Transitioning to University with a Mental Illness : Experiences of Youth and their Parent

TRANSITIONING TO UNIVERSITY WITH A MENTAL ILLNESS: EXPERIENCES OF YOUTH AND THEIR PARENT Kristy Boughton University of Guelph, 2017 Advisor: Dr. Margaret Lumley The transition to university is a significant life event that can be difficult for many youth. Making this transition while managing mental illness is likely to compound these difficulties for both youth, and their parents. This dissertation aimed to explore the experiences of first-year university students who reported a mental illness prior to attending university, as well as the experiences of their caregivers. Study 1 examined how youth perceptions of their relationship with their parents and their own personal strengths related to measures of well-being across a clinical and comparison sample of students. Results suggest that both relationship with parents and youth personal character strengths are important in predicting youth well-being to a similar degree across the clinical and comparison samples. Character strengths were not supported in serving as a moderator between parent relationship and youth well-being. Study 2 focused on a subset of the clinical sample and included information from a parent. This study examined parents’ experience of reward and burden associated with caring for a child with a mental illness, and how this related to parents’ virtues and student functioning. Further, parents’ role in supporting the transition to university was examined qualitatively from the perspective of youth and parents. Parent satisfaction with the parent-child relationship was associated with subjective caregiver burden. Parent virtues were not associated with caregiver reward, burden, or youth functioning in the ways expected, though some relations may be non-linear. Higher reported rewards of caregiving were associated with higher levels of youth reported life satisfaction, while higher levels of burden were related to depressive and anxiety symptoms and lower life satisfaction. Parents and youth described a number of similar themes regarding how parents may best support their youth as they manage university and mental illness. Implications for prevention and intervention efforts for post-secondary institutions are discussed.

students. Results suggest that both relationship with parents and youth personal character strengths are important in predicting youth well-being to a similar degree across the clinical and comparison samples. Character strengths were not supported in serving as a moderator between parent relationship and youth well-being. Study 2 focused on a subset of the clinical sample and included information from a parent. This study examined parents' experience of reward and burden associated with caring for a child with a mental illness, and how this related to parents' virtues and student functioning. Further, parents' role in supporting the transition to university was examined qualitatively from the perspective of youth and parents. Parent satisfaction with the parent-child relationship was associated with subjective caregiver burden. Parent virtues were not associated with caregiver reward, burden, or youth functioning in the ways expected, though some relations may be non-linear. Higher reported rewards of caregiving were associated with higher levels of youth reported life satisfaction, while higher levels of burden were related to depressive and anxiety symptoms and lower life satisfaction. Parents and youth described a number of similar themes regarding how parents may best support their youth as they manage university and mental illness. Implications for prevention and intervention efforts for post-secondary institutions are discussed.
iii Acknowledgements I would like to extend my gratitude to the many people who helped to make this project possible. I am so fortunate to have worked with, and been supported by, so many wonderful people throughout this process. I could not have done this without their guidance and encouragement.
First, to my supervisor, Dr. Margaret Lumley, thank you for believing in me so many years ago and for giving me the opportunity to join your research lab. You have been a part of my academic journey since the beginning and I could not have asked for a better mentor. Your insights, flexibility and unwavering support through all of the successes and setbacks have been invaluable in helping me to get to this point.
A sincere thank you to the members of my dissertation committee, Dr. John Beaton and Dr. Ian Newby-Clark, for contributing your insights and expertise. I would also like to thank the members of my examination committee, Dr. Nancy Galambos, Dr. Kimberley Wilson, and Dr. Heidi Bailey for taking the time to read my research and for your thoughtful comments, questions and feedback, which contributed to making this a stronger dissertation.
I am grateful to have been a part of the Resilient Youth Research Group. What an amazing group of insightful and supportive students and researchers! Thank you to Lindsey, Rachel, Sophia, Jordan, Hayley and Katie for all of your help, encouragement and friendship along the way. It has been a pleasure to work with each of you.
To my psychology cohort, I could not have gone through this program with a better group of people. You have all been incredibly kind, supportive and inspiring throughout every step of this journey. Thank you to each of you for always being so willing to listen, for providing so much validation, helping me work through challenges, and for being such wonderful friends.
Finally, thank you so much to my family and friends for all the direct and indirect ways in which you have supported me. I could not have done this without your love and support. To my parents, thank you for everything you have done to help me continue my education while also being mom. To Tim, Lily and Sierra, there are no words to express my appreciation for your unconditional love and the sacrifices you have made in supporting me to achieve my goals. Lily and Sierra, you bring so much joy to my life and I love you both more than anything! Tim, thank you for being my partner in life, for believing in me and for being there in the toughest moments. This truly would not have been possible without you. vii List of Tables   Table 1:

Transitioning to University with a Mental Illness: Experiences of Youth and their Parent
The transition to university is a major life event during which youth must adjust to changes in routines and relationships, and the new demands of being a university student (Wintre et al., 2011). Often as youth adjust to this university transition, they are simultaneously expected to be less dependent on parents and to take on increasing levels of responsibility and autonomy (e.g., living away from home, managing professor-student relationships, self-advocacy etc.). While this can be a positive experience of personal growth and maturity, for many students this transition can also be understandably difficult (Iarovici, 2014;Wintre & Yaffe, 2000). Many students experience a decrease in their ability to meet academic demands, as well as psychological distress (Dwyer & Cummings, 2001). For youth starting university with pre-existing mental illness, the ability to successfully manage the various tasks and challenges associated with taking on increasing independence and autonomy may be specifically impaired because of their symptoms and other difficulties (Davis & Vander Stoep, 1997). Additionally, because these youth are already struggling, they may be particularly vulnerable to further distress and to experiencing difficulty coping with the stressors of university life as they juggle their studies and managing a mental illness.
Research suggests that mental illness and related symptoms among university students may be increasing in number, complexity and severity (see Hunt & Eisenberg, 2010 for review; Iarovici, 2014) and that these difficulties persist over time, often into adulthood (Zivin, Eisenberg, Gollust & Goldstein, 2009). For parents, parenting during this period of development is likely to come with unique challenges (e.g., how to manage decision-making about course-work or social events), as youth are typically moving towards increasing independence. For any family, changes and challenges are likely to accompany a major life transition, such as attending university. However, making this transition while also managing mental illness may compound difficulties for both youth and their parents. For these families, research is needed to understand the factors that may promote best outcomes, including the youth's competency around autonomous decisionmaking and self-management of their academics, social relationships and physical and psychological health through appropriate and adaptive levels support from parents.
At present, most studies examine student mental illness and well-being during post-secondary education, with few studies specifically considering students who begin university with a pre-existing diagnosis. This is despite the fact that across campuses, it is consistently perceived that mental health needs of students have skyrocketed in the last few decades (Hunt & Eisenberg, 2010;Iarovici, 2014). Though it is important to understand the factors that confer risk and that can help to prevent psychological distress in all university students, it is also essential to understand the potentially unique needs of students who are already experiencing mental illness upon entry and how they can be best supported in the transition to university. Further, while university-aged youth are expected to have considerable independence from parents, research with this population suggests that relationship stressors, low levels of social support and living away from parents all contribute to poor mental health (e.g., Blanco et al., 2008), while positive family relationships may contribute to more positive student outcomes (Kenny & Donaldson, 1991;Wintre & Yaffe, 2000). This is juxtaposed with a campus environment that strives to promote autonomy of its students and often struggles with how to address the role of parents (e.g., parent requests for information and advocacy on behalf of their child). As such, understanding the role that family relationships may play in promoting positive outcomes for youth with mental illness transitioning to university would help elucidate best practice for campus services to draw on appropriate levels of parental support that facilitates, rather than hinders, youth transition towards increasingly autonomous university life.
Theory highlights the importance of not only examining the risk factors, difficulties and impairments associated with psychopathology, but also protective factors, resources and areas of strengths (e.g., Keyes et al., 2012;Seligman & Csikszentmihalyi, 2000;Wood & Tarrier, 2010). The Positive Youth Development Model suggests that all youth possess strengths and have the capacity for positive development (Lerner et al., 2005). The model also highlights that the inherent plasticity of human development (i.e., through the bidirectional interactions between individuals and the context in which they develop), suggesting that it is possible to promote changes that enhance development and improve well-being (Lerner et al., 2005). Incorporating this model with positive clinical psychology more generally (e.g., Wood & Tarrier, 2010), rather than defining positive development as simply the "absence of the negative", the "presences of the positive" is also essential (Lerner et al., 2005). Beyond risk and deficit, understanding the positive aspects of functioning (e.g., capacities, strengths, external resources) is essential for a comprehensive, holistic understanding of psychopathology, including resources that may be protective, buffer against stress, minimize symptoms and promote well-being (Clifton & Harter, 2003;Keyes et al., 2012;Seligman & Csikszentmihalyi, 2000;Wood & Tarrier, 2010). This is particular important for youth who are already experiencing psychopathology. Models suggest that mental health occurs on a continuum that is not synonymous with the presence and absence of pathology (Keyes et al., 2012). Taken together, this framework highlights the need to expand the research focus for youth with mental illness beyond psychopathology to considering the factors that contribute to, and can be enhanced to promote positive functioning and well-being.
To address these aims, this dissertation includes two studies that explore the experiences of first-year university students who reported having a diagnosed mental illness prior to attending university, as well as the experiences of their caregivers. A multi-method approach is used to incorporate youth and parent perspectives, as well as both quantitative and qualitative approaches to research inquiry and associated analysis.
This research seeks to provide a rich picture of the transition to university for individuals with mental illness and the role of parents in supporting their child through this transition.
The first study examines how youths' perception of their relationship with their parents and their own strengths relate to measures of well-being across a clinical and comparison sample of first year university students. The second mixed-method study focuses on a subset of the clinical sample including information from their parents. Parent experience of the reward and burden associated with caring for a child with a mental illness, and how this relates to parents' own virtues and student functioning is investigated. In addition, supporting the transition to university is explored in further depth qualitatively, specifically exploring youth and parent perspectives of parents' role in supporting their youth during university.

