247 ARTICLES Journal of Intercollegiate Sport, 2016, 9, 247 -267 © 2016 Human Kinetics, Inc. Toomey is with the Norton School of Family and Consumer Sciences, Family Studies and Human Development Program, The University of Arizona, Tucson, AZ. McGeorge and Carlson are with Human Development and Family Science, North Dakota State University, Fargo, ND. Address author correspondence to Russell B. Toomey at toomey@email.arizona.edu. http://dx.doi.org/10.1123/jis.2015-0049 A Mixed-Methods Pilot Study of Student Athlete Engagement in LGBTQ Ally Actions Russell B. Toomey University of Arizona Christi R. McGeorge and Thomas S. Carlson North Dakota State University Reports of sexual prejudice and heterosexism in sports are still pervasive, despite growing acceptance of lesbian, gay, bisexual, transgender, and queer (LGBTQ) persons in the U.S. Yet, few studies have investigated how and why college student- athletes engage in LGBTQ-focused ally actions (e.g., standing up and speaking out against sexual prejudice; showing support at LGBTQ events). This mixed-methods pilot study describes the development and psychometric testing of the Engagement in LGBTQ Ally Actions in Sports Scale, in addition to analyses of two open-ended questions that provided supplementary insight into what it means to collegiate athletes to be an ally to the LGBTQ community. The convenience sample included 159 college student-athletes from two public universities (M age = 20.67 years, SD = 1.56). Findings indicated that the Engagement in LGBTQ Ally Actions in Sports Scale demonstrated strong internal consistency, validity, and equivalence across gender. Qualitative analyses suggested that college student-athletes define being an ally to the LGBTQ community as involvement in advocacy behaviors and supporting LGBTQ teammates; yet, important gender differences emerged in these findings. Keywords: ally engagement; heterosexism; measurement; mixed methods; sexual prejudice; sports. Attitudes toward lesbian, gay, bisexual, transgender, and queer (LGBTQ) people in the United States have become more tolerant and accepting in recent years (McCormack, 2012; Pew Research Center, 2015). Yet, data still suggest that the collegiate sports context may be an unwelcoming environment for LGBTQ student-athletes (Denison & Kitchen, 2015; Gay, Lesbian and Straight Education http://dx.doi.org/10.1123/jis.2015-0049 248 Toomey, McGeorge, and Carlson JIS Vol. 9, No. 2, 2016 Network [GLSEN], 2013; Gill, Morrow, Collins, Lucey, & Schultz, 2010; Rankin, Merson, Sorgen, McHale, Loya, & Oseguera, 2011). For example, over half of U.S. athletes surveyed had heard disparaging jokes about LGBTQ people while participating in sports (Denison & Kitchen, 2015). Notably, while unwelcoming environments may be the status quo for LGBTQ athletes, several recent developments may represent a tipping point toward more acceptance of sexual minorities and gender diversity in the sporting context. First, in 2010, the National Collegiate Athletic Association (NCAA) adopted an enumerated nondiscrimination policy that included sexual orientation and gender expression, and released the first comprehensive resource focused on LGBTQ athletes for use by college athletic departments (Champions of Respect, Griffin & Taylor, 2013). Second, a handful of professional and collegiate athletes (for example, Kye Allums [U.S. college basketball], Megan Rapinoe [U.S. Olympic soccer], Michael Sam [National Football League], and Wade Davis [National Football League]) have pub- licly disclosed their sexual orientation while participating in their respective sports or after retirement. Finally, in the past ten years, several not-for-profit organizations have come to fruition with missions focused exclusively on fostering affirmative climates for LGBTQ athletes (e.g., Athlete Ally; Go! Athletes; You Can Play). A focus for some of these not-for-profit organizations is the training and identification of LGBTQ allies in sports (e.g., Athlete Ally, Taylor, 2015). Tradi- tionally, the term ally has been defined as a member of the “dominant or majority group who works to end oppression in his or her personal and professional life through support of, and as an advocate for, the oppressed population” (Washington & Evans, 1991, p. 195). By definition, the focus of allyship should be on actions (i.e., defining allyship as a verb) rather than identity (i.e., defining allyship as a noun). Thus, college student-athlete LGBTQ “allies” should theoretically be acting in ways that reduce oppression related to sexual orientation and gender diversity in the sports context (Taylor, 2015). This mixed-methods pilot study examined the psychometric properties of the Engagement in LGBQ Ally Actions in Sports Scale, which assesses college student-athletes’ level of engagement in LGBTQ ally-typed behaviors. Further, via qualitative data, this study examined what it means to be an ally to LGBTQ persons from the perspectives of college student-athletes as well as the antecedents for engaging in LGBTQ ally-focused behavior. For the purposes of this study, ally behaviors include a range of actions such as providing support to LGBTQ team- mates, friends, or family members to engagement in behaviors aimed at reducing oppression and promoting social justice for LGBTQ populations. The Unique Context of Collegiate Sports While research on LGBTQ ally identities and behaviors does exist (e.g., Asta & Vacha-Haase, 2013; Russell, 2011), it has not focused specifically on the experiences of college student-athletes. A focus on college, rather than high school or professional sports, is warranted given that college is a central context where civic development occurs (Flanagan & Levine, 2010). Yet, research has documented that student athletes, particularly those in prestigious Division I schools, have little time for campus activi- ties and academic responsibilities (Gayles & Hu, 2009). Further, Hoffman, Kihl, and Student Athlete Engagement in LGBTQ Ally Actions 249 JIS Vol. 9, No. 2, 2016 Browning (2015) found that while college student-athletes were more likely to engage in volunteerism than their nonathlete peers, they were reluctant to engage in political activism. However, none of these studies have focused on LGBTQ-specific ally behav- iors, but rather focused on engagement behaviors broadly (e.g., from volunteerism to overall political engagement to participation in college clubs or organizations). Notably, the level of stigma attached to LGBTQ communities, as well as LGBTQ ally identities and behaviors, likely varies across settings. An exclusive focus on LGBTQ ally engagement among college student-athletes is warranted given the unique dominance of hegemonic masculinity in the sports context, which has been framed as an underlying culprit for sexual prejudice (e.g., Anderson, 2008; Griffin, 2014). The sport industry, in particular, has a very narrow definition of what constitutes masculinity and femininity, which creates a culture that reinforces heteronormative behaviors in all participants, ranging from athletes to coaches to fans (Cunningham & Sagas, 2008; Knight & Guiliano, 2003; Sartore-Baldwin & Warner, 2012). Thus, given the unique culture of collegiate athletics and the general lack of inclusiveness within college athletic departments (see Cunningham, 2015b; Melton, 2015), it is likely that LGBTQ ally engagement by student-athletes differs from the extant literature on LGBTQ ally engagement among the general population. LGBTQ Ally Engagement The study of LGBTQ allies has largely focused on understanding the antecedents of one’s ally identity, rather than ally behaviors. That is, several rich, qualitative studies point to three main catalysts for why a person “adopted” an LGBTQ ally identity, including: (a) contact and relationships with LGBTQ persons (Fingerhut, 2011; Grzanka, Adler, & Brazer, 2015; Russell, 2011; Strotzer, 2009), (b) upbring- ing or family socialization (Duhigg, Rostosky, Gray, & Wilmsatt, 2010; Grzanka et al., 2015; Russell, 2011; Strotzer, 2009), and (c) a social justice-oriented ideology or belief system (Asta & Vacha-Haase, 2013; Broido, 2000; Duhigg et al., 2010; Rostosky, Black, Riggle, & Rosenkrantz, 2015; Russell, 2011). Empathy with the struggles of LGBTQ people has also been discussed as a reason why a person adopted an ally identity (Russell, 2011; Strozer, 2009). Yet, we are not aware of any study that has examined the meaning of LGBTQ ally identities among student- athletes or the corresponding LGBTQ ally behaviors engaged in by these students. Research on LGBTQ ally behaviors has posited that some allies, perhaps in earlier