Abstract
This article explores the complex relationship between an openly gay instructor, homophobia, and heteronormativity in a university classroom. The authors first tabulated the frequency with which the instructor used the lives of heterosexuals and homosexuals as examples of content or as content itself, and then they interviewed 32 students about their perceptions of these frequencies. They found that students significantly overestimated lesbian, gay, bisexual, and transgender (LGBT) frequencies and underestimated heterosexual ones. The authors develop two analytical concepts to highlight this form of heteronormativity: novelty attachment and content substitution. They explain these phenomena by suggesting that the novelty of using LGBT examples and discussing homosexuality as content results in the activation of stereotypes among otherwise gay-friendly students. They examine the cognitive underpinnings of this using social identity theory and call for further research to examine the applicability of their theory to other minority groups.
An extensive body of research focuses on homophobia in educational settings, highlighting the cultural and institutional discrimination that lesbian, gay, bisexual, and transgender (LGBT) individuals experience at both the student and faculty levels (Ferfolja 2007; Mills 2004; Rivers 1995). For this reason, there may be good motive for some teachers and lecturers to remain closeted. Interviewing 120 Australian LGBT educators (from the primary to university level), Irwin (2002) showed that there was overwhelming homophobia, harassment, and discrimination. Most of this came from administrators whose homophobia led to LGBT educators “being overlooked for promotion, not being offered the same opportunities as heterosexual staff and the sabotaging of work” (Irwin 2002:68). Consequently, many teachers remain closeted at work, and pupils are therefore bereft of exposure to sexual diversity during their formative years (Sands 2009).
While university instructors likely maintain better social and institutional support for coming out than other educators do, homophobic discrimination still occurs, particularly concerning teaching evaluations. Russ, Simonds, and Hunt (2002) evidenced the ways in which coming out in a university setting can be understood as what they call an “occupational hazard.” They hired a professional speaker to deliver an identical lecture to eight classes. Students were told that the guest speaker was applying for a post. To examine for attitudes toward homosexuality, the guest lecturer identified as homosexual by mentioning his partner, Jason, in four classes, and he revealed heterosexuality by mentioning his partner, Jennifer, in four other classes. After each class, students were asked to fill out an evaluation of the speaker. The students scored the speaker lower on a measure of credibility when he identified as homosexual, also rating him as less knowledgeable. Highlighting the vast difference in responses to the guest lecturer, 93% of students suggested that they would “unquestionably” hire him when they thought he was straight, while only 8% of students suggested that they would “unquestionably” hire him when he identified as gay. These findings were echoed by Anderson and Kanner (2011), who found that undergraduate students perceived gay and lesbian professors as having a political agenda when compared with heterosexual professors with the same syllabus. Given the importance placed on student evaluations in higher education, this is a serious issue in the promotion of equality and diversity within university settings.
Classroom homophobia has other serious consequences in the education of university students, as teachers concerned with being thought heterosexual must prove and reprove their sexuality (Taulke-Johnson 2010). Francis and Skelton (2001) showed how this can play out in the classroom, where male teachers wishing to be perceived as heterosexual use homophobic discourse in order to displace suspicion of homosexuality. They also found that closeted gay male teachers used sexual innuendos when talking to female students. Thus, institutional pressures to remain closeted (and act heterosexual) might ultimately result in orthodox forms of gender presentation as well as a diminution of gender and sexual orientation diversity. This limits the ability of educational settings to erode sexual inequalities, even at the university level (Jones 2007; Steeves et al. 2008).
There are, however, a number of cultural trends related to sexuality and gender that may influence the relationship between homophobia and the university. The most salient concerns the rapid reduction of cultural homophobia among undergraduate men (Anderson 2009; Kozloski 2010; Taulke-Johnson 2008). This has increased the social legitimacy of alternative categories of sexuality for college students and expanded their social and political landscapes (Anderson 2008; McCormack and Anderson 2010b). For example, Bush, Anderson, and Carr (forthcoming) have shown that even incoming male university athletes maintain near unanimous support for having a gay player on their university-based sport teams.
