Abstract
Purpose:
Our purpose was to assess the association between Gender-Sexuality Alliances (GSAs) advisors' self-efficacy to address transgender issues and their students' depressive symptoms, by students' gender identity (i.e., transgender vs. cisgender). We predict that higher advisor self-efficacy will be associated with decreases in student depressive symptoms for transgender students, though not necessarily for cisgender students.
Methods:
Data come from surveys of student members (n = 366) and advisors (n = 58) of 38 purposively sampled GSAs in Massachusetts high schools, in 2016–2017 and 2017–2018. We used a linear mixed-effects model to assess the association between advisor self-efficacy to address transgender issues and student change in Center for Epidemiological Studies Depression-10 scores between the beginning and end of the school year by gender identity, adjusting for student covariates.
Results:
Students were 10–20 years old (mean = 15, standard deviation [SD] = 1.4); 28% were transgender, 28% were students of color, and 86% were lesbian, gay, bisexual, or queer/questioning or other non-heterosexual identity. The GSA advisor self-efficacy scores ranged from 13 to 25 with a mean of 20.4 (SD = 3.0). Greater advisor self-efficacy to address transgender issues was associated with a decrease in depressive symptoms for transgender students (estimate = −0.47, p = 0.01), but not for cisgender students.
Conclusions:
GSA advisor self-efficacy to address transgender issues could be protective for transgender student depressive symptoms. Thus, increasing advisor self-efficacy to address transgender issues may help decrease depressive symptomatology for transgender youth, and intervention work in this area is needed to bolster this claim.
Keywords: depression, gender and sexuality alliances, mental health, supportive adults, transgender
Introduction
Transgender youth (i.e., youth whose gender identity does not align with their sex assignment at birth) experience poor mental health outcomes, including depression,1 which is related to substance use, suicidality, and other health concerns.2–4 Disparities in mental health are likely due to stigma and a lack of gender affirmation (i.e., the process of youth being affirmed and supported in living their gender identity and/or expression).4–7
Indeed, transgender students are more likely to report negative perceptions of school climate, violence victimization, and truancy than their cisgender peers.8,9 Limited research has investigated factors at the interpersonal level (e.g., supportive school staff) that could protect against depression among transgender youth.10
Youth resilience and program development literature emphasizes the role of adults in contributing to positive social and health outcomes among youth.11–13 Although parents are a source of resilience for transgender youth,14–16 many report not receiving parental support.17,18 Thus, non-parental adult support from school staff may help prevent mental health concerns like depression.19,20 Connections with adults at school is associated with feelings of safety,21 improved academic experience, and less absenteeism among transgender youth.10
Further, school connectedness and positive relationships with teachers are related to lower depressive symptoms among students broadly.22 Less is known about the relationship between characteristics of supportive school staff and depression among transgender youth. Transgender youth face specific challenges (e.g., transgender stigma) that may require specific supportive behavioral self-efficacy from school staff.23 However, studies of resilience have primarily highlighted how transgender youth advocate for themselves in school settings.24,25 Thus, understanding the supportive role of school staff for transgender students may bolster resilience for mental health.
Within schools, Gender-Sexuality Alliances (also known as Gay Straight Alliances, or GSAs) are youth-led, adult-advised clubs, which have been shown to be protective for lesbian, gay, bisexual, transgender, and queer/questioning (LGBTQ) youth.26,27 Youth in schools with GSAs report less violence victimization, truancy, smoking, drinking, suicide attempts, and sex with casual partners relative to those in schools without GSAs.28,29 Recently, there has been greater focus on components, activities, and functionality within GSAs that may be protective or health promoting for GSA members in general,30–32 and transgender members specifically.33–35 GSA advisors may foster members' mental health,12,36 but it is unclear how this applies to transgender youth.
Limited research has considered the role of GSA advisors' self-efficacy for supportive behaviors specific to transgender youth. Poteat and Scheer31 first examined advisor self-efficacy to address transgender issues, but they did not connect advisor self-efficacy with student outcomes. Although studies have examined GSA advisors' reasons for serving in this role and some of the efforts they undertake in this role (e.g., to provide social-emotional support or to advocate on behalf of students to other school personnel),37 most research on GSA advisors has not linked their data to data from the student members in their schools GSA. Poteat et al.30 used a combined measure of self-efficacy to address sexual and gender minority (SGM) issues, obfuscating gender identity specific factors. Thus, more work is needed to clarify GSAs' role in improving transgender GSA member mental health.
GSA advisor self-efficacy to address transgender issues may be defined as an advisor's belief in their capacity to describe or discuss gender diversity and the unique experiences and challenges faced by transgender youth.31 Multicultural counseling literature suggests that counselors' knowledge and skills are associated with their level of demonstrated competence in supporting diverse youth.38,39 Thus, GSA advisors with greater self-efficacy to address transgender issues may engage in more gender-affirming behaviors such as using inclusive language and advocating for transgender student rights. These behaviors may represent a source of support for transgender students that can foster resilience and positive mental health outcomes.
The purpose of this study was to assess the association between GSA advisors' self-efficacy to address transgender issues and changes in GSA members' depressive symptoms over the course of a school year and whether this association differed by GSA members' gender identity (i.e., transgender vs. cisgender). We hypothesized that greater advisor self-efficacy to address transgender issues would be associated with decreases in depressive symptoms over the school year for transgender students, but not for cisgender students.
Given the extent to which minority stressors contribute to mental health concerns for transgender youth, advisors' self-efficacy around transgender issues may be especially important in their ability to effectively support transgender youth and to address concerns underlying any depressive symptoms they are experiencing (whereas depressive symptoms among cisgender youth may stem from other stressors).
