LGBT+ Self‑Identification Among Youth in Ireland Aged 17‑18 Years: A Research Brief

In 2016, for the first time, a nationally representative survey of 6216 respondents, Growing Up in Ireland (GUI), included measurement of sexual orientation and gender identity (SOGI) among young people. Irish LGBTI+ youth and inclusion strategies identified collection of SOGI data as a policy objective, prioritizing analyses of GUI. In line with policy objectives, this research brief reports previously unpublished LGBT+ self-identification for youth in Ireland from Cohort ’98 born five years after decriminalization of homosexuality, with their coming-of-age coinciding with the marriage equality referendum, and legislation providing for greater gender recognition (2015). Ten percent of youth in Ireland, at 17-18 years, identify as lesbian, gay, bisexual, transgender, questioning, asexual or describe their gender as other. Female youth are significantly more likely to identify as a sexual minority, and to be bisexual than their male peers. A quarter of LGBT+ youth are questioning their sexual orientation. This finding supports the inclusion of a ‘questioning’ response in surveys collecting SOGI data, and also suggests sexual orientation self-identification is a dynamic process. A small percentage identify as asexual (0.2%). Just over 1% of young people identify as transgender or describe their gender as other. As the first generation in Ireland to grow up in an environment with increased LGBT+ visibility and social acceptance, these findings of a relatively high proportion of sexual and gender minorities among young people likely reflect improved social and legal climates. The importance of publication of SOGI estimates is under-scored. GUI are committed to ensuring visibility of LGBT+ youth populations in data collection. Suggestions for improved SOGI placement and phrasing will be considered in the questionnaire with Cohort ’08 at age 17, currently being drafted. The findings from these analyses will inform future policy and research. GUI offers rich possibilities for future comparative cross-cohort analyses.


Introduction
Ireland has witnessed significant progress in the advancement of lesbian, gay, bisexual, transgender (LGBT+) equality since 1993, when homosexuality was formally decriminalized, with particularly rapid progress in the last decade (Sloan, 2016). 1 In 2014, the Irish government became a signatory to the International Day Against Homophobia and Transphobia (IDAHOT) Declaration of Intent, and committed to implementing national policies to combat discrimination based on sexual orientation and gender identity (SOGI) (Department of Justice (DoJ), 2014). IDAHOT informed Irish policy measures including marriage equality, gender recognition by self-identification for adults, and publication of the LGBTI+ National Youth Strategy and LGBTI+ National Inclusion Strategy (DoJ, 2014); Department of Children and Youth Affairs (DCYA), 2018). These Strategies prioritized the development of research as a key goal, acknowledging the dearth of research nationally and internationally, and the urgent need for Irish-specific studies to ensure evidence-informed policy (Fig 1).
This advancement in policies coincides with marked increases in LGBT+ social acceptance in Irish society. Flores (2019) developed a Global Acceptance Index using studies that assess social acceptance to rank countries. Among 174 countries, Ireland rose from 20th place in 2000-2003 to 7th in 2014-2017. A recent Irish survey found that the majority of respondents expressed positive attitudes towards  1 The LGBTI + acronym refers to those identifying as lesbian, gay, bisexual, transgender and intersex. The use of this acronym evolved iteratively through consultative policy-making processes in Ireland, with the 'I' encompassing intersex variation, which was not measured in the Growing Up in Ireland survey. The LGBT + acronym is used in these analyses.
LGBT+ communities, understood gender as more complex than a binary view of male/female, and were broadly supportive of policy goals (Noone et al., 2022).
Despite improved social and legal climates for LGBT+ communities in Ireland, population-based SOGI data collection remains limited and Irish policymakers lack quantification of LGBT+ population size (DCYA, 2018;DoJ, 2019). Two National Academies of Science, Engineering and Medicine reports suggest this issue is not unique to Ireland (NASEM, 2020;. While data collection in some jurisdictions is improving, historically the invisibility of LGBT+ populations in data has hampered policy initiatives designed to improve the health and wellbeing of these communities (Guyan, 2022).
Without official data collection and published analyses, determining LGBT+ estimates in Ireland has relied on approximations, ranging between 1.2% to 10.8% of the population (Oireachtas Library and Research Service (L&RS), 2019). Despite limited data measuring LGBT+ communities, Ireland has collected high-quality survey data assessing the experiences of LGBT+ youth through the Growing Up in Ireland (GUI) study (McNamara et al, 2018). 2 Respondents from Cohort '98 were born five years after decriminalization of homosexuality, and came of age when two-thirds of the Irish electorate voted for marriage equality in 2015 (Sloan, 2016). This marks the first time that a publicly funded, population-based survey in Ireland gathered SOGI data (McNamara et al., 2018).

