Figure skating has undergone historic changes in terms of its gender identity. Popular conceptualizations of figure skating have shifted from it being seen as a masculine sport in the late 1800s to a so-called “girl’s sport” in contemporary times [12-14]. In the early 1990s, skating took a “macho turn”, where the heterosexuality of Canadian male skaters such as world champions Kurt Browning and Elvis Stojko was deliberately played up by the media. Concerted efforts were also made by figure skating officials in Canada and internationally to disassociate male participation in figure skating from the real or perceived associations with high rates of homosexuality within the sport during this same time period [12]. Both academic and autobiographical literature on figure skating has drawn attention to the various ways that the sport reproduces rigid and dominant gender norms and the implications such gender policing has on athletes [12-17].

Baiocco et al. (2020) reviewed the ways in which scholars have attempted to measure attitudes towards LGBTQ+ people in sporting organizations [10]. They argue that much of the work looking at sexual prejudice in sports has used scales which are not specific to the sporting environment. While there is still no agreed-upon measure to operationalize sexual prejudice in sports, which is a multidimensional process, several sports organizations have recently examined attitudes towards inclusion in their sports [9], using various scales. There have also been many other studies looking specifically at the experiences of discrimination faced by athletes [9, 18]. Many authors have commented that although attitudes towards inclusion have improved along with general societal improvements in the acceptance of gender and sexually diverse people, sports organizations still have much work to do to counter prejudicial attitudes [8].

We note that none of these studies has focused on the Canadian sporting context. In 2005, Canada was one of the first countries to legalize gay marriage, and the rights of gender and sexually diverse individuals are protected by Canadian law. However, research in the Canadian context has elucidated experiences of discrimination and of violence targeting gender diverse, two spirit, and gay, lesbian, and trans women [19] as well as the difficulties trans people encounter in relation to employment [20]. Additionally, national-level survey data indicates that discrimination based on gender and sexual orientation is still common, with gay men reporting particularly high levels of everyday discrimination based on their sexual orientation [21]. These and other studies reveal that while some progress has been made in relation to LGBTQI2S+ rights in Canada, it is far from perfect.

We wished to characterize attitudes toward LGBTQ+ participation in sports in Canada, and thus we attempted to develop a new scale that would be specific to the Canadian figure skating context. We partnered with the national figure skating organization, Skate Canada, to conduct this study. We distributed an online survey, and gave respondents 28 attitudinal questions, framing the questions to be specific to figure skating in Canada. The questions are listed in Appendix A. We hypothesized that some group differences with regards to attitudes towards LGBTQ+ inclusion in figure skating might be related to respondents’ self-identified gender and sexual orientation.

Following Allport’s intergroup contact theory [22, 23], we also hypothesized that attitudes towards LGBTQ+ inclusion in figure skating might be influenced by respondents’ experiences with gender and sexually diverse people both within and outside of the skating environment. Therefore, as well as asking respondents to self identify in terms of gender and sexuality, we added questions about intergroup contact to the survey. Intergroup contact theory is a social psychological theory, which posits that known contact amongst members of different social groups can assist in reducing both prejudice and intergroup conflict [22-26]. Contact may be viewed as any form of social interaction between persons, including via face-to-face, the media, or visualization.

Many researchers have employed Allport’s theory to examine whether prejudicial attitudes towards LGBTQ+ persons are reduced through increased intergroup contact [24, 27]. A review of this literature concludes that while intergroup contact has been shown to reduce prejudice against lesbians and gay men, the evidence for reductions in prejudice against bisexual and transgender people is less conclusive [28]. Following these studies, along with the assertion of other scholars who have argued that intergroup contact may serve to decrease homophobia in sports [29, 30], we sought to explore how intergroup contact theory might be useful in an analysis of current attitudes about LGBTQ+ inclusion amongst Skate Canada’s membership.

Skate Canada (the national governing body for figure skating in Canada) agreed to support a research project on LGBTQ+ inclusion in skating, for which funding was obtained in 2017. As one part of the larger project, the present paper reports on findings from a pilot survey of Skate Canada’s membership. The main objective of the survey was to develop measures with which to explore current attitudes toward LGBTQ+ participation in Canadian figure skating and to answer the following research questions:

• What is the range of current attitudes toward LGBTQ+ participation in figure skating in Canada?
• Do these attitudes differ depending on the gender and sexual identity of individuals?
• Do these attitudes differ depending on the role of individuals in the organization (athlete, coach, volunteer, official)?
• Are observed differences in attitudes explained by Allport’s (1954) intergroup contact theory?