To provide a context to our measure of advisor self-efficacy, we compared advisors' scores with their responses to an open-ended question on transgender-related GSA activities. Given self-efficacy research that suggests cross-domain relationships between self-efficacy for a particular behavior and performance on a distinct but related behavior, greater self-efficacy to describe or discuss gender diversity and the unique experiences and challenges faced by transgender youth may lead advisors to feel confident or competent to take on or support students' efforts that address transgender issues in their schools.40
Further, discussing these topics in the context of GSAs may be a supportive behavior, and at least may be a necessary skill for other supportive behaviors. This exploratory analysis of the open-ended survey question is used to widen the understanding of advisor self-efficacy, which can uncover new aspects of advisor self-efficacy for future investigations.41
Methods
Study design
We collected two waves of data from student members (n = 366) and advisors (n = 58) of 38 GSAs in Massachusetts high schools. In GSAs with multiple advisors (n = 17), all advisors were invited to respond to surveys. Schools were purposively sampled to ensure diversity in geography, population density, as well as the size, racial/ethnic, and socioeconomic makeup of each school. Advisor consent and student assent were obtained for student participation. Advisors also consented to complete an advisor survey. The Institutional Review Board (IRB) approved a waiver of parental consent in favor of advisor adult consent to avoid risks of inadvertently outing SGM youth to their parents. Participating schools and the IRB at Boston College approved all study procedures.
Data collection
For feasibility, data were collected from 19 GSAs/schools in year 1 (2016–2017) and from 19 different GSAs/schools in year 2 (2017–2018). No advisors moved between participating schools/GSAs during the study period, and we confirmed that there were no duplicate participants who may have transferred from one of our participating schools/GSAs in the first year to a participating school/GSA in the second year of our project. Although most of our participating GSAs were in high schools that included grades 9–12, two were located in schools that included grades 6–12. Surveys were distributed at GSA meetings during early fall (wave 1) and late spring (wave 2) of the academic school year, generally separated by 6 months, using identical procedures. Surveys took ∼30 minutes to complete. Students and advisors received gift cards ($10 at wave 1; $20 at wave 2) in remuneration for their participation.
Sample
All GSAs participated in both waves, and 73.9% of the original sample of students was retained at wave 2. Attrition was differential for race/ethnicity and sexual orientation, such that non-Latinx, White students were more likely to be retained than students of color (χ2 = 5.28, degrees of freedom [df] = 1, p = 0.02; 76.8% vs. 66.9%) and lesbian, gay, bisexual, and queer/questioning or other non-heterosexual identity (LGBQ) students were more likely to be retained than heterosexual students (χ2 = 13.56, df = 1, p < 0.001; 77.4% vs. 58.4%). Attrition was not differential for gender (χ2 = 1.84, df = 1, p = 0.17). Attrition was not associated with wave 1 depression scores (t = −1.0, df = 482, p = 0.32). As missing data for level 1 (i.e., student data) was <8% of the total sample, and at least one advisor from each school responded to each measure for wave 1, we used complete case analysis for a total sample of 337 students in 38 schools.
Measures
Student demographics
Gender identity was assessed with a single item, “Check all that apply to your gender identity” with response options male, female, transgender, genderqueer, gender fluid, non-binary, and other with a space to provide a written-in response. We used a dichotomous indicator for transgender (1 = participants who selected transgender, genderqueer, gender fluid, non-binary, or other, at either wave of data collection; 0 = participants who selected only male or female at both waves). Race/ethnicity was recoded as a binary indicator due to the small number of students of color (1 = student of color; 0 = non-Latinx, White). Sexual orientation identity was coded as a binary indicator (1 = LGBQ; 0 = heterosexual), as change in depression scores were similar across LGBQ identities.
Student depressive symptomatology
Depressive symptoms were assessed at both waves using the 10-item Center for Epidemiological Studies Depression 10 (CES-D 10), which has evidence supporting validity and reliability in adolescents.42 Likert scale response options for each symptom ranged from a score of 0 (rarely or none of the time) to 3 (all of the time). Scores were summed for each wave, and internal consistency was high at wave 1 (α = 0.85) and wave 2 (α = 0.87).
We modeled difference scores on this measure as our dependent variable in our analyses, as is recommended in nonrandomized studies, particularly when predictor variables are associated with baseline outcome measures.43–45 Scores above zero indicate an increase in depressive symptoms, whereas scores below zero indicate a decrease in depressive symptoms from wave 1 to wave 2.
Advisor self-efficacy to address transgender issues
Advisor self-efficacy to address transgender issues was assessed with a 5-item scale, which asked, “How competent do you feel to do the following: (a) Talk about the unique experiences that transgender students face, (b) Address issues and topics about transgender identities, (c) Explain the difference between gender identity and sexual orientation, (d) Talk about ways to support transgender students, and (e) Talk about laws and policies for transgender students (e.g., bathrooms, graduation gowns).” Response options for each item ranged from 1 (not at all) to 5 (very) and were summed, with higher scores representing greater self-efficacy to address transgender issues. A prior factor analysis found evidence for a single unidimensional factor for these items.31 Internal consistency was high (α = 0.89).
GSA activities related to gender identity
To explore further aspects of advisor self-efficacy,41 we conducted a content analysis of responses to the open-ended survey item, “Up to this point this year, please list any activities or events that your GSA has done to address the following: i) cultural issues, ii) issues related to gender identity or expression and transgender youth, iii) issues related to race or racism and iv) issues related to immigration” to understand how advisory self-efficacy influences the frequency of activities related to gender identity, expression, and transgender youth within GSAs.
Other advisor covariates
Advisors were asked how long they had been an advisor and how many hours per week they devoted to GSA-related work, both of which were fill in the blank responses and treated as continuous variables. Advisors also indicated if they had received any training specific to being a GSA advisor.
