Strengthening the Capacity of Education Staff to Support the Wellbeing of Indigenous Students in Boarding Schools: A Participatory Action Research Study

In meeting the social and emotional learning (SEL) needs of Aboriginal and Torres Strait Islander students, the capacities of school staff are critical. There is very limited evidence for relevant capacity development initiatives. This evaluation reports a multicomponent SEL training intervention delivered to staff of an Australian education service that operates independently of any particular school to assist with the transitions of students from remote communities to boarding schools. A participatory action research (PAR) approach was implemented over 13-months with 21 staff participants. Results from a pre-, mid- and six months post-training survey and staff interviews were analysed and fed back through reflective group discussions. The training was associated with improved staff attitudes to mental health and skills to support student wellbeing. Sixteen participants received a tertiary qualification. Despite ‘working in challenging environments’, staff were ‘dedicated to help’ students, and ‘acknowledged the need for change’ to better support student wellbeing. However, given the service's brokering role between families and schools, fewer staff members reported feeling empowered to influence issues in their workplace. The evaluation demonstrated the value of SEL training for education staff and potential utility for school teachers and boarding staff who have direct duty of care for Indigenous students. The multicomponent training described in this study would need to be condensed for school settings.

Evidence shows that such programs improve individual student competencies such as confidence about learning, a 'growth mindset' to persist when faced with challenges (Aronson, 2002), goal setting, stress management (Duckworth & Seligman, 2005) and problem-solving skills (Brooks, 2014). SEL programs have also influenced the culture, values and environments of schools, and the quality and nature of relationships between students and teachers and students with peers (Brooks, 2014). For students, these competencies can engender a sense of belonging and engagement in school that influences their readiness to learn (Flook, Repetti, & Ullman, 2005;Wang & Holcombe, 2010). School SEL programs also reduce the likelihood that psychosocial distress will progress mental illness through early identification of its signs and symptoms (Kalra et al., 2011). Early identification is important because 75% of mental illness manifests by age 24 (Kessler et al., 2005) and individualised primary healthcare and specialist service responses to mental illness are often poorly coordinated and resourced (McGorry, 2017). Important in supporting students' SEL are the capacities of school and education support staff. These capacities include the extent to which staff feels empowered and supported to undertake SEL initiatives, their attitudes to mental health, and their skills to support students. These environmental, attitudinal and skills-based factors influence how they interact with, provide opportunities for, and help recognise and support a student who is distressed (Centers for Disease Control and Prevention Substance Abuse and Mental Health Services Administration et al., 2012). Staff members' mental health literacy or attitudes toward mental illness also frame how they experience and express their own emotional problems and psychological distress, and whether they disclose these symptoms and seek care (Ajzen & Fishbein, 1980;Jorm et al., 1997). Supporting staff to feel confident and effective when engaging with students is central to instilling pedagogical change, building cultural competency and creating a welcoming, and inclusive school climate (Gower & Byrne, 2012). For example, an Australian whole of school SEL program to encourage engagement and interaction between teachers, parents, families, community and students demonstrated a significant increase in resilience for students experiencing depression, anxiety and emotional difficulties, which was maintained 12 months after the program was completed (Worsley, 2014).
Some schools attempt to provide SEL support to assist students to realise their full potential, but have inadequate knowledge of appropriate models, or resources and training to meet this goal (Heyeres, McCalman, Bainbridge, & Redman-MacLaren, 2016). For schools with culturally diverse populations, there is an even greater dearth of literature about how staff can best support SEL for students in culturally proficient ways. When working with Aboriginal and Torres Strait Islander (hereafter respectfully termed Indigenous) students, for example, as many as one in four Western Australian Aboriginal children (4-17 years) showed signs of serious emotional or behavioural difficulties compared to 15% for non-Aboriginal children (Zubrick et al., 2005). School staff then often struggle to meet the complex SEL needs of Aboriginal and Torres Strait Islander (hereafter respectfully termed Indigenous) students (Heyeres et al., 2016).
For boarding schools in particular, staff are required to support transitioning students who face separation from family at 10 or 11 years of age; often higher academic standards; changes in prescribed roles, responsibilities, and expectations; Western cultural and linguistic norms; and potential institutional discrimination and racism Stewart, 2015). Teachers at four Western Australian boys' boarding schools, for example, reported that while boarding schools presented Indigenous students with academic and social opportunities; student's also experienced challenges such as culture shock, homesickness, literacy and numeracy issues, prejudice and racism. Several of these teachers questioned whether their schools had the ability to support the learning needs of Indigenous boarding students; while some worried that educational aspirations for these students were lower than for non-Indigenous students. Positive relationships between staff and students, staff and parents, and school and community were perceived as critical to Indigenous students' success at school.
Many Indigenous students are compelled to attend secondary boarding schools because there is no secondary education provision in their home communities. There is a pressing need for whole of school culturally proficient SEL environments, yet many remote Indigenous students experience boarding school environments as a privileged educational and learning context within which they question their own belonging. Not only are there differences from their familiar primary school environments in school size; proportion of non-Indigenous students; uniforms; cultural cues and ways of being, knowing and doing; expectations; routines; food; weather; noise; population density; and language; the already difficult transitions to boarding school are made even harder by prejudice and overt and covert racism. For boarding schools to be places in which Aboriginal students feel safe and confident to express a sense of self and cultural identity, supportive environments and school practices need to be created that positively embrace complexity and diversity and where unacceptable attitudes and behaviours exhibited by staff and students are redressed. Hence, professional capacity development is important for boarding school staff to foster empathy with the significance of culture to Aboriginal boarding students and to develop skills for addressing their SEL needs.
A recent search of the Canadian, Australian, New Zealand and United States (CANZUS nations) literature found limited evidence of tailored staff capacity development initiatives to meet the SEL needs of Indigenous students' from CANZUS nations in boarding/ residential schools (Heyeres et al., 2016). An Australian Government report identified the need for additional support for already qualified staff as being of high importance (DEST, 2010). To improve culturally respectful education and practices, cultural awareness workshops programs have been provided in schools, led by experienced Indigenous educators and complemented by the involvement of Indigenous Elders and artists (Barr, 2009;DEST, 2010). Other strategies to enhance staff empowerment, attitudes to mental health and student support skills entailed professional feedback, supervision and counselling sessions (DEST, 2010;Meredith & Ryan, 2014), mentoring (Mander & Fieldhouse, 2009), networking (Mander & Fieldhouse, 2009), and professional supervision (Meredith & Ryan, 2014). But all studies were descriptive and none focused primarily on ways to enhance staff capacity to support Indigenous students (Heyeres et al., 2016).
This paper reports an evaluation of a multicomponent SEL staff capacity enhancement training intervention with members of an Australian education service that works independently from any particular school to broker the transitions of Indigenous students from remote communities to boarding schools. The aim of the SEL training was to inform the enhancement of the service's case management approach to better advocate for and support the wellbeing of Indigenous students who attend boarding schools. The research question was: Does SEL training influence the capacity of education staff to advocate for and support Indigenous student wellbeing?