Approval from the research ethics board at the research team’s university was obtained and followed; online implied consent was collected prior to respondents’ participation in the survey. Skate Canada was involved in the promotion, communication, and distribution of the study information through two different electronic newsletters, sent to subscribed members. The survey was available in both English and French. One hundred and six surveys were completed over the two months that the survey was available. The only inclusion criteria for the survey was that individuals be active members of Skate Canada and be 18 years of age and over. Because of the nature of the survey, identities were kept anonymous.

Initial univariate descriptive analyses were conducted on all relevant variables, including the demographic variables, the variables measuring intergroup contact, and the variables examining attitudes towards inclusion. We then used exploratory factor analysis to summarize the information gathered through the 28 attitudinal questions. Factor analysis examines the covariation among a set of variables and summarizes the relationships among the variables by producing factors [31]. Bivariate analyses were conducted to examine the relationship between respondents’ demographic variables and their scores on these factors and between demographic variables and intergroup contact. Using ANOVA, we compared group means on all of the factors, to examine which of the independent variables were significantly related to scores on the individual factors. Next, we ran multiple linear regression models predicting individual scores on each of the factors. After performing regression diagnostics to screen for collinearity, outliers, and non-normality, we ran the models with the demographic predictors first, and then ran the models adding intergroup contact as an additional independent variable. These nested models enabled us to assess whether group differences in attitudes towards inclusion could be explained by intergroup contact. All data analysis was conducted in IBM SPSS Statistics software, version 28.0.

Table 1 summarizes the demographic composition of the sample. Of the individuals who responded to the survey, 46.7% were coaches, while athletes, volunteers, and officials accounted for 19.0%, 20.0%, and 14.3%, respectively. The majority of survey respondents were over the age of 35 years, while most individuals had been active Skate Canada members for more than 20 years. In terms of gender identity, the sample was overwhelmingly comprised of women (76.9%), while men accounted for 14.4% and nonbinary persons accounted for 8.7% of respondents. Similarly, most respondents identified themselves as heterosexual (73.5%), with bisexual, gay, lesbian, and those of another identity making up the remaining 26.5%. Finally, with regards to youth living environment and current residence, most survey respondents grew up in urban environments and most were residing in Ontario at the time of the study.

Mean scores were calculated for each of the 28 Likert-scale survey questions specific to LGBTQ+ inclusion in figure skating, which depicted generally positive opinions (Fig. 1). This was somewhat expected given both the way in which the study was promoted as well as the types of people who were most likely willing to respond to the survey. The results for questions pertaining to trans participation were encouraging, as mean scores on negatively coded questions pertaining to this topic (e.g., “There should be rules that restrict trans participation in non-ISU events”) were amongst the lowest in the entire survey, while mean scores on items that reflected positively on trans participation were generally high. All questions regarding the current status of LGBTQ+ inclusion within figure skating had mean scores over 3.0 (in the “Agree” range).

Fig. (1). Mean Scores for Pilot Attitudinal Items. Mean strength of agreement with 28 Likert-scale questions pertaining to LGBTQ+ inclusion for 106 respondents. Respondents were asked to indicate their level of agreement with each question. Responses were coded in numerical values (Strongly disagree = 1; Disagree = 2; Neither agree nor disagree = 3; Agree = 4; and Strongly agree = 5). Determinants in light grey indicate statements pertaining to trans participation in figure skating. Determinants in black indicate perceptions related to the status of LGBTQ+ inclusion in figure skating. Determinants in dark grey indicate personal opinions toward inclusion.

A factor analysis was performed on the 28 Likert-scale survey questions in order to summarize the range of opinions expressed by the survey respondents. The Kaiser-Meyer-Olkin (KMO) Measure of Sampling Adequacy (KMO = .782) and Bartlett’s Test of Sphericity (p < .01) determined that a factor analysis was appropriate to perform on the data obtained. Varimax rotation was used, and coefficients below 0.50 were suppressed. Factors with eigen values greater than one were retained. Four factors emerged from the factor analysis, accounting for 49.35% of the total variation in the responses.