Perceptions of Parent Relationships and Personal Character Strengths
Mental illness affects one in four individuals during their lifetime (World Health Organization, 2001). The highest prevalence rate of mental illness is among those aged 15 to 24 years old (Statistics Canada, 2013), and 85% of Canadian university students fall within this age range (M = 22 years; Prairie Research Associates, 2011). Periods of stress and transition may increase risk of developing adjustment difficulties and symptoms of psychopathology and attending post-secondary education is a significant adjustment for most youth. Students must navigate a number of novel, and often stressful experiences, including adapting to new social and academic environments and expectations and taking on increased independence in their new roles and responsibilities (Crede & Niehorster, 2012;Hamaideh, 2011). With this increase in stressors, it is perhaps not surprising that among post-secondary student populations, as many as 25% endorse depressive symptoms (Lewinsohn et al., 1993;Mackenzie et al., 2011), more than 15% report engaging in self-injurious behaviours (Eisenberg, Hunt & Speer, 2013), as many as 50% struggle with at least one mental health issue (Zivin et al., 2009), and nearly 10% have seriously contemplated suicide (Eisenberg, Hunt & Speer, 2013;Iarovici, 2014). In 2016 at the University of Guelph specifically, it was reported that approximately 1 in 4 students report experiencing a mental health concern (American College Health Association, 2016). It is also not surprising that the first years of university confer the highest risk for stress and psychopathology (Bayram & Bilgel, 2008;Nelson, Karr & Coleman, 1995), given that this period involves the most significant transition and adjustment. Rates of mood and anxiety disorders, stress, alcohol-related problems and suicides are all evidenced to be higher among youth attending post-secondary education than among same aged youth in the general population (Bayram & Bilgel, 2008;Eisenberg, Gollust, Golberstein & Hefner, 2007;Mowbray et al., 2006;Small, Morgan, Abar & Maggs, 2011). Academic stress, pressure and conflict, relationship strain, lack of social support and living independently are factors found to relate to the development of psychological distress while attending post-secondary education (Blanco et al., 2008;Hirsch & Ellis, 1996;Ragheb & McKinney, 1993).
For youth beginning university with a pre-existing mental illness, the risk of dysfunction and distress may be even greater. The potential negative consequences of stressful life events are directly influenced by how effectively an individual is able to react and respond to these events (Hamaideh, 2011). One's ability to cope effectively with the challenges associated with the transition to university may be directly impacted by symptoms related to mental illness (e.g., Meyer, 2001) and indicators such as depression, negative emotionality, loneliness and affective state have all been shown to impact this adjustment (Beyers and Goossens 2002;Crede & Niehorster, 2012;Karasick 2004). Research indeed suggests that the experiences of youth attending university with and without a mental illness differ significantly. Students who have been diagnosed with a psychiatric disorder are less likely to graduate (Iarovici, 2014;Kessler, Foster, Saunders & Stang, 1995), experience more interpersonal and relationship difficulties (Iarovici, 2014), report a greater degree of academic impairment and are at greater risk of suicidal behaviour (Keyes et al., 2012) than those who have not been diagnosed. One study found that students with a mental illness had lower levels of engagement and poorer relationships with other students, faculty and administration on campus compared to students without a mental illness (Salzer, 2012). The author states that, in addition to facing the same challenges and barriers as other students, students with a mental illness are more likely to experience difficulties in areas that have been found to be associated with greater academic difficulties, poor retention, and higher level of degree noncompletion. This, in turn may place these individuals at greater risk for negative longterm outcomes (Kessler et al., 1995;Salzer, 2012). Given these particularly potent risks for students with mental illness, further delineating the experience of these students, and how this may differ from the challenges that all students may face during the transition to university is important for informing the improvement of outcomes for these students.

Mental Health in University
It has been increasingly recognized that understanding mental health extends beyond risk and psychopathology to examining factors associated with resilience, wellbeing and positive functioning, given that being free of mental illness does not indicate that an individual is experiencing good mental health (Keyes et al., 2012). Particularly for students with mental illness, facilitating a successful transition to university goes beyond understanding factors that may contribute to further risk and exacerbation of symptoms and requires a more holistic view of well-being. Positive factors and characteristics uniquely contribute to understanding mental health and disorder and may be protective in the face the stress associated with the university transition (Wood & Tarrier, 2010).
Incorporating a positive psychology lens allows for the identification of external and internal resources that will help these students not only minimize further symptoms of psychopathology, but to flourish and experience increased levels of emotional and social well-being and positive mental health (Keyes et al., 2012;Seligman & Csikszentmihalyi, 2000). For post-secondary institutions, a focus on promoting positive development and well-being is often more effective for increasing student success than approaches that seek to remediate deficits or areas of weakness (Clifton & Harter, 2003). Experiencing positive mental health is specifically protective against suicidal behaviour and academic impairment in college students with, and without, mental illness (Keyes et al., 2012). As such, beyond focusing on potential risk factors, shifting efforts to better capitalize on the external and internal capacities that students with mental illness already possess, may be an important strategy to better support a successful transition to university for these students. Two such protective factors that may be important for students with mental illness as they transition to university are relationships with parents and personal strengths. These have been consistently linked to functioning outcomes in other populations of youth and can be targeted in strategic planning of prevention/intervention initiatives.