Rationale
This paper addresses a recognized research gap, and specifically responded to the identified LGBTI+ National Youth Strategy objective for enhanced data quality, with specific reference to exploration of GUI (DCYA, 2018) ( Fig. 1). It also responded to the LGBTI+ National Inclusion Strategy to: "Ensure that disaggregated data on the LGBTI+ population and their needs is collected in national surveys where relevant" (DoJ, 2019: 25) (Fig. 1). This brief is twinned with a qualitative study with LGBT+ youth, in response to a query from the GUI study team of how best to collect SOGI data in future Waves (Ceatha et al., 2021) (Fig. 1). 3 The YuPP:) Project co-created a template for improved SOGI data collection and the experience of those providing that data, enhancing quality (Ceatha et al., 2023). As part of this process, preliminary findings were presented to the YuPP:) Panel who emphasized the valuable nature of GUI SOGI data and the importance of publication of such data, when collected and analyzed. SOGI data with youth samples is extremely rare and this Irish data has not been previously published. It provides the first available statistics on LGBT+youth, and the first SOGI Irish estimate (at any age).

Study Design and Setting
GUI is a nationally representative longitudinal study of children and young people who have grown up in Ireland (McNamara et al., 2018). Four waves have been collected from Cohort '98 (born between November 1, 1997 and October 31, 1998) at 9 years, 13 years, 17-18 years and 20 years (McNamara et al., 2018;O'Mahony et al., 2021). Wave 3 data (17-18 years) was collected between November 2015 and September 2016, marking a change where information was mainly collected from the young person, with parents/guardians providing additional data (Murphy et al., 2019). The attention to privacy and confidentiality by GUI during administration includes questionnaire self-completion by the young person using Computer Assisted Personal Interview complemented by Computer Assisted Self Interview, and extends to archiving and dataset access (Central Statistics Office (CSO), n.d.; GUI, n.d.; Murphy et al., 2019).
Respondents 8568 respondents and their families were first recruited from over 900 primary schools in Ireland, with Wave 3 representing 74% retention of the original sample (n = 6216) (Murphy et al., 2019). A set of weights and grossing factors were calculated for use in analysis to adjust the sample to ensure it was representative of the general population across key socio-demographic variables (Murphy et al., 2019). 6155 respondents who completed Wave 3 of the study and answered SOGI questions were included in these analyses.

Measures
The LGBT+ acronym comprises two dimensions: sexual orientation and gender identity. While the GUI survey relies primarily on identity measures of both sexual orientation and gender, it is important to note that there are other dimensions of both constructs. Sexual orientation encompasses identification, behavior and attraction, with higher life-time observations of same-gender behavior and attraction than identification (Gates, 2011). Gender identity includes someone's internal sense of gender, extends beyond a binary definition of male/female, may not accord with birthassigned sex (American Psychological Association (APA), 2015), and may include affiliation with a gender community (Spock et al., 2022). People have both a sexual orientation and gender identity.
In the GUI survey, phrasing and response options largely focus on respondent identity; however, sexual orientation uses attraction in clarifying language, and included: 'heterosexual/straight'; 'gay or lesbian'; 'bisexual'; 'questioning/not sure'; 'asexual'; 'don't know'; or 'prefer not to say' (GUI, 2016) (Fig 2). Inclusion of a 'questioning' response is recommended best practice (NASEM, 2020(NASEM, , 2022. Lorenz (2021) argues that inclusion of the 'asexual' option, a rarity in population-based data, may help improve data quality. This demonstrates a distinction that GUI can contribute to the international literature.
Respondents were also asked whether they describe themselves as 'male'; 'female'; 'other'; or 'prefer not to say'. Additionally, they were asked whether they would describe themselves as transgender, with possible responses of 'yes'; 'no'; or 'prefer not to say' (GUI, 2016) (Fig 2).