Table 2 shows the results of the factor analysis and the items that loaded on each factor. The four factors are summarized as follows: >Personal Advocacy (16% variance explained; Cronbach’s alpha = 0.88); Figure Skating As An Inclusive Environment (13% variance explained; Cronbach’s alpha = 0.87); Institutional Advocacy (10% variance explained; Cronbach’s alpha = 0.76); and Trans Acceptance (10% variance explained; Cronbach’s alpha = 0.57). Several survey items did not load on any factor and were thus eliminated from the remainder of the analysis.

To examine whether attitudes differed according to the gender and sexual orientation of respondents, we divided the respondents into two groups based on their responses to both the gender and sexual orientation questions. All respondents who indicated that they were either not heterosexual and/or not a man or a woman (non binary or other) were in one group which we labeled ‘gender and/or sexually diverse’. We compared this group (N = 35, 33% of the sample) to the ‘non-diverse’ group (N = 71, 67% of the sample). Respondents who identified as gender and/or sexually diverse scored higher than the non-diverse on three individual questions (concerning their knowledge about LGBTQ+ issues, their involvement in LGBTQ+ activities, and their opinion on whether skating in Canada used to be homophobic) and lower on three questions (concerning whether it was unfair for trans people to compete against cis individuals and whether the people they know in skating are open about their gender identity). These questions are listed in Table 3, along with average scores and standard errors for both groups.

Next, a one-way Analysis of Variance (ANOVA) was performed to compare factor means across all of the demographic variables (Table 4). For these analyses, we examined sexual orientation and gender identity separately. While there were no statistically significant differences in any of the average factor scores by age, the general trend was that the younger age groups scored higher on personal advocacy and trans acceptance, and lower on the belief that figure skating was already inclusive than the older age groups. There were also no statistically significant differences in factor scores by gender. However, there was a general pattern that non-binary people scored higher than those who identified as men or women on personal advocacy, institutional advocacy and trans acceptance.

We found significant differences in the Personal Advocacy factor with regard to sexual orientation. Results indicated that bisexuals (1.24 ± S.E. = .392) scored significantly higher than both heterosexuals (-0.07 ± S.E. = .119) and persons identifying with ‘another identity’ (-0.28 ± S.E. = .108) on this factor. Gays and lesbians also scored higher than other groups on this factor, but those differences were not statistically significant. Gays, lesbians and bisexuals also had the highest levels of trans acceptance.

Primary role was related to scores on the Figure Skating as Inclusive factor. Volunteers (0.54 ± S.E = .222) scored significantly higher than both coaches (-0.11 ± S.E. = .150) and athletes (-0.59 ± S.E. = .237), and officials (0.21 ± S.E. = .251) scored significantly higher than athletes on this factor. Athletes had the lowest perception of skating’s current level of inclusivity. Athletes and coaches also had the lowest levels of trans acceptance, with volunteers and officials scoring higher on this factor.

We asked about the frequency of contact with LGBTQ+ persons both inside and outside the skating community. The questions were worded as follows: “On average within/outside of the skating community, the amount of contact I have with LGBTQ+ persons is best described as: no interaction; less than once per month; one to three times per month; weekly; several times per week; daily.” Responses were coded so that higher numbers indicate more frequent contact. For the whole sample, known contact outside the skating community averaged between weekly and several times per week, while known contact within the skating community averaged at weekly.

Bivariate analyses were performed on the mean contact scores within and outside the skating community across the demographic variables (Table 4). Significant differences for Known Contact Outside the Skating Community occurred such that individuals aged 20-30 years (5.26 ± S.E. = .349) scored significantly higher than those over the age of 50 (50-60 years: 4.20 ± S.E. = .412; >60: 3.25 ± S.E. = .648) on this variable, and persons aged 30-40 years scored significantly higher than those over the age of 60. Furthermore, bisexuals (5.83 ± S.E. = .167) and gay people (5.80 ± S.E. = .490) reported significantly more contact with LGBTQ+ persons outside the skating community than did those with another identity sexual orientations (3.50 ± S.E. = .439).

Interestingly, no significant differences occurred for Known Contact Inside the Skating Community for any of the demographic variables. Along with the fact that average contact inside the skating community was less frequent than average contact outside the skating community, this may be indicative of the fact that gender and sexually diverse individuals are not comfortable disclosing their gender and/or sexual identity in the context of the world of figure skating.

