Theater and Dialogue to Increase Youth’s Intentions to Advocate for LGBTQQ People

Abstract

Objective: This study evaluates the effectiveness of an intervention using theater and dialogue to raise awareness about homophobia and transphobia and increase intentions to participate in macro-level change efforts around lesbian, gay, bisexual, transgender, queer, and questioning (LGBTQQ) issues. Methods: Using a pretest–posttest design, this study examines changes in advocacy intentions around LGBTQQ issues among middle school and high school students who participated in the intervention (n = 515). Results: Students reported a significant increase in intentions to advocate. Existing intervention behaviors and increased recognition of homophobia/transphobia as problems were associated with greater increases in intention to advocate. Respondents with higher existing intervention behaviors, as well as White students and cisgender women, reported higher advocacy intentions compared to students of color and cisgender men. Further, we found indication of greater awareness of homophobia/transphobia as a problem in school. Conclusions: Youth-led theater and dialogue-based interventions may be a promising strategy for addressing heterosexism and genderism in schools.

Keywords

lesbian gay bisexual transgender queer school climate advocacy arts dialogue

Systems of oppression, including heterosexism and genderism (i.e., structural, institutional, and interpersonal practices that marginalize individuals and communities based on sexual orientation and transgender identities, respectively), create a hostile school environment for lesbian, gay, bisexual, transgender, queer, and questioning (LGBTQQ) youth (Almeida, Johnson, Corliss, Molnar, & Azrael, 2009; Bontempo & D’Augelli, 2002; Dupper & Meyer-Adams, 2002; Graham, 2012; Graham, Braithwaite, Spikes, Stephens, & Edu, 2009; MacGillivray, 2004; Poteat & Espelage, 2007) and encourage strict gender and sexual norms that can negatively impact straight and cisgender students (Kimmel, 2001; Klein, 2006; Swearer, Turner, Givens, & Pollack, 2008). Cisgender refers to individuals who identify with the gender normatively associated with their assigned sex at birth (e.g., individuals who identify as women/girls who were assigned female sex at birth). Heterosexism and genderism are often manifested through both overt and subtle violence (verbal, physical, and sexual) against those who identify as or are perceived to be LGBTQQ (Centers for Disease Control & Prevention, 2011; Harris Interactive & GLSEN, 2005; Kosciw & Diaz, 2006; Nadal & Griffin, 2011; Silverschanz, Cortina, Konik, & Magley, 2008; Walls, Freedenthal, & Wisneski, 2008) as well as exclusion from school curricula, a lack of culturally sensitive resources and support from peers and adults, and other cultural norms that devalue lesbian, gay, bisexual, queer, transgender, and other similar identities (Currie, Mayberry, & Chenneville, 2012; Davis, Saltzburg, & Locke, 2009; Graham et al., 2014; McGuire, Anderson, Toomey, & Russell, 2010; Riot Youth, 2009; Sadowski, Chow, & Scanlon, 2009). Many communities continue to struggle with how to effectively transform school environments to better serve the academic, developmental, and social needs of all students, including LGBTQQ students (Ciardullo, 2005; MacGillivray, 2004; Wernick, Dessel, Kulick, & Graham, 2013).

Given the systemic nature of heterosexism and genderism, interventions aimed at institutional change are necessary in order to address these issues on a macro level. In addition, student engagement and youth-led interventions might engender the most effective approaches to addressing heterosexism and genderism (Christens & Dolan, 2011; Wernick, Dessel, et al., 2013; Wernick, Kulick, & Inglehart, 2013). While previous research has addressed ways to decrease homophobic and transphobic attitudes among students (Fuoss, Kistenberg, & Rosenfeld, 1992), decrease bullying behaviors (Graybill, Varjas, Meyers, & Watson, 2009; Toomey, Ryan, Diaz, & Russell, 2011), and increase anti-bullying intervention behaviors (Wernick, Dessel, et al., 2013), no research known to the authors has investigated how to increase student intentions to participate in and lead macro-level change efforts around LGBTQQ issues. Applied intervention research is needed in order to link knowledge to evidence-based practice with regard to macro-level changes addressing prejudice and negative school climate for LGBTQQ youth (Fraser, 2004; Fraser & Galinsky, 2010; Jenson, Dieterich, Brisson, Bender, & Powell, 2010; Kosciw, Palmer, Kull, & Greytak, 2013). The present study investigates an innovative intervention developed by LGBTQQ youth that uses theater and dialogue methods to raise critical awareness and advocacy intentions around LGBTQQ issues among middle school and high school students.

School Systems and Gay–Straight Alliances

LGBTQQ and heterosexual/cisgender youth actively shape their school environment (Wernick, Woodford, & Siden, 2010), including developing relationships across difference (Poteat, Espelage, & Koenig, 2009), which can encourage positive attitudes toward LGBTQQ individuals (Herek & Capitanio, 1996; Vonofakou, Hewstone, & Voci, 2007). However, heterosexism and genderism are pervasive in shaping the context in which youth can take these actions and form relationships (Poteat et al., 2009). In order to effectively support youth around issues of gender and sexuality and prevent bullying and violence, intervention research is critical (Jenson et al., 2010).

Many schools have been resistant to supporting LGBTQQ youth. This resistance is often manifested organizationally, in school districts (Rienzo, Button, Sheu, & Li, 2009), and through the actions of individual administrators and adults in schools (McCabe & Rubinson, 2008). Whether due to willful ignorance, indifference, or a lack of institutional support, teachers often do not engage critical pedagogy that can promote learning around power, inequality, and social justice (Mayo, 2013). Likely as a result of the lack of support and initiative on behalf of administrators and other power holders in schools, LGBTQQ youth have led change efforts in schools to create safe spaces, increase awareness about LGBTQQ issues, and publicly claim their rights as well as their gender and sexual identities (Cohen, 2005; Griffin, Lee, Waugh, & Beyer, 2004; Schindel, 2008; Walls, Wisneski, & Kane, 2013; Wernick, Kulick, & Woodford, 2014).

LGBTQQ-youth-led efforts in schools are often evident through gay–straight alliances (GSAs), and the extant research on school-based advocacy by LGBTQQ youth has focused almost entirely on these types of organizations. The capacity of GSAs to take action is often circumscribed by the attitudes and actions of administrators in schools; however, the tenacity and activism of youth and adults in schools as well as support through legal mandates and access to national networks (e.g., the Gay-Lesbian Straight Education Network) have helped sustain the activities of these types of organizations—even in hostile environments (Fetner & Kush, 2008; Mercier, 2010). Both participation in and the existence of a GSA or similar group in one’s school have been linked to positive outcomes for LGBTQQ youth (Goodenow, Szalacha, & Westheimer, 2006; Lee, 2002; Walls, Kane, & Wisneski, 2010). These organizations also promote school-based advocacy, including raising the visibility of LGBTQQ students and issues they face, using curricula and educational efforts, and developing social justice values among students (Goldstein & Davis, 2010; Graybill et al., 2009; Jeltova & Fish, 2005). These proactive approaches can be more productive than punitive strategies such as traditional forms of discipline (Graybill et al., 2009). However, greater research is needed to examine the efforts of these types of organizations to encourage systemic change (Currie et al., 2012) and to develop empirically tested interventions that can be replicated (Griffin et al., 2004; Riot Youth, 2009, 2013).

Awareness and Ally Development

The development of heterosexual and cisgender allies is a key component to addressing LGBTQQ bullying, harassment, and violence in schools (Goldstein & Davis, 2010; Mayo, 2013). As well, given the diversity of identities and experiences among LGBTQQ people, allyhood is a useful lens for understanding the myriad ways that LGBTQQ youth can support one another (Wernick, Dessel, et al., 2013). In order to engage in advocacy against sexual prejudice, bullying, and discrimination, youth must first recognize that problems exist, develop the motivations to address these problems, and have the resources and skills to take ongoing actions. Building awareness of negative climate for LGBTQQ youth is an important first step in the development of advocacy intentions. In particular, heterosexual/cisgender youth are often unaware of the experiences of gender and sexually marginalized youth or play a bystander role by overlooking incidents of bias and therefore actively contribute to a hostile climate (Mayo, 2013; Stueve et al., 2006).

In addition to recognizing the negative impacts of heterosexism and genderism on LGBTQQ students, recognition of one’s own heterosexual and/or cisgender privilege and heterosexual identity development are key areas of learning for LGBTQQ allies (Duhigg, Rostosky, Gray, & Wimsatt, 2010; Evans & Broido, 2005; Jordan, 2012; Mio, 2007; Worthington, Savoy, Dillon, & Vernaglia, 2002). Critical awareness linked to allyhood identification can be developed through early positive socialization and having positive social relationships with LGBTQQ-identified individuals (Broido, 2000; Dillon et al., 2004; Liang & Alimo, 2005; Pettigrew & Tropp, 2000; Roades & Mio, 2000). Straight/cisgender-identified students of color, who may also be questioning their sexual identity, can particularly benefit from role models in organizations such as GSAs (Mayo, 2013). By developing positive associations with LGBTQQ people, heterosexual and cisgender young people can come to recognize the injustice in inequality faced by these communities (Edwards, 2006).