Statistical Analysis
Analyses were conducted using Stata 17.0, using GUI provided population-based weights to analyze the proportion of LGBT+ youth. Drawing on best practice, 'don't know', 'prefer not to say' and 'refused' responses were not categorized as sexual and gender minority youth in these analyses (Dahlhamer et al., 2014). Following archiving, anonymized GUI Research Microdata Files were made available on a strictly confidential basis, subject to the authors' appointments as Officers of Statistics by the CSO (CSO, n.d.; GUI, n.d.). Data were recoded and descriptive statistics generated, with t-tests using p < 0.05 as our threshold for significance. All analyses ensured that youth in both categories, that is gender minority youth, who also identify as LGB+, were not counted twice.

Results
At 17-18 years of age, one-in-ten young people in Ireland (10.0%) identify as lesbian, gay, bisexual, transgender, questioning, asexual, or describe their gender as other (Table 1).

Sexual Orientation
Young women are significantly more likely (p < 0.05) to identify as lesbian, gay, bisexual, questioning, or asexual, than young men (10.5% vs 7.7% respectively) ( Table 2). Nearly half of sexual minority youth identify as bisexual, with a significant gender difference where young women are one-and-a-half times more likely as young men to identify as bisexual (5.8% vs 3.1%; p < 0.01). Young men are statistically almost twice as likely to identify as gay (2.7%) as young women are to identify as lesbian or gay (1.3%; p < 0.01). A quarter of sexual minority youth in Ireland describe themselves as questioning or unsure of their sexual orientation, with young women more likely to describe themselves as questioning (3.1%) than young men (1.8%). A small percentage of youth identify as asexual (0.2%).

Gender Identity
In relation to gender identity, just over 1% identify as gender minority youth (1.1%), with 0.7% identifying as transgender and 0.5% describing their gender as other (Table 3). Over half (59.1%) of gender minority youth identify their sexual orientation as lesbian, gay, bisexual, or asexual (LGB+). LGBTQAO (Sexual Orientation and Gender Identity)* All 10.0 6,155 9.0 11.0