It is perhaps understandable that research finds decreasing levels of homophobia at the undergraduate level, as recent research suggests this change in attitudes toward homosexuality among 16- to 18-year-old boys has been dramatic. McCormack (2011a, 2011b, 2012) not only finds high school boys espousing pro-gay attitudes, but he argues that students are also growing increasingly aware of the complex manner in which heteronormativity operates. Still, despite some theoretical work on heteronormativity in schools (Epstein, O’Flynn, and Telford 2003), there has been little empirical examination on the implicit ways in which heterosexuality is privileged within educational settings; thus, it is necessary to examine the mechanisms of heteronormativity in educational environments (McCormack and Anderson 2010a; Taulke-Johnson 2008).
The reproduction of heteronormativity occurs at both the institutional and interactional levels. At the institutional level this has occurred through formal policies, such as Section 28 in the United Kingdom, which resulted in a perception among many teachers that LGBT issues could not be discussed in schools (Nixon and Givens 2007). Until its repeal in 2003, teachers felt unable to address anti-gay language and homophobic bullying (Epstein and Johnson 1998). Institutional heteronormativity is also evidenced in educational textbooks where either LGBT people are erased (Schanz and Mitchell 2009; Snyder and Broadway 2004) or homosexuality is portrayed as something dangerous (Macgillivray and Jennings 2008). For example, Snyder and Broadway (2004) examined high school textbooks, finding that the term homosexuality was used only in discussions of AIDS. Furthermore, Zack, Mannheim, and Alfano (2010) suggest that the current content of teacher training programs continues to result in future educators not receiving adequate tools to critically discuss sexualities and therefore contest heteronormativity.
At an interactional level, many LGBT teachers conform to and therefore reproduce heteronormativity by remaining closeted (Epstein and Johnson 1998; Nixon and Givens 2007). But when educators have been open about their LGBT lives in the classroom, they often conform to heteronormative expectations of gender and sexuality (Rofes 2000). Tajfel and Turner’s (1979) social identity theory is useful for understanding these actions, as it might suggest that these educators are managing their behaviors and identities to conform to the norms of the wider heteronormative classroom in order to avoid out-group categorization (Griffin 1991).
Despite these theoretical and empirical examinations, however, there has been very little examination concerning the cognitive processes of heteronormativity in the classroom setting. Accordingly, in this research we first tabulated the frequency with which an openly gay instructor (Eric Anderson) used the lives of heterosexuals and homosexuals as examples of content or as content itself in a university classroom. We then interviewed the students about their perceptions of these frequencies, finding that students significantly overestimated LGBT frequencies and underestimated heterosexual ones.
Method
In this research, we combine observational data, questionnaires, and in-depth interviews. Data collection started when a heterosexual instructor conducted a survey on the total population (85) of a first-year cohort of students enrolled in a sports program at a British university during 2010 freshman orientation—students’ first day of attendance at this university. Students’ attitudes toward homosexuality were measured using several questions on a five-point Likert-type scale (Bush, Anderson, and Carr forthcoming). For example, students were asked to indicate their level of agreement with the statement “I think that a gay coach should not be allowed to coach heterosexual male athletes.”
This survey was conducted one week prior to students attending their first class of a unit led by the openly gay instructor, whom they had yet to meet. The rest of this department’s faculty were instructed not to reveal that one of their colleagues was openly gay, guarding against biasing the data through students’ knowledge that they would have a gay instructor. Unlike in the United States, where a schedule of classes indicates the assigned instructor for any given class, first-year modules at this English university are mandatory, and information on the assigned instructor is not available to incoming students.
The instructor (the second author) disclosed his homosexuality during the first 10 minutes of the first session of this sport sociology class by discussing his experiences as an openly gay coach. He then answered questions about being a gay man in sport before continuing with the primary content of the lesson. He was the sole conveyor of this 10-week class.
Two of the other authors observed the openly gay second author teach this weekly sport sociology class. The two researchers independently noted the number, style, and type of verbal examples (stories to illustrate content) given by the instructor during class. Attention was paid to whether the examples concerned the lives of LGBT or heterosexual people.