The awareness developed by potential allies is directly linked to their motivation to engage in ally practices, which can range from seeing themselves as an ally for other individuals, an ally for communities, to an ally to support social justice efforts that address macro-level social issues (Edwards, 2006). These motivations are then translated to action, which can range from making individual-level changes to participating in structural-level changes (Mayberry, 2012). Further, allies can benefit from ongoing development and refining of their skills to take action through an iterative process that involves: intentional opportunities to practice taking action (Broido & Reason, 2005); participating in spaces that allow them to reflect on how and why they take action (Duhigg et al., 2010; Jordan, 2012); and, seeing others take action (Wernick, Kulick, et al., 2013). Engaging in these types of practices might also help assuage ongoing issues that allies face, including fear of stigma by association with LGBTQQ students (Goldstein & Davis, 2010).

A wide breadth of research has documented how theater and dialogue can provide opportunities for youth to develop critical consciousness, reflect on their identities, and practice taking action (Bohm, 1996; Dessel, Rogge, & Garlington, 2006; Dessel, Woodford, Routenberg, & Breijack, 2013; Finn, Jacobson, & Campana, 2004; Saltzburg, 2008; Wernick et al., 2014; Wernick, Woodford, & Kulick, 2014). For instance, previous intervention research has highlighted that theater can be a creative and effective method for social change (Black, Weisz, Coats, & Patterson, 2000). Our previous research has investigated how these strategies can be effectively translated for use among LGBTQQ communities. We have found that theater and dialogue are effective tools for developing empowerment among LGBTQQ students and preparing them to take action at multiple levels (Dessel, Woodford, & Warren, 2011; Wernick, et al., 2014). Further, in examining the impact on the school system, we have found that LGBTQQ youth-led theater is effective in motivating adults to take action at multiple levels, including policy, organizational change, and individual action (Wernick, et al., 2014), and these theater efforts can develop students’ intentions and confidence to take individual-level action when witnessing anti-LGBTQQ bullying and harassment (Wernick, Dessel, et al., 2013).

The literature on awareness and allyhood development has focused on theoretical accounts and qualitative studies and usually among college students or adults. Further, these studies often involved retrospective accounts of those who have already become allies. Greater research is needed to establish empirically supported interventions that develop new allies (Wernick, Dessel, et al., 2013), particularly allies who will support macro-level change. The development and systematic study of purposive social change strategies are critical areas of social work research (Fraser, 2004; Fraser & Galinsky, 2010).

Research Aim

In the present study, we investigate if participation in an intervention designed and led by LGBTQQ youth using theater and dialogue is related to increases in participants’ intentions to advocate in support of LGBTQQ communities. In this case, participants include straight/cisgender and LGBTQQ youth; we include all respondents in the present study, given the ways that individuals might further develop advocacy intentions for intragroup advocacy (Wernick, Dessel, et al., 2013). In addition, we explore how awareness of homophobia and transphobia is constructed by youth and the relationship between participation in the intervention, homophobia and transphobia awareness, and LGBTQQ advocacy intentions.

Method

Intervention

Riot Youth is an LGBTQQ and allied (LGBTQQA) youth organization that focuses on providing safe space for and empowering LGBTQQA youth. Riot Youth uses a transformative organizing model, incorporating theater, participatory action research, and dialogic tools into their change-making strategies. The present study focuses on their use of performance and post-performance dialogues (Gayrilla Theater) to cultivate social justice activism intentions among other students. Other impacts of these theater interventions have been reported elsewhere (Wernick, Dessel, et al., 2013; Wernick, et al., 2014) as well as other impacts of their organizing model (Wernick, et al., 2014; Wernick et al., 2010). All aspects of the study were conducted in partnership with Riot Youth participants and adult advisors, and the University of Michigan Institutional Review Board approved the study as exempt from oversight. Data from the present study were collected by the organization and shared with the research team as de-identified secondary data.

Gayrilla Theater performances were developed by Riot Youth participants as a means to disseminate findings from their participatory action research project and catalyze multilevel change to support inclusion and safety for marginalized youth (Riot Youth, 2009, 2013). Scripted performances (∼35–40 minutes) were developed using personal experiences from youth participants, survey findings, and recommendations for change. The youth, in consultation with adult advisors, developed the script specifically to target the developmental needs and social location of peer audience members. Scenes used both seriousness and humor to depict experiences of heterosexism and genderism through a combination of storytelling interspersed with: quantitative findings; school scenes; internal monologues; and, depictions of ways to intervene when witnessing homo- and transphobia across various settings, particularly with friends/acquaintances. As well, for the performances for other students, Riot Youth developed a “common ground” activity and a curriculum for post-performance dialogues (∼35–40 min) to directly engage their peers in storytelling and conversations around issues related to identity, LGBTQQ communities, and bullying and harassment (Fisher & Checkoway, 2011; Nagda, McCoy, & Barrett, 2006). Common ground statements included whether they have seen or been a part of any of the scenarios depicted in the performance and how they did or did not react and why. Small group dialogues included questions related to reactions to the performances, connections between content presented in the performance and real-life experiences, connections with other social identities, and how they could challenge homo- and transphobia in their schools. These dialogues were conducted using a curriculum, including pre-written discussion questions, and co-facilitated by trained youth leaders and adult advisors and mentors. Further information about the development of these performances has been described in our earlier work (Wernick, Dessel, et al., 2013; Wernick, et al., 2014); as well, the curriculum has been published by the organization (Riot Youth, 2013).

In order to address the research aim of the present study, the research team employed a preexperimental, one-group, pretest–posttest design, supplemented by qualitative analysis of write-in responses. Both quantitative and write-in data were collected concurrently in the pretest and posttest survey instruments.

Participants

Participants were high school and middle school students who participated in one of eight intervention performances and post-performance dialogues conducted in five schools in Washtenaw County, MI (one middle school and four high schools). In the middle school and two of the high schools, participation was mandated by school officials. In the remaining two high schools, participation was optional. In one of these schools, students were mandated to attend a certain number of assemblies each school year, and Gayrilla was one option that could fulfill this requirement. In the other school, students were recruited through student clubs, leadership programming, and word of mouth. All students of each school were eligible for the study. Only those who participated in both a Gayrilla performance and dialogue were included in the final analysis. Each of the eight performances ranged in size from 40 to 250 participants, and dialogues were conducted in smaller groups (7–18 participants each in high school groups, 28–33 in middle school groups). Measures were taken to ensure confidentiality (e.g., posttests were collected in a sealed envelope), and all participants were notified that their completion of the survey (including individual questions) was voluntary.

Procedure

Audience members were provided with numerically coded and anonymously matched pretest and posttest surveys. Participants were asked to complete and return the pretest survey as they entered the performance space, prior to the start of the intervention. Posttests were completed and returned immediately following the post-performance dialogues. Surveys were one page and designed to take 5–10 min to complete in order to fit into school assembly time constrains.

Measures

Dependent variable: Intent to advocate

To measure changes in participant’s self-reported intent to join efforts to advocate for LGBTQQ people in the pretest and posttest surveys, respondents were asked the paired item: “I would join efforts to advocate for social justice, fairness and equality for sexual minorities” (1 = disagree completely and 5 = agree completely). This item was modified from the psychometrically validated Activism Intensions Scale (AIS; Moskalenko & McCauley, 2009), which assesses the intent to participate in nonviolent and legal political action. The full scale has shown predictive validity for activism actions (Moskalenko & McCauley, 2009). Given the time constraints placed on the time to complete evaluations, a single item was used. To ensure face validity, this item was reviewed (along with the rest of the survey instrument) by the research team as well as by community partners (Riot Youth teens and adult staff members). The item used for this study was adapted to specifically inquire about respondents’ intentions to participate in LGBTQQ-related activism.

Independent pretest control variables

Two measures were included on the pretest survey as controls: frequency of witnessing anti-LGBTQQ harassment and self-reported anti-bullying intervention behaviors. The first item assessed the self-reported frequency of students seeing other students engage in language or actions that may be perceived as anti-LGBTQQ harassment: “How often do you see/hear other students use language (such as “that’s so gay,” “no homo,” or “faggot”) or take actions that may be offensive toward lesbian, gay, bisexual, transgender, or questioning (LGBTQ) students?” The second item measured students’ self-reported frequency of existing intervention behaviors when witnessing perceived anti-LGBTQQ harassment or bullying prior to taking part in the performance and dialogue intervention: “How often do you intervene when students use language or take actions that may be offensive toward LGBTQ students?” (both questions: 1 = never and 5 = frequently). Both measures were adapted from items used previously by the organization (Riot Youth, 2009; Wernick, Kulick, et al., 2013) and from national LGBTQQ surveys (Kosciw, Greytak, Diaz, & Bartkiewicz, 2010), combining gender identity and sexual orientation issues into single items. These items were included in the multivariate analyses to control for prior exposure to perceived anti-LGBTQQ harassment and existing allyhood capacity on the individual level, given the impact these might have on students’ intent to advocate for social justice, fairness, and equality of LGBTQQ people (Wernick, Dessel, et al., 2013; Wernick, Kulick, et al., 2013).

