Psychological distress mediates the link between bullying and truancy in Australian LGBQ+ adolescents with experiences of homelessness

ABSTRACT

Homelessness is a high prevalence experience among lesbian, gay, bisexual, and queer (LGBQ+) youth, and potentiates both significant morbidity and future homelessness. While educational attainment is cited as a structural solution to homelessness, limited research exists on homeless LGBQ+ youths’ schooling experiences. A nationwide, cross-sectional survey involving 4,370 cisgender LGBQ+ participants aged 14–21 was conducted to address this gap. Regression analyses showed previous homelessness was associated with an increased odds of verbal, physical and sexual bullying and harassment within school contexts. Mediation analyses confirmed the mediating role of psychological distress on the associations between physical and sexual harassment and truancy, regardless of past homelessness. Previous experiences of homelessness are not associated with increased distress in relation to bullying or harassment. However, the indirect effect of harassment on truancy was significantly more pronounced for youth with past homelessness. Our findings crucially suggest that bullying presents both direct and indirect structural risks to cisgender LGBQ+ youth, and is particularly inconducive to continued engagement with schooling for youth with experiences of homelessness. Interventions aiming to address truancy among this group should aim to reduce the incidence of bullying within educational settings, whilst also providing assistance for managing the challenges associated with experiencing homelessness.

KEYWORDS:

• LGBTQ+
• young people
• adolescents
• truancy
• homelessness
• bullying
• Australia

Introduction

As defined by the Australian Bureau of Statistics, a person is considered homeless if they ‘do not have suitable accommodation alternatives and their current living arrangement is in a dwelling that is inadequate, has no tenure, or if their initial tenure is short and not extendable; or does not allow them to have control of, and access to, space for social relation’ (Australian Bureau of Statistics, 2014). In both Australian and other advanced economies, homelessness constitutes a persistent, burgeoning, yet largely preventable public health concern (Bassuk et al., 2020; Fransham & Dor“ing, 2018) that deleteriously impacts the health of individuals families and communities (Clifford et al., 2019). One pernicious aspect of homelessness relates to the difficulty associated with exiting homelessness; indeed, predominant perspectives suggest that negative early life experiences may initiate a trajectory that precipitates future homelessness risk (Milburn et al., 2009).

Regardless of societal context, the social concentration of homelessness frequently aligns with pre-existing marginality and disadvantage, and homelessness tends to disproportionately impact minoritized populations (Fowle, 2022). Experiences of homelessness are particularly prevalent among both cisgender LGBQ+ young people as well as transgender and gender diverse young people (LGBTQ+) (Rhoades et al., 2018; Shelton et al., 2018). Popular perceptions may suggest that homophobic parental rejection is the exclusive driver of youth homelessness among this group. In actuality, the reasons for this are myriad, as LGBTQ+ populations experience unique structural, institutional, and biographical factors which interact to form specific pathways to homelessness (McCarthy & Sadie, 0000). Consistent across many LGBTQ+ individuals’ experiences of homelessness, however, are a permutation of elements and life events which relate to parental rejection, discrimination within schools or other settings, and/or other sexual and/or gender minority stressors (Sara et al., 2022).

Irrespective of age, homelessness is a traumatic experience in and of itself (Jack et al., 2020). However, when experiences of homelessness occur during developmentally formative windows (e.g. adolescence), it is thought to add complexity to the risk factors which exacerbate one’s likelihood of future homelessness (Flatley et al., 2022). As homelessness is a high prevalence experience among LGBTQ+ young persons, it can be reasonably assumed that these developmental disturbances, and their associated negative outcomes disproportionately impact LGBTQ+ young persons. One important factor may be disruption to education – for instance, young people experiencing homelessness graduate, pass standardised tests and experience timely academic progression at lower rates than their peers (Forge et al., 2018; Parrott et al., 2022). For LGBTQ+ young persons, these challenges are experienced concurrently with other sexual and/or gender identity-related barriers to educational attainment, such as bullying and harassment (Ridings, 2020). While these obstacles are often experienced by LGBTQ+ pupils regardless of experiences of homelessness, there is presently little research that investigating the interface of these experiences within this population.