developmental stages of their identity, engage in more passive actions (e.g., displaying a rainbow on their social media profile to support marriage equality), while others, perhaps in later stages of development, engage in more active actions (e.g., advocating for institutional changes to disrupt heterosexism, sexual prejudice; Ji, Du Bois, & Finnessy, 2009). Indeed, Grzanka and colleagues (2015) found evidence for the distinction between passive and active LGBTQ-related activism. Importantly, the extant literature on ally action engagement has not been conducted within in the sports context, where, given the continued prevalence of sexual preju- dice, it may be important to perceive any LGBTQ ally-related actions as active rather than passive (Melton & Cunningham, 2012). Therefore, the current study intended to assess actions related to Standing Up to heterosexist behaviors and Showing Up to visibly support LGBTQ communities, rather than the passive-active dichotomy that has been conceptualized and used in the prior literature. Further, we theorize 250 Toomey, McGeorge, and Carlson JIS Vol. 9, No. 2, 2016 that, given the prestigious social status that athletes assume on college campuses, particularly in Division I schools, even attendance at an LGBTQ community event should be interpreted as a meaningful and an active action on the behalf of that athlete (Melton, 2015). That is, if the overall campus environment is hostile to LGBTQ populations, it is a daring act of leadership for an athlete to be seen with LGBTQ communities, as the stigma associated with being LGBTQ may be placed on that athlete (e.g., Goffman, 1963). Indeed, research does suggest that a hostile community or campus climate for LGBTQ communities can result in a lack of or hesitation to participate in ally behaviors (e.g., Melton & Cunningham, 2014a), particularly for collegiate athletes in conservative college towns (e.g., Melton & Cunningham, 2014b). Alternatively, when LGBTQ populations are affirmed on college campuses, the attendance of an athlete at LGBTQ events may be considered an act of leadership related to social change (Melton, 2015). Gender, LGBTQ Ally Engagement, and Sports Given the unique organizational climate and structure of collegiate sports (Cun- ningham, 2015b; Melton, 2015), as well as the documented rates of sexual prejudice in sports (Denison & Kitchen, 2015; GLSEN, 2013; Gill et al., 2010), engaging in LGBTQ ally-related behaviors may be distinct in the sporting context compared with other commonly studied contexts (e.g., workplace). Several scholars argue that the distinctiveness of LGBTQ issues in sports exists because of the salience and dominance of masculinity (e.g., Anderson, 2008; Griffin, 2014). Importantly, when integrating findings across studies, antigay-related hostility appears to manifest differently in women’s sports as compared with men’s sports (Anderson & Bullingham, 2013; Cunningham, Pickett, Melton, Lee, & Miner, 2013; Fink, Burton, Farrell, & Parker, 2012; Griffin 2012; Krane, 1996; Melton & Cunningham, 2012), suggesting that gender plays an important role in understanding how col- lege student-athletes address and deal with LGBTQ issues on and off the field. For example, Griffin and Taylor (2013) argue that one of the reasons for this differential manifestation is due to the influence of hegemonic masculinity which creates the erroneous assumption that all women athletes are lesbian or bisexual and no male athletes are gay or bisexual. Furthermore, given that women’s collegiate athletics is already marginalized in terms of funding and prestige due to gender bias (Krane, Choi, Baird, Aimar, & Kauer, 2004), the association of women athletes as sexual minorities may create a fear among coaches, administrators and athletes that such an association will further delegitimize women’s sports (Griffin & Taylor, 2013; Sartore-Baldwin & Cunningham, 2010). Thus, given that existing research has documented that attitudes toward the LGBTQ community differ between men and women athletes (e.g., Cunningham et al., 2013), we also examined how LGBTQ ally-related behaviors differed by gender among college student-athletes. Study Aims The overarching goal of this pilot study was to understand LGBTQ ally engagement in the context of collegiate athletics using both quantitative and qualitative methodolo- gies. A simultaneous mixed-methodology design (Creswell, Plano Clark, Gutmann, Student Athlete Engagement in LGBTQ Ally Actions 251 JIS Vol. 9, No. 2, 2016 & Hanson, 2003) was chosen to employ both a rigorous quantitative assessment of a new measure assessing LGBTQ ally engagement (i.e., the Engagement in LGBTQ Ally Actions in Sports Scale with two subscales—standing up and showing up), and to explore the meaning of these constructs with rich contextual understandings via qualitative methods. Based on existing research and recommendations (e.g., Griffin & Taylor, 2013; Russell, 2011), the first aim of the study was to establish the factor structure of and validate the newly created scale that assesses engagement in LGBTQ ally-related behaviors among college student-athletes. After adequate measurement structure was established, we examined the associations between the scale and five measures to assess the convergent (similarity between theoretically similar measures) and divergent validity (the absence of an association between theoretically unrelated constructs) of these measures (DeVellis, 2012). Given the important role that gender plays in understanding sexual prejudice, heterosexism and cissexism in the sports context (e.g., Griffin, 2014; Worthen, 2014), we also examined whether the measure was equivalent across gender. The second aim was to gain a deeper understanding of the meaning that college student-athletes associate with the term “athlete ally” and any antecedents of ally identity that are specific to the sports context. Method Procedure Two universities were selected using a convenience sampling approach for this pilot study. The Athletic Director at the first university was approached about having his athletes participate in this study as he was known to be an ally to the LGBT commu- nity. Upon the request of the researchers, this particular Athletic Director contacted an Athletic Director at another university who he thought would be willing to assist in the recruitment of student athletes for this study. Neither athletic department had implemented an LGBT ally training program; however, approximately 12% of the resultant sample had attended a university-sponsored LGBT ally training session. Importantly, a third university was initially recruited for inclusion in the study; however, the Athletic Director at that university withdrew participation in the study after one of their athletes engaged in an overt homophobic behavior via social media that received national media attention. College student-athletes at the two participating universities received recruit- ment emails that described the study in the spring semester of 2014. The emails were distributed to all student-athletes by athletic directors and academic advisors for athletics. After the initial e-mail was sent, student-athletes received three reminder emails, sent at two-week intervals. Participants completed a 15–20-minute web- based survey, which was hosted by Qualtrics. College student-athletes who agreed to participate in the study were provided the option to enter a drawing for one of ten $25.00 gift cards. The study’s protocol was approved by the institutional review boards at North Dakota State University and Kent State University. Sample The resultant convenience sample included 159 college student-athletes from two public universities. Both universities have NCAA Division I athletics programs, 252 Toomey, McGeorge, and Carlson JIS Vol. 9, No. 2, 2016 and each enroll approximately 15,000 students. One of the universities is located in the Mountain West region of the U.S., while the other is located in the Upper Midwest region. Nearly half (49.7%) identified as women, and the majority (95%) identified as heterosexual/straight (5% identified as lesbian or bisexual). A large percentage (86.2%) identified as White, non-Latino, and most were from upper middle class families of origin (reported median income range for family of origin was $75,000 to $100,000). Participants were, on average, 20.67 years old (SD = 1.56, Range = 18–28 years); 30.8% were freshman, 20.1% were sophomores, 23.9% were juniors, and 19.5% were fourth-year or fifth-year seniors; 1.9% were attending graduate school, and 3.8% did not answer the question about their current year of school