Repeated write-in responses

To gain a more nuanced understanding of the impact of the performance and dialogue intervention on participants’ intent to advocate and supplement quantitative items, we examined paired write-in items that assessed ways in which respondents understood perceived homophobia and transphobia as a problem at their school: “Do you think that homophobia/transphobia is a problem at your school?” This question followed the item about the frequency of witnessing anti-LGBTQQ harassment. This item was included to better understand respondents’ awareness of the problems that exist in their schools that would be addressed by LGBTQQ-related activism. The wording of the item invited a brief response, and many respondents wrote in simple responses (e.g., “yes” or “kind of”); however, a considerable number of respondents chose to elaborate, providing rich and nuanced data. To fully analyze responses to this question, we coded responses dichotomously, as well as thematically.

Demographic control variables

On the pretest, participants were asked to identify their race from a list of six options and instructed to “check all that apply.” Due to the small cell sizes for respondents who were not White-identified, the groups were collapsed into White and people of color (POC). The POC category included respondents who selected multiple identities that included at least one POC identity. Similar to race, for sexual orientation and gender, participants were instructed to “check all that apply,” and due to small sample sizes, response categories were collapsed for analysis. Gender was recoded into three groups: cisgender man, cisgender woman, and transgender/genderqueer (which included those who selected multiple responses, including “other”). Using the sexual orientation and gender variables, an LGBTQQ identity variable was constructed, in which participants who identified as both straight and a cisgender man/woman were coded “0” and respondents who indicated at least one transgender or sexual minority identity were coded “1.”

Data Analysis

All statistical analyses were conducted using the Statistical Package for the Social Sciences (SPSS), Version 21. Descriptive statistics for demographic and outcome measures were run. In order to test the primary research aim, repeated-measures analysis of covariance (ANCOVA) was run to test the intervention effects, with mean-centered covariates (Algina, 1982; Delaney & Maxwell, 1981) and independent control variables: race, gender, sexual orientation, grade, and school. Hierarchical linear modeling or similarly multilevel analyses were not used because five schools at the second level are insufficient to estimate between school effects (Maas & Hox, 2005). Post hoc pairwise comparison Bonferroni tests were conducted to investigate differences based on demographic variables. For post hoc tests on gender, which contained three pairwise tests, α = .02 was used. No interaction effects were found with the full factorial model, so only direct effects are reported.

Constant comparison method (Conrad, 1978) was used to analyze qualitative data. The analytic approach elaborated here is also reported in the previous study using the same data set (Wernick, Dessel, et al., 2013). This coding was primarily conducted by the second author. Throughout the coding process, memos were kept and reviewed with other authors to refine emergent analytic schema and themes.

Participants’ written responses to the two write-in questions were read line by line and iteratively grouped based on similarities in content and interpreted meaning (Singh, Hays, & Watson, 2009). Three of the authors reviewed these groupings and reached consensus on this coding scheme. Responses in each group were then summarized using, as much as possible, participants’ words, phrases, and language (Creswell, 2007). In constructing summaries, more weight was given to responses that included more context, were more descriptive, and exemplified a majority of responses in the group with greater intensity.

Next, these summaries were used to abstract themes based on relationships between groups, framing in response to the question, and relatedness of concepts to existing constructs previously articulated in the literature. Themes were prioritized for presentation based in part on the frequency of responses related to each theme. Responses were also assigned a numerical code based on whether they denied homophobia and transphobia as a problem (0) or affirmed that homophobia and transphobia was a problem, to any extent (1). Finally, themes were compared across demographic groups (race, gender, and LGBTQQ identity) and the newly created quantified measure of awareness, including by change across pretest and posttests.

An additional repeated-measures ANCOVA model was then run, which included this newly coded variable in order to explore the relationship between changes in retrospective homophobia and transphobia awareness and the impact of the intervention. Exploratory analyses were also run to examine differences in schools where participation was voluntary compared to nonvoluntary.

Results

Descriptive Statistics

For quantitative analyses, the overall sample (n = 832) was reduced to those who had no missing data for any of the variables of interest (n = 515); see Figure 1. There were no significant differences on the dependent variable, control measures, race, or gender between the full sample and the analytic sample. However, LGBTQQ students were significantly underrepresented in the analytic sample as compared to the overall sample, suggesting a higher dropout or partial completion rate among LGBTQQ respondents. Among LGBTQQ students, there were no other significant demographic differences between those who were removed and respondents in the analytic sample. Qualitative analyses included all respondents who answered legibly the pretest and/or posttest write-in item.

Figure 1.
Flow chart of participants.

Table 1 includes a full report of the demographics for the analytic sample. The sample was mostly straight and cisgender (n = 386; 75%) and White (n = 301; 58%). In terms of gender, the sample was mostly cisgender women (n = 273; 53%), with 41% cisgender men (n = 209) and 6% who identified as transgender/genderqueer (n = 33). For grade, the sample mean was 10th grade (M = 1.84; SD = 1.58). The mean score for the dependent variable, advocacy intentions (measured on a 5-point scale) was 4.09 (SD = 1.03) on the pretest, with a posttest mean of 4.24 (SD = 0.97).

Table 1.
Mean and Pairwise Comparisons for Within-Subject Effects: Repeated-Measures ANCOVA Results (Advocacy Intentions).

Time 1 Time 2 n F(1, 526)

M (SD) SE 95% CI M (SD) SE 95% CI

Dependent variable

Advocacy intentions 4.07 (1.59) 0.07 [3.39, 4.21] 4.28 (1.48) 0.07 [4.15, 4.40] 515 13.43***

Independent variables

LGBTQQ identity 0.03

Yes 4.12 (1.03) 0.09 [3.94, 4.30] 4.31 (0.97) 0.09 [4.15, 4.48] 386

No 4.03 (1.71) 0.09 [3.86, 4.19] 4.24 (1.59) 0.08 [4.08, 4.40] 129

Gender 0.40

Cisgender men 3.87 (1.24) 0.09 [3.70, 4.04] 4.02 (1.16) 0.08 [3.86, 4.18] 209

Cisgender women 4.25 (1.11) 0.07 [4.11, 4.38] 4.41 (1.02) 0.06 [4.29, 4.54] 273

Transgender/genderqueer 4.11 (1.01) 0.18 [3.76, 4.45] 4.40 (0.94) 0.16 [4.08, 4.72] 33

Race 1.17

POC 3.95 (1.23) 0.08 [3.78, 4.12] 4.11 (1.16) 0.08 [3.96, 4.27] 214

White 4.20 (1.41) 0.08 [4.04, 4.36] 4.44 (1.30) 0.08 [4.29, 4.59] 301

Grade 0.33

8th 5.16 (10.09) 0.75 [3.68, 6.63] 5.21 (9.40) 0.70 [3.84, 6.86] 181

9th 3.80 (1.56) 0.24 [3.32, 4.27] 4.14 (1.45) 0.23 [3.69, 4.58] 41

10th 3.80 (1.98) 0.22 [3.37, 4.23] 4.02 (1.84) 0.21 [3.61, 4.42] 80

11th 3.81 (2.17) 0.21 [3.39, 4.23] 4.00 (2.03) 0.20 [3.61, 4.39] 105

12th 3.81 (2.22) 0.21 [3.39, 4.23] 4.02 (2.07) 0.20 [3.63, 4.41] 108

School 0.14

Witnessing harassment^a 4.78*

Intervention behavior^b 0.77

Note. ANCOVA = analysis of covariance; LGBTQQ = lesbian, gay, bisexual, transgender, queer, and questioning; POC = people of color. Estimated marginal (adjusted) means are shown. Estimated marginal mean comparisons are computed using pairwise comparisons with Bonferonni tests.

^aEvaluated at 0.03 (mean centered).

^bEvaluated at 0.05 (mean centered).

p < .05. p < .01. p < .001.

Intervention (Within-Group) Effect and Differences Between Groups

The repeated-measures ANCOVA showed a significant increase in students’ reported intent to advocate between baseline and posttest F(1, 500) = 13.43, p < .001, Cohen’s d = .14. See Table 1 for a full report of this model. This suggests that average advocacy intentions were significantly higher in the posttest than in the pretest. These results controlled for covariate baseline reports of witnessing anti-LGBTQQ harassment and self-reported intervention and between-subjects differences of gender, LGBTQQ identity, race, grade, and school. Baseline reports of frequency of witnessing anti-LGBTQQ harassment were also positively significant in this within-subject (time) effects model, F(1, 500) = 4.78, p < .05, $η_{p}^{2} = .01$ . That is, increased frequency of witnessing anti-LGBTQQ harassment was associated with a greater increase in advocacy intentions from the pre- to posttest.