Study Design
This research responds to needs identified by an Australian education service for enhanced support for Indigenous students' physical, social, emotional and cultural wellbeing, including reducing the risk of suicide. Consistent with the definition of Aboriginal health from the National Health Strategy in 1989, wellbeing was considered broadly to be: 'not just the physical wellbeing of an individual but . . . the social, emotional and cultural wellbeing of the whole community in which each individual is able to achieve their full potential as a human being' (AH&MRC, 2017, p. 3).
A strengths-based participatory action research (PAR) approach founded on social constructivism was applied (Burr, 2003;McCashen, 2005). PAR has been established as an acceptable and feasible approach for research with Indigenous people; it promotes sustainability, mutual trust and respect in the relationship, partnerships, ownership and empowerment in the process, and benefits to the research population (Bainbridge, McCalman, Tsey, & Brown, 2011;McCalman, McEwan, & Tsey, 2009;McCalman et al., 2013;Tsey & Every, 2000;Tsey et al., 2004). Participation of the education staff was embedded in all aspects of the research to ensure user capacitydevelopment, engagement and ownership, as well as immediate translation of the research results into practice (Bainbridge et al., 2011;McCalman et al., 2013;Tsey & Every, 2000). The evaluation of the SEL training utilised a sequential, complementary mixed methods design (Teddlie & Tashakkori, 2009). Data for the evaluation was gathered from the following: (1) Three waves of surveys, administered pretraining, after the initial training modules, and six months after the final training modules had been delivered; (2) reflective group discussions; and (3) individual interviews with service staff.