Number of advisors
The median number of advisors was 1, with a mean of 1.5 and a max of 4. In cases where there were multiple advisors in a GSA, we averaged scores to obtain single measures of self-efficacy, length of being an advisor, and number of hours spent on GSA-related work. For training, we created an indicator such that GSAs received a 1 if any advisor had received training and 0 if no advisors had received training. We created a binary indicator for number of advisors (0 = 1 advisor; 1 = more than 1 advisor).
Analysis
Descriptive statistics were calculated using the full sample. We report correlations for continuous covariates and t-tests for binary covariates in association with student change in CES-D 10 scores.
We then used the PROC MIXED command in SAS version 9.4,46 with an unconditioned model to calculate the proportion of total variance in youth's change in CES-D 10 scores that existed across GSAs (i.e., the intraclass correlation [ICC]; ICC = 4.9%). As the amount of variation at the group level (i.e., GSAs) was nonzero, we used the mixed effects approach as is recommended given the data structure of youth nested within GSAs/advisor(s).47
Again, we used the PROC MIXED command in SAS to model GSA students' change in CES-D 10 scores. First, we specified the covariance structure by including a random intercept and a slope for the product of advisor self-efficacy and the lower-level predictor for gender as recommended due to the cross-level interaction.48 Then, we adjusted for student characteristics (i.e., age, sexual identity, race/ethnicity) and advisor characteristics as fixed effects. All models were estimated using the restricted maximum likelihood method.
Our main predictor of advisor self-efficacy was stable over time (t-paired = −0.08, df = 48, p = 0.93) and thus, we chose to use time 1 advisor self-efficacy scores to help preserve the temporal order between our predictor and outcome. Advisor characteristics that were not significant were removed from the final model for parsimony. Our primary independent variables were the main estimate of advisors' self-efficacy to address transgender issues and the cross-level interaction between this variable and students' gender identity.
To explain a significant interaction, we used the “estimate” statement in PROC MIXED to conduct post-model fit hypothesis tests to estimate the coefficient of advisor self-efficacy to address transgender issues separately for transgender and cisgender students. As a sensitivity analysis, we tested the final model adjusting for the number of advisors per GSA. Results were considered significant if p values were <0.05.
To explore the pragmatic implications of advisor self-efficacy, we conducted a content analysis of advisors' open-ended survey responses. The second author conducted an initial content analysis and categorized responses into three activity types: none/no-response, discussion (i.e., described discussion that took place in GSA meetings), or action (i.e., conducted activities other than discussion). Subsequently, three members of the research team conducted a secondary review of the analysis and identified and resolved discrepancies using a full consensus approach with 100% agreement on the final coding structure and application.
For the purpose of this article, we focus on the subset of responses specific to gender identity, expression, and transgender youth. An additional focused coding conducted by the first author examined the subset of responses specific to gender identity, expression, and transgender youth and coded them as specific to diverse genders (e.g., “transgender,” “gender neutral”) or not specific to diverse genders (e.g., “ally week,” “LGBTQ”). To examine how types of GSA activities related to gender might differ depending on advisor self-efficacy, we sorted responses to the activity variable based on advisor self-efficacy scores, broke the responses into thirds, and compared the frequency of activities across these groups.
Results
Sample characteristics
Approximately 28% of students were transgender, 28% were students of color, and 86% were LGBQ. The average student age was 15 years (range 10–20 years). Advisor self-efficacy to address transgender issues scores ranged from 13 to 25, with an average score of 20.4 (median = 21, and interquartile range = 3.5). Self-efficacy scores were broken into thirds, where the lower third (low self-efficacy) ranged from 13 to 19, for the middle third (moderate self-efficacy) 20–22.5, and for the upper third (high self-efficacy) ranged from 23 to 25. On average, advisors responded positively with regard to their self-efficacy to address transgender issues on each item. Additional student and advisor characteristics are reported in Table 1.
Table 1.
Gender-Sexuality Alliance Student and Advisor Characteristics at Wave 1
| n (%) | Mean (SD) | Range | |
|---|---|---|---|
| Students (n = 366) | |||
| Gender identity | |||
| Cisgender male | 59 (16.1) | — | — |
| Cisgender female | 202 (55.2) | — | — |
| Transgender | 31 (8.5) | — | — |
| Genderqueer | 8 (2.2) | — | — |
| Gender fluid | 7 (1.9) | — | — |
| Nonbinary | 22 (6) | — | — |
| Other | 36 (9.8) | — | — |
| Missing | 1 (0.3) | — | — |
| Sexual identity | |||
| Gay or lesbian | 61 (16.7) | — | — |
| Bisexual | 81 (22.1) | — | — |
| Questioning | 26 (7.1) | — | — |
| Heterosexual or straight | 52 (14.2) | — | — |
| Pansexual | 79 (21.6) | — | — |
| Asexual | 14 (3.8) | — | — |
| Queer | 18 (4.9) | — | — |
| Other | 32 (8.7) | — | — |
| Missing | 3 (0.8) | — | — |
| Race and ethnicity | |||
| Non-Latinx White | 264 (72.1) | — | — |
| Non-Latinx Black or African American | 8 (2.2) | — | — |
| Non-Latinx Asian | 10 (2.7) | — | — |
| Latinx | 39 (10.7) | — | — |
| Non-Latinx Middle Eastern, Arab or Arab American | 1 (0.3) | — | — |
| Biracial or multiracial | 40 (10.9) | — | — |
| Missing | 4 (1.1) | — | — |
| Age | |||
| Wave 1 (n missing = 3) | — | 15.5 (1.4) | 10 to 20 |
| CES-D 10 Score | |||
| Wave 1 (n missing = 9) | — | 14.1 (6.8) | 0 to 30 |
| Wave 2 (n missing = 15) | — | 13.9 (6.7) | 0 to 30 |
| Change score | — | 0.3 (5.7) | −19 to 21 |
| Advisors (n = 52) | |||
| Self-efficacy to address transgender issues | — | 20.4 (3.0) | 13 to 25 |
| Number of hours spent on GSA-related work per week | — | 2.8 (2.1) | 0.75 to 13.5 |
| Number of months serving as GSA advisor | — | 61.5 (67.0) | 2 to 302 |
| GSA specific training | 21 (55.3) | — | — |
Higher change scores represent an increase in depressive symptoms from wave 1 to wave 2.