Between-subject tests are used to investigate the differences on overall average scores for the dependent variable (i.e., the mean of advocacy intentions between pre- and posttest reports). Between-subject tests indicated students who reported a higher current intervention level when witnessing LGBTQQ harassment also reported a significantly higher intent to join advocacy efforts for LGBTQQ people F(1, 500) = 50.30, p < .001, $η_{p}^{2} = .09$ . This suggests that those who are already engaged in individual-level intervention behaviors are significantly more likely to report higher advocacy intentions. Further, White students, on average, reported greater advocacy intentions (M = 4.32, SD = 1.23) compared to students of color (M = 4.03, SD = 1.10), F(1, 500) = 8.98, p < .001, Cohen’s d = .25. In addition, between-subject pairwise tests suggest that cisgender women reported a significantly higher rate of advocacy intentions (M = 4.33, SD = 0.97) compared to cisgender men (M = 3.95, SD = 1.10), F(2, 500) = 11.40, p < .001, Cohen’s d = .37. See Figure 2 for changes over time by gender. These findings suggests that White and cisgender women respondents reported consistently higher advocacy intentions over the course of pre- and posttests than POC and cisgender men respondents. While there were significant differences across schools, the results are not reported here due to the small number of schools sampled (n = 5), preventing us from generalizing about differences between schools. There were no significant between-subject effects by LGBTQQ identity or grade.

Figure 2.
Change in intent to advocate by gender.

In the next model, using the constructed quantitative measure, when entered as a covariate with intent to advocate as the dependent variable, we found that student’s indication that homophobia and transphobia was a problem in their school was significantly positively associated with the change in intervention intentions F(1, 472) = 6.97, p < .01, $η_{p}^{2} = .02$ . That is, respondents’ affirmation that homophobia/transphobia is a problem at their school was associated with a greater increase in advocacy intentions from pre- to posttests. As well, the findings noted previously in the original model persisted in the same directions. The main (within-subjects) effect of the intervention remained significant F(1, 472) = 7.84, p < .01, Cohen’s d = .11. As well, baseline reports of witnessing anti-LGBTQQ harassment continued to be positively statistically significant F(1, 472) = 6.55, p < .05, $η_{p}^{2} = .01$ . Between-subject test continued to indicate an association between a reported higher current intervention level and a significantly higher intent to join advocacy efforts for LGBTQQ people, F(1, 472) = 50.32, p < .001, $η_{p}^{2} = .10$ . Similarly, Whites continued to report higher mean scores (M = 4.34, SD = 1.23) than POC (M = 4.06, SD = 1.08), F(1, 472) = 12.18, p < .001, Cohen’s d = 0.24. As well, cisgender women (M = 4.32, SD = 0.98) compared to cisgender men (M = 3.95, SD = 1.10) continued to report higher advocacy intentions, F(2, 472) = 10.46, p < .001, Cohen’s d = 0.36.

Finally, exploratory analyses suggested that in schools that had voluntary participation in the intervention, intention to advocate was significantly higher than respondents from those schools where participation was not voluntary in both models: without quantified write-in measure of awareness, F(1, 500) = 14.68, p < .001, Cohen’s d = .31; with quantified write-in measure of awareness, F(1, 472) = 13.41, p < .001, Cohen’s d = .30. The effect sizes and significance levels of within-subject and other between-subject indicators persisted. Given that this type of indicator is better suited for a hierarchal model, we do not report the full model here.

Analysis of the responses among participants who elaborated their awareness of homophobia and transphobia as problems revealed important findings related to the intervention. In the posttest, a greater number of students affirmed that homophobia/transphobia was a problem in their school. As well, a theme emerged in which participants described newfound knowledge and awareness about homophobia/transphobia, as well as a stronger desire to take action than evidenced in the pretest, which they directly attributed to Gayrilla. This theme existed among those both who affirmed and who denied the existence of homophobia/transphobia in the pretest. In addition, analysis unearthed nuance in how the intervention was associated with participants’ gathering of information related to homophobia/transphobia and the cognitive strategies used to interpret this information in making an assessment of homophobia/transphobia as a problem.

Information gathering

In both the pretest and posttest, participants relied on either a belief system or direct and vicarious experiences to gather information about homophobia and transphobia. For instance, one participant responded, “I don't think that if someone actually was gay/lesbian/transgender they wouldn't encounter much bullying, as most at our school are open minded.” Respondents referenced a belief about their school, town, friend group, or other community to signify whether or not homophobia/transphobia is a problem. Others directly cited specific instances of bullying or other actions (or lack thereof) in determining the prevalence of homophobia/transphobia. For instance, one respondent described their direct experiences of not observing homophobia/transphobia, “I do not think homophobia is a problem at our school at least from what I see and hear everyday.” Another respondent described having witnessed homophobic bullying, “Yes a lot of kids make fun of gays.” Among the posttest responses, an additional pattern emerged, in which participants indicated a lack of homophobic or transphobic experiences but a belief that despite this lack of personal experience, homophobia and transphobia did or could exist in their school. For instance, “I've never seen any here, but after the performance, I do think it's a problem.”

While no demographic differences emerged based on how respondents took in information about homophobia/transphobia, two interesting patterns emerged when the responses were analyzed by pretest and posttest within subjects’ differences. Participants who relied on direct experience as an indication of whether or not homophobia/transphobia existed in their school more frequently denied the existence of homophobia and transphobia on both the pretest and posttest, whereas those with preconceived or existential notions about whether or not homophobia and transphobia existed more frequently denied the existence of homophobia/transphobia on the pretest and affirmed it on the posttest.

Cognitive strategies affecting assessment of homophobia/transphobia

Participants demonstrated a wide breadth of cognitive strategies used to interpret the existence of homophobia and transphobia in their schools. Generally, among those who affirmed the existence of homophobia and transphobia, the most salient meaning-making strategies were assessing the impact on negatively affected students, holding one’s community in comparison to out-group others, contextualizing their assessment in a desire to take action to address homophobia and transphobia, and nuancing the definition of homophobia and transphobia. For instance, typical of respondents who assessed the impact of homophobia/transphobia, “Yes because people are effected [sic] by it everyday.” Those who described out-group others implicated a displacement of homophobia/transphobia onto those unlike them, for instance, “only among the ‘gangsta’ kids.” Others attributed homophobia/transphobia to age, social groups/networks, and masculinity/men/boys. Homophobia/transphobia was sometimes acknowledged as a malleable problem, with respondents indicating that their understanding of it was couched in a desire to create change, “better than most schools, but we still have work to do.” Those who responded by nuancing the definition of homophobia often responded with comments like, “not exactly ‘phobia’ but certainly discrimination. THAT is the problem.” In these instances, they often indicated another problem, such as discrimination, related to heterosexism/genderism that was not specifically related to attitudes or fear, as implied by “homophobia/transphobia.”

In contrast, among those who denied or diminished the existence of homophobia/transphobia, the most salient meaning-making strategies were assessing ambivalent intentions among perpetrators of anti-LGBTQQ harassment, minimizing the frequency or severity of anti-LGBTQQ incidents, and deploying an in-group identity associated with tolerance. Ambivalent intentions were often described like the following: “No but the people who say stuff like ‘that's so gay’ don't mean it to be offensive to LGBTQ students”; others similarly indicated that they believed those who participated in homophobia/transphobia were joking or otherwise didn’t mean to hurt anyone. The most common in-group identity invoked was of one’s school; for instance, “Not at my school but I've seen it at other schools.” “Other” schools sometimes explicitly included other local schools (e.g., larger schools were typified as more homophobic/transphobic) or other regions and areas of the country or world. Respondents associated the perceived or experienced hostility at “other” schools or in “other” environments to indicate that homophobia/transphobia was not a problem at their school. In these cases, it was not the strength of acceptance in their own localized context that predicated a denial of homophobia/transphobia but a belief in the relative acceptance or affirmation.

In comparing across demographic groups, patterns emerged along race, gender, and LGBTQQ identity. Participants with dominant identities (i.e., White respondents and straight/cisgender men and women) tended to rely more heavily on the perceived ambivalent intentions of perpetrators of anti-LGBTQQ behaviors and use of in-group and out-group assessments to claim a sense of tolerance among their own groups (e.g., friends, school, and region) and displace homophobia/transphobia onto others (e.g., other schools, “some people,” “gangsta kids”). In comparing pre- and posttest responses, assessments of the relative severity of these issues were more common in the posttest, whereas assessments of intent and claims of in-group tolerance decreased in frequency and emphasis from pre to post.