In all likelihood, experiences of homelessness are likely to confer a greater degree of social and economic vulnerability among LGBTQ+ individuals throughout life course, relative to their cisgender and/or heterosexual peers. Educational attainment is a key pre-requisite of future economic and civic enfranchisement (Parrott et al., 2022). Therefore, it is thought to offer individuals some protection against future experiences of homelessness (Heerde et al., 2020; Manfra, 2019). Educational attainment also reduces the likelihood that individuals will experience poverty (Thulitha et al., 2012), as well as outcomes associated with poverty that confer risk of homelessness, such as justice-involvement and incarceration (Lansford et al., 2016). Facilitating greater access to educational opportunity by encouraging sustained engagement with schooling is therefore often seen as an indispensable structural solution to the issue of homelessness (Parrott et al., 2022).

Understanding links between bullying and truancy

A significant contributing factor to disruptions in schooling is truancy – commonly defined within the reviewed literature as deliberate and unauthorised absences from class (Rocque et al., 2016). Contemporary scholarship suggests that truancy is associated with a range of negative academic and social outcomes (Escario et al., 2022), and may itself independently inform future vulnerability to homelessness. Truancy is a global issue and varies significantly across national and even jurisdictional contexts. Data from the Organization for Economic Co-Production and Development (Keppens & Spruyt, 2018) gathered across sixty-three member countries suggests that 27% of students surveyed reported skipping at least one class in the two prior two weeks. Within Australia, previous research suggests comparatively modest, yet still substantial rates of truancy – ranging from 7.5% to 17.7% (Australian Institute for Teaching and School Leadership AITSL, n.d.)

Truancy may be motivated by a plethora of factors, including boredom, avoidance of educational settings, academic burnout, low attachment to one’s school and even other students’ non-attendance (Escario et al., 2022; Gerth, 2020). However, peer victimisation or bullying is also an exceedingly common reason for truancy. School-based bullying is often a sustained and repeated experience for an affected individual (Shelley & Swearer, 2015). Bullying can jeopardise a young person’s attachment to their school, facilitating disciplinary problems and negative peer relationships, whilst constraining opportunities for positive social development (Cardwell et al., 2020). Bullying is also shown to motivate truancy, as students are likely to absent themselves from school to avoid further experiences of bullying (Cardwell et al., 2020).

Bullying is thought to be a common experience, with scholars estimating that anywhere from 10% to 33% of students within the general population experience peer bullying (Shelley & Swearer, 2015). Relative to these already-high incidence rates, LGBTQ+ pupils experience heightened rates of bullying due to their minority sexual identities (K. A. Tyler & Schmitz, 2018). This is often conceptualised as ‘discriminatory bullying’, or victimisation that includes a discriminatory component which targets an individual’s social characteristics and identities (Elamé, 2013). These concerns are salient to even those LGBTQ+ students who are not ‘out’ to their schoolmates, as they may nevertheless experience bullying that targets other characteristics – most notably, any expression of gender-nonconformity (Chan, 2022; Meyer, 2020).

For LGBTQ+ students, while certain forms of bullying (e.g. verbal, physical, etc.) may abate with age, others such as sexual harassment assume greater prevalence (H. Tyler et al., 0000). Discriminatory bullying contributes significantly to mental ill-health among LGBTQ+ student populations and is likely also linked to both truancy and disrupted schooling; these associations are even stronger among homeless LGBTQ+ youth, who appear more likely than their housed peers to report experiences of bullying and, subsequently, truancy and disrupted education (Cull et al., 2006; Gaetz et al., 2016; McCarthy & Sadie, 0000).

How might bullying shape experiences of truancy for homeless youth?