Illustrating our coding of a heterosexual example, while discussing the importance of regular exercise, the lecturer might have said, “Jason and his wife, Susan, go for a 30 minute jog every night after work.” Conversely, a homosexual example with the same content is illustrated with, “Jason and his husband, Mark, go for a 30 minute jog every night after work.”
These same researchers also recorded each time the lecturer discussed homosexuality or heterosexuality as content. For example, we coded “homosexuality” when discussing gay men in sport and “heterosexuality” when discussing how female athletes have fewer teenage pregnancies. The two researchers were positioned in the back two corners of the lecture hall so that they were out of student view for their note taking.
At the end of the second week, these researchers co-verified their results. It was determined that there was an 8 percent difference in the quantity of comments registered by each researcher. Each subsequent week saw lower differences between the two researchers. However, there was complete agreement concerning whether a topic concerned homosexuality or heterosexuality. Accordingly, this co-verification process permitted a level of mutual consistency as to what constituted a heterosexual or homosexual example, ultimately leading to more valid data (Denscombe 2002).
We next measured the students’ perceptions of the number of gay and straight examples and content that the instructor used each week. This was accomplished by interviewing 32 students throughout the next eight weeks of the course. We conducted four one-on-one interviews per week, asking two men and two women to volunteer for a 20-minute interview. We announced to the class that the interviews were about pedagogy, making no mention of sexuality. Interviews were conducted in private, immediately after the end of each class.
We had many more weekly volunteers than we could interview and so we randomly selected students from the pool of weekly volunteers. To avoid potential bias, the two note-taking researchers conducted half of the interviews each (May 2002). The 32 university students, age 18 to 21, all self-identified as heterosexual. They hailed exclusively from white, middle-class backgrounds.
A semistructured interview schedule was used to explore the participants’ perceptions of the quantity of homosexual and heterosexual examples the instructor gave in class. We asked for the examples the students could recall; we also asked them to suggest a ratio they thought best described the homosexual–heterosexual frequency (i.e., 1:1, 2:1). We asked similar questions concerning the number of times the instructor addressed homosexual and heterosexual content.
As part of the interview, the participants were given an explanation to help them distinguish between example and content. Interviewers said, “A couple, John and Jennifer, decided not to take their kids to a NHL game because they didn’t approve of the violence. The content is that professional sport is violent. He used an example of a heterosexual couple to illustrate that content.” Students were then given an example of sexuality—this time, a homosexual one—to illustrate sexuality as content: “Gay athletes who come out to their teams only do so after informally evaluating their teammates’ attitudes toward homosexuality.” Although we recognize that separating example from content can sometimes be a high-order analytic skill, we believe that our explanation helped students differentiate between them. 1
After discussing content and examples, we asked questions regarding their views on the quality of lecturing, whether they thought the instructor’s sexuality impacted this, and their attitudes toward homosexuality. Upon completion of the interview, students were directed not to tell their peers that they were asked about sexual examples versus content. Instead, they were instructed to tell their peers that the questions centered on the instructor’s teaching style. We have confidence that this was an effective strategy. Evidencing this, before beginning each new interview, students were asked if they had any idea what the interview was going to be about: None indicated they knew the interview was going to be about sexuality.
Finally, it should be noted that because part of this research was designed to assess whether undergraduate students conceptualize heteronormativity as a form of homophobia, the way McCormack (2012) shows British high school students do, it was necessary for interviewers to question participants’ perceptions when they were significantly inaccurate. This had the added benefit of highlighting (and hopefully reducing) heteronormativity. It should be noted that interviewers started to question participants only after data regarding perceptions of content and example were collected.
Transcriptions from the 32 completed interviews were then coded for predetermined categories and cross-checked by each researcher, including the instructor and third author. Anonymity was assured for all participants, with informed consent acquired from interviewees. All 85 students were told that the lecturer was being observed by two researchers for study purposes. There was no financial or educational reward for taking part in the research.