Discussion and Applications to Social Work

This performance and dialogue intervention is one way LGBTQQ youth in Michigan are challenging heterosexism and genderism in schools by developing support for advocacy efforts among young people. Whether or not students are aware of or acknowledge it, homophobia and transphobia in schools undergird the challenges many youth experience (Lugg, 2006; Meyer & Stader, 2009; Graham et al., 2014). In a context of relatively high overall support for LGBTQQ advocacy, participation in the Gayrilla Theater performance and post-performance dialogues were significantly associated with increased reports of willingness to advocate, overcoming a potentially steep regression to the main effect. Likely related to the high pretest mean score, we did find a relatively small main effect size. This intervention may have great potential to effectively increase students’ intentions to participate in macro-level change around LGBTQQ issues, mirroring our earlier findings that the intervention is associated with increased intentions to intervene at the interpersonal level (Wernick, Dessel, et al., 2013). Our findings also suggest key mechanisms that influenced the effectiveness of the intervention, providing important insights that can guide social work practice supporting advocacy for and by LGBTQQ students.

In our study, those who had more frequently witnessed anti-LGBTQQ harassment demonstrated a greater increase in intentions to advocate for social justice, fairness, and equality for LGBTQQ people. Verbal abuse and harassment related to actual or perceived marginalized sexual and gender identities are somewhat common in schools (Harris Interactive & GLSEN, 2005; Meyer & Strader, 2009). Our previous research suggests that witnessing homophobia and transphobia can communicate messages of hostility that impede intervention behaviors at the individual level, particularly in regard to harassment based on gender identity (Wernick, Kulick, & Inglehart, 2012). The intervention might interact or engage with previous exposure to harassment in a way that converts episodes of harassment into motivation to advocate for social justice, fairness, and equality. Further, we found that those who interpersonally intervened when they saw other students being harassed, had greater proclivity to advocate than those who did not, which suggests that interpersonal intervention may be an intermediate step toward joining advocacy efforts (Edwards, 2006). Social work practitioners interested in encouraging advocacy intentions among young people might consider focusing on students who already engage in interventions to support LGBTQQ students on an interpersonal level. As well, taking actions on a “smaller” scale might be beneficial in developing confidence, knowledge, and skills to participate in macro-level change.

Moreover, findings show that the personal stories shared in the performance and the dialogues likely influenced respondents to rethink their assessment of whether or not homophobia/transphobia was a problem in their school, as indicated by those who had indicated that homophobia/transphobia were not a problem on the pretest then affirmed the existence of these problems on the posttest. When predicting change in advocacy intentions, awareness of homophobia/transphobia as a problem was significant as a covariate, suggesting that these evaluations play an important role in the intervention’s relationship with increased intentions to advocate (Duhigg et al., 2010; Evans & Broido, 2005; Jordan, 2012; Mio, 2007; Worthington et al., 2002). Among students who did not initially consider homophobia and transphobia problems, the intervention appears to have curtailed considerations of the intent of perpetrators of LGBTQQ harassment and dis-identification with homophobic and transphobic networks. Among all participants, our findings also suggest Gayrilla may have bolstered considerations of the severity of homophobia/transphobia.

Personal stories shared in the performance and the dialogues might have encouraged respondents to reframe experiences, situations, and actions taken by their peers that they previously did not consider problematic; or, respondents may have generalized the stories shared to include their school even though they may not have had personal experiences to draw on to determine that homophobia/transphobia was a problem. While the intervention cannot change participants’ experiences, it appears that the intervention might be able to modify existing belief systems or introduce new belief systems that serve to build awareness about homophobia and transphobia. Students may have taken the stories shared during the performance and dialogues at face value, weighting the stories more substantially than their personal experiences or considering the stories shared and storytellers as more authoritative than their own experiences and assessments. The stories may affect students by prompting them to interpret their previous experiences in new or different ways.

These findings affirm previous research on the use of theater and dialogue by LGBTQQ people, suggesting that storytelling might be a powerful tool to move adults to action (Wernick, et al., 2014) and that dialogue can move peers to gain new awareness, motivating action (Dessel et al., 2013; Fisher & Checkoway, 2011). This study suggests that the combined impact of theater and dialogue in a single intervention engages the unique strengths of both theater and dialogue and may inspire peers to support macro-level change. In order to build knowledge, change beliefs, and encourage positive social action among young people, it may be beneficial to consider creative intervention strategies that incorporate the types of youth leadership, creative storytelling, and dialogue methods used by Gayrilla.

Our findings also suggest important differences by gender and race. We found that advocacy intentions were higher among White and cisgender women subsamples, although there were no significant differences in the rate of change over the course of the intervention. This might be partially explained by qualitative findings, which showed that White and cisgender students more strongly considered the intentions of perpetrators of anti-LGBTTQQ harassment and dis-identification with homophobic and transphobic networks in determining whether homophobia and transphobia were problems; and, these were the two processes the intervention seemed to influence most strongly in positive directions to produce increased willingness to advocate for social justice.

Cisgender women had higher advocacy intentions than cisgender men, which is not surprising. Previous research has found that women hold more affirming attitudes toward LGBTQQ civil rights than men (Herek, 2002). As well, homophobia/transphobia are central to masculinity in ways that are unparalleled with regard to femininity, and homophobia is used to police masculinity among cisgender men (Kimmel, 2001 Pascoe, 2005). Hence, challenging these systems might be threatening to straight cisgender men’s own sense of masculine self-worth, which is partially dependent on their identification with heterosexuality and cisgender masculinity. Showing support for LGBTQQ people might call one’s own heterosexual and masculine identities into question (Kimmel, 2001; Pascoe, 2005), creating an increased sense of risk in participating in this type of advocacy. The patriarchal systems that produce these barriers are closely tied into the systems that oppress LGBTQQ people (Pritchard, 2013); hence, support for LGBTQQ advocacy efforts must be conducted in tandem with efforts to transform sexism and gender norms. As well, interventions may need to specifically target men and boys to encourage critical interrogations of the relationships between masculinity, gender policing, and anti-LGBTQQ beliefs and actions. As well, future research should conduct interventions among large enough samples of transgender people to disaggregate this group in statistical analyses with sufficient power to detect any differences.

Our findings also suggested that White students had higher advocacy intentions than POC respondents. Although some have argued that homophobia and transphobia are more common among POC, we believe our findings are related to the racial–cultural contexts that shaped the intervention, measurement, and data analysis. First, Gayrilla creators and performers were a group of predominantly White students, and most of the audience members and dialogue participants were White, straight, and cisgender. These racial dynamics might impact the cultural orientation (Graham, Brown-Jeffy, Aronson, & Stephens, 2011) of the intervention and participation by POC respondents in ways that enable White students to participate and marginalize the participation of POC students (Wernick, Dessel, et al., 2013). Our qualitative analyses suggest that POC students’ considerations in assessing homophobia and transphobia were more evenly dispersed across the range of considerations outlined by all students in the results section. Perhaps if the intervention targeted more explicitly any unique considerations of POC students, it may be more effective among this subgroup.

It is unclear whether the experiences and sociocultural norms of particular ethnic groups included in the constructed POC category are similar enough to warrant grouping in this way. It is possible that the intervention is more or less effective for particular subgroups within the collapsed POC variable. Due to small subsample sizes, power is too low to disaggregate the POC variable. Related, it is unclear whether translational validity (face and content) of the measures are sufficiently high for subgroups of POC. As such, we are not sure how subgroups of POC are interpreting and assigning meaning to the items; for example, “I would join efforts to advocate for social justice fairness and equality for sexual minorities.” It is possible, for example, that subgroups of POC may view such framing, language, and efforts as culturally White and may therefore not be inclined to respond in the affirmative; but, they might respond differently if the statement read, for example, “I would work together with others in my community for civil rights for my gay and lesbian neighbors” (Parks, 2010). If the intervention took an explicitly intersectional (Crenshaw, 1991) approach to include more stories of particular subgroups of POC, messages might better resonate. This effort would benefit from additional formative research and critical race theory framings (Graham et al., 2011).

Finally, while exploratory results indicate that schools that had voluntary participation in the intervention are related to student’s higher intention to advocate, the overall model appeared relatively unchanged. This suggests that the intervention could be effective at increasing students’ intention to support LGBTQ advocacy efforts regardless of the whether or not students are mandated to attend. However, these results should be taken with caution as there were only five schools included in the model. Future research should replicate this study with samples including greater variance in schools.