Extant evidence provides further insight into why young people who experience homelessness may also be more likely to experience school-based bullying and associated effects including truancy. Social development models suggest that a child’s position within the social structure and hierarchy of any given educational environment is often a function of biographical factors such as socioeconomic status, race, and sexuality (Catalano et al., 2004). The same circumstances which predict or predispose certain populations to experiencing homelessness also denote their vulnerability to discriminatory bullying (Moore et al., 2019). LGBTQ+ young persons’ behavioural responses to discriminatory bullying may also intersect with parental rejection of their sexuality and/or gender identity. Specifically, the psychological distress resulting from these experiences may shape behavioural issues at home in ways that potentiate familial conflict and alienation (Heerde et al., 2020; Tunåker, 2015) – possibly accelerating these youths’ homelessness trajectories. Hence, social agents which facilitate LGBTQ+ youths’ initial experiences of homelessness may not be anti-social peers, as has been classically presumed (Milburn et al., 2009), but instead relate to negative peer relationships.

At present, however, the relationship between homeless LGBTQ+ youths’ experiences of bullying and truant behaviours are undertheorized. In general, policies relating to student homelessness have tended to place an emphasis on facilitating students’ access to and success in school, with little consideration given to their actual schooling experiences (Moore et al., 2019). Additionally, much of the available research investigates experiences of homelessness occurring within the U.S. and Canada, with data on LGBTQ+ homelessness in other societal contexts like Australia comparatively sparse (Côté & Blais, 2019; McNair et al., 2022). As the unique vulnerabilities of LGBTQ+ populations are presently rendered invisible within the context of Australian social policy on homelessness, with current policy not recognising LGBTQ+ populations as needing either targeted early intervention or housing assistance support (McNair et al., 2022). Hence, research investigating this issue within Australian contexts is nascent (McNair et al., 2022), and it is presently uncertain whether findings from other contexts are applicable to Australian LGBTQ+ youth (see, for example: Alberton et al., 2020).

The Australian context

Since 2016, Australian schools have been the site of significant controversy and moral panic centred upon attempts to make these environments more inclusive for LGBTQ+ pupils (Shannon & Smith, 2017). Safe Schools was a health promotion project started in Victoria in 2010 which also provided resources to educators to address homophobia and transphobia within classrooms. Following its success within the state, the initiative originally gained bipartisan support in Canberra, where it saw a federally funded rollout in 2014 (Shevlin & Gill, 2020), but was soon the subject of concerted and sustained vilification spearheaded by a local newspaper – the Australian, right-wing members of the then-governing party, as well as the Australian Christian Lobby (Baird & Reynolds, 2021). This backlash from conservative elements in the within the federal government eventually caused the program to be federally defunded and discontinued across the majority of Australian schools (Thompson, 2019). While research investigating the impact of this moral panic, and the discontinuation of Safe Schools is presently scarce, there is some limited evidence to suggest that experiences of bullying and harassment among LGBTQ+ youth remain consistently elevated within contemporary schooling environments (Parker et al., 2023; Shevlin & Gill, 2020)

The present study

This paper therefore broadly aims to address the above discussed gaps by bringing together structural and individual risk factors for cisgender LGBQ+ young people’s engagement in school to understand how experiences of homelessness might potentiate risk pathways for truanting. We tested a multiple-groups mediation model to understand how mental distress moderates the relationship between of physical and sexual harassment in school contexts and truanting among LGBQ+ young people with and without experiences of homelessness.

Methods

Sample and procedure

This study extracted data from Writing Themselves In 4 (WTI4), a cross-sectional, anonymous online survey that examined the health and wellbeing of LGBTQA+ persons aged 14–21 years in Australia. The survey was publicised via Facebook and Instagram advertisements, and was conducted in late 2019, prior to any known outbreaks of COVID-19 in Australia. Eligibility for participation was contingent upon a prospective participant: identifying as LGBTQA+, residing in Australia, and being 14–21 years of age. The study drew 6,418 legitimate responses, and the present article focuses on a subsection of participants (n = 6,114) enrolled in educational institutes such as secondary school, University or Technical and Further Education (TAFE¹) at the time of participation or within the past 12 months. Research ethics approval for the WTI4 study was granted by the La Trobe University Human Research Ethics Committee.