Setting of the Study
An Indigenous education support service (the service) was developed as a grass roots response from three remote communities to the needs of students for support in their transitions to boarding schools. The service now supports more than 500 students and families from 12 remote discrete Indigenous communities to transition to boarding schools. An evaluation of the service found that staff roles were demanding and complex, with some officers willing to go beyond what was expected to ensure that students' on-going and complex needs were met. The capacity of staff members to deal with situations as they arose was considered to be essential to students' succeeding and higher retention rates. It was recognised that on-going professional development was necessary, particularly for community support staff (Department of Education and Training, 2010).
At the time of study initiation (May 2015), the service had 21 staff members who were spread geographically across one state. They worked across three service areas: primary into secondary school transition; secondary school transition; and re-engagement for students who were excluded from boarding school. In the primary into secondary school area, staff worked with primary school (years 6/7) students and their families to assist them to apply to and take up placement at a boarding school, and support all student boarders in Years 8-12 to return to school each term on time. In the secondary school area, staff including youth mentors supported students enrolled at boarding schools to manage transition challenges and to develop opportunities that lead to Year 12 attainment (or equivalent) and to pathways beyond Year 12. One staff member supported students who were deenrolled from boarding schools to re-engage in learning or earning pathways. In addition to working with students, staff engaged with and between families, boarding schools and houses, and other services (e.g., social services, airlines and other transport services, health services) to support students' adjustment, orientation and ongoing stay at boarding schools.

Staff Training Intervention and Processes
The SEL training comprised three distinct but complementary training packages: Family Wellbeing Program (FWB), Aboriginal and Torres Strait Islander Mental Health First Aid (MHFA) and Resilience Training. These three components were chosen by the university researchers and co-researchers from the service in response to staff identified need to focus on both the risk and protective factors for mental health (Robinson, Silburn, & Leckning, 2012). Hence, as described below, FWB and Resilience training offered SEL competencies for navigating Indigenous wellbeing, while MHFA promoted awareness and early identification of risk. Training was followed by reflective PAR sessions with all staff members. The training was funded as part of the research study.

Family Wellbeing Training
The premise of FWB as a SEL program is that all human beings have basic physical, emotional, mental, and spiritual needs and failure to satisfy these needs results in behavioural problems (Tsey & Every, 2000). FWB was developed by an Aboriginal Education service to enhance the capacity and wellbeing of service staff and community members. The program was designed as a reflective and interactive small group training within which participants are able to raise their concerns, identify their strengths, build relationships and gain new skills through reflection on experience. The reflective nature of the program was ideally suited to participatory quality improvement processes. FWB has a sound evidence-base in group and community SEL efforts that work to collectively identify personal and community strengths and needs such as health and education improvement (Tsey & Every, 2000;Tsey et al., 2004;Tsey et al., 2010). Qualified facilitators were available within the research team.
The agreement between the service and the research team was to administer the introductory sections (30 h) of the accredited FWB program. The first two sections 'understanding self and improving personal interactions' and 'coping with grief and loss' were delivered in May 2015. Upon completion, the service requested training in the full certificate II in FWB Facilitation. The additional six sections plus a facilitators' training skill set were administered in the following months and ended in December 2015. Subsequent stages of FWB focused on managing emotions, managing stress, communicating effectively, deepening the understanding and recognition of psychological health. The first four sections of FWB training were delivered by an Indigenous researcher/facilitator who was supported by an Indigenous and a non-Indigenous researcher/facilitator; the last four were co-facilitated by staff members in order to build/demonstrate facilitation skills.

Aboriginal and Torres Strait Islander Mental Health First Aid Training
Aboriginal and Torres Strait Islander MHFA training was selected to increase service staffs' mental health literacy and knowledge of appropriate MHFA strategies that were sensitive to experiences of Indigenous people (MHFA Australia, 2008). The 14-h program covered topics on social and emotional wellbeing; mental health problems in community and Australian youth; adolescent development; anxiety; depression; suicide and nonsuicidal self-injury; psychosis and self-care. The training culminated in examination of the five-point Action Plan acronym ALGEE: Approach the person, assess and assist with any crisis; Listen nonjudgmentally; Give support and information; Encourage the person to get appropriate professional help; Encourage other supports. The training was delivered in May 2015 by two Indigenous facilitators through interactive presentations, video case studies, group exercises and the use of a manual.