There are a number of study limitations that should be noted. Neither schools nor students were randomly selected and findings are therefore not generalizable beyond the sample. Also, single items were used to measure constructs. Multidimensional measures may increase confidence in the validity of our construct operationalization. There was a drop-off in the LGBTQQ students from the overall sample to the study sample, so that may have influenced the findings; however, there was no significant difference between those who completed the full survey and those who dropped out or had missing data on either the pre- or posttest dependent variable (advocacy intentions). Further, while some students might have dropped out as a result of a negative reaction to the content of the performance, others might have dropped out because they felt they already had knowledge in these areas. Others yet may have dropped out because of other time commitments (interventions were typically conducted in the afternoon, therefore students involved in sports or other after school activities might have had to leave).

The question used to assess awareness, “Do you think homophobia/transphobia is a problem at your school,” could be considered double barreled and therefore problematic. However, given that many conceptualize LGBTQQ people as a single community (Moradi, Mohr, Worthington, & Fassinger, 2009), it might have effectively tapped into the way that young people conceptualize these interrelated forms of oppression. The question was approved by Riot Youth leaders who identify with a diverse range of gender and sexual identities and often conceptualize their oppressions as interconnected. However, future studies should ask about homophobia and transphobia separately as well as interrogating the relationship between awareness of both systems (Moradi et al., 2009). Finally, this pretest–posttest design does not allow us to assess the lasting impacts of the intervention over time, and whether intervention moved participants from intentions to concrete action. The only indication that there may be some impact is that school GSAs and Riot Youth experienced an increase in participation following each of these performances. Studies that include systematic follow-up components over longer periods of time should test the lasting effects of interventions like the one studied herein. As well, further research should explore which components of interventions are associated with particular action effects, including the mechanisms, processes, and means through which youth-led interventions most effectively led to advocacy.

In conclusion, the Gayrilla Theater performance and post-performance dialogues show promise as an intervention student groups, schools, community-based organizations, and others could use to increase willingness to join efforts to advocate for LGBTQ communities. Moving students to intervene at interpersonal and systemic levels are both important to address violence, harassment, and bullying in schools. Other performing arts interventions seeking to move people to action might target issues related to the impacts of harassment on LGBTQQ students, homophobia and transphobia in particular settings, variations in meaning of homophobia and transphobia, intentions of harassment perpetrators, and the frequency and severity of harassment. Interventions such as these should be funded and made accessible to students; and students should have the opportunity to develop their own performances and interventions. In addition, future interventions should engage intersectional frameworks that can engage students across a range of gender and racial identities to support social justice for gender and sexually marginalized communities.

	Time 1	Time 2	n	F(1, 526)
Dependent variable
Advocacy intentions	4.07 (1.59)	0.07	[3.39, 4.21]	4.28 (1.48)	0.07	[4.15, 4.40]	515	13.43***
Independent variables
LGBTQQ identity								0.03
Yes	4.12 (1.03)	0.09	[3.94, 4.30]	4.31 (0.97)	0.09	[4.15, 4.48]	386
No	4.03 (1.71)	0.09	[3.86, 4.19]	4.24 (1.59)	0.08	[4.08, 4.40]	129
Gender								0.40
Cisgender men	3.87 (1.24)	0.09	[3.70, 4.04]	4.02 (1.16)	0.08	[3.86, 4.18]	209
Cisgender women	4.25 (1.11)	0.07	[4.11, 4.38]	4.41 (1.02)	0.06	[4.29, 4.54]	273
Transgender/genderqueer	4.11 (1.01)	0.18	[3.76, 4.45]	4.40 (0.94)	0.16	[4.08, 4.72]	33
Race								1.17
POC	3.95 (1.23)	0.08	[3.78, 4.12]	4.11 (1.16)	0.08	[3.96, 4.27]	214
White	4.20 (1.41)	0.08	[4.04, 4.36]	4.44 (1.30)	0.08	[4.29, 4.59]	301
Grade								0.33
8th	5.16 (10.09)	0.75	[3.68, 6.63]	5.21 (9.40)	0.70	[3.84, 6.86]	181
9th	3.80 (1.56)	0.24	[3.32, 4.27]	4.14 (1.45)	0.23	[3.69, 4.58]	41
10th	3.80 (1.98)	0.22	[3.37, 4.23]	4.02 (1.84)	0.21	[3.61, 4.42]	80
11th	3.81 (2.17)	0.21	[3.39, 4.23]	4.00 (2.03)	0.20	[3.61, 4.39]	105
12th	3.81 (2.22)	0.21	[3.39, 4.23]	4.02 (2.07)	0.20	[3.63, 4.41]	108
School								0.14
Witnessing harassment^a								4.78*
Intervention behavior^b								0.77

Footnotes

Acknowledgment

The authors would like to thank the participants of the study, Riot Youth, Lori Roddy, Danny Brown, S. Jonah Thompson, Robbie Routenberg, and the Neutral Zone. A separate analysis using data from this study has been previously published (Wernick, Dessel, Kulick, & Graham, 2013).

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported in part by the Arcus Foundation, Liberty Hill Foundation, Mukti Fund, University of Michigan Arts of Citizenship, and the Community Foundation of SE Michigan.

References

Algina

(1982). Remarks on the analysis of covariance in repeated measures designs. Multivariate Behavioral Research, 17, 117–130.

Almeida

Johnson

R. M.

Corliss

H. L.

Molnar

B. E.

Azrael

(2009). Emotional distress among LGBT youth: The influence of perceived discrimination based on sexual orientation. Journal of Youth and Adolescence, 38, 1001–1014.

Black

Weisz

Coats

Patterson

(2000). Evaluating a psychoeducational sexual assault prevention program incorporating theatrical presentation, peer education, and social work. Research on Social Work Practice, 10, 589–606.

Bohm

(1996). On dialogue. New York, NY: Routledge.

Bontempo

D. E.

D’Augelli

(2002). Effects of at-school victimization and sexual orientation on lesbian, gay, or bisexual youths’ health risk behavior. Journal of Adolescent Health, 30, 364–374.

Broido

E. M.

(2000). The development of social justice allies during college: A phenomenological investigation. Journal of College Student Development, 41, 3–18.

Broido

E. M.

Reason

R. D.

(2005). The development of social justice attitudes and actions: An overview of current understandings. New Directions for Student Services, 2005, 17–28.

Centers for Disease Control and Prevention. (2011, 5 19). Lesbian, gay, bisexual, and transgender health: Youth. Retrieved from http://www.cdc.gov/lgbthealth/youth.htm

Christens

B. D.

Dolan

(2011). Interweaving youth development, community development, and social change through youth organizing. Youth & Society, 43, 528–548.

10.

Ciardullo

(2005). Advocates on both sides are as passionate as ever: SIECUS controversy report 2004-2005 school year. SIECUS Report, 33, 4–19.

11.

Cohen

(2005). Liberationists, clients, activists: Queer youth organizing, 1966-2003. Journal of Gay & Lesbian Issues in Education, 2, 67–86.

12.

Conrad

C. F.

(1978). A grounded theory of academic change. Sociology of Education, 51, 101–112.

13.

Creswell

J. W.

(2007). Qualitative inquiry and research design: Choosing among five approaches (2nd ed.). Thousand Oaks, CA: Sage.

14.

Crenshaw

(1991). Mapping the margins: Intersectionality, identity politics, and violence against women of color. Stanford Law Review, 43, 1241–1299.

15.

Currie

Mayberry

Chenneville

(2012). Destabilizing anti-gay environments through gay-straight alliances: Possibilities and limitations through shifting discourses. The Clearing House: A Journal of Educational Strategies, Issues and Ideas, 85, 56–60.

16.

Davis

T. S.

Saltzburg

Locke

C. R.

(2009). Supporting the emotional and psychological well being of sexual minority youth: Youth ideas for action. Children and Youth Services Review, 31, 1030–1041.

17.

Delaney

H. D.

Maxwell

S. E.

(1981). On using analysis of covariance in repeated measures designs. Multivariate Behavioral Research, 16, 105–123.

18.

Dessel

Rogge

Garlington

(2006). Using intergroup dialogue to promote social justice and change. Social Work, 51, 303–315.

19.

Dessel

A. B.

Woodford

M. R.

Routenberg

Briejak

D. P.

(2013). Heterosexual students’ experiences in sexual orientation intergroup dialogue courses. Journal of Homosexuality, 60, 1504–1080.

20.

Dessel

A. B.

Woodford

M. R.

Warren

(2011). Intergroup dialogue courses on sexual orientation: Lesbian, gay and bisexual student experiences and outcomes. Journal of Homosexuality, 58, 1132–1150.

21.

Dillon

F. R.

Worthington

R. L.

Savoy

H. B.

Rooney

S. C.

Becker-Schutte

Guerra

R. M.

(2004). On becoming allies: A qualitative study of lesbian-, gay-, and bisexual- affirmative counselor training. Counselor Education and Supervision, 43, 162–178.

22.

Duhigg

J. M.

Rostosky

S. S.

Gray

B. E.

Wimsatt

M. K.

(2010). Development of heterosexuals into sexual-minority allies: A qualitative exploration. Sexuality Research and Social Policy, 7, 2–14.

