bullying victimization.

Please cite this article in press as: B. Lehman, S.A. Dumais, Feminization of arts participation and extracurricular activities? Gender differences in cultural capital and bullying victimization, Poetics (2016), http://dx.doi.org/10.1016/j.poetic.2016.12.003

G Model POETIC 1262 No. of Pages 13

B. Lehman, S.A. Dumais / Poetics xxx (2016) xxx–xxx

3

In summary, research on cultural capital to date has demonstrated that it is beneficial – or at least not harmful – for academic and other outcomes, and there is support that it potentially benefits students from both higher and lower socioeconomic backgrounds. Given these benefits, it is important to empirically study the relationship between cultural capital and bullying because of the potential for students to be discouraged from participating in cultural activities if they are bullied for participating. Additional research has focused on cultural capital and gender and points to potential associations between cultural capital and bullying victimization, especially when we consider cultural capital’s role in allowing those who possess it to maintain positions of power and dominance. 2.2. Cultural capital, gender, and bullying in U.S. schools Dan Olweus, who is regarded as the first social scientist to systematically study bullying, notes that bullying involves an individual targeting a weaker victim (1993). Power or lack thereof is not limited to physical strength in this case as researchers have noted the importance of social status in U.S. high schools (Faris & Felmlee, 2014; Milner, 2004). Although adolescents do bully social outcasts, ranking higher in popularity is the primary means of achieving status. The dynamics of using bullying to maintain a high position in the adolescent social hierarchy and how cultural capital may be involved are described in this section. Bourdieu and Wacquant (1992: 97) describe a field as “a network, or a configuration, of objective relations between positions.” Different forms of capital may be more or less relevant depending on the field in question. In Bourdieu’s traditional conception of cultural capital, based on his research in France in the 1960s, schools could be conceived of as fields where upper class students are dominant and are able to use their embodied cultural capital (first acquired at home) to accumulate more cultural capital. Turning to the twenty-first century United States, schools can certainly be considered fields, but the prized resources and dominant groups have likely shifted. More generally, what constitutes cultural capital has changed from the time of Bourdieu’s first studies. Prieur and Savage (2013) note that highbrow culture does not exist in the same form that it did fifty years ago. They emphasize that for something to be considered cultural capital, “it should be linked to legitimacy, convertibility, and domination, and this link has to be shown” (Prieur & Savage, 2013: 262). As Klein (2012) notes, female students as well as male students are now subject to conforming to masculine ways of being, including the rejection of intimacy, compassion, cooperation, and other “feminine” qualities. Klein describes teachers’ reports that female students are increasingly violent, aggressive, and sexual. This shift may result in less cultural participation for both male and female students, as arts activities may be too closely associated with emotion and femininity. In contrast, competitive sports may be gaining popularity among both male and female students, although many males still consider organized sports to be a masculine domain (Kimmel, 2008). Using Prieur and Savage’s (2013) description of cultural capital, sports participation can be seen as a “legitimate” high school activity. Participation can be converted into social status, and it is often the case that students involved in the popular sports tend to dominate the hierarchies of popularity compared to students involved in academic activities (Coleman, 1961; Eder & Kinney, 1995; Milner, 2004). [124_TD$IF]Serre and Wagner (2015) suggest that cultural capital can be defined as it relates to a particular social space. In the present research, the particular social space is the school setting where adolescents spend many hours competing to move up in popularity hierarchies populated by their peers.1 Students who do not participate in the prescribed activities like competitive sports could have less legitimacy to claim valued positions in a school’s hierarchy, setting up the power imbalance (Olweus, 1993) that is a necessary condition for bullying to take place. There is some research on the relationship between particular cultural activities at school and bullying, though it is not discussed by the authors within the framework of cultural capital. Fredericks and Eccles (2005) found that high school arts participants did not have more of a sense of school belonging than non-participants, whereas students who were involved in sports and school activities such as student government did feel more of a sense of belonging than non-participants. Pascoe (2007) found that the teacher and students in a high school drama class mocked students who did not conform to heterosexual norms. On the other hand, participation in the school theater performance was actually an acceptable activity for male students. Pascoe (2007) hypothesized that the “drama nerds” were already so low in the social hierarchy that it was not necessary for other students to maintain their higher status by bullying them. Although Pascoe (2007) and Klein (2012) make compelling arguments about the primacy of enforcing cultural codes of masculinity in the relationship between cultural activities and bullying victimization, they only provide qualitative accounts. To add to this literature, our research examines this relationship through nationally representative quantitative data. Other studies have examined the relationship between EC activities and bullying but have not made gender a central focus or made distinctions between different types of cultural activities. Peguero (2008:75) included “band, orchestra, chorus, or choir” and “school play or musical” in his measure of classroomrelated extracurricular activities, along with student government, honor society, and school yearbook/newspaper. He found that high school students who participated in three or more of the activities had increased risk of bullying victimization compared to students who participated in no classroom-related activities. Because the activities were aggregated, it is not possible to tell whether the arts activities (band and school play) were more highly associated with bullying victimization

1 For a theoretical discussion linking sports and the original conception of cultural capital, see LaBerge (1995). For an empirical example of a positive relationship between competitive sports participation in high school and future college attendance, see Kaufman and Gabler (2004).

Please cite this article in press as: B. Lehman, S.A. Dumais, Feminization of arts participation and extracurricular activities? Gender differences in cultural capital and bullying victimization, Poetics (2016), http://dx.doi.org/10.1016/j.poetic.2016.12.003