CES-D 10, Center for Epidemiological Studies Depression 10; GSA, Gender-Sexuality Alliance; SD, standard deviation.
Bivariate relationships
CES-D 10 scores, on average, were relatively stable over time, as displayed in Table 2. Transgender students reported higher depressive symptoms than cisgender students at wave 1 (mean difference [MD] = 4.8) and at wave 2 (MD = 4.0). Table 3 reports further information about demographic differences.
Table 2.
Gender-Sexuality Alliance Student Center for Epidemiological Studies Depression 10 Scores at Wave 1 and Wave 2 and Matched Pairs t-Tests Between Wave 1 and Wave 2 by Gender Identity
| CES-D 10 Wave 1, mean (SD) | CES-D 10 Wave 2, mean (SD) | t | df | p | |
|---|---|---|---|---|---|
| All students | 13.96 (6.79) | 13.75 (6.96) | 0.93 | 341 | 0.3521 |
| Cisgender students | 12.36 (6.25) | 12.33 (6.80) | 0.03 | 226 | 0.9755 |
| Transgender students | 17.29 (6.67) | 16.53 (6.43) | 1.42 | 114 | 0.1578 |
df, degrees of freedom.
Table 3.
Bivariate Relationship Between Gender-Sexuality Alliance Student Characteristics and Center for Epidemiological Studies Depression 10 Scores
| Mean difference (SE) | t | df | p | |
|---|---|---|---|---|
| CES-D 10 wave 1 | ||||
| Transgender versus cisgender | 4.8 (0.7) | 6.55 | 335 | <0.0001 |
| Sexual minority identity versus heterosexual | 4.8 (1.0) | 4.65 | 335 | <0.0001 |
| Student of color versus non-Latinx White | −0.4 (0.8) | −0.42 | 335 | 0.6719 |
| CES-D 10 wave 2 | ||||
| Transgender versus cisgender | 4.0 (0.8) | 5.18 | 335 | <0.0001 |
| Sexual minority identity versus heterosexual | 3.2 (1.1) | 3.0 | 335 | 0.0033 |
| Student of color versus non-Latinx White | −1.1 (0.9) | 1.32 | 335 | 0.1865 |
| CES-D 10 change score | ||||
| Transgender versus cisgender | 0.9 (0.7) | 1.31 | 335 | 0.1906 |
| Sexual minority identity versus heterosexual | 1.6 (0.9) | 1.83 | 335 | 0.0683 |
| Student of color versus non-Latinx White | 0.8 (0.7) | 1.09 | 335 | 0.2786 |
Higher change scores represent an increase in depressive symptoms from wave 1 to wave 2.
SE, standard error.
Advisor self-efficacy to address transgender issues was associated with their reported number of hours spent on GSA work per week (r = 0.11, p < 0.05). Among transgender students, advisor self-efficacy to address transgender issues was associated with a change in CES-D scores (r = −0.21, p < 0.05). Student change in depression scores was not associated with number of hours spent on GSA work per week or number of months served as advisor.
Generalized linear mixed model
Results for the fully conditioned linear mixed model are reported in Table 4. After adjusting for fixed effects, the G matrix was no longer positive definite due to the random intercept being equal to zero. Therefore, we dropped the random intercept from our final model. The fixed-effect interaction between gender identity and advisor self-efficacy to address transgender issues was significant (estimate = −0.52, p = 0.04). Post-model fit hypothesis tests revealed that among transgender students, the group association of a 1 U increase in advisor self-efficacy to address transgender issues was an average decrease of 0.47 in student CES-D 10 scores (p = 0.01). This relationship did not hold for cisgender students (estimate = 0.06, p = 0.73). Figure 1 displays the simple slopes of the interaction between advisor self-efficacy to address transgender issues and students' gender identity on students' change in CES-D 10 scores.
Table 4.
Generalized Linear Mixed Model Regressing Gender-Sexuality Alliance Students' Change in Center for Epidemiological Studies Depression 10 Scores on Student and Advisor Characteristics
| Estimate | SE | df | z | t | p | |
|---|---|---|---|---|---|---|
| Covariance parameters (random effects) | ||||||
| Advisor self-efficacy to address transgender issues × transgender | 0.0044 | 0.0037 | — | 1.20 | — | 0.1154 |
| Residual | 30.66 | 2.64 | — | 11.63 | — | <0.0001 |
| Level 1 fixed effects (students; n = 337) | ||||||
| Intercept | −5.09 | 4.95 | 115 | — | −1.03 | 0.3061 |
| Transgender (cisgender as reference) | 11.72 | 5.10 | 82.1 | — | 2.30 | 0.0242 |
| Student of color (White as reference) | 0.89 | 0.72 | 328 | — | 1.24 | 0.2167 |
| Sexual minority (heterosexual as reference) | 1.48 | 0.93 | 330 | — | 1.59 | 0.1127 |
| Age | 0.16 | 0.23 | 304 | — | 0.69 | 0.4917 |
| Level 2 fixed effects (advisor; n = 38) | ||||||
| Advisor self-efficacy to address transgender issues | 0.06 | 0.16 | 40.6 | — | 0.35 | 0.7279 |
| Interaction fixed effects | ||||||
| Advisor self-efficacy to address transgender issues × transgender | −0.52 | 0.24 | 70.5 | — | −2.14 | 0.0354 |
| Estimates of advisor self-efficacy by gender identity fixed effects | ||||||
| Advisor self-efficacy to address transgender issues among transgender youth | −0.47 | 0.18 | 116 | — | −2.55 | 0.0121 |
| Advisor self-efficacy to address transgender issues among cisgender youth | 0.06 | 0.16 | 40.6 | — | 0.35 | 0.7279 |
Higher change scores represent an increase in depressive symptoms from wave 1 to wave 2. Intraclass correlation for the unconditional model = 4.9%.