enrollment. Finally, the sports played by participants ranged widely (e.g., basketball, cross country, football, soccer, wrestling), and most (76.7%) were funded by a full or partial athletic scholarship. Measures Ally Engagement. The Engagement in LGBTQ Ally Actions in Sports Scale was created to assess how frequently college student-athletes engaged in actions that are considered to be supportive of the LGBTQ community. Items were reviewed for inclusion in the scale by a panel of ally-identified college students and two ally-identified athletic directors to ensure that the items had face validity. Eight items (see Table 1) were created based on the recommendations and results of the Champions of Respect: Inclusion of LGBTQ Student-Athletes and Staff in NCAA Programs report (Griffin & Taylor, 2013). Items were assessed using a 4-point Likert-scale, ranging from Never to Often. The response option, Not Applicable, was also provided for participants if, for instance, their school did not hold campus events intended to support the LGBTQ community. The first hypothesized subscale, named Standing Up Engagement, assessed the degree to which athletes engage in ally actions that include speaking out against heterosexist behaviors in the sports context and was comprised of three items (Items 1, 2, and 6; see Table 1). The second hypothesized subscale, named Showing Up Engagement, assessed the degree to which athletes engage in ally actions that include attending events to show support of the LGBTQ community and was comprised of four items (Items 3, 4, 5, and 8; see Table 1). Higher scores indicate more frequent LGBTQ ally engagement. Open-Ended Questions. Two open-ended questions were used to further explore college student-athletes’ understandings of what it means to be an ally to the LGBTQ community. In particular, participants were asked to respond to the fol- lowing questions: 1) What does it personally mean to you to be an “athlete ally” to the LGBTQ community? and 2) Looking back on your life experiences, what do you think prompted you to identify as an ally to the LGBTQ community? Convergent Validity Measures. Four measures were used to assess the convergent validity of the Engagement in LGBTQ Ally Actions in Sports Scale including (a) attitudes toward social justice, (b) common fate with LGBTQ people ideology, (c) affirming beliefs about LGBTQ persons, and (d) intervention in LGBTQ- related bias in sports. First, attitudes toward social justice were measured using an adapted version of Kizer’s (2011) Attitudes Toward Social Justice Scale. The original scale was used to assess counselors’ attitudes toward social justice and Student Athlete Engagement in LGBTQ Ally Actions 253 JIS Vol. 9, No. 2, 2016 was adapted for the current study to assess the attitudes of student-athletes. The adapted scale consists of nine out of the ten original items; one item could not be adapted for use with a noncounselor population as it was specific to beliefs about the ethics of the counseling profession. Participants rated the items on a 6-point Likert-type scale, ranging from Strongly Disagree to Strongly Agree, with higher scores indicating greater social justice attitudes and beliefs. Exploratory and Table 1 Original Items of the Engagement in LGBTQ Ally Actions in Sports Scale 1. I let teammates know that I don’t like to hear them use anti-LGBTQ slurs and explain why they are problematic. (Standing Up) 2. I stand up for teammates who are LGBTQ if I hear other teammates talking badly about them behind their backs because of their sexual orientation or gender identity and expression. (Standing Up) 3. I participate in my university’s LGBTQ-straight alliance group to make my school safer for all students. (Showing Up) 4. When offered, I regularly attend campus events that are intended to support the LGBTQ community (for example activities associated with Coming Out Week, Trans- gender Awareness Week, Pride Week). (Showing Up) 5. When offered, I regularly attend community events that are intended to support the LGBTQ community (for example activities associated with Coming Out Week, Trans- gender Awareness Week, Pride Week). (Showing Up) 6. I communicate my identity as an ally to the LGBTQ community through the use of symbols (for example, I have worn a rainbow sticker, button, or t-shirt that is supportive of the LGBTQ community or have used an LGBTQ equality symbol as my Facebook profile picture). (Standing Up) 7. I have attended rallies or political demonstrations in support of LGBTQ rights (for example, same-sex marriage, employment nondiscrimination acts, or hate crime legis- lation). (Showing Up) 8. I have written emails or sent letters to my political (for example, local council mem- bers, state governors, or national Congressional representatives) or campus representa- tives (for example, university president or provost) to urge them to pass LGBTQ rights legislation (for example, same-sex marriage, employment nondiscrimination acts, hate crimes legislation) or policies (for example, domestic partner benefits for faculty and staff). (Standing Up) Note. Item 8 was dropped from the scale because it lacked substantial variance (item is italicized). 254 Toomey, McGeorge, and Carlson JIS Vol. 9, No. 2, 2016 confirmatory factor analyses revealed that the factor structure was unidimensional and that items 5 (“It is okay for my family members or friends to make jokes about marginalized groups in the privacy of their own homes”) and 7 (“Involvement in social justice activities is a matter of personal preference”) did not significantly load onto the latent factor, and thus were removed from the scale. The final scale had seven items, which had excellent reliability (α = .90). Second, common fate with LGBTQ people ideology was assessed by three items (e.g., “How much do you feel you have in common with most LGBTQ people?”). These three items were adapted from Curtin’s (2011) measure that assessed common fate related to gender identity. Items were rated on a 5-point Likert-scale, ranging from Hardly at All to Very Much; higher scores reflect a higher sense of common fate with LGBTQ people. The measure had acceptable reliability (α = .73). Third, affirmative beliefs about LGBTQ people were assessed by a single item: “On a scale from 1 to 7, where 1 is extremely not accepting and 7 is extremely accepting/affirming, how would you describe your own beliefs about LGBTQ people?” Finally, participants responded to a single item that assessed how often they actively intervened in LGBTQ-related bias in sports: “How often do you stop others from making negative comments or slurs based on sexual orientation or gender identity and expression?” This item was rated on a 4-point Likert-scale, ranging from Never to Often; higher scores reflected more frequent intervention. Divergent Validity Measure. Given that a participant’s birth month conceptually should not be associated with engaging in LGBTQ ally behaviors, it was chosen to assess divergent validity for the newly developed measure. Birth month was assessed by one item: “What is your date of birth?” participants selected their month of birth from a dropdown box, coded from 1 (January) to 12 (December). At least one participant was born in each month of the year (Range per month: n = 6–22). Analytic Method for Quantitative Data Three steps were used to examine the psychometric properties of the Engagement in LGBTQ Ally Actions in Sports Scale. Step one examined the factor structures of the scale using confirmatory factor analyses (CFA). Three fit indices were used to examine overall model fit: the chi-square statistic, the comparative fit index (CFI), and the standardized root-mean-square residual (SRMR). A nonsignificant chi-square statistic suggests that the model fits the data well; however, this statistic is prone to bias related to sample size (Kenny & McCoach, 2003). Model fit was considered to be good (acceptable) if the SRMR was less than or equal to .05 (.08), and if the CFI was greater than or equal to .95 (.90) (Kline, 2016). In step two, we evaluated whether the factor structures of the newly developed measures were equivalent across gender identity groups. Factor invariance would suggest that the items functioned similarly for women and men, and was tested with a series of nested multiple group CFAs that included gender identity as the group variable. The tenability of invariance was tested at three sequential levels: configural, loading, and intercept invariance (e.g., Little, Card, Slegers, & Ledford, 2007). The tenability of the constraints was evaluated by using a chi-square difference test, in which