Discussion
This is the first nationally representative LGBT+ estimate in Ireland. The finding that more than one-in-ten youth in Ireland aged 17-18 years identify as LGBT+ is broadly consistent with the upper end of the Irish approximations (L&RS, 2019). However, the sexual orientation question in the GUI survey conflates sexual orientation identity terminology with definitions of sexual attraction (Fig 2). Many surveys record higher proportions of same-gender attraction than LGBT+ identification (Gates, 2011;National Center for Health Statistics, 2022). Further, the responses only allow for a gender binary, for example, the definition of bisexual in the GUI survey assumes two genders (male/female) (Fig 2). There is a slight possibility that some respondents did not understand the question, particularly if they themselves do not self-identify as heterosexual/straight (Dahlhamer et al., 2014). YuPP:) Project consultations frequently mentioned the marriage equality referendum as a reference point, suggesting Cohort '98 were more likely to understand SOGI questions (Ceatha et al., 2023). However, they observed that questions placed immediately prior to those on SOGI ask about ease of talking openly about sex, firstly with respondents' mothers, then repeated regarding fathers, and may have affected respondents' comfort in answering these questions accurately (Ceatha et al., 2023). (Fig 2). This has resonance for a sample population living at home and attending secondary school at the time of data collection (McNamara et al., 2018), especially for those not 'out' at home or school (Bry et al., 2017;Watson et al., 2015). Despite these caveats, the finding of 10.0% LGBT+ youth self-identification reflects international trends. Jones (2022) analyzed Gallup data from 2017, (one year after GUI data was collected), and found 10.5% of young adults in the US, born after 1997, identified as LGBT+.
In relation to sexual orientation identification of 7.1% (lesbian, gay, bisexual, asexual), similar patterns have been observed in Census 2021 findings for England and Wales, with 6.91% of younger age groups, aged 16-24 years, identifying as sexual minority youth (Office for National Statistics (ONS), 2023). The 2015 Healthy Kids Colorado Study found 9.3% of youth with a mean age of 15.8 years identified as LGB (Walls et al., 2019). Research using the US Youth Risk Behavior Survey (YRBS) found that in 2015, 8.3% of students between 14 and 18 were LGB (Olsen et al, 2020), rising to 10.4% in 2017 (Kann et al, 2018). The Irish findings are also consistent with gendered patterns reported internationally, with more young women identifying as LGB, and the majority describing themselves as bisexual (ONS, 2023;Jones, 2022;Kann et al., 2018). For example, while 8.0% of 2017 YRBS respondents identified as bisexual, this was higher for female youth (13.1%) than their male peers (2.8%) (Kann et al., 2018).
The relatively high numbers of young people describing themselves as questioning (2.5%) accords with estimates of 2.3% from the 2015 YRBS (Olsen et al., 2020), rising to almost a third of sexual minority youth in the 2017 YRBS (4.2%) (Kann et al., 2018). The inclusion of a questioning response may provide nuance, important for the understanding of sexual orientation self-identification as a process, rather than a once-off event which assumes "well-defined starting and ending points" (Savin-Williams, 2005:81). Rather, sexual orientation may be regarded as dynamic, with a subsequent rejection of rigid identity labels by young people (Ceatha et al., 2021;Coleman-Fountain, 2014;Diamond, 2016;Formby, 2017;Lorenz, 2021).
The percentage of youth in Ireland identifying as asexual (0.2%) is consistent with the Census 2021 findings of 0.18% (ONS, 2023). However, both findings are smaller than the 1.6% of asexual self-identification found by Rothblum et al. (2020), the first study to assess asexual identification in a US representative sample. The GUI description of asexual as 'not attracted to either sex' does not distinguish between types of attraction, for example emotional, romantic or sexual, which may have affected findings (Fig 2). NASEM (2022) emphasizes the need for response categories which recognize broader conceptualizations of attraction, and the absence of attraction. Robbins et al. (2016) found that scepticism from family and friends, lack of acceptance and misunderstanding may result in the non-disclosure of an asexual identity. The AVEN Community Survey found a majority of asexual respondents were aged 18-27 years, with a median age of 21 years (Ginoza et al. 2014). Inclusion of the 'asexual' response underscores the uniqueness and future benefit the GUI longitudinal survey offers.
In relation to gender identity, the percentage of 1.1% gender minority youth is broadly similar to that observed in the US. Herman et al. (2022) found that 1.3% of those aged 18-24 years identified as transgender or gender non-conforming, as did 1.4% of those aged 13-17 years. De Pedro et al. (2018) found that 0.8% of respondents aged 12-17 years identified as transgender in the 2015-2017 California Healthy Kids Survey. Analyses of the 2019-2020 California Health Interview Survey found that 1.5% of 14-to 17-year-olds either described themselves as transgender or reported a gender different to that assigned at birth. 4 Gender identity proportions may be an underestimation if it fails to capture gender minority youth who don't use the terminology of transgender and identify as male or female (Kronk et al., 2021). The 2015 Minnesota Student Survey used 2 consecutive questions on birthassigned sex and gender, with 2.7% of 14-to 17-year-olds identifying as transgender or reporting a non-binary gender (Eisenberg et al., 2017;Gower et al., 2018). Within the GUI context, the YuPP:) Project confirmed that we have less knowledge about how those who describe their gender as 'other' identify (Ceatha et al., 2023). The use of 'other' may be a literal 'othering', and imply that genders beyond the binary of male and female are abnormal (Ceatha et al., 2023;Kronk et al., 2021;Cameron & Stinson, 2019). APA (2015) recognize gender as non-binary and call for accurate and inclusionary measurement.