Materials and measures

Demographic measures

Participant gender identity was assessed through an item which prompted participants to select the term most descriptive of their gender identity. This list contained 17 items and included the option not to answer. Participants were categorised as cisgender if their assumed gender at birth was identical to their chosen gender identity. Due to the specificity of trans and gender diverse individuals’ experiences of homelessness, only participants categorised as cisgender were included in the present analysis. Analyses which specific examine trans and gender diverse individuals’ experiences are forthcoming.

Participant sexual identity was captured via an item asking them to select from a list of terms. Options included: (i) Lesbian, (ii) Gay, (iii) Homosexual, (iv) Bisexual, (v) Pansexual, (vi) Heterosexual, (vii) Queer, (viii) Asexual, and (ix) Prefer not to have a label, (x) Prefer not to answer, (xi) Don’t know and (xii) Something Different. A small minority of participants selected (iii) Homosexual, (ix) Prefer not to have a label, (x) Prefer not to answer, (xi) Don’t know and (xii) Something Different. These groups were merged under (xii) Something Different. Participants who identified as ‘Homosexual’ were not re-categorised as either ‘Gay’ or ‘Lesbian’ to preserve their decision not to identity with these labels.

Experiences of homelessness

Borrowing from a previous study of youth homelessness in the U.S. (Perlman et al., 2014), homelessness was defined to participants as ‘not having a stable or safe place to live and can include things like sleeping outside, and living or sleeping in a car, shelter, hostel, or refuge’. Lifetime prevalence of homelessness was measured through a binary ‘yes’/’no’ response to an item asking participants whether they had ever experienced this circumstance, with ‘no’ set as the reference outcome.

Physical and sexual harassment in educational settings

Participants’ experiences of sexual and/or gender minority identity-related bullying within educational settings (e.g. school, university, or TAFE) in the previous 12 months were classed into two modalities: physical (e.g. being shoved, punched, or injured with a weapon) and sexual (e.g. unwanted touching, sexual remarks, sexual messages, or being forced to perform any unwanted sexual act). Given our sample’s relatively high mean age (17.29 years), we investigated both physical and sexual harassment to account for shifting profile of bullying behaviours that is expected for LGBQ+ individuals as they progress into middle and late adolescence. Responses were recorded on a ‘yes’/’no’ binary for each modality, with ‘no’ set as the reference outcome.

Kessler psychological distress scale (K10)

The K10 is a self-rated instrument measuring non-specific psychological distress (Kessler et al., 2002). It demonstrates high internal consistencies (Onegeri et al., 2022), unidimensional factor loadings (Smout, 2019), and consistent performance across socio-demographic groups (Furukawa et al., 2003). The instrument contains 10 items regarding previous affective experience in the last 4 weeks. Participants responded to items such as ‘(In the past 4 weeks) About how often did you feel depressed?’ on a 5-point scale ranging from: (i) None of the Time to (v) All of the time. Responses are scored and summed to arrive at a total score ranging from 10–50 points, with higher scores denoting greater psychological distress.

Australian normative data has been used to establish standard cut-offs indicating low (10–15), moderate (16–21), high (22–30) and very high (31–50) psychological distress (Andrews & Slade, 2001). For the purposes of the current analyses, however, the K10 was used as a continuous variable. At the time of writing, the K-10 has yet to be validated for use with LGBTQ+ populations, and normative data for sexual and gender minority populations is unavailable at the time of writing. However, the K-10 is generally considered valid and appropriate for use with these populations (Tan et al., 2022). The Cronbach’s alpha coefficient for the K10 instrument in the present sample was 0.92, indicating that the scale had good internal consistency within this study’s sample.