Results
Classroom Inclusivity
Survey results of the 85 students showed strong support for gays and lesbians. For example, on the statement “I think that a gay coach should not be allowed to coach heterosexual male athletes,” no student marked strongly agreed or agreed with the statement. One student had a neutral opinion, one disagreed, and the remaining 83 students strongly disagreed. Similar results were found with the other three questions to measure homophobia. This survey therefore suggests that students who volunteered to be interviewed were not biased in their support of homosexuality.
Data from our interviews also evidenced pro-gay attitudes. Carol said, “Even though I’m straight, I feel strongly about supporting gay issues. I don’t understand how it’s still an issue with some people.” Ben said, “I have gay and bisexual friends; it’s never really been an issue with me.” Katy agreed: “There are two lesbians on my debate team; I have never considered them as anything but equal.” Most of these students indicated that they had either gay friends, neighbors, or family members. Highlighting the familiarity these students had with LGBT people, Colleen said, “I’ve got gay friends. Don’t we all?”
Views supporting homosexuality were also evidenced by students’ positive opinions of the openly gay instructor. Given that research has highlighted homophobic responses to openly gay university instructors (Russ et al. 2002), it is significant that there were no negative comments about his sexuality. While it is possible that students feared speaking harshly of him to the observing researchers (both of whom were heterosexual), the quality of comments illustrates evidence of their sincerity. For example, Claire said,
His approachableness is unlike any other of my lecturers. You can talk to him about anything. I mean, he even gives you his mobile number and encourages you to text him. And he is incredible to watch teach. This class is the only class that every student attends. I’ve even brought my housemates to see his lectures.
In further contrast to Russ et al.’s (2002) study, where students used the perceived homosexuality of an instructor to discount his legitimacy, this lecturer’s sexuality was not used against him in formal student evaluations. In fact, the evaluation scores for this course averaged 4.7 out of 5.0 (the average at this university is 3.6).
Some students argued that sexuality was irrelevant to the instructor’s quality. Claire said, “He’s a good lecturer, and it makes no difference if he’s gay, straight, or whatever; he’s just a good lecturer.” However, other students indicated that they were pleased to have the second author teach this class because of his sexuality. For example, Kieran said, “I think it’s great that I have a gay lecturer.” Tom agreed:
You don’t get this in sixth form [high school]. It’s about time I had a gay teacher. . . . His sexuality is a benefit. . . . It has allowed me to think about issues surrounding a minority that I had not thought about before. I haven’t got any other gay friends myself but I like to consider him [the instructor] as my mate now.
In discussing their views of the instructor, however, it also became apparent that students attached onto the instructor’s sexuality, seeing him primarily as “gay” (see also Steeves et al. 2008). When we asked Kieran how he would describe the lecturer, he laughed and answered, “As gay.” He added, “It’s strange, I have no issues with gay people at all, but I just can’t stop seeing him as gay.”
Kieran was not the only student who viewed the instructor primarily as gay. When asked to describe the instructor, 28 of the 32 participants first described him as “gay,” while the other 4 described him as “American.” When asked how she would describe the lecturer, Carol said, “I would describe him as gay.” The researcher asked, “Like he’s really flamboyant or something?” “No,” Carol answered. “I don’t think he is more or less masculine than the guys in class, but he is the gayest.”
In contrast to this, students do not attach a master identity of “heterosexual” to other instructors who teach in this department. Accordingly, when Carol was asked how she saw Dr. Bush (the teacher who introduced the survey and teaches another module), she replied, “I don’t know. He’s just Dr. Bush. . . . He’s nice.” This labeling of homosexuality and normalizing of heterosexuality comes from an attachment to the “exotic,” which comes to define the gay instructor. However, seeing their instructor’s master identity as gay was not the only way students attached to homosexuality.
Novelty Attachment
During this 10-week course, the second author used more examples to illustrate the lives of heterosexuals than to illustrate the lives of homosexuals. Out of 115 examples, 39% (n = 45) were coded by the two researchers as “gay,” while 61% were coded as “straight” (n = 70). However, students estimated a far greater number of homosexual examples. When we asked interviewees to quantify the number of heterosexual to homosexual examples, their average score did not reflect this reality. Although the ranges varied from 1:1 to 10:1, most students suggested that the instructor used significantly more homosexual examples, with the average student suggesting that the ratio was 4:1. No student thought the instructor used more heterosexual examples than homosexual examples.