a significant change in chi-square indicated that the constraints were not tenable (i.e., the items did not function similarly across gender identity groups). Student Athlete Engagement in LGBTQ Ally Actions 255 JIS Vol. 9, No. 2, 2016 Finally, in step three, we examined the reliability and validity of the Engage- ment in LGBTQ Ally Actions in Sports Scale. Reliability was examined using Cronbach’s alpha, and convergent and divergent validity were tested using structural equation modeling, controlling for age and gender. To test for convergent validity, the examination of whether measures are correlated in conceptually meaningful ways, it was hypothesized that both ally engagement subscales would be positively associated with attitudes toward social justice, common fate with LGBTQ people ideology, affirming beliefs about LGBTQ persons, and intervention in LGBTQ- related bias in sports (e.g., Russell, 2011; Taylor, 2015). To test for divergent validity, when measures are unrelated to constructs and no conceptual reason for an association exists, we examined the associations between ally engagement and birth month. All analyses were conducted in Mplus version 7.3 (Muthén & Muthén, 2014), and full information maximum likelihood was used to handle missing data (Schlomer, Bauman, & Card, 2010). Analytic Method for Qualitative Data Thematic analysis was used to code the qualitative data to identify categories, themes, and subthemes that might exist within and across the participants’ responses (Braun & Clarke, 2006). Specifically, Braun and Clarke (2006) defined thematic analysis as “a method for identifying, analyzing, and reporting patterns (themes) within the data set in (rich) detail” (p. 79). Categories represent broad or overarch- ing descriptions of the data, while themes are unique representations of the data within the larger category (Braun & Clarke, 2006). Subthemes are “themes-within- a-theme” that highlight varying descriptions or understanding within a theme (Braun & Clarke, 2006, p. 92). Before we (the second and third authors) began the coding process, we read the qualitative data several times to familiarize ourselves with the participants’ responses. After this review, we identified meaningful patterns in the data and worked to identify codes or common ideas that existed across participants’ responses. We then grouped the data in terms of these initial codes in an effort to explore the extent to which these codes were represented in participants’ responses and looked for the emergence of categories. Once categories were defined, we coded the data within each category to identify any possible themes and subthemes. After initially identifying the categories, themes, and subthemes, we met for peer debrief- ing with the first author to add more credibility to the coding process (Daly, 2007; Lincoln & Guba, 1985). Finally, after consensus was reached on the categories, themes, and subthemes, we selected verbatim quotations as illustrations of the data. Results Engagement in LGBTQ Ally Actions in Sports Scale Before CFA analyses, we examined the descriptive statistics of each of the eight items to ensure that each item was normally distributed and contained sufficient variance. All of the items, with the exception of item 8, met these criteria, sug- gesting that participants varied on their levels of ally engagement. Item 8 (“I have written emails or sent letters to my political [for example, local council members, 256 Toomey, McGeorge, and Carlson JIS Vol. 9, No. 2, 2016 state governors, or national Congressional representatives] or campus representa- tives [for example, university president or provost] to urge them to pass LGBTQ rights legislation [for example, same-sex marriage, employment nondiscrimination acts, hate crimes legislation] or policies [for example, domestic partner benefits for faculty and staff]”) was dropped because it lacked substantial variance (only two participants answered that they had ever engaged in this type of action). A CFA of the resultant 2-factor, 7-item scale revealed that it had good fit: χ2 (df =13) = 19.09, p = .12; CFI = 0.98; SRMR = .04. Factor loadings, residuals, and latent intercepts are shown in Table 2. Tests of measurement invariance revealed that the items functioned similarly for women and men. That is, criteria for configural, loading (Δχ2(df = 5) = 7.593, p = .18), and intercept invariance (Δχ2(df = 5) = 5.87, p = .32) were not significant. Importantly, Standing Up and Showing Up subscales were significantly associated with one another (r = .73, p < .001); however, a test of redundancy (Kline, 2016), indicated that the two subscales were unique (Δχ2 (df = 1) = 27.47, p < .001). Finally, we examined the reliability and validity of the Standing Up and Show- ing Up subscales. Acceptable reliability was demonstrated for the Standing Up subscale for all participants (α = .80), men (α = .81), and women (α = .73). Tests of convergent validity were consistent with hypotheses (see Table 3 for summary of latent correlations), and model fit of this structural model was acceptable: χ2 (df =165) = 231.18, p < .05; CFI = 0.939; SRMR = .06. As expected, Standing Up was positively associated with attitudes toward social justice (Latent r = .56, p < .001), affirmative beliefs about LGBTQ people (Latent r = .71, p < .001), having a common fate with LGBTQ people ideology (Latent r = .59, p < .001), and inter- vention in LGBTQ-related bias in sports (Latent r = .60, p < .001). Evidence for divergent validity was found, in that the Standing Up subscale was not associated with birth month (Latent r = -.06, p = .46). Of note, men were less likely to engage in Standing Up actions compared with women (Latent r = -.43, p < .001). Acceptable reliability was also demonstrated for the Showing Up subscale for all participants (α = .83), men (α = .83), and women (α = .82). Tests of convergent validity were consistent with hypotheses, such that Showing Up was positively associated with attitudes toward social justice (Latent r = .36, p < .001), affirmative beliefs about LGBTQ people (Latent r = .38, p < .001), having a common fate with LGBTQ people ideology (Latent r = .28, p < .05), and intervention in LGBTQ- related bias in sports (Latent r = .38, p < .001). Finally, evidence for divergent validity was found, in that the Showing Up subscale was not associated with birth month (Latent r = -.11, p = .23). Of note, men were also less likely to engage in Showing Up actions compared with women (Latent r = -.23, p < .05). What it Means to Simultaneously be an Ally and an Athlete As mentioned previously, the first open-ended question asked participants to describe what it means to them personally to be an “athlete ally” to the LGBTQ community. The thematic analysis lead to the identification of three categories: 1) Actions, 2) Attitudes, and 3) I am Not an Ally. The Action category encompassed participants’ responses that defined being an ally in terms of certain behaviors they felt were central to being an ally for the LGBTQ community. Within this category, three themes were identified. The first 257JIS Vol. 9, No. 2, 2016 Ta b le 2 F a c to r L o a d in g s a n d R e s id u a ls o f th e E n g a g e m e n t in L G B T Q A ll y A c ti o n s i n S p o rt s S c a le S ta n d ar d iz ed F ac to r L o ad in g U n st an d ar d iz ed Fa ct o r L o ad in g s (S ta n d ar d E rr o r) U n st an d ar d iz ed R es id u al L at en t In te rc ep t S ta nd in g U p I le t te am m at es k no w t ha t I do n’ t li ke t o he ar t he m u se a nt i- L G B T Q s lu rs a nd e xp la in w hy t he y ar e pr ob le m at ic . .7 47 .7 09 ( .0 9) ** * .3 98 ( .0 8) ** * 1. 93 I st an d up f or t ea m m at es w ho a re L G B T Q i f I he ar o th er t ea m - m at es t al ki ng b ad ly a bo ut t he m b eh in d th ei r ba ck s be ca us e of th ei r se xu al o ri en ta ti on o r ge nd er i de nt it y an d ex pr es si on . .7 47 .8 15 ( .1 3) ** * .5 28 ( .1 4) ** * 2. 44 I co m m un ic at e m y id en ti ty a s an a ll y to t he L G B T Q c om m un it y th ro ug h th e us e of s ym bo ls ( fo r ex am pl e, I h av e w or n a ra in bo w st ic ke r, b ut to n, o r t- sh ir t th at i s su pp or ti ve o f th e L G B T Q c om - m un it y or h av e us ed a n L G B T Q e qu al it y sy m bo l as m y F ac eb oo k pr ofi le p ic tu re ). .8 06 .6 94 ( .0 8) ** * .2 59 ( .0 6) ** * 1. 53 S ho w in g U p I pa rt ic ip at e in m y un iv er si ty ’s L G B T Q -s tr ai gh t al li an ce g ro up t o m ak e m y sc ho ol s af er f or a ll s tu de nt s. .7 09 .4 87 ( .0 7) ** * .2 35 ( .0 4) ** * 1. 