Truancy due to feeling unsafe or uncomfortable

Participants were asked about the number of days of schooling they had missed in the past 12 months due to feeling unsafe or uncomfortable. Response options provided were ‘No, I have not missed any days’, ‘part of a day’, ‘1 day’, ‘2–3 days’, ‘4–5 days’, ‘6–7 days’, ‘1–2 weeks’ and ‘more than 2 weeks’. This data was coded from 0–7 in the present analyses.

Statistical analysis

All statistical analyses were computed using STATA (Version 17 SE; StataCorp, College Station, TX). To handle cases of missing data, we first conducted multiple data imputations over 40 imputations using fully conditional specifications with homelessness as a grouping variable, including as auxiliary variables, gender, sexual orientation, disability, family religiosity, location (capital city, regional city, remote/rural) and ethnicity. After testing the relationship between experiences of homelessness and harassment via multivariable logistic regressions, we then estimated two multiple groups mediation models, analysing the mediational pathway between a form of harassment, mental distress, and truancy, stratified by homelessness. Because of the non-normality of residuals in the regression models, we used 1,000 bootstrap replications in our mediation analyses to generate percentile-based 95% confidence intervals for model parameters.

Results

Sample characteristics

Table 1 summarises descriptive statistics for all study variables.

Table 1. Sample characteristics (N = 4262).

								n	%
Sexual Orientation
Lesbian								567	13.3%
Gay								804	18.9%
Bisexual								1,653	38.8%
Pansexual								325	7.6%
Queer								254	6.0%
Asexual								162	3.8%
Something Else								487	11.4%
Gender Identity
Cis Man								1,221	28.6%
Cis Woman								2,904	68.1%
Age
14–17								2,773	65.1%
18–21								1,489	34.9%
Ethnicity
Anglo-Celtic Only								1,713	40.2%
Multicultural								2,372	55.7%
Country of Birth
Australia								3,754	88.1%
Other								507	11.9%
Location
Capital city, inner/outer suburban								2,831	66.4%
Regional city or town								1,010	23.7%
Rural/Remote								418	9.8%
Educational Institute
Secondary School								2800	65.7%
Alternative Education Program								93	2.2%
Special Needs School								8	0.2%
Private College								52	1.2%
University								1,103	25.9%
TAFE								156	3.7%
Other								50	1.2%
Lifetime experience of homelessness
No								3,442	80.8%
Yes								787	18.5%
Homelessness during the past 12 months
No								3,965	93.0%
Yes								393	9.2%
Currently experiencing homelessness?
No								4,164	97.7%
Yes								62	1.5%
Recent (<12 months) experience of Physical Harassment within School Settings
No								3,829	89.8%
Yes								397	9.3%
Recent (<12 months) experience of Verbal Harassment within School Settings
No								3,296	77.3%
Yes								800	18.8%
Recent (<12 months) experience of Sexual Harassment within School Settings
No								3,438	80.7%
Yes								253	5.9%
School days missed in the last 12 months
None								3,050	71.6%
Part of a day								260	6.1%
1–3 days								441	10.3%
4–7 days								185	4.3%
1–2 weeks								87	2.0%
More than two weeks								230	5.4%
K10 Score Band
Low or Moderate								937	22.0%
High or Very High								3,297	77.4%

Sample Characteristics (N = 4262).

The average participant age was 17.29 years (SD = 2.05 years), and most participants (89.0%, n = 5,558) reported being born in Australia. Most participants identified as either cisgender women (50.6%, n = 3,162) or cisgender men (22.3%, n = 1,394). Bisexual participants (33.8%, n = 2,164) comprised the largest group within the sample population, followed by gay (16.6%, n = 1,063), lesbian (12.0%, n = 771), pansexual (11.2%, n = 717), queer (8.4%, n = 540) and asexual (4.60%, n = 295) participants. Most participants resided in outer suburban (57.8%, n = 3,705) areas of capital cities and almost one-quarter in regional cities or towns (24.9%, n = 1,598). Most participants were students at either a secondary school (60.0%, n = 3,850) or university (24.1%, n = 1,545) during the previous 12 months. Almost a quarter of all participants (23.6%, n = 1,501) reported experiencing homelessness at some point in their life. With respect to harassment in an educational setting, 4.71% (n = 245) of participants reported experiencing physical harassment within school settings during the last 12 months, and 71% (n = 358) reported experiences of sexual harassment in the past 12 months.