G Model POETIC 1262 No. of Pages 13

4

B. Lehman, S.A. Dumais / Poetics xxx (2016) xxx–xxx

than the other classroom-related activities. Peguero (2008) also did not test for different effects by gender. In a more recent study, Popp and Peguero (2011) found that participation in classroom-related activities (using the same variables as Peguero’s 2008 study) was associated with increased violent victimization, but that this association did not vary significantly by gender. Again, because of the aggregated activity measure, the effects of the specific arts activities on bullying victimization cannot be determined. Empirical research is needed in order to establish whether arts participation exacerbates, diminishes, or has no effect on the likelihood of experiencing bullying, and whether the relationship between cultural capital and bullying victimization is different for male and female students. This article contributes to these areas of need by examining the relationship between cultural capital and bullying victimization in multiple ways. First, cultural capital is measured as arts-related courses that students take in school and we distinguish between which kinds of classes the students have taken. Second, cultural capital is also measured in a broader sense (beyond Bourdieu’s definition) as participation in various EC activities that are common in the United States. In addition to examining multiple definitions of cultural capital and their relationships to bullying victimization, the final contribution of the article is to examine gender differences in these relationships among a nationally representative sample of students. This lends insight into recent definitions of masculinity as well as insight into how bullying can suggest what the dominant (or marginalized) forms of cultural capital are according to U.S. adolescents. Given our focus on the cultural significance of masculinity to U.S. adolescents, cultural activities, and EC activities, we can suggest four hypotheses to be tested through analyses of the 2002 Education Longitudinal Study data. Hypothesis 1. Given the evidence that boys are penalized for being associated with femininity and that femininity is typically associated with art, drama, and music, when male students take courses in these areas they should be more likely to report bullying victimization. Hypothesis 2. Given the same statement about femininity and cultural activities, when male students participate in cultural EC activities they should be more likely to report bullying victimization. Hypothesis 3. Given the evidence that athletic EC activities (sports) are associated with the dominant form of masculinity in U.S. high schools, when male students participate in club/academic EC activities, they should be more likely to report bullying victimization. Hypothesis 4. Given the evidence that sports are increasingly valued among students regardless of gender, all students who participate in sports should be less likely to report bullying victimization. 3. Methods 3.1. Data and sample The ELS base-year questionnaire was chosen as the data source for this study. The questionnaire was administered during students’ sophomore year in high school. Data collection took place in 2002 after the National Center for Education Statistics (NCES) selected a sample of 750 schools from across the United States. From each school, a sample of students was selected to participate. Although the scope of the ELS project included gathering information about each school and administering surveys to students, parents, and school administrators, the large majority of the data analyzed in this study consist of student responses to surveys and school characteristics. During the base-year ELS data collection, students were asked questions referring to their individual characteristics, inschool and out-of-school activities, and experiences of being bullied in various ways. Based on anonymous identification codes, students’ individual responses can be linked to the demographic and local characteristics of their schools. Those without identification codes, valid sampling weights, or those with missing data related to bullying victimization were excluded from the sample. This left us with a final sample size of 13,024 students. Missing data were replaced through multiple imputation analyses which used existing data to create prediction equations to estimate the values of all missing responses (Allison, 2001).2 Individual and school-level weights were applied in all analyses to account for the ELS stratified sampling design. 3.2. Measures of bullying victimization Bullying victimization was measured as a count of the times a student reported being bullied in various ways. These incidents included being hit, being threatened, having money or items forcefully taken away, or being “bullied.” Reponses

2 Ten datasets free of missing data were created through this process. These datasets were analyzed simultaneously to produce the average coefficients and standard errors after the analysis of each data set. These procedures produce reliable estimates of the findings that would have been reached in the absence of missing data (Allison, 2001).

Please cite this article in press as: B. Lehman, S.A. Dumais, Feminization of arts participation and extracurricular activities? Gender differences in cultural capital and bullying victimization, Poetics (2016), http://dx.doi.org/10.1016/j.poetic.2016.12.003

G Model POETIC 1262 No. of Pages 13

B. Lehman, S.A. Dumais / Poetics xxx (2016) xxx–xxx

5

that an incident never happened were coded with a 0; responses that an incident happened once or twice were coded with a 1; any time an incident was reported to happen more than twice, it was coded with a 2. Adding each item created a scale ranging from 0 to 8 with an alpha coefficient of 0.63. Although the first three items are physical in nature, the fourth item refers to bullying as defined by the victim. While the ELS data do not have information about relational bullying or other forms of bullying more subtle than physical attacks, including this broad interpretive item helps to address this issue. 3.3. Course taking and extra-curricular activities The predictor variables of greatest interest in this study include measures of students’ cultural capital. Cultural capital can take the form of participation in cultural activities as conceptualized by Bourdieu (1986) or participation in gendernormative activities that indicate an understanding and willingness to participate in cultural gender codes of the given time period (Klein, 2006; Lamont & Lareau, 1988). The variables were separated into two sets one for course-taking and another for EC activities. In the set of course-taking variables, we measured participation in cultural activities with three variables indicating how many courses (since 9th grade) students have taken in drama, the fine arts, and music.3 In the set of EC activities, students were asked if they took music, art, or language classes out of school. These are distinct from courses taken in school because although they have the potential to transmit academic and cultural skills to students, they take place away from an environment saturated with peers who could be potential in-school bullies. Students could respond that they took these classes rarely or never, less than once a week, once or twice a week, or everyday/almost every day. This variable was measured as a 1–4 Likert scale where 1 indicated the least amount of participation in the out-of-school classes. The two other measures of EC activity participation included the number of varsity athletic sports participated in and the number of in-school academic activities and clubs participated in. The kinds of sports included baseball, softball, basketball, football, soccer, individual sports, other sports, or cheerleading. The kinds of academic activities/clubs included musical or theatrical activities, student government, honors societies, school yearbook/newspaper/magazines, hobby, academic, or service clubs. Rather than duplicate the focus on cultural activities, this measure includes many activities that if participated in would not seem to fit into a hegemonic definition of masculinity. Although there were 8 possible sports and 9 possible academic/club activities, less than 1% of students reported participating in more than 5 of either kind of EC activity.4 3.4. School characteristics Although the previously mentioned cultural activities are the central focus of our study’s analysis of bullying victimization, characteristics of students’ schools modify the extent to which students are bullied as well. Thus, we controlled for these characteristics. The location of students’ schools were controlled for with two binary measures indicating that schools were in either urban or rural areas (reference group, suburban areas) as determined by the NCES Common Core of Data (CCD). The CCD also collected data related to school size measured as the number of students in grades 9–12 enrolled in the school. Our measure of school size is categorical and ranges from 1 to 9 where 1 indicates the smallest enrollments and 9 indicates the largest enrollments. As a proxy for school socioeconomic status, a categorical measure of the percent of students receiving free and reduced price lunches was controlled for. The percentile categories were 0–5, 6–10, 11–20, 21–30, 31–50, 51–75, and 76–100 as determined by the NCES. Public schools in the U.S. are rarely single-sex, though many private schools are either all-boy or all-girl schools. We controlled for school type with a binary variable where 1 indicated that the school was private. 3.5. Individual-level control variables Several binary and composite variables served to control for student characteristics. For example gender (female reference group) and race (white reference group) were controlled for. A binary variable also indicated whether or not a student identified with multiple racial/ethnic groups. Any students in this category were not included in the other racial/ ethnic categories. Disability status was measured as a binary variable indicating that a student was a member of one or more federal disability categories. ELS compiled this information from school personnel. Socioeconomic status (SES) was controlled for through a composite and weighted measure. As part of the base-year data collection, ELS surveyed students’ parents/guardians and compiled information concerning their educational attainment, income, and occupations. This information was linked to individual students to produce the composite measure of student SES. The measure was weighted to have a mean of 0. The final control variable tapped the extent to which a student misbehaved at school to account for other

3 The ELS data also contained information on courses taken in design, crafts, dance, film, and “other” classes, however for each of these categories less than 5% of the sample took any of these courses. 4 In separate analyses presented in Appendix A, course-taking and EC activities are measured with binary variables in a more conservative estimate of the effects of cultural capital and bullying victimization. We decided to present analyses with the count and Likert measurements in the text because they capture a cumulative effect of courses and activities as well as greater variation among our units of analysis. Similar measures have been used in other studies of in-school victimization (Peguero, 2008; Popp & Peguero, 2011).