38 W he n of fe re d, I r eg ul ar ly a tt en d ca m pu s ev en ts t ha t ar e in te nd ed to s up po rt t he L G B T Q c om m un it y (f or e xa m pl e ac ti vi ti es a ss oc i- at ed w it h C om in g O ut W ee k, T ra ns ge nd er A w ar en es s W ee k, P ri de W ee k) . .8 69 .5 78 ( .0 5) ** * .1 08 ( .0 2) ** * 1. 34 W he n of fe re d, I r eg ul ar ly a tt en d co m m un it y ev en ts t ha t ar e in te nd ed t o su pp or t th e L G B T Q c om m un it y (f or e xa m pl e ac ti vi - ti es a ss oc ia te d w it h C om in g O ut W ee k, T ra ns ge nd er A w ar en es s W ee k, P ri de W ee k) . .9 17 .5 61 ( .0 5) ** * .0 6 (. 02 )* ** 1. 31 I ha ve a tt en de d ra ll ie s or p ol it ic al d em on st ra ti on s in s up po rt o f L G B T Q r ig ht s (f or e xa m pl e, s am e- se x m ar ri ag e, e m pl oy m en t no nd is cr im in at io n ac ts , o r ha te c ri m e le gi sl at io n) . .6 74 .5 16 ( .0 7) ** * .3 21 ( .0 5) ** * 1. 23 N ot e. * ** p < .0 01 . 258 JIS Vol. 9, No. 2, 2016 Ta b le 3 L a te n t C o rr e la ti o n s A m o n g t h e E n g a g e m e n t in L G B T Q A ll y A c ti o n s i n S p o rt s S c a le a n d V a li d it y C o n s tr u c ts 1. 2. 3. 4. 5. 6. 7. 8. 9. 1. S ta nd in g U p — 2. S ho w in g U p .7 4* ** — 3. S oc ia l Ju st ic e A tt it ud es .5 6* ** .3 6* ** — 4. A ffi rm at iv e B el ie fs .7 1* ** .3 8* ** .4 7* ** — 5. C om m on F at e Id eo lo gy .5 9* ** .2 8* .4 5* ** .7 5* ** — 6. I nt er ve nt io n in L G B T Q B ia s .6 0* ** .4 0* ** .4 9* ** .5 0* ** .3 4* ** — 7. G en de r -. 43 ** * -. 23 * -. 19 * -. 28 ** * -. 26 ** -. 39 ** * — 8. A ge .1 2 -. 04 .0 6 .1 1 .2 0* -. 14 .2 1* * — 9. B ir th M on th -. 06 -. 11 -. 02 -. 04 .0 3 -. 02 .0 7 .1 2 — N ot e. G en de r w as c od ed 0 = W om en a nd 1 = M en . * ** p < .0 01 . * * p < .0 1. * p < .0 5. Student Athlete Engagement in LGBTQ Ally Actions 259 JIS Vol. 9, No. 2, 2016 theme was labeled Standing and Speaking Up and represented participants’ views that being an ally involves defending the rights of LGBTQ teammates. This theme is illustrated by the following quotation by a woman athlete: Personally, it means that I stand up for those who may have a hard time speak- ing for themselves in a potentially harsh environment. Athletics is not known for being the most supportive of LGBTQ community, but I do what I can to make sure that others don’t attack them just because of their sexual orienta- tion. If a person attacks my teammate for being gay I immediately stand by their side and defend them. This theme of Standing and Speaking Up had one subtheme called Political Action. This subtheme represented participants’ beliefs that being an LGBTQ ally also involves broader involvement at the political level, beyond the microsystem of sport, and is represented by the below quotation from a man athlete: Being an ally to this community certainly means that one would stick up for sexual equality in orientation and prevent clear bullying of individuals that are a part of the LGBTQ community. To me, it would also mean that there is a strict obligation to vote on the matter when it finally comes to that political stage. The second theme within the Action category was labeled as Supporting, which included responses that reflected the importance of providing active support to members of the LGBTQ community and other athletes. The following quote by a woman athlete is an illustration of this theme: “To me an athlete ally means providing physical, emotional and psychological support.” The third and final theme in this category was named Role Modeling, which reflects participants’ beliefs that college student-athletes have a special responsi- bility to be an LGBTQ ally given their privileged and often influential status on college campuses. The following quotation from a man athlete is representative of this theme: “As an athlete I am held to a higher standard and people think more of our stances on issues. For me to voice my opinions on this matter means more than someone else in the community in my opinion.” The second category, Attitude, was comprised of three themes and represented participants’ views that being an athlete ally was associated with certain types of beliefs or understandings about the LGBTQ community. The first theme in this category was labeled Accepting and highlights participants’ beliefs that acceptance plays a vital role in being an athlete ally. It seems important to note that all of the responses coded in this theme were from women student-athletes. A sample quotation from this theme is: “It means that all my teammates can ‘Like’ whoever they want without being judged. I want everyone to feel comfortable and be with who they want to be.” The second theme in the Attitude category was named Equality and Human Rights and reflected the notion that athletes who are allies to the LGBTQ com- munity should believe LGBTQ individuals deserve the same rights and treatment as heterosexual persons. The following quotations by two women student-athletes are representative of this theme: “I believe that wanting everyone to have the same rights as everyone else should be a common goal for everyone. It is a basic human right” and “Everyone should be treated equally, no matter who they love.” One 260 Toomey, McGeorge, and Carlson JIS Vol. 9, No. 2, 2016 subtheme was identified within this theme, which was labeled Sexual Orientation Should Not Matter. This subtheme highlights the belief that a person’s sexual orientation is not relevant in athletics and should not influence how teammates are treated. The following two quotations from women student-athletes provide good illustrations of this theme: “Accepting and respecting every athlete that I see whether it’s a competitor or a teammate. Not classifying fellow athletes as ‘gay’ or ‘straight’ but rather as one unified classification of ‘athlete’” and “It means that I support any person in their athletics regardless of their sexual orientation or gender identification.” The third, and last theme, in the Attitude category was called Non-Judgmental, which describes participants’ views about the importance of not judging LGBTQ teammates. A women athlete shared: “I feel like I am doing the right thing, and I have no right to judge who others love,” while a man athlete wrote “I feel like it is everyone’s job to make everyone feel comfortable around each other and not feel judged.” While the survey question asked participants to describe what it meant to them to be an athlete ally, 11 of the men student-athletes (but no women athletes) responded that they were not an ally to the LGBTQ community. These responses representative our final category, I am Not an Ally. Antecedents of Ally Identity The second open-ended question asked participants to share what prompted student athletes to identify as an ally to the LGBTQ community. The thematic analysis led to the identification of five categories: 1) Personal Beliefs, 2) Relationships with LGBTQ People, 3) Family Socialization, 4) Discrimination, and 5) I am Not an Ally. The first category, Personal Beliefs, reflects participants’ responses about how their own values and morals influenced their decision to be an ally. The following quotations from men student-athletes illustrate this category: “That I believe in equality for all people” and “I believe in equality and think LGBTQ [equality] is in many ways similar to race equality.” There were no themes identified within this category. The second category, Relationships with LGBTQ People, was comprised of three themes and represented participants’ views about the importance of having personal relationships or significant contact with members of the LGBTQ community. The first theme in this category was labeled Family and included participants’ descriptions of the important role that having a family member who identified as LGBTQ played in their development of an identity as an ally. This following quotation by a man athlete is a good illustration of this theme: “Mostly having two cousins that have come out and a couple people I know in high school coming out made me become more of an ally! Knowing someone personally I think changes your whole mindset on that LGBTQ community.” The second theme in this category, Friends and Teammates, describes the significance of having an LGBTQ friend or teammate. Two women student-athletes provided responses that are indicative of this theme: “Hearing stories from LGBTQ friends and teammates and seeing the struggle they face on being judged” and “Personally, there are a lot of LGBT players on my team. Those teammates are not only my friends, but my family and I would love to support them in Student Athlete Engagement in LGBTQ Ally Actions 261 JIS