Relationship with Parents and the Transition to University
Parenting experiences and parent-child relationships are some of the key developmental influences on youth and research repeatedly supports their relation to the development of youth psychopathology as well as the promotion of resilience and wellbeing (e.g., Armsden & Greenberg, 1987;Flouri, Midouhas, Joshi & Tzavidis, 2015;Masten et al., 1999;McLeod, Weisz & Wood, 2007;Pettit, Laird, Dodge, Bates & Criss, 2001;Wood, McLeod, Sigman, Hwang & Chu, 2003). Having a positive parent-child relationship in which parents are aware of, and able to effectively respond to their child's needs and emotions, and in which children are able to openly communicate with their parents has been found to buffer against stressors, promote emotional and behavioural resilience and positive developmental outcomes (Armsden & Greenberg, 1987;Flouri, Midouhas, Joshi & Tzavidis, 2015;Masten et al., 1999). In contrast, negative parent-child relationships in which the child's needs and emotions are not responsively met, or in which the child feels alienated or emotionally detached from their parents (Armsden & Greenberg, 1987) may confer risk for poor outcomes including lowered self-esteem, emotional dysregulation, negative beliefs about the self, and depression (e.g., Hipwell et al., 2008;McLeod, Weisz et al., 2007).
As youth develop through adolescence and towards adulthood, they are expected to move toward increasing independence and autonomy from parents. Though it was once believe that the importance of parental influence wanes across late adolescence, current research suggests that older adolescents continue to do best when they are able to maintain a close and supportive relationship with their parents that balances connection and independence (e.g., Allen, Hauser, Bell & O'Conner, 1994;Kenny & Donaldson, 1991). Attending university is a time of significant change and challenge, and having support from positive family relationships is one factor that may help to buffer this stress and promote positive adjustment (e.g., Crede & Niehorster, 2012;Henton, Lamke, Murphy & Haynes, 1980;Kenny & Donaldson, 1991;Small et al., 2011;Wintre & Yaffe, 2000). General social support from parents during university has been related to higher grades (Cutrona, Cole, Colangelo, Assouline, & Russell, 1994), less experience of crisis (Henton, Lamke, Murphy & Haynes, 1980), and better psychological adjustment (Holahan, Valentiner, & Moos, 1994). An authoritative parenting style, which balances developmentally appropriate demands with responsiveness, has been associated with lower levels of depressive symptomatology and perceived stress, and higher levels of self-esteem in first-year university students (Wintre & Yaffe, 2000). In contrast, maternal authoritarianism, which is characterized by a demanding parenting style with little responsiveness, has been associated with higher levels of depressive symptoms and perceived stress, and lower academic achievement (Wintre & Yaffe, 2000). Parental closeness is related to higher levels of social competence and psychological well-being, while conflictual or maladaptive family functioning is associated with higher psychological symptoms in university students (Hoffman & Weiss, 1987;Kenny & Donaldson, 1991). Communication with parents has been associated with decreased problematic drinking behaviours during the first semester of university (Small et al., 2011) and positive adaptation to university in female students (Wintre & Yaffe, 2000). Some research has demonstrated that students' reports of their current relationship with their parents is a better predictor of functioning in university than recalling parenting style from earlier in childhood (Wintre & Yaffe, 2000), which suggests that parents continue to have an important role to play, even as their children transition to increasingly autonomous university life. In spite of this evidence, the importance of parents' continued influence on their children during this transition towards independence is often discounted (Agliata & Renk, 2008). Further, there is concern that parental overinvolvement decreases students' ability to successfully cope autonomously with challenges (Iarovici, 2014). Taken together, it is important to highlight the need for appropriate parental involvement during post-secondary studies and resources should be available to educate parents on topics about university life, common challenges they may encounter and how to best support (yet not stifle) their child (Wintre & Yaffe, 2000).
Despite the increasing number of students with mental illness attending university and the importance of parental support, there is a dearth of research that specifically considers the experiences of students who are beginning post-secondary studies with a diagnosed mental illness and how parental relationships impact this transition.
Additionally, research studies that provide recommendations regarding mental health on campus do not typically address potential inclusion of family support systems in promoting student well-being. Further, the conflicting views regarding the appropriate level of parental involvement during late adolescence and emerging adulthood can lead to discrepant recommendations for how to best support these transitional youth (Kenny & Donaldson, 1991). For youth who are already experiencing psychological distress this is of particular concern. Given that parents can be an important form of social support for promoting positive youth adjustment and that for post-secondary institutions, encouraging and promoting positive parent-student involvement is relatively easy to implement as a protective process (Small et al, 2011), further research in this area is warranted.