Please cite this article in press as: B. Lehman, S.A. Dumais, Feminization of arts participation and extracurricular activities? Gender differences in cultural capital and bullying victimization, Poetics (2016), http://dx.doi.org/10.1016/j.poetic.2016.12.003

G Model POETIC 1262 No. of Pages 13

6

B. Lehman, S.A. Dumais / Poetics xxx (2016) xxx–xxx

Table 1 Descriptive statistics for all variables. Measurement

Mean

S.D.

Dependent Variable Bullying Incidents

0–8

0.87

1.41

Individual Male Asian Black Hispanic Multiracial SES Disability Misbehavior

[12_TD$IF]0,1 0,1 [12_TD$IF]0,1 [12_TD$IF]0,1 [12_TD$IF]0,1 2.11 to 1.82 0.1 0–6

0.49 0.10 0.15 0.16 0.05 0.02 0.08 1.17

– – – – 0.74 – 1.38

School Size Private Urban Rural % Free Lunch

1–9 [12_TD$IF]0,1 [12_TD$IF]0,1 [12_TD$IF]0,1 1–7

5.09 0.22 0.32 0.18 3.45

2.55 – – – 1.93

Cultural Capital Drama Courses Fine Arts Courses Music Courses

0–8 0–9 0–17

0.14 0.53 0.67

0.52 0.89 1.39

Out of School Classes Varsity Sports Clubs/Acad. Activities

1–4 0–8 0–9

1.65 0.95 0.99

1.07 1.30 1.30

N = 13,024.Source: ELS, NCES 2002.

forms of conflict that may be associated with being bullied. Such behaviors included getting into physical fights at school, skipping school, and otherwise getting “in trouble” at school as reported by the student. Each type of misbehavior was coded as either 0 = the event never happened, 1 = it happened once or twice, 2 = it happened more than twice. All three items were then combined into one scale ranging from 0 to 6 with an alpha coefficient of 0.84. Table 1 reports measurements, means, and standard deviations for all variables in the analysis. A gender-disaggregated comparison of the dependent and cultural capital variables is also provided in Table 2. 3.6. Preliminary analyses and multivariate analyses Before multivariate analyses could be conducted, we had to consider data structure, variation in the dependent variable across schools, and the distribution of the dependent variable. With students nested within schools, the ELS data structure is multilevel. This suggests the possibility that traditional single-level regression techniques may be inappropriate. With multilevel data, one cannot assume that regression coefficients will apply equally across various contextual-level units (Luke, 2004). In the case of the current study, the contextual-level units are the schools. The diagnostic statistic which assesses the need for multilevel regression is the Intra-Class Correlation (ICC). This tells us the amount of variation in the dependent variable that can be attributed to differences across schools (Luke, 2004; Raudenbusch & Bryk, 2002). This is particularly important to know in the current study because we are interested in variation in the dependent variables that can be associated with students’ course-taking and EC activities. The ICC tells us whether or not variation across schools may interfere with the relationship of primary interest, and thus whether multilevel modeling is necessary. In the preliminary analysis of bullying incidents, a model predicting variation in reporting bullying incidents with no independent variables was assessed. This null model indicated how much variation in the dependent variable was attributed to differences between schools. For this variable the level-2 (school level) variance was equal to 0.938. After using the equation for calculating an ICC with a non-normal distribution O’Connell (2010),5 we calculated an ICC of 0.22. In other words, 22% of the variation in bullying incidents can be attributed to differences between schools. Multilevel modeling

5 The intra-class correlation for a variable with a non-normal distribution can be found by substituting p2/3 = 3.29 for the level-1 variance component (O’Connell, 2010). Then, the calculation can be completed by dividing the sum of 3.29 and the intercept variance component by the intercept variance component (Raudenbusch [123_TD$IF]and Bryk, 2002; Luke, 2004). In our case: (0.938 + 3.29)/9.38 = 0.22

Please cite this article in press as: B. Lehman, S.A. Dumais, Feminization of arts participation and extracurricular activities? Gender differences in cultural capital and bullying victimization, Poetics (2016), http://dx.doi.org/10.1016/j.poetic.2016.12.003

G Model POETIC 1262 No. of Pages 13

B. Lehman, S.A. Dumais / Poetics xxx (2016) xxx–xxx

7

Table 2 Means Tests for Cultural Capital by Gender. Variable

Female

Male

Bullying Victimization Drama Courses Fine Arts Courses Music Courses Out of School Classes Varsity Sports Clubs/Acad. Activities

0.62 (1.14) 0.18 (0.59) 0.54 (.89) 0.90 (1.56) 1.76 (1.13) 0.91 (1.21) 1.21 (1.37)

1.03 (1.50)*** 0.11 (0.48)*** 0.53 (0.90) 0.57 (1.34)*** 1.41 (0.91)*** 1.07 (1.30)*** 0.77 (1.21)***

N

6588

6436

*p < 0.05, ** p < 0.01, *** p < 0.001 Note: standard deviations in parentheses.Source: ELS, NCES.