Vol. 9, No. 2, 2016 gaining equality.” The third and final theme in the category of Relationships with LGBTQ People was labeled as Community and School, which highlights the role that interacting with members of the LGBTQ community had in shap- ing these participants’ ally identities. The following quotations by two women student-athletes are representative of this theme: “Meeting people in college that are gay has opened my eyes to how they feel” and “Knowing people in the LGBTQ community.” The third category, Family Socialization, describes the role that college student- athletes’ families of origin played in their development of an identity as an ally to the LGBTQ community. As an example of this category, a woman athlete shared: “My parents raised me to be tolerant of others. They exposed me to the LGBTQ community at a relatively young age, and after that I felt that I was responsible for making sure they got treated equally,” while a man athlete explained “I was taught to love everyone and keep an open-mind with everyone who I encounter.” No themes were identified within this category. Discrimination is the fourth category related to what prompted participants to develop an identity as an LGBTQ ally. This category highlights how experiencing, witnessing, or a sensitivity to discrimination shaped student-athletes’ develop- ment of ally identity. Example quotations from this category include: “I was also treated unfairly at a point in my life” (man athlete), “Being friends with people who identify as LGBTQ, and seeing them be mistreated. It sparks the same anger and injustice as the mistreatment and sexual objectification of women” (woman athlete), and “Seeing kids get bullied and not treated the same upset me” (woman athlete). There were no themes identified within this category. As in the previous section, there were 12 participants (an equal number of women and men) who responded to this question by stating they were not allies to the LGBTQ community and these responses represent the fifth and final category, which was labeled I am Not an Ally. Discussion The results of this pilot study provide initial support for the validity and factor structure of the Engagement in LGBTQ Ally Actions in Sports Scale as a measure of the frequency with which college student-athletes engaged in actions that are con- sidered to be supportive of the LGBTQ community. In addition, the results provide initial support for the validity of the two subscales of the Engagement in LGBTQ Ally Actions in Sports Scale: Standing Up and Showing Up. The establishment of this measure is significant given that this is the first study focused on LGBTQ ally engagement in a sample of college student-athletes. Moreover, the results of this study suggest that the Engagement in LGBTQ Ally Actions in Sports Scale is an equally valid measure for both women and men. In addition to establishing the psychometric properties of the Engagement in LGBTQ Ally Actions in Sports Scale, the findings from the qualitative data lend further support to the items com- prising this scale as well as providing supplemental insight into the LGBTQ ally development process for college student-athletes. One finding about the development of an identity as an ally to the LGBTQ community that seems uniquely relevant for college student-athletes is the belief 262 Toomey, McGeorge, and Carlson JIS Vol. 9, No. 2, 2016 that they have a responsibility to be role models on their campuses for the equal treatment of LGBTQ people. Notably, other studies have found that an important aspect of being an ally to the LGBTQ community involved role modeling behav- iors (e.g., Cunningham, 2015a; Melton & Cunningham, 2014b; Rostosky et al., 2015); however, those actions were not framed by the same heightened status that college student-athletes mentioned in the current study. Of importance, however, the work by Cunningham (2015a) and Melton and Cunningham (2014b) did find the importance of role modeling leader expectations of allyship in the context of employees working in college athletics. The higher status associated with being a college student-athletes may allow them to be more “effective advocates for social change” (Melton, 2015, p. 2). This finding is particularly important given that research posits that interventions may be most successful if they are led by key individuals who hold power within their social networks (e.g., Dearing, 2009). Thus, future work to enhance campus climate could consider the role of college student-athletes in promoting more affirming, safer environments for LGBTQ students, faculty, and staff. It seems important to acknowledge the gender differences that emerged in both our qualitative and quantitative findings. First, when asked about what it means to be an ally, twice as many men athletes reported that they were not allies, while only women athletes discussed the importance of accepting all teammates regardless of sexual orientation. Second, and similarly, in the quantitative data, women athletes were more likely to engage in Standing Up and Showing Up LGBTQ ally-related behaviors. These findings are consistent with research on heterosexuals in the general population suggesting that men have more negative attitudes about sexual minorities than women (Herek, 2002). Further, they are consistent with studies of sport organization employees that found that women were more likely to be sup- portive of LGBT issues (Melton & Cunningham, 2014b). Thus, it will be important for future research to unpack gender differences in athletes’ ally identification and engagement in ally-related actions. Based on the findings of this study, there are important implications for the future use of the Engagement in LGBTQ Ally Actions in Sports Scale. For example, athletic departments could use this scale to assess the types and frequencies with which their student-athletes are engaging in LGBTQ ally behaviors. Further, the scale could be used as a team-level assessment to inform discussions between coaches and players and to help encourage LGBTQ ally behavior among teammates. Limitations and Implications for Future Research As with any research there are limitations to this study that may impact the gener- alizability of the findings. First, while the sample size was more than adequate for our statistical analyses, the sample size was relatively small and lacked diversity in terms of race, economic class, and sexual orientation. It would be important that future researchers seek to recruit college student-athletes from minority groups to ensure broader representation, particularly given research noting the importance of attending to the intersectionality of race, gender, and sexual orientation (Walker & Melton, 2015). Relatedly, our convenience sample was limited to two universities that participate in NCAA Division I-level sports. Given that the sports contexts and campus climates vary widely based on region, importance of sports to the Student Athlete Engagement in LGBTQ Ally Actions 263 JIS Vol. 9, No. 2, 2016 university and surrounding community, these results are limited in terms of their generalizability to athletes in other sport and university contexts. As noted in the Methods section, we had originally hoped to gather data from a third school that would have added substantial diversity to the sample; however, when an athlete in that school engaged in public forms of sexual prejudice via social media the Athletic Director withdrew his support for the study. This example documents the considerable difficulty that still remains in gathering empirical evidence on LGBTQ issues in athletics, and clearly was a contributor to the limitations inherent to this pilot study. Third, it is possible that the student-athletes who chose to participate in this study may have had strong opinions, either negative or positive, about the research topic which may have influenced the findings. Moreover, this study relied on self- report measures and it is possible that the participants either over or under-reported their actions related to being an ally for the LGBTQ community. Finally, with regard to the qualitative data, while the on-line survey methodology allowed for the gathering of a wide range of responses and provided participants with increased anonymity, the asking of follow-up or clarifying questions was not possible. Given the exploratory nature of this study, the research was focused on gathering a breadth rather than depth of responses. The findings of this study suggest a number of directions for future research. For example, to expand on this study, researchers could conduct in-depth interviews