Character Strengths and Virtues and the Transition to University
In addition to exploring how external support may buffer the stress associated with the transition to university, understanding how to capitalize on students' internal capacity to promote positive adjustment is another important area of study. Indeed, a number of personal traits and characteristics have been theorized to facilitate students' adjustment to post-secondary education, including individual differences in personality, self-evaluative beliefs (Crede & Niehorster 2012), coping ability (Dwyer & Cummings, 2001), cognitive characteristics (Hirsch, Conner & Duberstein, 2007), and emotional well-being (Keyes et al., 2012).
Personal character strengths are a particular aspect of internal capacity related to a number of positive outcomes (Niemiec, 2013). Character strengths are positive traits that individuals possess and which are evidenced in their thoughts, feelings, and behaviours.
In one prominent taxonomy, strengths have been classified under six overarching virtues including, Wisdom and Knowledge, Courage, Humanity, Justice, Temperance and Transcendence (Park, Peterson & Seligman, 2004). Considerable research suggests that character strengths or virtues can be used or further developed to promote positive functioning, improved quality of life and well-being (e.g., Biswas-Diener, Kashdan, & Minhas, 2011;Seligman, Steen, Park, Peterson, 2005;Wood et al., 2011). Character strengths are particularly useful as a construct to assess positive personal characteristics, as they have been supported as universally relevant across cultures and diverse belief systems (Niemiec, 2013;Park, Peterson & Seligman, 2006). These strengths are also relatively stable, but can be targeted and enhanced (Niemiec, 2013). Fostering strengths can be applicable and beneficial to individuals across the lifespan and in a variety of contexts and experiences. See Appendix A for Peterson and Seligman's classification of character strengths and virtues.
There has been a proliferation of research supporting the application of character strengths to positive education with elementary and high school students (e.g., Park & Peterson, 2009) and some research is emerging on character strengths in post-secondary students (e.g., Bowers & Lopez, 2010;Lounsbury, Fisher, Levy & Welsh, 2009;Park & Peterson, 2009;Stebleton, Soria & Albecker, 2012). Generally, strengths relate positively to college student satisfaction regarding schooling and grade point average (Lounsbury, Fisher, Levy & Welsh, 2009). The strengths most strongly endorsed among post-secondary and graduate students include humor, love, kindness, honesty, curiosity, and social intelligence (Fialkov & Haddad, 2012;Karris & Craighead, 2012). The specific strengths of perseverance, prudence, hope, love of learning, kindness, humor, fairness, gratitude and love, drawn from all of the virtues mentioned above, are associated with higher academic achievement in post-secondary studies (Gilman, Dooley, & Florell, 2006;Lounsbury, Fisher, Levy & Welsh, 2009;Park & Peterson, 2009;. Additionally, academic programing focused on enhancing students' strengths and virtues is becoming increasingly common in colleges and universities (Bowers & Lopez, 2010;Seoane, Tompkins, De Conciliis & Boysen, 2016). A number of benefits and positive outcomes of these programs have been found, including increased self-awareness, personal and academic confidence, motivation, interpersonal skills and higher grade point averages (Bowers & Lopez, 2010;Williamson, 2002). Students who are more aware of their personal strengths might be better able to make academic decisions, have more accurate views of their own academic and vocational abilities and more realistic expectations for the future (Stebleton, Soria & Albecker, 2012). Coursework focused on virtues is associated with improved well-being, prevention of burnout and improved engagement and patient care for medical students (Seoane, Tompkins, De Conciliis & Boysen, 2016) Beyond the impact personal strengths and virtues can have on academic success, strengths have important implications for mental health. In the general population and among those seeking therapy, adults who learn to use their top strengths have been found to evidence increased levels of happiness, well-being and self-esteem and decreased stress and decreased depressive symptoms (Seligman, Rashid & Parks, 2006;Seligman, Steen, Park, & Peterson, 2005;Wood et al., 2011). In a study of suicidal adult inpatients, positive psychology interventions were found to improve hopelessness and optimism, with personal strengths exercises among the most efficacious (Huffman et al., 2013). An overall higher level of reported strengths is also related to higher levels of life satisfaction (Ruch et al., 2007). The character strengths most related to life satisfaction include love, hope, gratitude, curiosity and zest (Park, Peterson & Seligman, 2004). Among children and youth, the character strengths of hope, leadership and zest are significantly associated with lower levels of depression and anxiety (Park & Peterson, 2008).
Researchers are also beginning to examine strengths and the mental health and well-being of post-secondary student populations. In general, character strengths (Lounsbury, Fisher, Levy & Welsh, 2009) and the use of more general personal strengths (Stander, Diedericks, Mostert & de Beer, 2015) have been associated with higher life satisfaction among college students. Strengths and virtues associate with the using more positive coping strategies and being less likely to engage in avoidant coping in college students (Gustems-Carnicer & Calderon, 2016). Higher levels of optimism relate to decreased levels of suicidal ideation in college students, and suggest that enhancing optimism may have potential for decreasing risk of suicide in this population (Hirsch, Conner & Duberstein, 2007). In college students, the virtues of transcendence and justice are associated with abstinence from alcohol, while the virtue of temperance is associated with abstinence, lower-risk drinking, lower blood alcohol levels, and fewer negative consequences of drinking, even among heavy drinkers (Logan, Kilmer, & Marlatt, 2010).
One study found that law school students, who used their personal strengths on a daily basis evidenced lower levels of depression and stress and higher levels of life satisfaction . In this study, the character strengths most strongly correlated with student well-being were hope, love and zest ).
Another study examined the efficacy of online strengths-based intervention for first year university students that focused on identifying and cultivating character strengths (Koydemir & Sun-Selışık, 2015). Over an 8-week period, students who participated in the intervention demonstrated significant improvements in subjective happiness, life satisfaction, psychological health and emotional well-being relative to the control group (Koydemir & Sun-Selışık, 2015).
Character strengths and virtues are emerging as an important construct relevant to positive functioning across a number of contexts. To date, research examining virtues in post-secondary students remains limited, and based on this review, no studies have examined character strengths or virtues of students who are beginning post-secondary studies with a diagnosed mental illness and how this relates to their well-being. An understanding of these strengths among youth who are already experiencing mental illness is important for a variety of reasons. All students possess relative strengths and these are identifiable variables that can be capitalized on and enhanced to create conditions for effective learning and academic success (Kuh et al., 2005). A strengths focus may provide students with opportunities for self-reflection and a more positive selfview and can be used to overcome specific challenges associated with being a college student (Schreiner, 2015). Furthermore, building personal strengths has been proposed as a potential avenue for counselling services to address the rising mental health needs across university campuses (Schreiner, 2015). Indeed, research suggests that strengths and virtues may play an important role in improving mental health and that strengthfocused interventions in post-secondary institutions may have important implications on both academic and psychological functioning in students.