Resilience Training
A one-day resilience-training workshop was facilitated for service staff and partner organisations, including schools and health services, in October 2015. The workshop was facilitated by international resilience research leader, Professor Michael Ungar from the Resilience Research Center at Dalhousie University, Canada, and encompassed handouts, group exercises and video recordings of case studies to demonstrate ways to nurture resilience in young people (Ungar, 2012). Resilience is defined: 'in the context of exposure to significant adversity' . . . as 'both the capacity of individuals to navigate their way to psychological, social, cultural, and physical resources that sustain their wellbeing, and their capacity individually and collectively to negotiate for these resources to be provided and experienced in culturally meaningful ways' (Ungar, 2012). Resilience theory has increasingly explored understandings of why some youth who experience adversity are able to overcome hardship and display positive developmental outcomes (Wilson & Gauvin, 2012;Zimmerman et al., 2013). School-based outcomes of resilience programs have demonstrated increased knowledge, improved attitudes to mental illness and suicide, lowered suicide attempt rates, and enhanced adaptive attitudes about depression and suicide postintervention (Closing the Gap Clearinghouse, 2013; Mann et al., 2005;Tsey et al., 2004). Studies have also found that improving problem solving, coping with stress, and increasing resilience enhanced the protective factors against suicide (Mann et al., 2005;Zimmerman, 2010). For Indigenous students, having the resilience to make healthy adjustments in times of high vulnerability is vital to maintaining their wellbeing.

Service Role and Participants
The role of the education support service was to support remote-dwelling Indigenous families to gain a boarding Post-training Survey administered school placement for their child and to broker the transitions of students to take up a placement and to complete the secondary phase of schooling. As a brokering service, staff members had no direct mandate over student wellbeing; rather, their influence was enacted indirectly and informally through advocacy via established relationships with families, students and schools. All 21 service staff participated in the multicomponent SEL training. The cohort was diverse in roles, ethnicity, gender, age and capacity. It comprised: the service manager, 14 field staff, 6 youth mentors. Eight were Indigenous and 13 non-Indigenous participants. Their ages included 10 under 35 years and 11 over 35 years; and 18 were female and 3 were male. Most of the field staff were trained teachers, vocational guidance counsellors or public servants; all had considerable experience in working with Indigenous students, families and schools. None of the youth mentors and two other staff members yet had formal tertiary qualifications. Staff turnover during the seven-month training period (May-December 2015) meant that five participants discontinued the training and two new staff members joined the cohort at different stages. Catch-up classes were provided to ensure maximum participation and graduation from the training.

Data Collection and Analysis
Data were co-generated with all service staff who participated in the training through a survey, reflective group discussions, and individual interviews. The survey was administered pre-, mid-and six months post-training to determine changes in: (1) staff empowerment; (2) staff attitudes to mental health and (3) staff skills to support students' wellbeing (Table 1). It was also intended that the survey findings be used in reflective group discussions with service staff to identify aspects of service practice that could be enhanced and strategies for enhancing the service model of student support.

Survey Items
Survey development was informed by a brief review of the literature, conducted to identify appropriate scales and instruments to measure the service environment (workplace empowerment), staff attitudes to mental health, and staff skills to support student wellbeing. Potential items were assessed utilising Searles, Whiteside, McCalman, Tsey, and Doran (2012) criteria; that they should represent the issues that they were supposed to measure, have reasonable sensitivity to change, be based on evidence, easily interpreted and allow comparison over time.
A further requirement to constrain the length of the questionnaire meant that only a limited number of items were included.

Workplace empowerment
To assess the service environment, questions relating to workplace empowerment were adapted from the Growth and Empowerment Measure (GEM) (Haswell et al., 2010). The GEM is a validated tool for measuring the outcome of empowerment (Emotional Empowerment Scales, α = .891). It had particular relevance to the staff training because it originated from evaluations of the FWB program (Haswell et al., 2010). Respondents were asked: Thinking about your work in the service, to what extent are you able to: feel safe; feel respected in my workplace; feel empowered in my workplace; speak out and be heard; understand my role; and undertake further learning. Items were assessed on a five-point Likert scale.