allowed us to account for that variation in order to have unbiased estimates of the effects of cultural capital (at the individual level) on bullying incidents reported by students. Bullying victimization was measured as a count of the number of times a student reported being bullied in various ways. This variable was skewed due to a majority of ‘0’ responses and a tail of a smaller proportion of students reporting several victimizations. This distribution violates the assumption of a normally distributed, continuous dependent variable in hierarchical linear modeling and necessitated the use of hierarchical over-dispersed Poisson modeling instead (Raudenbusch et al., 2011). The estimates for independent variables’ effect on bullying incidents are reported as log-odds coefficients and can be interpreted as the odds that a student will score high on the scale of bullying incidents net of the effects of other variables in the model.6 After assessing the ICC, the multilevel analysis proceeded in the following steps. Individual level control variables were entered into the first model followed by school-level variables in the second model. Measures of cultural capital were added in model 3, but analyses of course-taking were separated from analyses of EC activities. This strategy provided for two separate analyses of cultural capital and bullying victimization. Thus the effects of cultural capital in the form of arts-related courses could be compared to cultural capital in the form of EC activities in a parsimonious way. In either case, the fourth and final model added interaction terms measuring the effect of each respective measure of cultural capital for male students only. This strategy aligns with our interests in the extent to which boys specifically are penalized for possessing nondominant and non-masculine cultural capital. 4. Findings The means tests in Table 2 indicate that male students on average were more likely to report bullying victimization compared to female students. There were also significant differences in both types of cultural capital between male and female students. Male students on average took significantly fewer drama and music courses compared to female students while the average number of fine arts courses did not vary significantly between the groups. In terms of extracurricular cultural capital, male students had less frequent participation in out-of-school cultural classes and also participated in fewer academic/club-related EC activities on average. Female students participated in fewer varsity sports on average. These findings provide preliminary evidence of the gendering of cultural and EC activities among U.S. high school students. However, to assess the study’s hypotheses related to cultural capital, gender, and bullying victimization, multivariate analyses were required. Table 3 describes our findings in the multivariate analyses of bullying victimization. Models 1 and 2 of the multilevel Poisson model estimate the effects of individual and school-level control variables on bullying incidents. In Model 1, even after controlling for race, disability status, and behavior in school, being a male student increased one’s odds of scoring high on the scale of bullying victimization. Specifically, after converting the log-odds coefficient to an odds ratio, being male was associated with a 41% increase in the odds of scoring high on the bullying victimization scale.7 As student socioeconomic status increased, the odds of scoring high on the bullying victimization scale decreased. When school effects were added in Model 2, students in private schools had decreased odds of reporting bullying incidents relative to students in public schools while students in urban schools had decreased odds relative to students in suburban schools. In Model 3A, cultural capital was introduced in the form of arts course-taking. After controlling for individual and school characteristics, for every additional music course, the odds of scoring high on the scale of experiencing bullying incidents

6 All multilevel regression coefficients are fixed effects with robust standard errors to allow for a strict test of the hypotheses. Independent variables were centered on their group means to allow for meaningful interpretations of the coefficients. 7 All coefficients are interpreted and discussed as odds ratios in the text.

Please cite this article in press as: B. Lehman, S.A. Dumais, Feminization of arts participation and extracurricular activities? Gender differences in cultural capital and bullying victimization, Poetics (2016), http://dx.doi.org/10.1016/j.poetic.2016.12.003

G Model POETIC 1262 No. of Pages 13

8

B. Lehman, S.A. Dumais / Poetics xxx (2016) xxx–xxx

Table 3 Hierarchical Poisson Modeling of Bullying Incidents. Model 1 Individual Characteristics Male 0.344*** (0.047) Asian 0.167* (0.074) Black 0.290*** (0.081) Hispanic 0.364*** (0.073) Multiracial 0.195* (0.094) SES 0.113*** (0.032) Disability 0.372*** (0.076) Misbehavior 0.294*** (0.015) School Characteristics Size Private Urban Rural % Free Lunch

Model 2

Model 3A Courses Model 4A Courses Model 3B EC Activities Model 4B EC Activities

0.345*** (0.046) 0.116 (0.076) 0.226** (0.079) 0.309*** (0.076) 0.218* (0.095) 0.074* (0.033) 0.340*** (0.076) 0.297*** (0.015)

0.370*** (0.047) 0.103 (0.077) 0.214*** (0.079) 0.295*** (0.076) 0.225* (0.095) 0.091** (0.034) 0.349*** (0.075) 0.302*** (0.014)

0.380*** (0.062) 0.101 (0.076) 0.213** (0.079) 0.288*** (0.076) 0.230* (0.095) 0.091* (0.033) 0.349*** (0.074) 0.302*** (0.014)

0.386*** (0.049) 0.139 (0.078) 0.242** (0.075) 0.307*** (0.076) 0.210* (.098) 0.083* (.034) .326*** (.075) 0.303*** (0.015)

0.318*** (0.066) 0.138 (0.078) 0.247*** (0.074) 0.306*** (0.076) 0.207* (0.097) 0.082* (0.033) 0.330*** (0.075) 0.303*** (0.015)

0.001 (0.009) 0.165* (0.070) 0.139* (0.056) 0.054 (0.053) 0.011 (0.012)

0.002 (0.009) 0.153* (0.070) 0.144* (0.056) 0.038 (0.053) 0.010 (0.012)

0.003 (0.009) 0.155* (0.070) 0.143* (0.056) 0.038 (0.053) 0.010 (0.012)

0.001 (0.008) 0.179* (0.071) 0.149* (0.054) 0.044 (0.054) 0.012 (0.012)

.002 (.008) 0.180* (0.071) 0.152** (0.053) 0.045 (0.054) 0.012 (0.012)

0.042 (0.046) 0.018 (0.024) 0.058*** (0.014)

0.009 (0.058) 0.005 (0.039) 0.056** (0.019)

0.029 (0.019) 0.052** (0.016) 0.065*** (0.018)

0.036 (0.029) 0.045 (0.028) 0.025 (0.023)

Arts Courses # Drama # Fine Arts # Music Interaction Terms Male*Drama Male*Fine Arts Male*Music EC Activities Culture class Sports Clubs/Academic Interaction Terms Male*Culture Male*Sports Male*Clubs Variance Component 0.758***

0.103 (0.089) 0.042 (0.049) 0.003 (0.027)

0.011 (0.037) 0.012 (0.034) 0.076* (0.033) 0.768***

0.766***

0.765***

0.762***

0.760***

*p < 0.005, **p < 0.01, ***p < 0.001 Note: log odds coefficients (standard errors)Source: ELS, NCES.