As the bivariate analysis of the factor means yielded no significant relationships between the demographic variables and scores on either the Institutional Advocacy or the Trans Acceptance factors (p < .05), we focus the rest of the paper on the remaining two factors (Personal Advocac and Figure Skating as Inclusive). We ran two multivariate regression models. The first model individually regressed the factor scores on the demographic variables of age, gender, sexual orientation, primary role, residence, and youth living environment. The second model added known contact with LGBTQ+ persons outside of the skating community and known contact with LGBTQ+ persons inside the skating community as the fifth and sixth independent variables, respectively. Tables 5 and 6 show the multiple regression results for the two factors tested.

Table 5 shows the multiple regression results for the Personal Advocacy factor. Model 1 indicates that, when other demographic variables were controlled, older individuals tended to score lower on the Personal Advocacy factor than younger persons (b = -.013; S.E. = .405). Additionally, bisexual persons (b = .925; S.E. = .469) still scored significantly higher than heterosexuals on this factor, controlling for all other demographic variables. Volunteers (b = -.495; S.E. = .269) scored lower than coaches, controlling for all other demographic variables.

With the addition of the known contact variables to the regression (Model 2), older individuals no longer demonstrated significantly lower scores than younger persons on the Personal Advocacy factor (b = -.008; S.E. = .008), but those identifying as bisexual still scored higher than people with other sexual and gender identities. Known contact outside the skating community had a significant independent effect on this factor (b = .175; S.E. = .072), controlling for all the demographic variables in the model. Known contact with LGBTQ+ persons inside the skating community did not have a significant effect on this factor.

Table 6 shows the multiple regression results for the Figure Skating As Inclusive factor. Model 1 indicates that, when other demographic variables were controlled, lesbians (b = .914; S.E. = .476) scored significantly higher than heterosexuals on this factor. This difference disappeared in Model 2, when the known contact variables were added to the analysis. However, another identity individuals (b = .666; S.E. = .386) tended to score higher than heterosexuals on the Figure Skating as Inclusive factor when both demographic and the known contact variables were considered. The effects of primary roles varied for Model 1 such that volunteers and athletes scored significantly higher and lower than coaches on the Figure Skating as Inclusive factor, respectively, controlling for other demographic variables. These results did not change once known contact was added to the model (Model 2). Volunteers continued to score higher and athletes continued to score lower than coaches on the factor, even when both the demographic and the known contact variables were considered.

Current attitudes toward LGBTQ+ participation and inclusion in Canadian figure skating amongst Skate Canada members who responded to the online survey were mostly positive. The results indicate, mean scores on each of the survey questions showed that participants were generally in favour of inclusion and opposed to discrimination, particularly relating to transgender persons that have been excluded from the sport. Given the previous research on LGBTQ+ inclusion in sports and specific literature regarding homophobia and marginalization in figure skating [12, 32, 33], these results were at least somewhat encouraging. However, it is important to remember that those who identified as gender and/or sexually diverse felt a little less positive about the inclusivity offered by skating. In particular, respondents who identified as gender and/or sexually diverse were more likely to believe that skating in Canada used to be homophobic and were less likely to report that the people they know in skating are open about their gender identity.

Although there were no large demographic differences in reports of contact with gender and/or sexually diverse people, the finding that all respondents typically report having less contact with LGBTQ+ individuals within the skating community than outside the skating community is potentially concerning. This result could be interpreted as meaning that there are fewer people who identify as LGBTQ+ people in skating than in the general public, or that LGBTQ+ people in skating are not comfortable disclosing their gender and/or sexual identity to others in the skating community.

The factor analysis revealed that attitudes toward LGBTQ+ inclusion clustered around four major themes: Figure Skating as Inclusive, Personal Advocacy, Institutional Advocacy, and Trans Acceptance. These themes reflect theoretical work on diversity and inclusion in organizations, which argues that attitudes towards diversity lie on a continuum from passive to active both at the individual and the organizational level [34]. We note that while there were almost no demographic differences in scores forInstitutional Advocacy, sexual and gender identity affected scores on Personal Advocacy and Trans Acceptance. Respondents who identified as gender and/or sexually diverse scored higher on both personal advocacy and trans acceptance. These findings support Worthen’s (2013) assertions that research on inclusion should look at attitudes held by equity deserving groups separately, even if they constitute a small portion of the sample [35].