University
Both relationships with parents and youth character strengths may play important roles in the well-being of students transitioning to university, and these factors may be of particular importance for promoting positive experiences in university for youth already living with mental illness. For some families, acknowledging the importance of continued parental support and providing resources regarding how to best support transitional youth may be a relatively easy to implement protective process that effectively supports a more positive university transition (Small et al, 2011;Wintre & Yaffe, 2000). Yet for many youth, parents may be unable or unwilling to provide appropriate parental support, or they may have a strained, negative or maladaptive parent-child relationship. In these cases, students' internal capacities, which are within their direct control and potentially more malleable than their relationships with parents, may be more important to target and enhance as a protective factor. Research investigating whether personal strengths have the potential to reduce the impact of a negative parent-child relationship and facilitate positive well-being in this population would be informative.
Though character strengths have not previously been examined as potential moderators between parent-child relations and well-being, research has considered whether other similar internal characteristics may mediate and/or moderate the relation between parenting and youth outcomes. For instance, considerable research has focused on how parenting and youth internal characteristics including negative cognitive and personality variables are involved in the development of psychopathology (e.g., Fanti & Munoz Centifanti, 2014;Gibb, 2002;Lumley & Harkness, 2007, 2009). Other research considers how more positive personal characteristics may play a protective role and contribute to a resilient process in the face of negative parenting experiences. Youth religiosity (Kim-Spoon, Farley, Holmes & Longo, 2014), beliefs and interpretations regarding parenting practices (Camras, Sun, Li & Wright, 2012), temperament (Gallitto, 2015), personality characteristics (O'Connor & Dvorak, 2001) and positive schemas (Lumley & McArthur, 2016) are all internal factors that have been identified as potential moderators in the relation between parent-child relationships and youth outcomes.
Given research suggesting that the current parent-child relationship is a better predictor of functioning in university (Wintre & Yaffe, 2000), in the present study, youth character strengths will be examined as a potential moderating influence in the relation between current negative parental relationship and youth well-being.

The Current Study
The central focus of this study is to examine the unique experiences of university students with a pre-existing mental illness during the transition to university, with a particular interest in how relationship with caregivers and youth's own personal character strengths and virtues may associate with youth well-being during this time. This study seeks to contribute to the literature on the experience of university for youth with a mental illness in a number of novel ways. First, though understanding mental health issues during post-secondary education is an area of considerable research, much less is known about the experiences of students who begin university with a mental illness.
Specifically, little research informs the potential impact that parental relationships or character strengths may have in ameliorating the risks and buffering the stress faced by these students as they adjust to university life. Further, no research to my knowledge investigates whether the parental context interacts with personal strengths to predict functioning in university students.
This quantitative study seeks to elucidate the potentially unique relations between student perceptions of their relationship with their parents, youth reported personal character strengths, and student functioning and well-being in youth with mental illness by including a comparison group of first year students without a pre-existing mental illness. Specifically, this study seeks to address: (1) How the quality of parental relationships relate to student symptoms of anxiety and depression and life satisfaction and whether the parenting context differs in predicting student well-being across the clinical and comparison samples. (2) Whether virtues relate to student symptoms of anxiety and depression and life satisfaction across the clinical and comparison samples, and which virtues are most associated with these outcomes. (3) Whether personal strengths moderate the relation between the parent-child relationship and youth functioning in youth attending university with a pre-existing mental illness and comparison group. Based on previous research, I expected that more positive relationship quality (i.e., higher levels of trust and communication; lower levels of alienation) with parents would be associated with more positive student well-being (i.e., fewer symptoms of anxiety and depression; higher life satisfaction), and that students in the clinical group would evidence patterns of more sensitivity to the parental context compared to the comparison control (i.e., evidence stronger relations between parenting and outcomes). As reviewed above, character strengths across a number of virtues are evidenced to be important predictors of post-secondary student well-being. An exploratory approach was taken to examine whether strengths relate to student depressive and anxiety symptoms and life satisfaction to the same degree across the clinical and comparison control students, and whether differences emerge in which particular virtues are most associated with those outcomes. Finally, total character strengths were hypothesized to moderate the relation between parent relationship and youth well-being for youth in both samples, such that when youth experience more negative relationships with parents, higher levels of strengths may help buffer against negative outcomes, associating with lower symptomology and higher life satisfaction for all youth.