Attitudes to mental health
To determine mental health literacy, participants were asked to indicate their level of agreement with nine statements. Four statements were drawn from the depression stigma scale (Griffiths, Christensen, & Jorm, 2008): mental illness is a sign of personal weakness; people with a mental illness could snap out of their illness, people with mental illness are dangerous, and people with mental illness tend to be unpredictable. Responses were compared to 'correct' answers obtained through experts' consensus (Hadlaczky, Hökby, Mkrtchian, Carli, & Wasserman, 2014  . All items were assessed on a five-point Likert scale, ranging from strong agreement to strong disagreement, with a neutral option as the midpoint.

Student support skills
As mentioned, service staff had no direct responsibility to support student wellbeing, but did so through their established relationships with families, students and schools. Questions related to staff members' skills to support student wellbeing were also adapted from GEM scenarios (Haswell et al., 2010). Respondents were asked to rate the extent to which they were able to make changes to the way they did their work, had good relationships with colleagues, students, families and other people; and dealt with crises. Additional dimensions of wellbeing were adapted from the standard FWB evaluation survey (Searles et al., 2012). These were the ability to: cope with day-to-day things, deal positively with conflict, be a leader, and support student resilience. Questions were modified to focus on students specifically, rather than people generally, to suit the context of the participants.
For workplace empowerment and attitudes regarding mental health, participants responded using a five-point Likert scale with five anchors: no effect, little effect, some effect, much effect or major effect. Dichotomous (yes/no) responses were used to indicate the actions used to support students talking about mental health concerns, and student support skills.
Draft versions of the questionnaire were circulated to the service management for feedback; they indicated that the questionnaire addressed their needs. The staff survey was administered pretraining (May 2015); immediately after the initial five days of FWB stage one and MHFA training (May 2015); and then completed by 13 remaining participating staff six months-post completion of the full training course in June 2016.

Reflective Group Discussions and Interviews
Preliminary findings from the survey for workplace empowerment, staff attitudes to mental health and skills to support students, were shared in two reflective group discussions at the completion of the first five-day training period and a month thereafter (June 2015). Service staff were asked to reflect on the meaning of their responses to the survey statements in terms of what was working well in the provision of student support, what was not working well, and how student support strategies could be improved. Five individual interviews with service staff members were also conducted three months following the completion of training; this added to the richness of qualitative data.

Data Analysis
Survey data were manually entered into Microsoft Excel, then exported to SPSS Version 24 (IBM Organisation) for analysis. Descriptive and nonparametric tests were used to examine the data because of the nonnormal distribution of the variables. Differences between the subsamples who left the service and those who remained were examined.
Reflective group discussions and interviews were digitally recorded and transcribed. The transcripts were thematically analysed. Following Braun and Clarke (2006), a six-phased process was used: (1) familiarisation with data; (2) generating initial codes; (3) searching for repeated patterns of meaning or themes; (4) reviewing themes; (5) defining and naming themes and (6) producing the paper. These phases were not applied lineally, but rather, there was a constant moving back and forward between the data sets, coded extracts of data, and the analysis. Two of the authors read all data; one generated initial codes using open coding methods in NVIVO 8 qualitative research software. Preliminary themes were identified and compared by discussion between three authors until consensus on key themes was reached. These themes were then defined and named, and used as subheadings to write the findings of the paper. Because of the small numbers and potential identifiability of individual staff, no descriptors for staff are provided for quotes. A comparison of the two data sources then enabled focus on consistencies and inconsistencies, interpretation and identification of ways forward toward improved student support.

Workplace Empowerment
Workplace empowerment was assessed by 14 items using a five-point Likert scale. No significant differences were found between participants who remained at the service and those who left prior to the final survey. Pretraining scores across most indicators indicated the workplace having a positive effect (much effect), and subsequently potential effect changes were modest. Overall scores for each indicator are shown in Table 2. Small sustained improvements (at post survey time period) were achieved for coping, planning for the future, confidence in supporting students' resilience and understanding their role within the organisation. Scores remained steady for dealing positively with conflict, leadership ability, relationships, crisis management and feeling heard in the workplace. Small declines in effect occurred for feeling safe in their employment, interest in further learning, respect in the workplace, empowerment and making improvements.