Despite these figures, the students were adamant that gay examples were more frequent. For example, Claire said, “Homosexuality. By far . . . I’m not complaining about that. But definitely homosexuality.” When Claire was asked for her estimate of the ratio of gay to straight examples, she answered, “I’d guess 4 to 1?” Dom suggested that the ratio was much higher: “10 to 1. It’s easily 10 to 1. I’ve never heard so much gay stuff talked about in my life. Don’t get me wrong, I’ve enjoyed it and I’ve never had such a good lecturer; but he’s all about gay.”
Dom was unable to recall an example of heterosexuality from the previous lecture, yet he was readily able to discuss a homosexual one. Similarly, when Jordan was asked how many times the lecturer gave gay examples (compared with straight examples), he said, “They’re all gay.” The researcher interviewing him asked, “Are you saying that every single example he gave in class was about the lives of homosexuals?” Jordan answered, “I don’t remember him talking about heterosexuals. He talked about gay people a lot though.”
This highlights that although students maintained respect for their openly gay lecturer, they nonetheless viewed his actions through a heteronormative lens. Accordingly, students remembered homosexual examples because of their novelty, but they failed to notice (or remember) heterosexual examples—even though this was the numerically dominant category. We describe this process as novelty attachment.
Novelty attachment not only existed in the giving of examples but also was manifest in lecture content. Just as students overestimated the frequency of homosexuality as examples, they also overestimated the frequency of discussions about homosexuality. For example, Martin said, “I’d say that he talked about homosexuality a lot. Like at least 10 times a lecture. Maybe he talks about heterosexuality sometimes, but I can tell you for sure that he talked about homosexuality all the time.” Karen agreed, saying that she brought her flatmates to class with her (friends not enrolled in the class) precisely because he “constantly talks about homosexuality.” But when Karen was asked how frequently the lecturer also discussed heterosexuality, she replied, “To be fair, he does. Not as much, but every now and then he does. Like I remember one time he was talking about . . . [provides example] . . . so he does. But nothing like he does about gay stuff.” Again, we highlight that the operation of ordinary is that it goes unnoticed and therefore unremembered.
We do not provide a ratio concerning the average of students’ estimations of the number of times the instructor discussed homosexuality as content compared with heterosexuality as we did with the examples of homosexuality and heterosexuality. This is because most students did not perceive that the lecturer ever talked about heterosexuality. For example, when asked to provide a ratio, Tom said, “What do you mean? He doesn’t talk about heterosexuality.”
Content Substitution
One of the reasons why students overestimated the number of times that the instructor talked about homosexuality as content is because the students interpreted examples of homosexuality as content. For example, one time the instructor was discussing the expense of buying tickets to a professional sporting match, giving an example of how “Rob and his husband” were unable to afford them. Three of the four students interviewed after this lecture erroneously listed this as an example of a time in which the instructor talked about homosexuality as content. They did not recognize the content as about the ability of sport to highlight economic matters.
Another example comes from Liam. When asked to give examples of when the lecturer discussed homosexuality as content, Liam said, “Yeah, I can tell you about that. He was lecturing about gay footballers today.” When the researcher said, “Yes, he mentioned a gay football league, but what was the point of that part of the lecture?” Liam responded, “What do you mean? It was about gay players.” The researcher responded,
Actually, I think that the point was made about gay leagues in context to how there are actually multiple types of sporting leagues. . . . The point he was trying to make was that when we think of sport we immediately think of men’s professional sport, and not these other types of sports.
Upon hearing the actual context of the gay example, Liam replied, “I guess I didn’t see it that way. To me he was talking about gay players, but now that you mention it, I guess I can see your point.”
This morphing of examples into content did not occur with heterosexual examples. Highlighting this, one of the researchers said to a student, “Do you remember when Eric was talking about how he knew this couple [who were described as heterosexual in the lecture] who made it their mission to run a race every weekend?” The student indicated that he recalled. “What was the point of that example?” the interviewer asked. The student responded (correctly), “It had to do with issues around overdedication, didn’t it?”