23.

Dupper

Meyer-Adams

(2002). Low-level violence: A neglected aspect of school culture. Urban Education, 37, 350–364.

24.

Edwards

(2006). Aspiring social justice ally identity development: A conceptual model. Journal of Student Affairs Research and Practice, 43, 1235–1256.

25.

Evans

N. J.

Broido

E. M.

(2005). Encouraging the development of social justice attitudes and actions in heterosexual students. New Directions for Student Services, 2005, 43–54.

26.

Fetner

Kush

(2008). Gay-Straight Alliances in high schools: Social predictors of early adoption. Youth & Society, 40, 114–130.

27.

Finn

Jacobson

Campana

J. D.

(2004). Popular education and participatory research. In Garvin

Galinsky

Gutierrez

(Eds.), Group work handbook (pp. 326–343). New York, NY: Guilford Publications.

28.

Fisher

Checkoway

(2011). Intergroup dialogue facilitation for youth empowerment and community change. In Maxwell

K. E.

Nagda

Thompson

(Eds.), Facilitating intergroup dialogues: Bridging differences, catalyzing change (pp. 133–146). Sterling, VA: Stylus.

29.

Fraser

(2004). Intervention research in social work: Recent advances and continuing challenges. Research on Social Work Practice, 14, 210–222.

30.

Fraser

Galinsky

(2010). Steps in intervention research: Designing and developing social programs. Research on Social Work Practice, 20, 459–466.

31.

Fuoss

K. W.

Kistenberg

C. J.

Rosenfeld

L. B.

(1992). “OUT ART”: Reducing homophobia on college campuses through artistic intervention. Text and Performance Quarterly, 12, 349–361.

32.

Goldstein

Davis

(2010). Heterosexual allies: A descriptive profile. Equity & Excellence in Education, 43, 478–494.

33.

Goodenow

Szalacha

Westheimer

(2006). School support groups, other school factors, and the safety of sexual minority adolescents. Psychology in the Schools, 43, 573–589.

34.

Graham

L. F.

(2012). Psychosocial health of black sexually marginalized men. In Treadwell

H. M.

Xanthos

Holden

K. B.

(Eds.), Social determinants of health among African American men (pp. 63–81). San Francisco, CA: Wiley & Sons.

35.

Graham

L. F.

Braithwaite

Spikes

Stephens

C. F.

Edu

U. F.

(2009). Exploring the mental health of Black MSM. Journal of Community Mental Health, 45, 272–284.

36.

Graham

L. F.

Brown-Jeffy

Aronson

R. E.

Stephens

(2011). Critical race theory as theoretical framework and analysis tool for population health research. Critical Public Health, 21, 81–93.

37.

Graham

L. F.

Crissman

H. P.

Tocco

Lopez

Snow

R. C.

Padilla

(2014). Navigating community institutions: Black transgender women’s experiences in schools, the criminal justice system, and churches. Sexuality Research and Social Policy. Available Online First.

38.

Graybill

Varjas

Meyers

Watson

(2009). Content-specific strategies to advocate for lesbian, gay, bisexual, and transgender youth: An exploratory study. School Psychology Review, 38, 570–584.

39.

Griffin

Lee

Waugh

Beyer

(2004). Describing roles that Gay-Straight Alliances play in schools. Journal of Gay & Lesbian Issues in Education, 1, 7–22.

40.

Harris Interactive & GLSEN. (2005). From teasing to torment: School climate in America, a survey of students and teachers. New York, NY: GLSEN. Retrieved from http://www.glsen.org/cgi-bin/iowa/all/research/index.htm

41.

Herek

G. M.

(2002). Gender gaps in public opinion about lesbians and gay men. Public Opinion Quarterly, 66, 40–66.

42.

Herek

G. M.

Capitanio

J. P.

(1996). “Some of my best friends”: Intergroup contact, concealable stigma, and heterosexuals' attitudes toward gay men and lesbians. Personality and Social Psychology Bulletin, 22, 412–424.

43.

Jeltova

Fish

M. C.

(2005). Creating school environments responsive to gay, lesbian, bisexual, and transgender families: Traditional and systemic approaches for consultation. Journal of Educational and Psychological Consultation, 16, 17–33.

44.

Jenson

J. M.

Dieterich

W. A.

Brisson

Bender

K. A.

Powell

(2010). Preventing childhood bullying: Findings and lessons from the Denver public schools trial. Research on Social Work Practice, 20, 509–517.

45.

Jordan

M. L.

(2012). Heterosexual ally identity development: A conceptual model. Journal of the Indiana University Student Personnel Association, 2012, 67–78.

46.

Kimmel

M. S.

(2001). Masculinity as homophobia: Fear, shame, and silence in the construction of gender identity. In Cook

(Ed.), Gender relations in global perspective: Essential readings (pp. 266–287). Toronto, ON: Canadian Scholars Press.

47.

Klein

(2006). Sexuality and school shootings: What role does teasing play in school massacres? Journal of Homosexuality, 51, 39–62.

48.

Kosciw

J. G.

Diaz

E. M.

(2006). The 2005 National School Climate Survey: The experiences of lesbian, gay, bisexual, and transgender youth in our nation’s schools. Washington, DC: Gay Lesbian and Straight Education Network.

49.

Kosciw

J. G.

Greytak

E. A.

Diaz

E. M.

Bartkiewicz

M. J.

(2010). The 2009 National School Climate Survey: The experiences of lesbian, gay, bisexual, and transgender youth in our nation’s schools. Washington, DC: Gay Lesbian and Straight Education Network.

50.

Kosciw

Palmer

Kull

Greytak

(2013). The effect of negative school climate on academic outcomes for LGBT youth and the role of in-school supports. Journal of School Violence, 12, 45–63.

51.

Lee

(2002). The impact of belonging to a high school gay/straight alliance. High School Journal, 85, 13–26.

52.

Liang

C. T. H.

Alimo

(2005). The impact of white heterosexual students’ interactions on attitudes toward lesbian, gay, and bisexual people: A longitudinal study. Journal of College Student Development, 46, 237–250.

53.

Lugg

C. A.

(2006). Thinking about sodomy: Public schools, legal panopticons, and queers. Educational Policy, 20, 35–58.

54.

Maas

C. J. M.

Hox

J. J.

(2005). Sufficient sample sizes for multilevel modeling. Methodology: European Journal of Research Methods for the Behavioral and Social Sciences, 1, 85–91.

55.

MacGillivray

I. K.

(2004). Gay rights and school policy: A case study in community factors that facilitate or impede educational change. International Journal of Qualitative Studies in Education, 17, 347–370.

56.

Mayberry

(2012). Gay-Straight Alliances: Youth empowerment and working toward reducing stigma of LGBT youth. Humanity & Society, 37, 35–54.

57.

Mayo

J. B.

Jr (2013). Critical pedagogy enacted in the Gay-Straight Alliance: New possibilities for a third space in teacher development. Educational Researcher, 42, 266–275.

58.

McCabe

Rubinson

(2008). Committing to social justice: The behavioral intention of school psychology and education trainees to advocate for lesbian, gay, bisexual, and transgendered youth. School Psychology Review, 37, 469–486.

59.

McGuire

J. K.

Anderson

C. R.

Toomey

R. B.

Russell

S. T.

(2010). School climate for transgender youth: A mixed method investigation of student experiences and school responses. Journal of Youth and Adolescence, 39, 1175–1188.

60.

Mercier

M. T.

(2010). Fighting to fit in: Gay–Straight Alliances in schools under United States jurisprudence. In Chan

P. C. W.

(Ed.), Protection of sexual minorities since Stonewall: Progress and stalemate in developed and developing countries (pp. 47–61). New York, NY: Routledge/Taylor & Francis.

61.

Meyer

E. J.

Stader

(2009). Queer youth and the culture wars: From classroom to courtroom in Australia, Canada and the United States. Journal of LGBT Youth, 6, 135–154.

62.

Mio

J. S.

(2007, 5). Allies and their motivations to intervene. Invited address delivered to the 87th Annual Meeting of the Western Psychological Association Convention, Vancouver, BC.

63.

Moradi

Mohr

J. J.

Worthington

R. L.

Fassinger

R. E.

(2009). Counseling psychology research on sexual (orientation) minority issues: Conceptual and methodological challenges and opportunities. Journal of Counseling Psychology, 56, 5–22.

64.

Moskalenko

McCauley

(2009). Measuring political mobilization: The distinction between activism and radicalism. Terrorism and Political Violence, 21, 239–260.

65.

Nadal

K. L.

Griffin

K. E.

(2011). Microaggressions: A root of bullying, violence, and victimization toward lesbian, gay, bisexual, and transgender youths. In Paludi

M. A.

(Ed.), The psychology of teen violence and victimization (Vol. 1, pp. 3–22). Santa Barbara, CA: ABC-CLIO, LLC.