FIG. 1.
Simple slopes of the regression of student change in CES-D 10 scores on the interaction between student gender identity and advisor self-efficacy to address transgender issues. The simple slope for transgender students is −0.47 (p = 0.01), and the simple slope for cisgender students is 0.06 (p = 0.73). Simple intercepts were calculated using the mean values for age, sexual orientation, and race/ethnicity. Changes in CES-D 10 scores above zero represent an increase in depressive symptoms from wave 1 to wave 2, whereas scores less than zero represent a decrease in depressive symptoms from wave 1 to wave 2. CES-D 10, Center for Epidemiological Studies Depression 10.
Ancillary analysis of advisor activities
In advisors' reports of past activities to support transgender students (n = 52), 48.1% were coded action (e.g., “working on creating a gender neutral bathroom”), 30.8% were coded as discussion (e.g., “discussions of topics/news related to gender identity”), 13.5% as none or no response, and 7.7% as unclear in terms of action or discussion. About 46.2% of responses used language specific to diverse genders (e.g., “transgender,” “gender identity,” or “two-spirit”).
Findings from the ancillary analysis are summarized across levels of advisor self-efficacy in Table 5. Among the seven responses that represented no activity to support transgender students, self-efficacy among six advisors was low whereas one was moderate. Further, among the advisors with low self-efficacy, only 27.7% used language specific to diverse genders compared with advisors with moderate to high self-efficacy, where about 54% of responses included language specific to diverse genders. Advisors with moderate self-efficacy tended to reflect actions, whereas advisors with high self-efficacy reflected discussion.
Table 5.
Results of the Ancillary Analysis Comparing Qualitative Findings of Gender-Sexuality Alliance Advisors' Responses with the Open-Ended Question About Activities That Took Place During the Past Year to Support Transgender Students Across Levels of Advisor Self-Efficacy to Address Transgender Issues
| Advisor self-efficacy level | Findings from qualitative analysis |
|---|---|
| Low self-efficacy (score of 13–19) | Includes six responses that did not indicate any activity to support transgender students. About one quarter of responses included language specific to diverse genders. |
| Moderate self-efficacy (score of 20–22.5) | Includes one response that did not indicate any activity to support transgender students. Activities tended to reflect actions other than discussion to support transgender students. About half of the responses used language specific to diverse genders. |
| High self-efficacy (score of 23–25) | All responses reported activities to support transgender students. Activities tended to reflect discussions about gender. About half of the responses used language specific to diverse genders. |
Language specific to diverse genders implies the use of terms specific to transgender youth such as “transgender” or “gender neutral” rather than terms that may be considered inclusive of but not specific to transgender youth such as “LGBTQ” or “ally week.” Actions other than discussion refers to activities that are not only discussion based such as “conducting a review of school policies related to gender.” Discussion activities refers to discussion that took place during regular GSA meetings.
LGBTQ, lesbian, gay, bisexual, transgender, and queer/questioning.
Discussion
Our findings suggest that higher GSA advisor self-efficacy to address transgender issues is associated with decreases in depressive symptoms across the academic year for transgender students, but not for cisgender students. These findings contrast with previous work that found no association between advisor self-efficacy to address SGM issues and student depression,30 emphasizing the need to consider sexual orientation and gender identity as distinct constructs and not to conflate them in health research and educational messages.49 Self-efficacy to discuss or talk about transgender-related topics may reflect the gender-affirming behaviors of GSA advisors more broadly. These findings contribute to a broader program of research that focuses on aspects of GSAs,30–35,50 and they underscore the important role that adult advisors play in contributing to student mental health.
Research suggests that trusted adults are critical for creating a safer school environment for transgender youth.10,21,51,52 However, research has yet to describe how supportive adult characteristics are related to health outcomes for transgender youth and our findings begin to fill this gap. Interventions aimed at increasing GSA advisor self-efficacy to address transgender issues may improve the mental health of transgender students.
Findings from the ancillary analysis suggest that advisor self-efficacy to address transgender issues may be related to the ability of advisors to use transgender-specific language, and the content of the GSA activities specific to gender identity reported by advisors. Although discussion was the primary domain in our measure of self-efficacy, it is noteworthy that many GSA advisors reported actions beyond discussion that supported transgender members in open-ended responses (e.g., advocating at school committee meetings for transgender students).
As part of ongoing scale development, future research on advisor self-efficacy should include items that ask about confidence in leading or supporting other gender-affirming efforts in addition to facilitating conversations. Development of additional sub-constructs could strengthen our understanding of how GSA advisors' self-efficacy for supportive behaviors is protective for transgender student mental health.
Additional advisor characteristics, including receipt of training specific to GSA advising, months spent in advisory role, and number of hours spent on GSA work per week, were not associated with change in student depression. This contrasts with work that found that these factors were important predictors of advocacy over the school year.49 In our study, more hours spent on GSA work per week was associated with higher self-efficacy to address transgender issues in bivariate analyses.
Additional research is needed to further explore the relationships between these variables, as these characteristics may be indirectly related to transgender student mental health. Although competency-based frameworks have been developed for classroom education,53–56 it is unclear to what extent these frameworks are relevant to school club settings and we are not aware of any such framework specific to student diversity club advisors. Nevertheless, advisor characteristics and competencies continue to be a point of inquiry in student club settings.57
Limitations
Current gender identity was assessed using a single item, rather than a recommended two-step method.58 Although we observed changes in student depressive symptoms in association with advisor self-efficacy to address transgender issues, it is possible that there are other stronger predictors of students' depressive symptoms that may underlie both advisor self-efficacy and student depressive symptoms (e.g., school policies). However, we adjusted for multiple student and advisor covariates with robust findings. Longer periods of follow-up and intervention work could strengthen this evidence.