with college student-athletes who identify as allies for the LGBTQ community to gain greater insight into the processes by which they developed their identities as allies and how those processes were (or were not) informed by their participation in athletics. Future studies could also more systematically examine whether attend- ing a formal ally training was associated with ally engagement and development among college student athletes. In addition, it may be important to interview or survey LGBTQ college student-athletes about the ally actions that they observe from their teammates as well as the actions and behaviors they would find par- ticularly helpful to improve the climate for LGBTQ student-athletes. Likewise, future research could also examine how LGBTQ ally behaviors are related to other types of civic engagement behaviors engaged in by student-athletes (Weerts, Cabrera, & Perez Mejias, 2014). For example, do student athletes who engage in LGBTQ ally behaviors also engage in other social justice related activities, like issues related to race-ethnicity (e.g., the 2015 protest by University of Missouri football players in response to racial injustices on campus; Tracy & Southall, 2015)? This latter point is particularly intriguing, given the time demands placed on student athletes, particularly those in a Division I sports context (Gayles & Hu, 2009). Beyond studying ally behaviors in college student-athletes, future research could also focus on high school athletics given the increasingly early age at which LGBTQ youth are coming out (Grov, Bimbi, Nanín, & Parsons, 2006; Poteat & Russell, 2013). With slight modifications (e.g., replacing “campus” or “university” with “school”), our items can be made relevant to other school-aged populations (e.g., middle or high schools); these modifications, however, will need to be empiri- cally assessed. A final recommendation is to focus on the attitudes and behaviors of professional athletes, coaches, and sports fans as it relates to creating a safe and welcoming environment for LGBTQ athletes. 264 Toomey, McGeorge, and Carlson JIS Vol. 9, No. 2, 2016 Conclusions Given that the existing research suggests that the sports context may be an unwel- coming environment for LGBTQ student-athletes (GLSEN, 2013; Gill et al., 2010), the development of measures to assess LGBTQ ally actions among college student- athletes is particularly important. To address this need, the current study established the reliability and validity of the Engagement in LGBTQ Ally Actions in Sports Scale. Qualitative analyses suggested that many participants identified as allies to the LGBTQ community, and consistent with prior research, that these identities stemmed from their personal beliefs about equality and fairness, relationships with LGBTQ people, family socialization, and perceptions of discrimination as unjust. Author Note We thank the student-athletes who participated in this study and the athletic department staff who helped disseminate study recruitment announcements. References Anderson, E. (2008). “Being masculine is not about who you sleep with…:” Heterosexual athletes contesting masculinity and the one-time rules of homosexuality. Sex Roles, 58, 104–115. doi:10.1007/s11199-007-9337-7 Anderson, E., & Bullingham, R. (2013). Openly lesbian team sport athletes in an era of decreasing homohysteria. International Review for the Sociology of Sport; Advance online publication. Asta, E.L., & Vacha-Haase, T. (2013). Heterosexual ally development in counseling psy- chologists: Experiences, training, and advocacy. The Counseling Psychologist, 41(4), 493–529. doi:10.1177/0011000012453174 Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3, 77–101. doi:10.1191/1478088706qp063oa Broido, E.M. (2000). The development of social justice allies during college: A phenomeno- logical investigation. Journal of College Student Development, 41(1), 3–18. Creswell, J.W., Plano Clark, V.L., Gutmann, M.L., & Hanson, W.E. (2003). Advanced mixed methods research designs. In A. Tashakkori & C. Teddlie (Eds.), Handbook of mixed methods in social and behavioral research (pp. 209–240). Thousand Oaks, CA: Sage. Cunningham, G.B. (2015a). Creating and sustaining workplace cultures supportive of LGBT employees in college athletics. Journal of Sport Management, 29, 426–442. doi:10.1123/jsm.2014-0135 Cunningham, G.B. (2015b). LGBT inclusive athletic departments as agents of social change. Journal of Intercollegiate Sport, 8, 43–56. doi:10.1123/jis.2014-0131 Cunningham, G.B., & Sagas, M. (2008). Gender and sex diversity in sport organizations: Introduction to a special issue. Sex Roles, 58(1), 3–9. doi:10.1007/s11199-007-9360-8 Cunningham, G.B., Pickett, A., Melton, E.N., Lee, W., & Miner, K. (2013). Psychological safety and expression of sexual orientation personal identity. In J. Hargreaves & E. Anderson (Eds.), Routledge handbook of sport, gender and sexuality (pp. 406–416). New York: Routledge. Curtin, N. (2011). The role of experiences of discrimination, collective identification, and structural awareness in own-group and ally activism (Unpublished doctoral disserta- tion). The University of Michigan, Ann Arbor, MI. Daly, K.J. (2007). Qualitative methods for family studies and human development. Thousand Oaks, CA: Sage Publications, Inc. http://dx.doi.org/10.1007/s11199-007-9337-7 http://dx.doi.org/10.1177/0011000012453174 http://dx.doi.org/10.1191/1478088706qp063oa http://dx.doi.org/10.1123/jsm.2014-0135 http://dx.doi.org/10.1123/jis.2014-0131 http://dx.doi.org/10.1007/s11199-007-9360-8 Student Athlete Engagement in LGBTQ Ally Actions 265 JIS Vol. 9, No. 2, 2016 Dearing, J.W. (2009). Applying diffusion of innovation theory to intervention development. Research on Social Work Practice, 19, 503–518. PubMed doi:10.1177/1049731509335569 Denison, E., & Kitchen, A. (2015). Out on the fields: The first international study on homophobia in sport. Retrieved from http://www.outonthefields.com DeVellis, R.F. (2012). Scale development: Theory and applications (3rd ed.). Thousand Oaks, CA: Sage Publications, Inc. Duhigg, J.M., Rostosky, S.S., Gray, B.E., & Wimsatt, M.K. (2010). Development of het- erosexuals into sexual-minority allies: A qualitative exploration. Sexuality Research & Social Policy, 7, 2–14. doi:10.1007/s13178-010-0005-2 Fingerhut, A.W. (2011). Straight allies: What predicts heterosexuals’ alliance with the LGBT community? Journal of Applied Social Psychology, 41, 2230–2248. doi:10.1111/j.1559- 1816.2011.00807.x Fink, J., Burton, L., Farrell, A., & Parker, H. (2012). Playing it out. Journal for the Study of Sports and Athletes in Education, 6(1), 83–106. doi:10.1179/ssa.2012.6.1.83 Flanagan, C., & Levine, P. (2010). Civic engagement and the transition to adulthood. The Future of Children, 20, 159–179. PubMed doi:10.1353/foc.0.0043 Gay, Lesbian, and Straight Education Network (GLSEN). (2013). The experiences of LGBT students in school athletics (Research Brief). New York: Author. Gayles, J.G., & Hu, S. (2009). The influence of student engagement and sport participation on college outcomes among Division I student athletes. The Journal of Higher Educa- tion, 80, 315–333. doi:10.1353/jhe.0.0051 Gill, D.L., Morrow, R.G., Collins, K.E., Lucey, A.B., & Schultz, A.M. (2010). Perceived climate in physical activity settings. Journal of Homosexuality, 57, 895–913. PubMed doi:10.1080/00918369.2010.493431 Goffman, E. (1963). Stigma: Notes on the management of spoiled identity. New York: Simon & Schuster, Inc. Griffin, P. (2012). LGBT equality in sports: Celebrating our successes and facing our chal- lenges. In G.A. Cunningham (Ed.), Sexual orientation and gender identity in sport: Essays from activists, coaches, and scholars (pp. 1–13). College Station, TX: The Center for Sport Management Research and Education. Griffin, P. (2014). Overcoming sexism and homophobia in women’s sports: Two steps forward and one step back. In J. Hargreaves & E. Anderson (Eds.), Routledge hand- book of sport, gender, and sexuality (pp. 265–274). New York, NY: Routledge., 10.4324/9780203121375.ch28. Griffin, P., & Taylor, H. (2013). Champions of respect: Inclusion of LGBTQ student-athletes and staff in NCAA programs. Indianapolis, IN: National Collegiate Athletic Association. Grov, C., Bimbi, D.S., Nanín, J.E., & Parsons, J.T. (2006). Race, ethnicity, gender, and generational factors associated with the coming-out process among gay, les- bian, and bisexual individuals. Journal of Sex Research, 43(2), 115–121. PubMed doi:10.1080/00224490609552306 Grzanka, P.R., Adler, J., & Blazer, J. (2015). Making up allies: The identity choreography of straight LGBT