increased by 6%. Model 4A assessed the effect of gender and course-taking on bullying victimization through the interactions of being male multiplied by the number of courses taken. However, none of the interaction terms related to male students’ drama, fine arts, and music course-taking were significant. The effect of taking music courses which applies to the entire sample did remain significant nonetheless. In the next set of analyses, cultural capital in the form of EC activity participation replaced the courses taken at school. This provided a second version of Model 3 (termed Model 3B above) which is identical to the previous model 3 except for the change in cultural capital variables. The results here show that for the entire sample, participating in club or academicrelated EC activities was associated with an increase in odds of scoring high on the bullying incidents scale. Model 4B added the interaction terms again. Here, the results in model 3B were explained by accounting for the interaction between being male and extra-curricular cultural capital. After accounting for male students’ club/academic EC participation, female participation was not significant. For each additional activity that male students participated in, their odds of scoring high on the bullying incidents scale increased by 8%. Moving back to the top of Table 3, adding these interaction terms also reduced the effect of being male (regardless of cultural capital) by 10%. In other words, adding the interaction between gender and cultural capital partially helped to explain the gender difference in reporting bullying incidents. 5. Discussion In this study we reviewed cultural capital’s role in social reproduction and suggested that it also may play a role in the extent to which adolescent students are victims of bullying in school. Much of this evidence revolves around studies that have pointed to how cultural, academic, and athletic activities are gendered in U.S. schools (Christin, 2012; DiMaggio, 1982; Dumais, 2002; Klein, 2006; Pascoe, 2007). Participation in cultural activities such as drama, the fine arts, and music tends to be associated with femininity while participation in athletics tends to be associated with masculinity. Though cultural capital can have long-term benefits, male students may risk being bullied in the short term if they embody the cultural capital of previous generations compared to current and dominant masculinity-oriented cultural capital. It is important to remember the numerous consequences of being bullied when interpreting these results. Some of the consequences most

Please cite this article in press as: B. Lehman, S.A. Dumais, Feminization of arts participation and extracurricular activities? Gender differences in cultural capital and bullying victimization, Poetics (2016), http://dx.doi.org/10.1016/j.poetic.2016.12.003

G Model POETIC 1262 No. of Pages 13

B. Lehman, S.A. Dumais / Poetics xxx (2016) xxx–xxx

9

pertinent to students’ educational futures include depression, increased anxiety, and lower self-esteem (Jordan & Austin, 2012) as well as decreased academic achievement (Arseneault et al., 2006; Beran, Hughes, and Lupart 2008; Nakamoto & Schwartz, 2010), classroom participation (Ladd, Herald-Brown, and Weiser 2008), and attachment to school (Popp & Peguero, 2011). Through a quantitative analysis of nationally representative data from U.S. high school sophomores, we estimated the relationship between cultural capital and bullying victimization and examined how this relationship varied by gender. In addition, we estimated the effects of two different measurements of cultural capital. Our first is related to Bourdieu’s (1986) initial conceptualization of cultural capital in the form of participation in “high-brow” activities as taken in courses including drama, fine arts, music, etc. The second consisted of EC activities which may indicate membership in dominant or nondominant groups within a school, especially for U.S. high school students at the turn of the 21st century. For males in U.S. schools, membership in competitive athletic activities tends to be a dominant position compared to membership in club or academic EC activities (Klein, 2006; Milner, 2004; Pascoe 2007). Based on this framework, we examined four hypotheses. The first suggested that if cultural activities were associated with femininity, then male students’ participation in drama, fine arts, and music courses should be associated with increased risk of bullying victimization. This hypothesis was not supported because although there was a significant and positive association between the number of music courses taken and bullying victimization, the relationship was significant for both male and female students. The second hypothesis focused on the same relationship but for participation in out-of-school cultural activities such as language, music and art. With no significant results related to this form of participation, this hypothesis was not supported either. The third hypothesis suggested that male students’ participation in club/academic EC activities should be associated with increased bullying victimization. Although it first appeared that all students were bullied in association with participation in these activities, when we examined male students’ number of activities via an interaction term, only male students’ participation was associated with increased bullying victimization. Male students’ victimization seems to be driving the relationship between club/academic EC activities and bullying victimization. This provides some support for the hypothesis. Finally, we expected that all students who participated in interscholastic athletics should be less likely to report bullying victimization in [125_TD$IF]Hypothesis 4. This proved to be the case in Model 3B before the interaction terms were introduced. Here, we found mixed results in terms of the hypothesis. Overall, the results show us that studying the relationship between cultural capital and bullying victimization provides more detail into how gendered inequalities are expressed in adolescent social contexts. Further, we have additional evidence that the dominant forms of cultural capital in U.S. schools have changed over time from high-brow activities to activities that signal masculinity. For example, although we expected male students alone to be penalized by peers for taking arts-related courses, it appears that both male and female students are more likely to be bullied when they take music courses in school. Although learning music can be a personally and academically enriching experience and although Bourdieu would have argued that refined tastes in music would provide advantages in advancing through the stratification system, U.S. students today may choose to avoid the stigma associated with music courses. Examining students’ future educational outcomes was outside the scope of the current study, but future research could examine the extent to which being stigmatized for participating results in dropping out of cultural activities. Further, because the human capital returns to cultural capital tend to be higher for boys from lower-SES backgrounds ([126_TD$IF]DiMaggio, 1982; Dumais, 2002) future research should examine the role this process plays in the reproduction of educational inequalities. Regardless of social class, male students in the United States already lag behind female students in a host of achievement and attainment metrics (Buchmann, DiPrete, and McDaniel 2008). Related to this issue, our results showed that male students’ participation in academic/club-oriented activities explains why there is a relationship between those activities and bullying victimization. Once again, serving in student government, writing for a school newspaper, working on the yearbook, etc. can all be enriching activities that supplement students’ education. The means tests showed that female students were already more likely to participate in these activities relative to male students. If male students are being stigmatized for participating, future research could determine the extent to which the avoidance of stigma and bullying victimization during adolescence contributes to the gender gap in education. When examining our final hypothesis, we found some evidence to support the claim that competitive sports is a dominant cultural activity for U.S. adolescents regardless of gender (Klein, 2006). Although male students who participate in out-of-school cultural activities and/or academic/club-related EC activities are subject to the enforcement of gender codes, competitive sports appears to be a realm where both boys and girls are generally accepted. Participating in competitive sports protects female students from bullying victimization just as it does for male students. Rather than a code of conforming to one’s gender roles, this pattern is more indicative of the privileging of masculinity in the current U.S. adolescent society among both male and female students. Klein (2012) notes that female students too are expected to engage in aggression, competition, and the putting down of others who do not embody masculine cultural capital. This shows that the study of bullying cannot necessarily be reduced to the study of who is singled out for being different. Although that process certainly does take place, bullying also involves the competition for status and legitimacy (Faris & Felmlee, 2014; Milner, 2004). Our results can serve as a reminder that students who do not participate in competitive sports and male students who gravitate towards club or academic activities are at higher risk for the many consequences of bullying victimization, even though working class boys could benefit the most from cultural enrichment. Future research should investigate what pathways male and female students take to avoid bullying and how this affects gender and social class

Please cite this article in press as: B. Lehman, S.A. Dumais, Feminization of arts participation and extracurricular activities? Gender differences in cultural capital and bullying victimization, Poetics (2016), http://dx.doi.org/10.1016/j.poetic.2016.12.003