Participants
Participants were university students recruited for a larger study through the University of Guelph Introductory Psychology course participant pool. Potential participants were screened for those who reported receiving a mental illness diagnosis prior to attending university and those who did not identify as having a mental illness prior to attending university. A total of 438 university students completed the larger study. Of those, 82 participants who were outside of their first year of study were excluded, leaving a final sample of 356 first year university students (59 males, 296 females, 1 participant who indicated "Other" gender), aged 17 -25 (M = 18.14, SD = .69). Consistent with the demographics in this southern Ontario community, the participants were mainly Caucasian (n = 284), but also included Aboriginal/First Nations/Metis (n = 4), Black/African/Caribbean (n = 9), Southeast Asian (n = 19), South Asian (n = 21), West Asian (n = 1), Arabic (n = 5), Latin American (n = 4) and 11 participants who indicated "Other" ethnicity (n = 17). The majority of participants indicated that they were single (n = 248) or in a relationship (n = 104) and one was married.

Measures
Mental Illness Screen. The Mental Health Questionnaire (Lumley & McArthur, 2016) is a nine-item questionnaire examining mental health history and was used to screen potential participants related to symptoms and history of mental illness. This questionnaire examines self-reported recent history (ranging from the past month to the past year) of contact with mental health professionals, use of medication to treat symptoms of mental illness, and diagnosed mental illness. Eight items are answered in a "yes" or "no" format, and one question asks to indicate frequency of contact with a mental health professional.
The BDI-II is a widely used 21-item self-report scale designed for individuals 13 years of age and older to assess the number and severity of depression symptoms over the preceding two weeks. Each item contains four statements scored on a four-point scale (e.g., 0-I do not feel sad to 3: I am so sad or unhappy that I can't stand it) and scores were calculated by summing the rating of all items with higher scores indicating higher severity of depression symptoms. The BDI evidences high internal consistency, test-retest reliability, and construct validity (Beck et al., 1996). In the present study the Cronbach's alpha for the BDI was .93.
To assess self-reported anxiety symptoms, participants completed the Beck Anxiety Inventory (BAI; Beck & Steer, 1993). The BAI is a widely used 21-item selfreport scale designed for individuals 17 years of age and older to assess the number and severity of anxiety symptoms over the preceding week. Each item is scored on a fourpoint scale assessing how bothered the individual has been by each symptom, ranging from Not At All to Severely (e.g., Unable to relax; Hands trembling). Scores were calculated by summing all items with higher scores indicating higher severity of anxiety symptoms. The BAI evidences good internal consistency and test-retest reliability (Beck, Epstein, Brown & Steer, 1988 were not administered to participants. For each participant, scores for these items were prorated using the other two items from each of their respective subscales. Caregiver Relationship. The quality of the student's relationship with their parents was assessed using the Inventory of Parent and Peer Attachment (IPPA; Armsden & Greenberg, 1987). The mother and father subscales were used, with each consisting of 28 items rated on a five-point scale ranging from Almost Never or Never True to Almost Always or Always True (e.g., My father accepts me as I am; I feel angry with my mother).
In addition to providing an overall score for relationship with mother and relationship with father, the IPPA is scored along the parenting dimensions of trust, communication and alienation. Scores were calculated by taking the sum of the ratings on the items comprising each scale, with higher scores indicating student perception of higher levels of those qualities within their relationship with their parent. The IPPA evidences good reliability and validity (Armsden & Greenberg, 1987). In the present study, the Cronbach's alphas were .96 for relationship with mother and .96 for relationship with father.

Procedure
Participants were recruited through the University of Guelph Introductory Psychology course participant pool. All students in this participant pool were first screened regarding their mental health history, using the Mental Health Screen measure described above, and those meeting study criteria (i.e., answered "yes" to the screening question "Before starting university, have you ever received a mental health diagnosis") were automatically given access to the study through the participant pool system and invited to participate online. The comparison group of students was recruited following the same procedure and these participants were given access to the study if they met control criteria (i.e., answered "no" to the screening question "Before starting university, have you ever received a mental health diagnosis"). Information regarding the study was provided online and those interested in participating were directed to an online consent form outlining the study's purpose, procedures and potential risks. Consent was assumed if participants choose to continue with the survey.
The online survey took approximately 1 hour to complete and included the Mental Health Screen, VIA-72, IPPA, BDI, BAI and BMSLSS, among other measures (see Study 2) as these data were a part of a larger study. Following completion of the survey, an electronic debriefing form was provided and students were automatically granted 1.5 course credits for participating.