activism. Sexuality Research & Social Policy; Advance online publica- tion. doi:10.1007/s13178-014-0179-0 Herek, G.M. (2002). Gender gaps in public opinion about lesbians and gay men. Public Opinion Quarterly, 66, 40–66. doi:10.1086/338409 Hoffman, J., Kihl, L., & Browning, A. (2015). Civic engagement and activism: Do college students and college athletes differ? Journal of College and Character, 16, 75–89. doi :10.1080/2194587X.2015.1028315 Ji, P., Du Bois, S.N., & Finnessy, P. (2009). An academic course that teaches heterosexual students to be allies to LGBT communities: A qualitative analysis. Journal of Gay & Lesbian Social Services, 21(4), 402–429. doi:10.1080/10538720802690001 Kenny, D.A., & McCoach, D.B. (2003). Effect of the number of variables on measures of fit in structural equation modeling. Structural Equation Modeling, 10(3), 333–351. doi:10.1207/S15328007SEM1003_1 https://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=20976022&dopt=Abstract http://dx.doi.org/10.1177/1049731509335569 http://www.outonthefields.com http://dx.doi.org/10.1007/s13178-010-0005-2 http://dx.doi.org/10.1111/j.1559-1816.2011.00807.x http://dx.doi.org/10.1111/j.1559-1816.2011.00807.x http://dx.doi.org/10.1179/ssa.2012.6.1.83 https://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=20364626&dopt=Abstract http://dx.doi.org/10.1353/foc.0.0043 http://dx.doi.org/10.1353/jhe.0.0051 https://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=20665330&dopt=Abstract http://dx.doi.org/10.1080/00918369.2010.493431 https://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=16817058&dopt=Abstract http://dx.doi.org/10.1080/00224490609552306 http://dx.doi.org/10.1007/s13178-014-0179-0 http://dx.doi.org/10.1086/338409 http://dx.doi.org/10.1080/2194587X.2015.1028315 http://dx.doi.org/10.1080/2194587X.2015.1028315 http://dx.doi.org/10.1080/10538720802690001 http://dx.doi.org/10.1207/S15328007SEM1003_1 266 Toomey, McGeorge, and Carlson JIS Vol. 9, No. 2, 2016 Kizer, B. (2011). Construction of the lesbian and gay affirming social justice competency scale (Unpublished doctoral dissertation). University of Missouri – Kansas City, Kansas City, MO. Kline, R.B. (2016). Principles and practices of structural equation modeling (4th ed.). New York: The Guilford Press. Knight, J.L., & Giuliano, T.A. (2003). Blood, sweat, and jeers: The impact of the media’s heterosexist portrayals on perceptions of male and female athletes. Journal of Sport Behavior, 26, 272–284. Krane, V. (1996). Lesbians in sport: Toward acknowledgment, understanding, and theory. Journal of Sport & Exercise Psychology, 18(3), 237–246. doi:10.1123/jsep.18.3.237 Krane, V., Choi, P.L., Baird, S.M., Aimar, C.M., & Kauer, K.J. (2004). Living the paradox: Female athletes negotiate femininity and muscularity. Sex Roles, 50(5-6), 315–329. doi:10.1023/B:SERS.0000018888.48437.4f Lincoln, Y.S., & Guba, E.G. (1985). Naturalistic inquiry. Newbury Park, CA: Sage Publica- tions, Inc. Little, T.D., Card, N.A., Slegers, D.W., & Ledford, E. (2007). Representing contextual effects in multiple-group MACS models. In T.D. Little, J.A. Bovaird, & N.A. Card (Eds.), Modeling contextual effects in longitudinal studies (pp. 121–148). New York: Routledge. McCormack, M. (2012). The declining significance of homophobia: How teenage boys are redefining masculinity and heterosexuality. New York, NY: Oxford University Press. doi:10.1093/acprof:oso/9780199778249.001.0001 Melton, E.N. (2015). Creating social change in and through intercollegiate sport. Journal of Intercollegiate Sport, 8, 1–13. doi:10.1123/jis.2015-0017 Melton, E., & Cunningham, G. (2012). When identities collide. Journal for the Study of Sports and Athletes in Education, 6(1), 45–66. doi:10.1179/ssa.2012.6.1.45 Melton, E.N., & Cunningham, G. (2014a). The experiences of LGBT sport employees: A social categorization perspective. Journal of Sport Management, 28, 21–33. doi:10.1123/ jsm.2011-0157 Melton, E.N., & Cunningham, G. (2014b). Who are the champions? Using a multilevel model to examine perceptions of employee support for LGBT-inclusion in sport organizations. Journal of Sport Management, 28, 189–206. doi:10.1123/jsm.2012-0086 Muthén, B.O., & Muthén, L.K. (2014). Mplus Version 7. Los Angeles, CA: Authors. [Com- puter software]. Pew Research Center. (2015, June). Support for same-sex marriage at record high, but key segments remain opposed. Retrieved from http://www.people-press.org/2015/06/08/ support-for-same-sex-marriage-at-record-high-but-key-segments-remain-opposed/ Poteat, V.P., & Russell, S.T. (2013). Understanding homophobic behavior and its implica- tions for policy and practice. Theory into Practice, 52(4), 264–271. doi:10.1080/004 05841.2013.829729 Rankin, S., Merson, D., Sorgen, C. H., McHale, I., Loya, K., & Oseguera, L. (2011, June). Student-athlete climate study (SACS) final report. Indianapolis, IN: National Collegiate Athletic Association. Rostosky, S.S., Black, W.W., Riggle, E.D.B., & Rosenkrantz, D. (2015). Positive aspects of being a heterosexual ally to lesbian, gay, bisexual, and transgender (LGBT) people. The American Journal of Orthopsychiatry, 85, 331–338. PubMed doi:10.1037/ort0000056 Russell, G.M. (2011). Motives of heterosexual allies in collective action for equality. The Journal of Social Issues, 67(2), 376–393. doi:10.1111/j.1540-4560.2011.01703.x Sartore-Baldwin, M.L., & Cunningham, G.B. (2010). The lesbian label as a component of women’s stigmatization in sport organizations: An exploration of two health and kinesiology departments. Journal of Sport Management, 24(5), 581–581. Sartore-Baldwin, M.L., & Warner, S. (2012). Perceptions of justice within intercollegiate athletics among current and former athletes. Journal of Issues in Intercollegiate Athlet- ics, 5, 269–282. http://dx.doi.org/10.1123/jsep.18.3.237 http://dx.doi.org/10.1023/B:SERS.0000018888.48437.4f http://dx.doi.org/10.1093/acprof:oso/9780199778249.001.0001 http://dx.doi.org/10.1123/jis.2015-0017 http://dx.doi.org/10.1179/ssa.2012.6.1.45 http://dx.doi.org/10.1123/jsm.2011-0157 http://dx.doi.org/10.1123/jsm.2011-0157 http://dx.doi.org/10.1123/jsm.2012-0086 http://www.people-press.org/2015/06/08/support-for-same-sex-marriage-at-record-high-but-key-segments-remain-opposed/ http://www.people-press.org/2015/06/08/support-for-same-sex-marriage-at-record-high-but-key-segments-remain-opposed/ http://dx.doi.org/10.1080/00405841.2013.829729 http://dx.doi.org/10.1080/00405841.2013.829729 https://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=25798894&dopt=Abstract http://dx.doi.org/10.1037/ort0000056 http://dx.doi.org/10.1111/j.1540-4560.2011.01703.x Student Athlete Engagement in LGBTQ Ally Actions 267 JIS Vol. 9, No. 2, 2016 Schlomer, G.L., Bauman, S., & Card, N.A. (2010). Best practices for missing data man- agement in counseling psychology. Journal of Counseling Psychology, 57(1), 1–10. PubMed doi:10.1037/a0018082 Strotzer, R.L. (2009). Straight allies: Supportive attitudes toward lesbians, gay men, and bisexuals in a college sample. Sex Roles, 60, 67–80. doi:10.1007/s11199-008-9508-1 Taylor, H. (2015). Activating change through allyship. Journal of Intercollegiate Sport, 8, 37–42. doi:10.1123/jis.2015-0025 Tracy, M., & Southall, A. (2015, November 8). Black football players lend heft to protests at Missouri. The New York Times. Retrieved from http://www.nytimes.com Walker, N.A., & Melton, E.N. (2015). The tipping point: The intersection of race, gender, and sexual orientation in intercollegiate sports. Journal of Sport Management, 29, 257–271. doi:10.1123/jsm.2013-0079 Washington, J., & Evans, N.J. (1991). Becoming an ally. In N.J. Evans & V.A. Wall (Eds.), Beyond tolerance: Gays, lesbians and bisexuals on campus (pp. 215–224). Alexandria, VA: American Association for Counseling and Development. Weerts, D.J., Cabrera, A.F., & Perez Mejias, P. (2014). Uncovering categories of civically engaged college students: A latent class analysis. The Review of Higher Education, 37, 141–168. doi:10.1353/rhe.2014.0008 Worthen, M.G. (2014). Blaming the jocks and the greeks?: Exploring collegiate athletes’ and fraternity/sorority members’ attitudes toward LGBT individuals. Journal of College Student Development, 55, 168–195. doi:10.1353/csd.2014.0020 https://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=21133556&dopt=Abstract https://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=21133556&dopt=Abstract http://dx.doi.org/10.1037/a0018082 http://dx.doi.org/10.1007/s11199-008-9508-1 http://dx.doi.org/10.1123/jis.2015-0025 http://www.nytimes.com http://dx.doi.org/10.1123/jsm.2013-0079 http://dx.doi.org/10.1353/rhe.2014.0008 http://dx.doi.org/10.1353/csd.2014.0020