G Model POETIC 1262 No. of Pages 13

10

B. Lehman, S.A. Dumais / Poetics xxx (2016) xxx–xxx

stratification in academic achievement and attainment. We would not have been able to identify these new questions without studying the relationship between cultural capital and bullying victimization. Despite arriving at some clarification of what kinds of cultural capital are relevant to bullying in U.S. schools and the role of gender in this area of study, there are several shortcomings of the research to be noted. Many of these shortcomings point to opportunities to conduct new research and refine the study of cultural capital and bullying. For example, although this study made use of data from high school sophomores, the relationship between measures of cultural capital and bullying victimization may vary across different age groups. Future research could address similar questions among middle school students who already tend to have higher rates of bullying victimization (Pellegrini & Bartini, 2000). Related to our measures of cultural capital, we also cannot rule out that some of the arts related courses were mandatory; such courses may not be perceived as an individual expression of masculinity or femininity if all students are required to take them. Nonetheless, one can still argue that students who took several arts related courses chose many of them as electives. Another potential area of concern is related to the use of secondary data. Although the ELS data provided us with ample measures of the courses students took and their activities outside of class, we recognize that these are still proxy measures for cultural capital. Given the findings presented here, future research could nonetheless examine students’ academic and club-related EC activities more closely. Are there particular activities that drive the increased odds of boys’ bullying victimization? Although we decided to separate curricular course-taking from extra-curricular activities, such an analysis would also address the issue that EC activities such as band and theater may overlap with arts-related courses.8 In addition, the ELS data did not survey students about their sexual orientation or provide non-cisgender identities as options to identify with. Future research should address cultural capital and bullying victimization while including sexual orientation and gender identity as additional relevant student characteristics. 6. Conclusion Despite the shortcomings of the research, the analyses of the ELS data presented here have provided the opportunity for a nationally representative analysis of the national concern of bullying victimization. Further, we have extended the literature on cultural capital by measuring it in multiple ways while considering potential problematic effects of cultural capital. Finally, we have considered the extent to which gender is relevant in modifying the relationship between cultural capital possession and bullying victimization. The gendering of cultural capital has been a frequent concern among previous researchers (DiMaggio & Mukhtar, 2004; DiMaggio, 1982; Dumais, 2002; Klein, 2006; LaBerge, 1995); we found instances where the relationship did depend on gender and other instances where the relationship did not vary regardless of gender. Our findings lend insight into cultural capital that is dominant in U.S. schools in the 21st century. In summary of the findings expected by the study’s hypotheses, male students’ participation in academic/club-related EC activities explains why there is a positive relationship between those activities and bullying victimization among U.S. students. Further, both female and male students’ participation in interscholastic sports drives the negative relationship between sports and bullying victimization. In terms of predicting the risk of bullying victimization, some cultural, academic, and club activities depend on the gender of the participant while sports does not. Overall, approaching various activities based on their association with masculinity and contemporary forms of legitimate/illegitimate cultural capital as reported by previous research (Klein, 2006; Morris, 2008; Pascoe, 2007) has been useful for predicting variation in bullying victimization. The findings that we arrived at after considering the interaction terms focused on boys’ participation are reminiscent of Kimmel’s (2008) detailed view into the lives of adolescent boys at the turn of the 20th century. He describes a desperation among males to prove one’s masculinity to each other and echoes previous accounts of how critical it has been for males to dissociate themselves from anything feminine (Francis, 1999; Morris, 2008; Pascoe, 2007). Kimmel (2008) argues that this way of life where masculinity is supreme is a largely white middle class phenomenon. We have already speculated on the risk of working-class boys missing out on culturally and academically enriching opportunities, but given Kimmel’s (2008) observations future research could examine the extent to which middle class boys avoid activities culturally defined as feminine. Although we emphasized instances where boys could have been bullied for breaking cultural codes of masculinity, analyzing instances where all participants are at risk of increased bullying provides an opportunity to further clarify the roles of cultural capital and gender in explaining why high school students bully each other. For example there appears to be a general (non-gendered) stigma against taking music courses given that the analysis of bullying victimization indicated music courses predicted increased odds of victimization for all students. Participation in varsity sports and students’ treatment of athletes does not appear to be gendered either. As sports are still associated with masculinity we can interpret this finding as an example of the increased value of masculinity for all students regardless of gender (Klein, 2012). Performing indicators of masculinity has value for achieving social status and protecting oneself from being bullied or pushed further down in the social status hierarchy. Although we might traditionally associate competition and violence with boys in the U.S., Klein (2012) makes it clear that girls are increasingly engaging in this behavior too.

8 In separate analyses which included all six measures of cultural capital, we checked for multicollinearity to ensure any qualitative overlap in the variables was not accompanied by statistical overlap that would bias our findings. All variance inflation factors were < 1.25.

Please cite this article in press as: B. Lehman, S.A. Dumais, Feminization of arts participation and extracurricular activities? Gender differences in cultural capital and bullying victimization, Poetics (2016), http://dx.doi.org/10.1016/j.poetic.2016.12.003

G Model POETIC 1262 No. of Pages 13

B. Lehman, S.A. Dumais / Poetics xxx (2016) xxx–xxx

11

This makes further sense after considering the passage of Title IX which gave female students greater access to interscholastic sports participation. In 2002 when the ELS data were collected, the effects of Title IX may have carved out a space for female participation to be respected. In contrast to athletics, the average musically inclined high school student is not treated well by his or her peers. Taking music classes does not appear to be a dominant form of gaining status among female or male students. The same can be said for male students who participated in academic/club EC activities. Although extra exposure and training in these areas of study are culturally and academically enriching, the participants still face the backlash of their peers that enforce dominant cultural codes. This is another piece of evidence to consider in the future as educators simultaneously aim to provide more access to such activities as well as reduce bullying in U.S. schools. Drawing from the findings in the relationship between cultural capital, gender, and bullying can provide insight into reducing bullying. For example, although it used to be abnormal for female students to participate in athletics (Eder & Kinney, 1995; Messner, 1988), the findings here suggest that granting more access can normalize the participation of a group in an activity over time. The same effort could be applied to music, foreign languages, academic activities, clubs, etc. Given the many social, psychological, and academic benefits of EC activity participation (Feldman & Matjasko, 2005), this is already a goal of many schools across the country. Schools could also couple such efforts with anti-bullying programs and programs designed to include underprivileged students in cultural activities. Teachers, parents, administrators, etc. can promote the message that participation in the arts is valuable and socially acceptable regardless of gender. They can also seek to elevate the status of students showing promise and success in the arts. In other words, community-wide praise and acclaim should not be reserved for athletes alone. Now that we have learned more about cultural capital and its potential negative effects, we can use this information to not only promote academic enrichment but also to make schools safer and more inclusive. [127_TD$IF]Appendix A. Hierarchical Poisson Modeling of Bullying Incidents when Cultural Capital is Measured with Binary Variables