The findings that over half of transgender and gender other youth also identify as LGB+ (59.1%), is broadly comparable with Walls et al. (2019) who found that transgender youth were equally likely to identify as LGB as heterosexual. This accords with Meyer, Wilson, & O'Neill (2021) who note a greater diversity of sexual minority identification among transgender adults.
GUI respondents may experience a broader expression of their orientations and identities than reflected in survey response options (Ceatha et al. 2021;Coleman-Fountain, 2014;Diamond, 2016;Formby, 2017;Kronk et al., 2021;Lorenz, 2021;Spock et al., 2022;Ward, 2021). Ceatha et al. (2021) found that the '+' symbol from the LGBT+ acronym included: 23 terms used by sexual minority youth, alongside lesbian, gay and bisexual; 16 terms for transgender youth; with 16 descriptors encompassing broader gender minority identities. The YuPP:) Project concurs that the findings reported in this brief suggest Cohort '98 respondents negotiated SOGI conceptual and measurement issues within GUI (Ceatha et al., 2023).
GUI are committed to ensuring visibility of LGBT+ youth populations in data collection, and that young people "see themselves reflected in the questions put to them" (GUI, 2022:25). The study team has welcomed the template developed by the YuPP:) Project as a resource and will consider its inclusion in the draft GUI questionnaire, due for piloting with Cohort '08 at age 17. This is encouraging in a rapidly changing environment, with a consistent trend of increased LGBT+ identification observed with younger generations, as they reach adulthood. For example, in the US in 2017, 10.5% of youth identified as LGBT+, rising to 20.8% in 2021, and remaining steady at 19.7% in 2022 (Jones, 2022;. While in England and Wales, youth who identified as LGB+ rose from 4.4% to 6.6% between 2018 and 2019 (ONS, 2021). This rise was confirmed by Census 2021 (6.9%) with youth aged 16-24 years more than twice as likely to identify as LGB+ than the overall population (ONS, 2023). In Northern Ireland, Census 2021 found 2.1% of the overall population identify as LGB + , however, this more than doubles for youth aged 16-24 years (4.6%) (Northern Ireland Statistics and Research Agency, 2023). Statistics Canada (2021) found almost a third of the sexual or gender minority population (LGBTQ2+) were youth aged under 25 (compared with 14% of the non-LGBTQ2+ population).
This brief advocates for SOGI data collection across national surveys and in the Census in line with stated Irish policy objectives. It concludes with an appeal for the analyses and publication of SOGI data when collected. The CSO requirement of strict statistical disclosure control measures is commended, with SOGI data considered special category data (CSO, n.d.; GUI, n.d.). Such protections align with recommended strategies to maintain respondent confidentiality to improve SOGI measurement (Spock, et al., 2022). The case-by-case consideration of statistical outputs produced with small group sizes as vital statistics accords with CSO best practice (Linehan & Dineen, n.d.).
While these analyses report data from one point in time, GUI, as a population-based study, offers unique insights. Cohort '98 are the first generation in Ireland to grow up in an environment where social acceptance coincided with a time when many LGBT+ youth begin the process of coming out and sharing their sexual orientation and gender identity with others. The self-identification of onein-ten LGBT+ youth in Ireland likely reflects improved social climates, increased visibility, and changes in social attitudes. On this basis, we expect LGBT+ identification to rise. The planned inclusion of SOGI questions with Cohort '08 offers rich possibilities for future cross-cohort comparative analyses. study team for SOGI data collection in future waves. Many thanks to the YuPP:) Project members who provided pivotal feedback on the preliminary findings, particularly Ayrton Kelly, Tara Killeen, Katie McCabe, James Murray, Jayson Pope and Niamh Scully for their rich insights. The authors would like to express appreciation for the suggestions from the peer reviewers which strengthened this manuscript. Funding Open Access funding was provided by the IReL Consortium. This work was conducted as part of the SPHeRE programme under Grant No. SPHeRE/2013/1. The first author (NC) was awarded a fee waiver from University College Dublin, Ireland. The first author (NC) was a member of the Oversight Committee for the Irish LGBTI+ National Youth Strategy which specified analyses of Growing Up in Ireland as a policy objective.

Author Contributions
Data Availability Results in this research brief are based on analyses of Wave 3 data from Growing Up in Ireland, with access to Research Microdata Files only by application to the Irish Central Statistics Office (CSO). The authors requested an exemption from making the data available, as the analyses are based on a proprietary data set, which is considered sensitive.

Declarations
Ethical Approval The authors were appointed as Officers of Statistics by CSO to access the Research Microdata Files of the GUI Wave 3 dataset. Ethical approval exemption was granted by the Human Research Ethics Committee, University College Dublin,.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.