Preliminary analyses

Prior to conducting mediation analyses, a series of univariable logistic regressions were conducted to understand the relative prevalence of experiences of harassment among participants with and without experiences of homelessness. These results suggested that experience of homelessness was associated with greater odds of both physical harassment (OR = 4.08, 95% CI [3.08, 5.41]) and of sexual harassment (OR = 2.82, 95% CI [2.18, 3.64]). Pearson’s correlational analyses were also conducted on the different forms of harassment and are detailed in Table 2.

Table 2. Pearson correlation.

		Physical Harassment	Sexual Harassment	Verbal Harassment	K10 Score
Physical Harassment		1.000
Sexual Harassment		0.258	1.000
Verbal Harassment		0.359	0.296	1.000
K10 Score		0.122	0.110	0.164	1.000

Mediation analyses

Multiple-groups mediation models for physical harassment (see Figure 1) suggested that the direct effects between experiences of harassment to psychological distress were similar in magnitude between those without (β = 3.69, 95% CI [1.99, 5.35]) and with (β = 4.06, 95% CI [2.47, 5.63]) experience of homelessness; the direct effect between harassment and truancy was also numerically, but not statistically significant, greater in those with experience of homelessness (β = 1.40, 95% CI [0.90, 1.90] vs β = 1.31, 95% CI [0.85, 1.78]). However, the direct effect linking mental distress and truancy was substantially higher in those with experience of homelessness (β = 0.12, 95% CI [0.10, 0.14]) than in those without experience of homelessness (β = 0.08, 95% CI [0.07, 0.09]). Similarly, multiple-groups structural equation models for sexual harassment revealed a numerically, but not statistically significant, greater path between harassment and mental distress, but a statistically significant greater relationship between mental distress and truancy.

Figure 1. Multiple-groups structural equation models for physical and sexual harassment experienced by participants with experiences of homelessness.

Thus, we re-estimated multiple-groups models fixing paths that were not significantly different between groups to be equal (see Figure 2 and Figure 3). In models for physical harassment, indirect effects were significant for those with experiences of homelessness (α₁β₁ = 0.45, 95% CI [0.30, 0.61]) and without experience of homelessness (α₂β₂ = 0.30, 95% CI [0.21, 0.39]); moreover, the indirect effect was significantly greater for those with experience of homelessness (difference in paths 0.15, 95% CI [0.07, 0.26]). Similarly, in models for sexual harassment, comparing indirect effects for those with experience of homelessness (αβ = 0.19, 95% CI [0.10, 0.28]) and without experience of homelessness (αβ = 0.30, 95% CI [0.16, 0.44]) generated significantly different indirect effects (difference in paths 0.11, 95% CI [0.05, 0.19]).

Figure 2. Comparison of initial (I) and re-estimated I multiple-groups model for physical harassment.

Figure 3. Comparison of (I) and re-estimated(R) multiple-groups model for sexual harassment..

Discussion

The current findings add to a small but growing body of literature investigating homelessness among LGBQ+ youth within Australia and elaborates upon how the schooling experiences of this group informs their educational outcomes. Past research had consistently demonstrated that LGBQ+ individuals are at a higher risk of experiencing both physical and sexual violence (Atteberry-Ash et al., 2020), and separately, that homeless persons are similarly vulnerable to these experiences (Heerde et al., 2020). Our findings add to these findings by highlighting how these vulnerabilities may overlap in the experiences of LGBQ+ young people experiencing homelessness. More specifically, they show that young LGBQ+ people who are homeless are at more likely to have experienced school-based harassment which contributed to higher rates of truancy among this group; moreover, that these compounded vulnerabilities do not simply reflect the greater opportunity for violent experiences that are associated with homelessness (Mackie et al., 2019).