Our coding structure for the open-ended advisor question about GSA activities related to gender identity does not account for the descriptive quality or valence of those activities (e.g., the overall positivity or negativity of a conversation or extent to which there was discord or resolution to disagreements). This limits our understanding of how advisor self-efficacy may play a role beyond whether activities do or do not occur.
Further, our findings may not generalize beyond those GSAs included in this study. In schools with multiple advisors, we averaged advisor scores to model their association with student outcomes, which may overlook nuances in GSAs with co-advisors. Due to a limited sample size, we were unable to assess differential effects across diverse gender identities.
Conclusions
In GSAs, advisor self-efficacy to discuss issues pertaining to transgender students' lived experiences was protective against transgender students' depressive symptoms. Thus, increasing advisor self-efficacy to address transgender issues may help decrease depressive symptomatology for transgender youth, and intervention work in this area that engages and centers the experiences of transgender youth is needed to bolster this claim.
In addition, self-efficacy to discuss transgender issues may also apply to trusted adults in diverse settings (e.g., other school staff or community health workers) and researchers in these areas may draw upon this work to inform future investigations of the characteristics of trusted adults and transgender youth mental health.
Acknowledgments
The authors wish to thank Jeff Perrotti for his invaluable assistance in identifying and recruiting GSAs to be involved in this project, the GSA advisors for the substantial amount of time they generously devoted to ensuring their GSA's ability to participate, and all the young people who participated and shared their experiences.
Authors' Contributions
J.A. conceptualized this study, led the analysis, helped interpret the findings, and was the primary author drafting the article and incorporating co-author feedback. L.S. helped conduct the ancillary analysis and participated in drafting and editing of the article. V.P.P., H.Y., and J.P.C. conceptualized and designed the larger study on which the current article is based, oversaw data collection, provided input on the analysis, helped interpret findings, and participated in drafting and editing the article. All authors approved the final article as submitted and agree to be accountable for all aspects of the work.
Disclaimer
The funder did not participate in the research. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health (NIH).
Author Disclosure Statement
No competing financial interests exist.
Funding Information
Research reported in this publication was supported by a grant from the National Institute on Minority Health and Health Disparities of the NIH under award number R01MD009458 (Principal Investigator: V.P.P.; Co-Investigators: J.P.C. and H.Y.); J.A. is supported by the National Institute on Drug Abuse (NIDA), NIH under Award Number T32DA023356.
References
- 1. Connolly MD, Zervos MJ, Barone II CJ, et al. The mental health of transgender youth: Advances in understanding. J Adolesc Health 2016;59(5):489–495; doi: 10.1016/j.jadohealth.2016.06.012 [DOI] [PubMed] [Google Scholar]
- 2. Johns MM, Lowry R, Andrzejewski J, et al. Transgender identity and experiences of violence victimization, substance use, suicide risk, and sexual risk behaviors among high school students—19 States and large urban school districts, 2017. MMWR Morb Mortal Wkly Rep 2019;68(3):67–71; doi: 10.15585/mmwr.mm6803a3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Day JK, Fish JN, Perez-Brumer A, et al. Transgender youth substance use disparities: Results from a population-based sample. J Adolesc Health 2017;61(6):729–735; doi: 10.1016/j.jadohealth.2017.06.024 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Mustanski B, Liu RT. A longitudinal study of predictors of suicide attempts among lesbian, gay, bisexual, and transgender youth. Arch Sex Behav 2013;42(3):437–448; doi: 10.1007/s10508-012-0013-9 [DOI] [PubMed] [Google Scholar]
- 5. Hendricks ML, Testa RJ. A conceptual framework for clinical work with transgender and gender nonconforming clients: An adaptation of the Minority Stress Model. Prof Psychol Res Pr 2012;43(5):460–467; doi: 10.1037/a0029597 [DOI] [Google Scholar]
- 6. Testa RJ, Michaels MS, Bliss W, et al. Suicidal ideation in transgender people: Gender minority stress and interpersonal theory factors. J Abnorm Psychol 2017;126(1):125–136; doi: 10.1037/abn0000234 [DOI] [PubMed] [Google Scholar]
- 7. Sevelius JM. Gender affirmation: A framework for conceptualizing risk behavior among transgender women of color. Sex Roles 2013;68(11):675–689; doi: 10.1007/s11199-012-0216-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Pampati S, Andrzejewski J, Sheremenko G, et al. School climate among transgender high school students: An exploration of school connectedness, perceived safety, bullying, and absenteeism. J Sch Nurs 2020;36(4):293–303; doi: 10.1177/2F1059840518818259 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Day JK, Perez-Brumer A, Russell ST. Safe schools? Transgender youth's school experiences and perceptions of school climate. J Youth Adolesc 2018;47(8):1731–1742; doi: 10.1007/s10964-018-0866-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Johns MM, Beltran O, Armstrong HL, et al. Protective factors among transgender and gender variant youth: A systematic review by socioecological level. J Prim Prev 2018;39(3):263–301; doi: 10.1007/s10935-018-0508-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Masten AS, Barnes AJ. Resilience in children: Developmental perspectives. Children 2018;5(7):98–114; doi: 10.3390/children5070098 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. National Research Council, Institute of Medicine. Community Programs to Promote Youth Development. The National Academies Press: Washington, DC; 2002. [Google Scholar]