Attitudes to Mental Health
Staff attitudes to mental health were assessed by nine items using a five-point Likert scale. Some items should be considered to be reverse coded, in that a lower score indicates an attitude more consistent with knowledge of mental health issues. No significant differences were found between participants who remained in the service and those who left prior to the final survey.  Overall indicators for attitudes to mental health are shown in Table 3. Small improvements in understanding and attitudes toward mental health were noted for most indicators. Social distance assessment, willingness to interact socially with a person with a mental health issue, remained static at a strong agreement of 4.5 from preto post-training. Confidence in approaching and helping a student that seems to have a mental health problem increased from 3.4 to 4.1. Items from the depression stigma scale, three of which are reverse coded, also improved with declines in the mean level of agreement for three of the items.

Skills to Support Students
Student support skills were reported using a selfassessment of participants' current capacity in fourteen areas with a dichotomous response. Actions taken when talking about mental health concerns were similarly reported using a dichotomous response. No significant differences were found between participants who remained in the service and those who left prior to the final survey.

THE AUSTRALIAN JOURNAL OF INDIGENOUS EDUCATION
Overall indicators for student support skills are shown at Table 4. A number of skills showed a small decrease in confidence immediately following the initial training, with a return to the pretraining level of confidence, or close to it, in the post-training assessment. This occurred for skills relating to supporting students' resilience in terms of their relationships, identifying trustworthy adults or peers, independence, acceptance, managing feelings, personality and structure of boarding house. Confidence in some skills diminished across the three time periods. This occurred for skills relating to supporting students' resilience in terms of their self-esteem, communication, problem solving, and empathy.
Participants also reported on the actions they had taken while talking about mental health concerns in the previous six months. Details of the actions for the six months prior to the initial training, and the six months after the completion of training are shown in Table 5. No significant differences were found between participants who remained in the service and those who left prior to the final survey. Overall there was an increase in the use of a number of strategies when talking with students about mental health concerns. Increased use of self-help strategies, reference to books and websites, provision of general information and provision of service information were reported. Time spent listening, and making appointments maintained a similar level.

Results from Thematic Analysis
Thematic analysis of service staff responses through reflective group discussions and individual interviews to the baseline and postinitial training survey data revealed four key issues and suggestions for actions to address them. The four key themes were: working in challenging environments; being dedicated to help; acknowledging the need for change; and negotiating individual, team and service responsibilities.

Working in Challenging Environments
The theme 'working in challenging environments' referred to the brokering role of the service as a link between community and school. While this brokering role meant that service staff had close relationships with families, students and schools, at times of crisis, staff members often had Gave them information about their problem 19 23 Gave them information about local services 33 62 Made an appointment for them with services 24 23 limited direct influence. Staff spoke of some boarding school environments in which conflict was a common feature of some students' schooling experiences. A staff member commented: 'It becomes very, very stressful with students, how they are dealt with at schools.' The challenges of being an advocate for a student in some conflict situations, where neither the student nor service staff member felt that their voices are heard, created consequent stress for staff members. One said: 'The reality is that the situation is so out of your control.' Another staff member said: 'it's hard to get outcomes, it's hard to get improvement and that can be quite draining, doing that kind of work' . Service staff suggested that they could have more positive impact in such situations by working more directly with boarding schools to enhance strengths-based practice in their support of remote Indigenous students. Staff members also spoke of the needs of family members for support to gain and take up boarding school placements for their children. Staff members were faced with high demand for their services provided when in remote communities. One said: 'When we go into community, particularly when we are only there for two or three or four times a term, the inundation of requests! . . . We say yes to basically everybody and deal with it as we go' . Staff members also noted the rewarding nature of their relationships with family members.

Being Dedicated to Help
Staff members spoke of their personal and professional dedication to the work of supporting remote Indigenous students. Staff members knew the families and students, and saw their roles as being an advocate; a support; a link between student and school and family. The roles of staff members were therefore broad and the nature of support that was offered to students did not fit a formal prescribed program. Rather, staff members spoke of being available: 'One said: ' . . . it's not talking at kids, its being with kids . . . when it's with young people, I'm there helping their families and being with them' . One staff member described an element of her role as 'supporting students to accept responsibilities and believe that their actions can make a difference . . . feel that they have what it takes to achieve and feel strong themselves . . . ' . Within their duty of care to student needs, staff members were daily or weekly called upon to manage crises, but could only respond within the bounds of their defined relationships with individual schools. While staff members spoke of the desire to use best practice such as skills developed within the FWB program, in practice, they recounted that they were not necessarily given the opportunity to respond as such, because the duty of care lies with schools.