This finding illustrates the significant pedagogical implications of our research. While delineating content from examples can sometimes be somewhat complex, the success of doing this for heterosexual examples demonstrates that these students were intellectually capable of completing this task. Accordingly, this research indicates that the students, who are so used to hearing examples about the lives of heterosexuals, cognitively “attach” to the content of heterosexual examples, therefore remembering it and gaining the educational objectives. Conversely, when this instructor uses examples of homosexuals, students in this class instead attach to the example, often forgetting what concept the example was designed to illustrate in the first place. This demonstrates a need for more discussion of homosexuality in educational settings, something we address in the Discussion section.
Interpreting Heteronormativity as Homophobia
At the end of each interview, the researcher disclosed the topic of investigation. On hearing that the lecturer actually gave examples in a ratio of two heterosexual examples to every one homosexual example, participants first expressed surprise. Joanne exclaimed, “I can’t believe it! I was certain that he used more gay examples than straight ones.” Tom asked, “Are you sure? I wasn’t that wrong, was I?”
After expressing their surprise, the majority of the students feared that they were being homophobic by having misjudged the ratio of examples. Accordingly, the participants were quick to reestablish their support for homosexuality. Carol said, “I feel awful. Trust me, I’m not homophobic.” Without an accusation of homophobia, Grant also defended himself, “Great, so now I’m a homophobe? . . . I’m not.” Gary responded as if the researcher (both other classroom researchers were openly straight) was offended, “I’m so sorry. I didn’t mean it in a homophobic way. I love Eric.” Jack, who gave us a ratio of 10:1, said, “No way, I can’t believe I got it so wrong. I hope you don’t think I’m in any way homophobic because I said that.”
Many of the students explained their error as a result of the lack of exposure to homosexuality. “My sixth form [high school] was pretty homophobic. . . . The students weren’t, but the teachers were.” Anthony continued, “I haven’t heard any other lecturer mention homosexuality even once, in any of their lectures, ever. That’s pretty much homophobia.” Accordingly, these students not only were shocked by the degree to which they were wrong but attributed their errors to the lack of diversity in their earlier education. Our students’ understanding of heteronormativity as homophobia was also found by McCormack (2012) in his study of students in three British high schools.
Discussion
In this study we investigated the presence of heteronormativity among gay-friendly university students. First, we formally surveyed them for their attitudes about homosexuality. This occurred during their first day of university attendance of their freshman year. A week later they were enrolled in a 10-week sport sociology unit, which they were compelled to take. Here, we tallied the instructor’s examples and content based on heterosexuality or homosexuality. We then interviewed 32 students about their perceptions of these frequencies, finding that students unintentionally overestimated the numbers of homosexual examples and content while simultaneously underestimating those regarding heterosexuality.
We conceptualize this perception error through novelty attachment, arguing that students remember the homosexual examples because of their rarity. Similarly, the students also overestimated the number of times the instructor spoke of LGBT content. This is because they morphed examples of LGBT individuals into content, a process we call content substitution. Seemingly, homosexual examples were so novel to these students that they falsely assumed the instructor was talking about homosexuality.
Both of these findings can be understood through social identity theory (Tajfel and Turner 1979). This theory argues that people categorize others through stereotypes associated with the othered group. This is as a way of maintaining group distinctiveness, serving to enhance the individual’s self-perception of those in the in-group (Hogg, Terry, and White 1995). We argue that novelty attachment and content substitution are two cognitive processes that enable these students to protect their in-group identities from the “threat” of a gay instructor whose very existence challenges the heteronormative cohesion of the group.
To explain this in more detail, we turn to the social psychological literature on bias and stereotyping (Brewer and Harasty 1996; Kunda et al. 2002). In examining these processes, we first argue that novelty attachment occurs because these students are not used to hearing gay examples in class: They therefore attach onto them, effectively being distracted from the wider point. While this may have pedagogical implications in and of itself, the problem is exacerbated because these students have been socialized within a heteronormative society: When the instructor talks about gay examples, he is effectively placing himself as part of a different social group to the students and simultaneously challenging their heteronormative worldview.