- 13. Fergus S, Zimmerman MA. Adolescent resilience: A framework for understanding healthy development in the face of risk. Annu Rev Public Health 2005;26(1):399–419; doi: 10.1146/annurev.publhealth.26.021304.144357 [DOI] [PubMed] [Google Scholar]
- 14. Katz-Wise SL, Rosario M, Tsappis M. Lesbian, gay, bisexual, and transgender youth and family acceptance. Pediatr Clin North Am 2016;63(6):1011–1025; doi: 10.1016/j.pcl.2016.07.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Andrzejewski J, Pampati S, Steiner RJ, et al. Perspectives of transgender youth on parental support: Qualitative findings from the Resilience and Transgender Youth Study. Health Educ Behav 2021;48(1):74–81; doi: 10.1177/1090198120965504 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Katz-Wise SL, Ehrensaft D, Vetters R, et al. Family functioning and mental health of transgender and gender-nonconforming youth in the trans teen and family narratives project. J Sex Res 2018;55(4–5):582–590; doi: 10.1080/00224499.2017.1415291 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Eisenberg ME, Gower AL, McMorris BJ, et al. Risk and protective factors in the lives of transgender/gender nonconforming adolescents. J Adolesc Health 2017;61(4):521–526; doi: 10.1016/j.jadohealth.2017.04.014 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Johnson KC, LeBlanc AJ, Sterzing PR, et al. Trans adolescents' perceptions and experiences of their parents' supportive and rejecting behaviors. J Couns Psychol 2020;67(2):156–170; doi: 10.1037/cou0000419 [DOI] [PubMed] [Google Scholar]
- 19. Allen LR, Watson LB, VanMattson SB. Trans young adults' reflections on adolescent sources of extra-familial support. J LGBT Youth 2020;17(1):1–23; doi: 10.1080/19361653.2019.1591323 [DOI] [Google Scholar]
- 20. Mulcahy M, Dalton S, Kolbert J, et al. Informal mentoring for lesbian, gay, bisexual, and transgender students. J Educ Res 2016;109(4):405–412; doi: 10.1080/00220671.2014.979907 [DOI] [Google Scholar]
- 21. McGuire JK, Anderson CR, Toomey RB, et al. School climate for transgender youth: A mixed method investigation of student experiences and school responses. J Youth Adolesc 2010;39(10):1175–1188; doi: 10.1007/s10964-010-9540-7 [DOI] [PubMed] [Google Scholar]
- 22. Joyce HD, Early TJ. The impact of school connectedness and teacher support on depressive symptoms in adolescents: A multilevel analysis. Child Youth Serv Rev 2014;39:101–107; doi: 10.1016/j.childyouth.2014.02.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Stieglitz KA. Development, risk, and resilience of transgender youth. J Assoc Nurses AIDS Care 2010;21(3):192–206; doi: org/10.1016/j.jana.2009.08.004 [DOI] [PubMed] [Google Scholar]
- 24. Singh AA. Transgender youth of color and resilience: Negotiating oppression and finding support. Sex Roles 2013;68(11):690–702; doi: 10.1007/s11199-012-0149-z [DOI] [Google Scholar]
- 25. Johns MM, Zamantakis A, Andrzejewski J, et al. Minority stress, coping, and transgender youth in schools—Results from the resilience and transgender youth study. J Sch Health 2021;91(11):883–893; doi: 10.1111/josh.13086 [DOI] [PubMed] [Google Scholar]
- 26. Heck NC, Flentje A, Cochran BN. Offsetting risks: High school gay-straight alliances and lesbian, gay, bisexual, and transgender (LGBT) youth. Sch Psychol Q 2013;26(2):161–174; doi: 10.1037/2329-0382.1.S.81 [DOI] [Google Scholar]
- 27. Toomey RB, Ryan C, Diaz RM, et al. High school gay–straight alliances (GSAs) and young adult well-being: An examination of GSA presence, participation, and perceived effectiveness. Appl Dev Sci 2011;15(4):175–185; doi: 10.1080/10888691.2011.607378 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Poteat VP, Sinclair KO, DiGiovanni CD, et al. Gay–straight alliances are associated with student health: A multischool comparison of LGBTQ and heterosexual youth. J Res Adolesc 2013;23(2):319–330; doi: 10.1111/j.1532-7795.2012.00832.x [DOI] [Google Scholar]
- 29. Marx RA, Kettrey HH. Gay-straight alliances are associated with lower levels of school-based victimization of LGBTQ+ youth: A systematic review and meta-analysis. J Youth Adolesc 2016;45(7):1269–1282; doi: 10.1007/s10964-016-0501-7 [DOI] [PubMed] [Google Scholar]
- 30. Poteat VP, Calzo JP, Yoshikawa H, et al. Greater engagement in gender-sexuality alliances (GSAs) and GSA characteristics predict youth empowerment and reduced mental health concerns. Child Dev 2020;91(5):1509–1528; doi: 10.1111/cdev.13345 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Poteat VP, Scheer JR. GSA advisors' self-efficacy related to LGBT youth of color and transgender youth. J LGBT Youth 2016;13(4):311–325; doi: 10.1080/19361653.2016.1185757 [DOI] [Google Scholar]