66.

Nagda

McCoy

Barrett

(2006). Mix it up: Crossing social boundaries as a pathway to youth civic engagement. National Civic Review, 95, 47–56.

67.

Parks

C.W.

Jr (2010). A window illuminating the reservations of Black men who have sex with men in fully embracing the institution of same-sex marriage. Journal of Gay & Lesbian Social Services, 22, 132–148.

68.

Pascoe

C. J.

(2005). “Dude, you’re a fag”: Adolescent masculinity and the fag discourse. Sexualities, 8, 329–346.

69.

Pettigrew

T. F.

Tropp

L. R.

(2000). Does intergroup contact reduce prejudice: Recent meta-analytic findings? In Oskamp

(Ed.), Reducing prejudice and discrimination (pp. 93–114). Mahwah, NJ: Erlbaum.

70.

Poteat

V. P.

Espelage

D. L.

(2007). Predicting psychosocial consequences of homophobic victimization in middle school students. The Journal of Early Adolescence, 27, 175–191.

71.

Poteat

V. P.

Espelage

D. L.

Koenig

B. W.

(2009). Willingness to remain friends and attend school with lesbian and gay peers: Relational expressions of prejudice among heterosexual youth. Journal of Youth and Adolescence, 38, 952–962.

72.

Pritchard

(2013). For colored kids who committed suicide, our outrage isn’t enough: Queer youth of color, bullying, and the discursive limits of identity and safety. Harvard Educational Review, 83, 320–402.

73.

Rienzo

Button

Sheu

J-J.

(2009). The politics of sexual orientation issues in American schools. In Schroeder

Kuriansky

(Eds.), Sexuality education: Past, present, and future (Vol. 3, pp. 472–481). Westport, CT: Praeger Publishers/Greenwood Publishing Group.

74.

Riot Youth. (2009). The Riot Youth Climate Survey of Ann Arbor public high schools: Report of findings. Ann Arbor, MI: Neutral Zone. Retrieved from http://www.neutral-zone.org/programs/679/riot-youth-presents-its-school-climate-survey

75.

Riot Youth. (2013). Becoming visible: A project of Riot Youth at the Neutral Zone. Ann Arbor, MI: Neutral Zone. Retrieved from http://www.neutral-zone.org/programs/1070/gayrilla-becoming-visible

76.

Roades

L. A.

Mio

J. S.

(2000). Allies: How they are created and what are their experiences? In Mio

J. S.

Awakuni

G. I.

(Eds.), Resistance to multiculturalism: Issues and interventions (pp. 63–82). Philadelphia, PA: Brunner/Mazel.

77.

Sadowski

Chow

Scanlon

C. P.

(2009). Meeting the needs of LGBTQ youth: A “relational assets” approach. Journal of LGBT Youth, 6, 174–198.

78.

Saltzburg

(2008). Mentoring beyond homophobia: Reauthoring for cultural competence. Journal of Baccalaureate Social Work, 13, 35–48.

79.

Schindel

J. E.

(2008). Gender 101—Beyond the binary: Gay-Straight Alliances and gender activism. Sexuality Research and Social Policy, 5, 56–70.

80.

Silverschanz

Cortina

L. M.

Konik

Magley

V. J.

(2008). Slurs, snubs, and queer jokes: Incidence and impact of heterosexist harassment in academia. Sex Roles, 58, 179–191.

81.

Singh

A. A.

Hays

D. G.

Watson

L. S.

(2009). Strength in the face of adversity: Resilience strategies of transgender individuals. Journal of Counseling and Development, 89, 20–27.

82.

Stueve

O’Donnell

Tehranifar

Wilson-Simmons

Slaby

Link

Bruce

(2006). Rethinking the bystander role in school violence prevention. Health Promotion Practice, 7, 117–124.

83.

Swearer

S. M.

Turner

R. K.

Givens

J. E.

Pollack

W. S.

(2008). “You're so gay!”: Do different forms of bullying matter for adolescent males? School Psychology Review, 37, 160−173.

84.

Toomey

Ryan

Diaz

Russell

(2011). High school Gay–Straight Alliances (GSAs) and young adult well-being: An examination of GSA presence, participation, and perceived effectiveness. Applied Developmental Science, 15, 175–185.

85.

Vonofakou

Hewstone

Voci

(2007). Contact with out-group friends as a predictor of meta-attitudinal strength and accessibility of attitudes toward gay men. Journal of Personality and Social Psychology, 92, 804.

86.

Walls

Wisneski

Kane

(2013). School climate, individual support, or both? Gay Straight Alliances and the mental health of sexual minority youth. School Social Work Journal, 37, 88–112.

87.

Walls

N. E.

Freedenthal

Wisneski

(2008). Suicidal ideation and attempts among sexual minority youths receiving social services. Social Work, 53, 21–29.

88.

Walls

N. E.

Kane

S. B.

Wisneski

(2010). Gay-Straight Alliances and school experiences of sexual minority youth. Youth & Society, 41, 307–332.

89.

Wernick

L. J.

Dessel

A. B.

Kulick

Graham

L. F.

(2013). LGBTQQ youth creating change: Developing allies against bullying through performance and dialogue. Children and Youth Services Review, 35, 1576–1586.

90.

Wernick

L. J.

Kulick

Inglehart

M. H.

(2013). Factors predicting student intervention when witnessing anti-LGBTQ harassment: The influence of peers, teachers and climate. Children and Youth Services Review, 35, 296–301.

91.

Wernick

L. J.

Kulick

Inglehart

(2012, 11). Factors predicting student intervention when witnessing anti-LGBTQ bullying. Paper presented at the Annual Program Meeting of the Council on Social Work Education, Washington, DC.

92.

Wernick

L. J.

Kulick

Woodford

M. R.

(2014). How theater within a transformative organizing framework cultivates individual and collective empowerment among LGBTQQ youth. Journal of Community Psychology, 42, 838–853.

93.

Wernick

L. J.

Woodford

M. R.

Kulick

(2014). LGBTQQ youth using participatory action research and theater to effect change: Moving adult decision-makers to create youth-centered change. Journal of Community Practice, 22, 47–66.

94.

Wernick

L. J.

Woodford

Siden

J. Y.

(2010). Youth-led participatory action research: Fostering effective youth-adult partnerships–a case study. In Harter

Hamel-Lambert

Millesen

(Eds.), Participatory partnerships for social action and research (pp. 165–186). Dubuque, IA: Kendall/Hunt Publishing.

95.

Worthington

R. L.

Savoy

H. B.

Dillon

F. R.

Vernaglia

E. R.

(2002). Heterosexual identity development: A multidimensional model of individual and social identity. The Counseling Psychologist, 30, 496–531.

	Time 1			Time 2			n	F(1, 526)
	M (SD)	SE	95% CI	M (SD)	SE	95% CI	n	F(1, 526)
Dependent variable
Advocacy intentions	4.07 (1.59)	0.07	[3.39, 4.21]	4.28 (1.48)	0.07	[4.15, 4.40]	515	13.43***
Independent variables
LGBTQQ identity								0.03
Yes	4.12 (1.03)	0.09	[3.94, 4.30]	4.31 (0.97)	0.09	[4.15, 4.48]	386
No	4.03 (1.71)	0.09	[3.86, 4.19]	4.24 (1.59)	0.08	[4.08, 4.40]	129
Gender								0.40
Cisgender men	3.87 (1.24)	0.09	[3.70, 4.04]	4.02 (1.16)	0.08	[3.86, 4.18]	209
Cisgender women	4.25 (1.11)	0.07	[4.11, 4.38]	4.41 (1.02)	0.06	[4.29, 4.54]	273
Transgender/genderqueer	4.11 (1.01)	0.18	[3.76, 4.45]	4.40 (0.94)	0.16	[4.08, 4.72]	33
Race								1.17
POC	3.95 (1.23)	0.08	[3.78, 4.12]	4.11 (1.16)	0.08	[3.96, 4.27]	214
White	4.20 (1.41)	0.08	[4.04, 4.36]	4.44 (1.30)	0.08	[4.29, 4.59]	301
Grade								0.33
8th	5.16 (10.09)	0.75	[3.68, 6.63]	5.21 (9.40)	0.70	[3.84, 6.86]	181
9th	3.80 (1.56)	0.24	[3.32, 4.27]	4.14 (1.45)	0.23	[3.69, 4.58]	41
10th	3.80 (1.98)	0.22	[3.37, 4.23]	4.02 (1.84)	0.21	[3.61, 4.42]	80
11th	3.81 (2.17)	0.21	[3.39, 4.23]	4.00 (2.03)	0.20	[3.61, 4.39]	105
12th	3.81 (2.22)	0.21	[3.39, 4.23]	4.02 (2.07)	0.20	[3.63, 4.41]	108
School								0.14
Witnessing harassment^a								4.78*
Intervention behavior^b								0.77