Acknowledging the Need for Change
Staff members discussed the need for change at three levels: individually, as a team and systemically. A staff member reflected: 'it feels like there is multiple levels in which there can be responses to these things.' Staff members considered that individual responsibility for changes to work practices required reflection on how to build resilience and mental health awareness into practice. Staff members considered that while it was not appropriate or feasible to deliver the lengthy FWB, MHFA and resilience training as a whole to boarding school staff or students, they did contain useful elements and frameworks. Staff members also considered that capacity enhancement was better suited within individual schools for staff members, as they are well placed with direct and daily access to students.
Staff members also considered that the service's individual case management approach could be enhanced to better support student wellbeing. A staff member suggested: 'there are solutions or strategies in what people are . . . already doing well, we can do more of that . . . together' . Another staff member said: 'hopefully this framework, FWB training, will help us to plan, focused activities with our kids' . Specific service-level responses suggested included the identification and communication of reasonable expectations of the service role to external stakeholders.

Negotiating Individual, Team and Service Responsibilities
The theme of 'negotiating individual, team and service responsibilities' referred to staff members' consideration of the extent to which their coping was an individual versus collective or service-level responsibility, and the attitudinal and physical changes that needed to be enacted to better support the coping of individual staff members. Discussions highlighted the need for routine debriefing, connecting and checking in mechanisms. For example, one staff member reflected appreciatively on a FWB activity where: 'we had to say what we were going to do to help ourselves in our daily life, and then we had a buddy to follow up on . . . I really appreciated it.' Another suggested: 'some third party to have a yarn with, outside of work' .
Service staff members also discussed how they could pool their collective practice knowledge to better support student resilience. One said: 'if you look at the principles and goals of Indigenous Leadership Program that they resonate so much in the goals of FWB and also with some of the key principles of Australian Stronger Smarter. So it's like there are synergies.' The value of encompassing learnings from other sources was acknowledged.
Formal service policy responses to stressful circumstances were also considered. Application of learnings from FWB, MHFA and resilience training resulted in further PAR planning workshops to drive the service's model of student support from a case management to a resilience approach. The aim was to provide support and capacity enhancement to schools to provide more supportive environments for students, and with students and families to encourage negotiation of their day-to-day challenges. As mentioned above, the constraints for service staff members in their indirect brokerage roles and resultant limited capacity to support student wellbeing in the environments within which students lived and learned were clearly identified during the SEL training. It was considered that boarding school staff who work on the ground every day with students would benefit more from receiving SEL training.

Outcomes
The training resulted in 16 staff members, some of whom did not have prior tertiary qualifications, obtaining a certificate two qualification in FWB. This qualification allowed for facilitation of the program to others. However, service staff members considered that it would not be feasible for further SEL training in schools to encompass the full 120 h certificate two training. Subsequently, service staff members worked to adapt the service's individual case management approach. SEL capacity development initiatives were developed for boarding school teachers and residential staff. This initiative comprised half day planning sessions, identification of 'Step Up' plans whereby schools chose a strengths-based initiative for better supporting remote Indigenous boarding school students, and online professional development training in trauma-based care as well as strengths-based approaches.

Strengths and Limitations
A strength of the study was the high participation rate of service staff members -all agreed to participate. However, the small staff numbers, and subsequent sample size meant that there was insufficient power to undertake a number of statistical analyses. There was also a high attrition of participants, which meant that the sample at six months differed from that of the consistent pre-and mid-training sample. Such high attrition rates have been noted in other studies of the effect of MHFA training (Jensen, Morthorst, Vendsborg, Hjorthøj, & Nordentoft, 2016;Kitchener & Jorm, 2004). Results from the items for 'use of actions when talking to students' should be interpreted with caution as there was no measure of the opportunity (number of students participants spoke to about mental health concerns) or appropriateness of individual actions to each situation. A further limitation of the study is the impact of broader structural changes to the program environment that occurred at a point during the study period that may have had a negative impact on some measures.