- 32. Poteat VP, Yoshikawa H, Calzo JP, et al. Contextualizing gay-straight alliances: Student, advisor, and structural factors related to positive youth development among members. Child Dev 2015;86(1):176–193; doi: 10.1111/cdev.12289 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Poteat VP, Calzo JP, Yoshikawa H, et al. Discussing transgender topics within gay-straight alliances: Factors that could promote more frequent conversations. Int J Transgend 2018;19(2):119–131; doi: 10.1080/15532739.2017.1407983 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Chong ES, Poteat VP, Yoshikawa H, et al. Fostering youth self-efficacy to address transgender and racial diversity issues: The role of gay–straight alliances. Sch Psychol 2019;34(1):54–63; doi: 10.1037/spq0000258 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Ioverno S, Russell ST. Homophobic bullying in positive and negative school climates: The moderating role of gender sexuality alliances. J Youth Adolesc 2021;50(2):353–366; doi: 10.1007/s10964-020-01297-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Poteat V, Heck NC, Yoshikawa H, et al. Gay-straight alliances as settings to discuss health topics: Individual and group factors associated with substance use, mental health, and sexual health discussions. Health Educ Res 2017;32(3):258–268; doi: 10.1093/her/cyx044 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Valenti M, Campbell R. Working with youth on LGBT issues: Why gay–straight alliance advisors become involved. J Community Psychol 2009;37(2):228–248; doi: 10.1002/jcop.20290 [DOI] [Google Scholar]
- 38. Spanierman LB, Poteat VP, Wang YF, et al. Psychosocial costs of racism to white counselors: Predicting various dimensions of multicultural counseling competence. J Couns Psychol 2008;55(1):75–88; doi: 10.1037/0022-0167.55.1.75 [DOI] [Google Scholar]
- 39. Worthington RL, Mobley M, Franks RP, et al. Multicultural counseling competencies: Verbal content, counselor attributions, and social desirability. J Couns Psychol 2000;47(4):460–468; doi: 10.1037/0022-0167.47.4.460 [DOI] [Google Scholar]
- 40. Paunonen SV, Hong RY. Self-efficacy and the prediction of domain-specific cognitive abilities. J Pers 2010;78(1):339–360; doi: 10.1111/j.1467-6494.2009.00618.x [DOI] [PubMed] [Google Scholar]
- 41. Allen M (ed.) The SAGE Encyclopedia of Communication Research Methods. SAGE Publications, Inc.: Thousand Oaks, CA; 2017. [Google Scholar]
- 42. Bradley KL, Bagnell AL, Brannen CL. Factorial validity of the Center for Epidemiological Studies Depression 10 in adolescents. Issues Ment Health Nurs 2010;31(6):408–412; doi: 10.3109/01612840903484105 [DOI] [PubMed] [Google Scholar]
- 43. Van Breukelen GJ. ANCOVA versus change from baseline had more power in randomized studies and more bias in nonrandomized studies. J Clin Epidemiol 2006;59(9):920–925; doi: 10.1016/j.jclinepi.2006.02.007 [DOI] [PubMed] [Google Scholar]
- 44. Castro-Schilo L, Grimm KJ. Using residualized change versus difference scores for longitudinal research. J Soc Pers Relat 2018;35(1):32–58; doi: 10.1177/2F0265407517718387 [DOI] [Google Scholar]
- 45. Eriksson K, Häggström O. Lord's paradox in a continuous setting and a regression artifact in numerical cognition research. PLoS One 2014;9(4):e95949; doi: 10.1371/journal.pone.0095949 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46. SAS Institute, Inc. Base SAS 9.4 Procedures Guide. SAS Institute, Inc.: Cary, NC; 2015. [Google Scholar]
- 47. Musca SC, Kamiejski R, Nugier A, et al. Data with hierarchical structure: Impact of intraclass correlation and sample size on type-I error. Front Psychol 2011;2(74):1–6; doi: 10.3389/fpsyg.2011.00074 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48. Heisig JP, Schaeffer M. Why you should always include a random slope for the lower-level variable involved in a cross-level interaction. Eur Sociol Rev 2019;35(2):393–401; doi: 10.1093/esr/jcy053 [DOI] [Google Scholar]
- 49. Andrzejewski J, Rasberry CN, Mustanski B, et al. Sexual and reproductive health web sites: An analysis of content for sexual and gender minority youth. Am J Health Promot 2020;34(4):393–401; doi: 10.1177/2F0890117119899217 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50. Poteat VP, O'Brien MD, Yang MK, et al. Youth advocacy varies in relation to adult advisor characteristics and practices in gender-sexuality alliances. Appl Dev Sci 2021;26(3):460–470; doi: 10.1080/10888691.2020.1861945 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51. Seelman KL, Forge N, Walls NE, et al. School engagement among LGBTQ high school students: The roles of safe adults and gay–straight alliance characteristics. Child Youth Serv Rev 2015;57:19–29; doi: 10.1016/j.childyouth.2015.07.021 [DOI] [Google Scholar]
- 52. Bird JDP, Kuhns L, Garofalo R. The impact of role models on health outcomes for lesbian, gay, bisexual, and transgender youth. J Adolesc Health 2012;50(4):353–357; doi: 10.1016/j.jadohealth.2011.08.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53. Szucs LE, Andrzejewski JD, Robin L, et al. The Health Education Teacher Instructional Competency Framework: A conceptual guide for quality instruction in school health. J Sch Health 2021;91(10):774–787; doi: 10.1111/josh.13076 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 54. Darling-Hammond L. Teacher quality and student achievement. Educ Policy Anal Arch 2000;8(1):1–44; doi: 10.14507/epaa.v8n1.2000 [DOI] [Google Scholar]
- 55. Wei RC, Darling-Hammond L, Andree A, et al. Professional Learning in the Learning Profession: A Status Report on Teacher Development in the U.S. and Abroad. Technical Report. National Staff Development Council: Dallas, TX; 2009. [Google Scholar]
- 56. Stronge JH, Ward TJ, Grant LW. What makes good teachers good? A cross-case analysis of the connection between teacher effectiveness and student achievement. J Teach Educ 2011;62(4):339–355; doi: 10.1177/2F0022487111404241 [DOI] [Google Scholar]
- 57. Parry B, Cornelius-White JH, MacGregor C, et al. High school diversity clubs: Advisor qualifications, resources, and curriculum. High Sch J 2020;104(1):54–74. [Google Scholar]
- 58. GenIUSS Group. Best Practices for Asking Questions to Identify Transgender and Other Gender Minority Respondents on Population-Based Surveys. Herman JL. (ed.) The Williams Institute: Los Angeles, CA; 2014. [Google Scholar]