It is well know that stereotypes are invoked to understand an individual of a different social group (Brewer and Harasty 1996), but of importance to this study, Kunda et al. (2002) demonstrate that stereotypes are often invoked when people of one group are challenged in their thinking by someone from another group. In accordance with Kunda et al.’s (2002) findings that stereotypes of others reactivate after someone goes against the group’s belief, we suggest the instructor’s use of gay examples reactivated the students’ stereotype of a gay man whom they suppose not only has “an agenda” but is obsessed by homosexuality (Russ et al. 2002). This unrecognized bias resulted in the students perceiving their instructor to be repeatedly discussing homosexuality, substituting his examples into content because they were interpreting his arguments through a lens blurred by stereotype.
Given that the literature on stereotyping and social identity theory is generalizable across social groupings, it is important to explore the applicability of our findings to other socially marginalized groups. It is quite possible that there are parallel effects in relation to how women are viewed within predominantly male academies (Cameron 2007), as well as similar issues for race, disability, or other categories of an instructor’s difference. We believe it likely that novelty attachment and content substitution may occur in analogous ways, dependent on the novelty of the minority group and the extent to which the instructor challenges the group’s collective identity.
There are several important sociological and policy implications to this research; however, care is needed in determining what these implications are. For example, ostensibly it could be argued that using examples of sexual minorities to illustrate curricula could be considered poor pedagogy. This is because we have shown that novel examples (in this case, those of LGBT people) can distract students from the primary objective of learning curricular content, particularly given the relative absence of openly gay teachers in both the United Kingdom and the United States. However, failure to contest the dominance of heterosexuality also permits the continual erasure of sexual minorities—fostering a form of oppression. And because LGBT people are underrepresented in schools, failure to use examples of the lives of LGBT people essentially makes education antidemocratic (Dewey 1916). In the United Kingdom, this is in contravention of equalities law and against the guidelines of the national school inspectorate.
Furthermore, the continued erasure of homosexuality should be, and increasingly is, interpreted as, an intentional act of homophobia. As McCormack (2012) and McCormack and Anderson (2010a) find with high school students, the denial of homosexual exposure is not read as a privileging of heterosexuality but, rather, as an intentional act of homophobia.
Another important implication to this research is that because these students (who hail from throughout the United Kingdom) had not heard their teachers use gay examples or gay content before arriving to university, heteronormativity remains high at the primary and secondary levels of education. One reason for this might be the failure of teacher education programs to provide student teachers with information concerning how to talk about sexual minorities (Mills 2004; Robinson and Ferfolja 2007; Zack et al. 2010). This finding therefore makes salient that even in a country with a relatively positive outlook on homosexuality, heterosexuality still maintains hegemony in classroom discourse. Accordingly, teachers who use examples of homosexuality in the classroom subject themselves to their students’ attachment bias.
The last implication of this research is that because the combination of novelty attachment and content substitution provides the illusion that the instructor was discussing homosexuality even when he was not, it raises concerns as to how homophobic students will evaluate an instructor who uses examples of LGBT individuals in his or her classroom. This has the potential of harming a teacher, both in terms of unit evaluations and possible complaints about an LGBT educator “having an agenda” (Russ et al. 2002).
The solution to the balance between teaching inclusively and unacknowledged prejudices of heteronormativity should therefore come from those invested with institutional power. And to avoid LGBT teaching staff members being discriminated against for inserting their personal narratives to illustrate classroom content, it is imperative that heterosexual lecturers use gay examples and discuss issues pertaining to sexual minorities in their classes, too, something that should be mirrored in secondary and primary schools (see Atkinson and DePalma 2009). Finally, LGBT educators should also be encouraged to be open about their sexualities and use the lives of LGBT people to illustrate curricular content. Perhaps if they discuss the operations of novelty attachment and content substitution with their students, it might help them learn to see and remember heterosexuality as well as homosexuality.