Discussion
Overall there was a consistency of findings between the quantitative and qualitative elements of the evaluation. Both elements of the evaluation highlighted the challenging nature of service staff roles in advocating for students' educational and wellbeing outcomes across community and school environments in which the service has only indirect influence. The service staff, while confident in their abilities, recognised the limitations of their positions within the broader context of the students' lives. The students' themselves are navigating complex environments and service staff are restricted, since they are staff members of a brokering service with no direct jurisdiction or duty of care over student education or and have limited access and influence in how they can support the students. Furthermore, they identified potential changes at individual, service and system levels that could enhance the development of student's resilience.
Staff reported consistently high scores for their leadership, relationships and ability to deal with conflict. Their already high levels of coping, confidence in supporting students, ability to deal with crisis and confidence in supporting student resilience increased; as did their understandings of their role. This is consistent with previous evaluations that have found FWB and Resilience training components were associated with improved skills to better support student wellbeing (Tsey & Every, 2000;Tsey et al., 2010). However, there was a reduction in the level of support for the statements: 'feel safe' , 'feel respected' , 'feel empowered' , 'able to speak out and be heard' , and 'would like to undertake further learning' . In a rapidly changing work environment, these changes were attributed to the broader structural factors at play in the work environment.
While the level of confidence staff had in their overall ability to support students' resilience increased posttraining and was maintained to the post-training survey, a number of individual skills showed a reduction in confidence immediately following the initial training. This is unsurprising as participants' understanding of the complexity and challenges of the situation were increased by the interactive training sessions. Over the ensuing 12 months, the level of confidence recovered for a number of skills, but declined further for some. The pattern of recovery and decline for these items supports an increased understanding by participants of the challenges of the broader structural environment and limitations of their roles. The statements that showed recovery in levels of confidence reflected supports that the service staff could have more influence over, such as developing the students' support networks, skills, attitudes and practical knowledge, while the statements that declined in levels of confidence reflected supports that the service staff had little control or influence over, such as empathy.
SEL training was associated with overall strong improvements in service staff attitudes to mental health and skills to support students. It is highly likely that improvements to staff attitudes can be attributable to the MHFA component of the training. Improvements from pre-to six months post-training were found in the proportion of staff members who reported 'correct' answers to eight of the nine attitudinal statements. The only statement where ambivalence was found was 'I am willing to socially interact with persons with a mental illness' . The results suggest that like studies of the effects of MHFA for other participant groups (Jensen et al., 2016;Kitchener & Jorm, 2004), for service staff, the MHFA training was associated with a reduction in the stigma of mental illness and improved confidence in providing support for students suffering from mental illness.
Qualitative data from staff reflective group discussions and interviews demonstrated that for service staff, working in challenging environments to advocate for Indigenous student wellbeing was personally stressful. Staff members acknowledged the need for change. However, some staff members felt that negotiations of individual, team and service responsibilities were required. Other studies have also documented challenging and complex change processes within Indigenous service organisations, which go through processes of stagnation, preparation, action and review (e.g. McEwan, Tsey, McCalman, & Travers, 2010). This paper outlines the effects of a staff capacity development and quality improvement approach in one service. The multicomponent training program represented a considerable investment of 140 h of training for service staff members. Such extensive training is unlikely to be practicable for boarding school staff members, but an abbreviated version, which incorporates only the first stage of the FWP along with the MHFA and Resilience training components, could be tailored for the teaching, residential and pastoral staff of destination boarding schools. International evidence suggests that improving staff capacity is likely to improve the prevalence and quality of SEL approaches targeting students in schools (Durlak et al., 2011). However, there is very little published literature about how education staff can navigate the challenges of working across diverse environments and cultures to provide optimal SEL support for Indigenous students. Hence, there is little evidence to guide schools and related services about what types of training are likely to be effective in improving workplace empowerment and support, attitudes to mental health and student support skills for supporting Indigenous students' transitions. Given the complexity of issues associated with implementing staff SEL training, as outlined in this paper, schools and support services need to be well resourced to implement such approaches.

Conclusions
The staff SEL training was associated with improved staff attitudes to mental health and skills to support student wellbeing. However, there was an overall decrease in the proportion of service staff members who felt empowered in their workplace over the 13-month study. Qualitative research findings showed that despite 'working in challenging environments' , service staff members are 'dedicated to help' students, and 'acknowledged the need for change' to better support student wellbeing. The improved staff attitudes to mental health and student support skills associated with the multicomponent training program will provide a necessary resource for developing and implementing service improvement processes. Issues associated with planning and implementing change included the need to ensure 'negotiation of responsibilities at individual, team and service levels' . Continual improvement of the evidence-informed SEL approach contributed to a subsequent shift in the focus of the service's case management approach to a formalised resilience approach that provides schools with enhanced SEL support for Indigenous students' wellbeing.