It is not possible to tell from these data what factors increase young people’s vulnerability to school-based harassment on the basis of being homeless. It is possible that the increased prevalence of harassment within this group reflects broader societal stigma attached to homelessness and homeless persons (Pernilla et al., 2020). Concurrently, given that homelessness is associated with factors such as low socio-economic status, and holding marginalised identities or experiences (e.g. relating to racial minority identity or disability), the elevated incidence of bullying among homeless participants may also reflect experiences of victimisation attributable to these identities. Additionally, these results shed some light upon the lasting impact potentiated by experiencing homelessness. The current analyses investigated lifetime experiences of homelessness instead of recent (<12 months) experiences due to low N in the latter category. However, insofar as previous experiences of homelessness are generally predictive of future homelessness, our findings nevertheless suggest that homelessness may confer a higher likelihood of experiencing violence, or at least denotes a stable raft of factors which are associated with experiences of violence.

Crucially, our findings also suggest that experiences of homelessness neither sensitises nor desensitises LGBQ+ young persons to the psychological impact of bullying – irrespective of prior experiences of homelessness. Specifically, while the mediational models linking harassment, mental distress and truancy were similar in several respects between those with and without experience of homelessness, previous experiences of homelessness uniquely potentiated the associations between school-based harassment and truancy. This accordingly suggests that though adolescents with experiences of homelessness are more likely to experience bullying, they are not more affected by bullying compared to other LGBQ+ youth. With respect to wellbeing, participants with experiences of homelessness were not more likely to report greater psychological distress in relation to bullying than those who were not homeless, but they were more likely to report truancy. In other words, the findings suggest that people with experiences of homelessness are more likely to respond to harassment or bullying by absenting themselves from school, even if this victimisation is not experienced as more distressing by this cohort compared to their counterparts without previous experiences of homelessness.

Previous evidence demonstrates that homeless students may experience considerable obstacles to school attendance – including those relating to mobility and stability (Pavlakis & Nelson Pryor, 2021). However, our findings additionally suggest that homeless students’ schooling experiences themselves may constitute a further obstacle to school attendance. Truanting among homeless students who experience bullying may be primarily motivated by a desire to avoid these experiences – however, it is likely that these decisions are at least partly informed by the difficulty that regular attendance presents to these pupils. More specifically, a student whose engagement with schooling already disincentivized these difficulties may have a lower threshold for disengagement – one easily crossed when the resulting schooling experience involves bullying and harassment.

These findings also speak to the unique but oftentimes paradoxical position that educational institutes occupy in improving the health and well-being of LGBQ+ students experiencing homelessness. Schools are often a crucial source of stability and support and may well represent the last social institution that homeless youth are involved with (Moore et al., 2019), However, continual exposure to harassment or bullying can incur a considerable emotional cost for LGBQ+ students (Bidell, 2014). Several implications for truancy policies are evident here – particularly as they relate to students experiencing homelessness.

Specifically, our findings add to an increasingly critical consensus on the ineffectiveness (or damaging impact) of disciplinary or putative measures to deter truancy (Weathers et al., 2021). This is likely to be of particular salience to the Australian socio-legal context, where previous attempts to address truancy has historically involved the use of heavy-handed social policies. These policies previously attempted to coerce parental cooperation to address truanting by withholding welfare payments to socioeconomically disadvantaged families of truant pupils (Collingwood & Mazerolle, 2022). Given the links between truancy and homelessness suggested by previous and present findings, it is likely that such policies will simply increase pupils’ risk of homelessness, which in turn will constitute a challenge to further engagement in schooling.

In the present context, disciplinary responses may inadvertently penalise students’ prioritisation of their own emotional safety – particularly when implemented without any attempt to address the factors which motivate truanting. Worse, where these responses include some form of disciplinary exclusion (e.g. out of school suspensions or expulsions), they will likely run counter to the goal of improving schooling engagement and may accelerate marginalised students’ disengagement from the school system (Anderson et al., 2019). Instead, our findings imply that improving the schooling experiences of LGBQ+ students experiencing homelessness is a key component of improving educational outcomes for LGBQ+ students and for attenuating future vulnerability to homelessness.