Individual Characteristics Male Asian Black Hispanic Multiracial SES Disability Misbehavior School Characteristics Size Private Urban Rural % Free Lunch

Model 1

Model 2

0.344*** (0.047) 0.167* (0.074) 0.290*** (0.081) 0.364*** (0.073) 0.195* (0.094) 0.113*** (0.032) 0.372*** (0.076) 0.294*** (0.015)

0.345*** (0.046) 0.383*** (0.047) 0.116 (0.076) 0.102 (0.077) 0.226** (0.079) 0.206** (0.078) .275*** (.076)

Model 4A Courses

Model 3 B EC Activities Model 4 B EC Activities

0.419*** (0.069) 0.101 (0.076) 0.205** (0.078)

0.394*** (0.048) 0.144 (0.077) 0.230** (0.078)

0.271*** (0.076)

0.296*** (0.077)

0.336*** (0.082) 0.145 (0.077) 0.231** (0.078)

0.309*** (0.076) 0.218* (0.095) 0.074* (.033) .340*** (.076) 0.297*** (0.015)

0.234* (0.095) 0.094** (0.034) .356*** (.074) 0.302*** (0.014)

0.236* (0.095) 0.094** (0.034) 0.353*** (0.074) 0.302*** (0.014)

0.216* (0.098) 0.087* (0.034) 0.318*** (0.075) 0.304*** (0.016)

0.214* (0.098) .085* (.034) 0.318*** (0.075) 0.303*** (0.015)

0.001 (0.009) .165* (0.070) 0.139* (0.056) 0.054 (0.053) 0.011 (0.012)

0.001 (0.011) 0.165* (0.074) 0.145** (0.056) 0.042 (0.056) 0.021 (0.014)

0.001 (0.011) 0.166* (0.074) 0.144** (0.056) 0.043 (0.056) 0.020 (0.014)

0.002 (0.011) 0.184* (0.074) 0.141* (0.056) 0.054 (0.056) 0.019 (0.014)

0.003 (0.011) 0.186* (0.074) 0.141* (0.056) 0.055 (0.055) 0.020 (0.014)

0.064 (0.080) 0.002 (0.045) 0.230*** (0.049)

0.016 (0.109) 0.034 (0.070) 0.258*** (0.068)

0.113* (0.050) 0.131** (0.044) 0.162*** (0.046)

0.082 (0.070) 0.141* (0.068) 0.133 (0.074)

Arts Courses # Drama # Fine Arts # Music Interaction Terms Male*Drama Male*Fine Arts Male*Music EC Activities Culture class Sports Clubs/Academic Interaction Terms Male*Culture Male*Sports Male*Clubs Variance Component

Model 3A Courses

0.298*** (0.078)

0.116 (0.160) 0.065 (0.090) 0.056 (0.096)

0.064 (0.100) 0.021 (0.088) 0.047 (0.092) 0.758***

0.768***

0.762***

0.762***

0.763***

0.763***

* p <0.05, ** p <0.01, *** p < 0.001. Note: log odds coefficients (standard errors). Source: ELS, NCES.

Please cite this article in press as: B. Lehman, S.A. Dumais, Feminization of arts participation and extracurricular activities? Gender differences in cultural capital and bullying victimization, Poetics (2016), http://dx.doi.org/10.1016/j.poetic.2016.12.003