Significant differences were found in scores on the Figure Skating as Inclusive factor amongst varying sexual orientations and primary roles, when controlling for other demographic variables. Regarding sexual orientation, lesbians scored significantly higher than heterosexuals on the Figure Skating as Inclusive factor. In terms of primary roles, athletes and volunteers scored significantly lower and higher than coaches on Figure Skating as Inclusive, respectively. The results for athletes could be due to more personal experiences in skating, or an increased awareness amongst these persons of LGBTQ+ issues in the greater society, which as a “carryover effect” has influenced their perceptions of inclusion in figure skating similar to what scholars have suggested about sport more generally [36]. However, these findings are more likely the result of athletes witnessing events that other individuals in the sport simply do not have access to; that is, homonegative incidents that occur during practice or in the locker room. According to Martens and Mobley (2005), most homonegative taunts and slurs take place in these times and spaces, often when athletes are beyond the scope of direct adult supervision (and thus may not be witnessed by coaches or volunteers) [37].

Contact appears to mitigate the effects of sexual orientation on the feeling that skating is already inclusive. Once contact is accounted for, persons who identified as another identity did tend to score significantly higher than heterosexuals, believing that skating is more inclusive. By contrast, while lesbians tended to score significantly higher than heterosexuals on the Figure Skating as Inclusive factor when only demographic variables were controlled for, this finding disappeared when contact with LGBTQ+ persons inside/outside the skating community was taken into consideration. Individuals who experienced greater amounts of contact with LGBTQ+ persons outside of figure skating were also significantly more likely to score higher on the Personal Advocacy factor than those who experienced less contact. This result is directly in line with the central premise of Allport’s (1954) intergroup contact theory.

This finding implies that lesbians and persons of another identity may have different “baselines” of attitudes toward the Figure Skating as Inclusive factor, when other variables (specifically, known contact inside the skating community) are controlled. The baseline for lesbians is high because they reported little known contact with LGBTQ+ persons in figure skating; the baseline for persons of another identity is low because they reported high known contact with LGBTQ+ persons in the sport. These differences might be explained by how each population experienced or conceptualized “known contact” and thus answered the questions pertaining to it.

As noted previously, the analysis did not employ random sampling methods. The title of the survey, the manner through which it was distributed and promoted, and the nature of the project all could have influenced the types of people who were most willing to participate in the study. As such, results pertaining to overall attitudes concerning LGBTQ+ inclusion in figure skating are almost certainly not representative of the entire figure skating community. The results may be biased in favour of inclusivity. The results of this particular survey may under-estimate the extent to which varying forms of exclusion occur in the sport.

We acknowledge that for intergroup contact to be “realized” in terms of gender identity and/or sexuality, individuals must first feel comfortable disclosing their gender identity and/or sexuality. Therefore, it is possible that more tolerant attitudes lead to more “known” contact or disclosure of gender identity and sexuality, rather than the reverse (Harrison & Michelson, 2019).

Our questions about contact concerned contact with “LGBTQ+ individuals,” treating gender and/or sexually diverse individuals as one group. However, recent literature suggests that inter-group contact has less effect on reducing bias against bisexual and transgender individuals than it does on reducing bias against lesbians and gay men (Harrison & Michelson, 2019). In the context of sports, in particular, there tends to be increased prejudice against transgender persons than other LGBTQ+ individuals. Analysis of the attitudinal variables in our data demonstrated that Transgender Acceptance emerges as a separate factor from the other attitudinal factors relating to advocacy and inclusivity in skating in general. Therefore, in the future, we suggest that researchers ask about contact with transgender persons separately.

The nature of this study as a cross-sectional, exploratory analysis limited the extent to which contact could be said to directly influence attitudes toward LGBTQ+ inclusion in figure skating. While increased known contact may have led to improvements in personal advocacy of LGBTQ+ persons, already inclusive individuals may have been more willing to interact with LGBTQ+ persons, thus increasing the awareness of contact. It is also possible that already inclusive individuals were more welcoming and accommodating to the divulgence of sexual and/or gender identity from LGBTQ+ persons, which would have further improved the awareness of contact. Because the study was only able to analyse the relationship between known contact and attitudes at a single point in time, contact could not be proclaimed to cause improvements in attitudes concerning LGBTQ+ inclusion.