Finally, during intervening period between data collection and the publication of these findings, the cost of both home ownership and rental housing in Australia have becoming increasingly prohibitive (Morris, 2023). The close alignment of social marginalisation with homelessness risk generally means that LGBTQ+ populations in Australia are disproportionately impacted by homelessness (McNair et al., 2022). In conjunction, this suggests that our findings are highly relevant to the contemporary context, and emphasise the need for educators to address these extracurricular factors. This is essential to ensuring continued engagement with schooling among this critically vulnerable group of students.

Limitations

These findings are circumscribed by several limitations. Firstly, experiences of being harassed (Strøm et al., 2018) and of being homeless (Markowitz & Syverson, 2021) are highly stigmatised, biasing participants against self-reporting these experiences. As such, these experiences are likely underreported within our sample, such that the impact of bullying on school truancy in the population of interest may be understated. Secondly, the present analyses did not differentiate between so-called ‘shor’-term‘ and “long”-term’ experiences of homelessness (Cutuli & Janette, 2019); current evidence convincingly demonstrates that protracted and/or recurrent experiences of homelessness both proportionately and detrimentally inform (Seastres et al., 2020) health outcomes for persons experiencing homelessness. However, other scholars have argued that the notion of a well-defined dose–response relationship between homelessness duration and associated negative health outcomes may be overly simplistic, and these effects may instead be tied to housing insecurity – potentially beginning long before individuals are homeless (Clark et al., 2019). Thirdly, as most of our sample was comprised of bisexual-identifying cisgender women, further exploration of the differences in the prevalence and correlates of homelessness between various identity subgroups may be an important direction for future research.

Lastly, the way that data about participants’ experiences of bullying was captured disallowed attributions of these experiences to specific institutional settings. The item in question made no distinction between secondary and post-secondary educational settings, and participant enrolment dates and year levels were not captured. Given the 12-month time frame specified in this item, responses may reflect experiences from either secondary school or post-secondary contexts, further complicating contextual analyses. Previous research suggests that both the prevalence, nature, and impact of bullying within secondary and post-secondary institutions is highly distinct among LGB+ pupils (Day et al., 2020; Koehler & Copp, 2021). However, the fundamental premise of the current investigation – that experiences of bullying and harassment are particularly associated with a higher likelihood of disengagement for adolescents with past experiences of homelessness – is likely to hold true across these different settings. Future research should nevertheless adopt a more granular approach to investigating this topic that accounts for institutionally specific forms of victimisation.

Conclusions

For LGBQ+ students, being homeless constitutes a structural risk for experiencing either physical and/or sexual bullying within school settings, and further potentiates the relationship between psychological distress and truancy. Additionally, as continual engagement in schooling is a prerequisite for educational attainment, and therefore predicts future susceptibility to homelessness, bullying may present a significant obstacle to exiting homelessness for LGBQ+ young people. Schools and educators are uniquely positioned to improve both the long- and short-term outcomes of LGBQ+ students experiencing homelessness and should adopt a holistic approach to addressing truancy that accounts for these pupils’ schooling experiences.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Participants in the present study have not given written consent for their data to be shared publicly. Due to the sensitive nature of the research supporting data is not available.

Additional information

Funding

The data presented above was derived from Writing Themselves In 4, a study of LGBTQ+ young people funded by: (i) the Victorian Department of Premier and Cabinet, (ii) the Australian Capital Territory Government Office for LGBTIQ+ Affairs, (iii) the New South Wales Department of Health and (iv) SHINE SA, with support from (v) the Office of the Chief Psychiatrist in South Australia.

Notes

1. Technical and Further Education refers to tertiary level vocational education and training in Australia (Rice et al., 2021)