G Model POETIC 1262 No. of Pages 13

12

B. Lehman, S.A. Dumais / Poetics xxx (2016) xxx–xxx

When cultural capital is measured with binary variables, the effect of participating in competitive athletics reduces the odds of reporting bullying victimization for both male and female students. The finding in our analysis in the main text about male students’ club/academic EC activities does not remain when using the recoded measures of cultural capital here. There is no difference in the relationships between athletics, clubs, or course-taking and bullying victimization for female students compared to males using the binary variables. We argue that using the measures of the raw number of sports or club activities in the main text is preferable because participation in multiple activities makes it more likely for peers to notice this participation. This cumulative effect is lost when the variation in participation is collapsed into a binary variable. References Allison, P. D. (2001). Missing data. Thousand Oaks, CA: Sage. Andersen, I. G., & Jaeger, M. M. (2015). Cultural capital in context: Heterogeneous returns to cultural capital across schooling environments. Social Science Research, 50, 177–188. Arseneault, L., Walsh, E., Trzesniewski, K., Newcobe, R., Caspi, A., & Moffitt, T. E. (2006). Bullying victimization uniquely contributes to adjustment problems in young children: A nationally representative cohort study. Pediatrics, 118, 130–138. Beran, T. N., Hughes, G., & Lupart, J. (2008). A model of achievement and bullying: Analyses of the Canadian National Longitudinal survey of children and youth data. Educational Research, 50, 25–39. Bourdieu, P., & Wacquant, L. (1992). An invitation to reflexive sociology. Chicago: University of Chicago Press. Bourdieu, P. (1973). Cultural reproduction and social reproduction. knowledge, education, and cultural change. In Richard Brown (Ed.), Papers in the sociology of education (pp. 71–112).London: Tavistock. Bourdieu, P. (1986). The forms of capital. In J. Richardson (Ed.), Handbook of theory and research for the sociology of education (pp. 241–258).New York: Greenwood. Buchmann, C., DiPrete, T. A., & McDaniel, A. (2008). Gender inequalities in education. Annual Review of Sociology, 34, 319–337. Carrera, M. V., DePalma, R., & Lameiras, M. (2011). Toward a more comprehensive understanding of bullying in school settings. Educational Psychology Review, 23, 479–499. Christin, A. (2012). Gender and highbrow cultural participation in the United States. Poetics, 40, 423–443. Coleman, J. (1961). The adolescent society: The social life of the teenager and its impact on education. New York: The Free Press. DeVoe, J. F., & Bauer, L. (2011). Student victimization in U.S. schools: Results from the 2009 school crime supplement to the national crime victimization survey (NCES 2012-314). U.S. Department of Education, National Center for Education Statistics. Washington, DC: U.S. Government Printing Office. DiMaggio, P., & Mukhtar, T. (2004). Arts participation as cultural capital in the United States, 1982–2002: signs of decline? Poetics, 32, 169–194. DiMaggio, P., & Useem, M. (1978). Social class and arts consumption: The origins and consequences of class differences in exposure to the arts in america. Theory and Society, 5, 141–161. DiMaggio, P. (1982). Cultural capital and school success: The impact of status culture participation on the grades of U.S. high school students. American Sociological Review, 47, 189–201. Dumais, S. A. (2002). Cultural capital, gender, and school success: The role of habitus. Sociology of Education, 75, 44–68. Eder, D., & Kinney, D. A. (1995). The effect of middle school extracurricular activities on adolescents’ popularity and peer status. Youth and Society, 26, 298– 324. Faris, R., & Felmlee, D. (2014). Casualties of social combat: School networks of peer victimization and their consequences. American Sociological Review, 79, 228–257. Feldman, A. F., & Matjasko, J. L. (2005). The role of school-based extracurricular activities in adolescent development: A comprehensive review. Review of Educational Research, 75, 159–210. Francis, B. (1999). Lads, lasses, and (new) labour: 14–16-year-old students’ responses to the ‘Laddish behaviour and boys’ underachievement' debate. British Journal of the Sociology of Education, 20, 355–371. Fredericks, J. A., & Eccles, J. S. (2005). Developmental benefits of extracurricular involvement: Do peer characteristics mediate the link between activities and youth outcomes? Journal of Youth and Adolescence, 34, 507–520. Jaeger, M. M. (2011). Does cultural capital really affect academic achievement? New evidence from combined sibling and panel data. Sociology of Education, 84, 281–298. Jordan, K., & Austin, J. (2012). A review of the literature in bullying in U. S. schools and how a parent-educator partnership can be an effective way to handle bullying. Journal of Aggression, Maltreatment, and Trauma, 21, 440–458. Kaufman, J., & Gabler, J. (2004). Cultural capital and the extracurricular activities of girls and boys in the college attainment process. Poetics, 32, 145–168. Kimmel, M. (2008). Guyland. New York: HarperCollins. Klein, J. (2006). Cultural capital and high school bullies: How social inequality impacts school violence. Men and Masculinities, 9, 53–75. Klein, J. (2012). The bully society. New York: New York University Press. LaBerge, S. (1995). Toward an integration of gender into bourdieu’s concept of cultural capital. Sociology of Sport Journal, 12, 132–146. Ladd, G. W., Herald-Brown, S. L., & Reiser, M. (2008). Does chronic classroom peer rejection predict the development of children’s classroom participation during the grade school years? Child Development, 79, 1001–1015. Lamont, M., & Lareau, A. (1988). Cultural capital: Allusions, gaps, and glissandos in recent theoretical developments. Sociological Theory, 6, 153–168. Luke, D. A. (2004). Multilevel modeling. Thousand Oaks, CA: Sage. Messner, M. A. (1988). Sports and male domination: The female athlete as contested ideological terrain. Sociology of Sports Journal, 5, 197–211. Milner, M. Jr. (2004). Freaks, geeks, and cool kids: American teenagers, schools, and the culture of consumption. New York: Routledge. Morris, E. W. (2008). ’Rednecks, ‘Rutters’, and ‘Rithmetic’: social class, masculinity, and schooling in a rural context. Gender and Society, 22, 728–751. Morris, E. W. (2012). Learning the hard way: Masculinity, place, and the gender gap in education. New Brunswick, NJ: Rutgers University Press. Mueller, A. S., Wesley, J., Seth, A., & Martin, L. L. (2015). Suicide ideation and bullying among U. S. adolescents: Examining the intersections of sexual orientation, gender, and Race/Ethnicity. American Journal of Public Health, 105, 980–985. Nakamoto, J., & Schwartz, D. (2010). Is peer victimization associated with academic achievement? a meta-analytic review. Social Development, 19, 221–242. O’Connell, A. A. (2010). An illustration of multilevel models for ordinal response data. In C. Reading (Ed.), Data and context in statistics education: Towards an evidence-Based societyVoorburg, The Netherlands: International Statistics Institute. Olweus, D. (1993). Bullying at school: What we know and what we can do. Oxford: Cambridge University Press. Pascoe, C. J. (2007). Dude, you’re a fag: Masculinity and sexuality in high school. Berkeley: University of California Press. Peguero, A. A. (2008). Bullying victimization and extracurricular activity. Journal of School Violence, 7, 71–85. Pellegrini, A. D., & Bartini, M. (2000). A longitudinal study of bullying, victimization, and peer affiliation during the transition from primary school to middle school. American Educational Research Journal, 37, 699–725. Popp, A. M., & Peguero, A. A. (2011). Routine activities and victimization at school: The significance of gender. Journal of Interpersonal Violence, 26, 2413–2436. Prieur, A., & Savage, M. (2013). Emerging forms of cultural capital. European Societies, 15, 246–267. Raudenbusch, S. W., & Bryk, A. S. (2002). Hierarchical linear models: Applications and data analysis methods, 2nd ed. Thousand Oaks, CA: Sage. Raudenbusch, S. W., Bryk, A. S., & Congdon, R. T. (2011). Hierarchical linear and nonlinear modeling. Lincolnwood, IL: Scientific Software International.

Please cite this article in press as: B. Lehman, S.A. Dumais, Feminization of arts participation and extracurricular activities? Gender differences in cultural capital and bullying victimization, Poetics (2016), http://dx.doi.org/10.1016/j.poetic.2016.12.003

G Model POETIC 1262 No. of Pages 13

B. Lehman, S.A. Dumais / Poetics xxx (2016) xxx–xxx

13

Serre, D., & Wagner, A. C. (2015). For a relational approach to cultural capital: A concept tested by changes in the French social space. The Sociological Review, 63, 433–450. Tramonte, L., & Willms, J. D. (2010). Cultural capital and its effects on education outcomes. Economics of Education Review, 29, 200–213. Warrington, M., Younger, M., & Williams, J. (2000). Student attitudes, image, and the gender gap. British Educational Research Journal, 26, 393–407. Brett Lehman is an Assistant Professor of Sociology at Auburn University at Montgomery. His research interests include bullying, race/class/gender inequality, and social psychology. He has taught courses in Statistics, Social Stratification, and Sociology of Education.

Susan A. Dumais is an Associate Professor of Sociology at CUNY-Lehman College. Her research is in the areas of education, inequality, and culture. She currently teaches courses in Education and Society, Education in Global and Cross-National Perspectives, and Research Methods.

Please cite this article in press as: B. Lehman, S.A. Dumais, Feminization of arts participation and extracurricular activities? Gender differences in cultural capital and bullying victimization, Poetics (2016), http://dx.doi.org/10.1016/j.poetic.2016.12.003