Data Screening
To manage missing data, when less than 25% of items on a subscale were missing, the missing items were prorated using the mean of that subscale. Participants who did not report age were assigned the mean (M = 18.14). Data were examined for outliers using scatterplots in SPSS. No outliers were observed for age, the six virtues, total strengths, depressive or anxiety symptoms life satisfaction, or overall relationship quality with mother or father. Data were also examined for skew and kurtosis. All variables above were below an absolute value of 1.5 for both skew and kurtosis, with the exception of age (skew = 2.13, kurtosis = 10.24), which is expected given that the sample was restricted to first year university students. Multicollinearity was examined by regressing the independent variables (overall relationship quality with mother and father, virtues, total strengths) on each of the dependent variables (depressive symptoms, anxiety symptoms, life satisfaction). In all analyses, tolerance was greater than 0.1 and VIF was less than 3.0, indicating no issues with multicollinearity.

Descriptive Characteristics of Samples
Descriptive statistics and inter-correlations of all study variables for participants in the clinical and comparison group are presented in Table 1 and Table 4. The effects of gender and age in relation to the parent relationship and current functioning measures used in the study were examined for both groups. Given the limited diversity of the present sample, ethnicity did not prove a particularly informative variable and thus was not examined here.
Of the 356 participants, 200 identified as having a mental illness diagnosis prior to attending university (clinical sample; 33 males, 166 females, 1 participant who indicated "Other" gender; M age = 18.27, SD = .76), while 156 participants identified that they did not have a mental illness diagnosis prior to attending university (comparison sample; 26 males, 130 females; M age = 18.00, SD = .46). Diagnoses that participants self- 41, such that males reported higher strengths than did females. As such, age and gender were included as covariates in subsequent analyses.

Relationship with Parents and Functioning
Relations among parent-child relationship variables and student functioning were first explored by examining the bivariate correlations between these variables for both the clinical and comparison samples (Table 1). In both the clinical and comparison samples, all parent-child relationship variables were found to relate to students' symptoms of anxiety and depression, such that positive indicators (i.e., overall relationship, trust, communication) were associated with lower symptomology and the negative indicator (i.e., alienation) was associated with higher symptomology. The only exception was that in the clinical sample, communication with father was not associated with symptoms of anxiety. In both samples, each parent-child relationship variable also related to students' life satisfaction, such that positive indicators were associated with higher life satisfaction and negative indicators were associated with lower life satisfaction.
To further explore the relations between parenting variables and youth outcomes across groups, structural equation modeling in AMOS 23 was used. Five participants were excluded from this analysis as they provided incomplete data on measures of parenting. The overall ratings of relationship quality with mother and father were used as predictors (composite of trust, communication and alienation) and each regressed on youth depressive symptoms, anxiety and life satisfaction. Given that gender was found to relate to anxiety symptoms, this was added as a covariate in this analysis. Age was found to relate to depression and life satisfaction, and was added as a covariate in these analyses.
All pathways reported in the models are the standardized coefficients ( Figure 1). In the clinical group, this model accounted for 26% of the variance in depressive symptoms, 15% of the variance in anxiety symptoms and 29% of the variance in life satisfaction.
Relationship with mother and father were both significant predictors of all youth outcomes. In the comparison group, this model accounted for 24% of the variance in depressive symptoms, 10% of the variance in anxiety symptoms and 32% of the variance in life satisfaction. Relationship with mother and father were both significant predictors of all youth outcomes, with the exception of father relationship not significantly predicting youth anxiety (Table 2).
To examine whether the strength of these relations differed for students in the clinical versus comparison groups, AMOS 23 multi-group analysis function was used to calculate chi squared difference tests for each pathway between mother and father relationship quality and each youth outcome to examine whether they were significantly different across groups. There were no significant differences across the clinical and comparison group for any of the relations between parent relationships and youth outcomes (all ps > .05; Table 3).

Virtues Among First Year Students With Mental Illness
The

Virtues and Student Well-Being
Relations among students' virtues and student well-being (i.e., depressive and anxiety symptoms, life satisfaction) were explored by examining the bivariate correlations between these variables for both the clinical and comparison samples ( Table   4) To further explore the relations between strengths and youth outcomes across groups, structural equation modeling in AMOS 23 was used. The total rating of all strengths was used as the predictor and regressed on youth depressive symptoms, anxiety symptoms and life satisfaction (with gender and age controlled). All pathways reported in the models are the standardized coefficients ( Figure 2). In the clinical group, this model accounted for 18% of the variance in depressive symptoms, 7% of the variance in anxiety symptoms and 23% of the variance in life satisfaction. Total strengths was a significant predictor of all youth outcomes (Table 5). In the comparison group, this model accounted for 11% of the variance in depressive symptoms, 6% of the variance in anxiety symptoms and 12% of the variance in life satisfaction. As in the clinical sample, total strengths was a significant predictor of all youth outcomes.
To examine whether the strength of these relations differed for students in the clinical versus comparison groups, AMOS 23 multi-group analysis function was used to calculate chi squared difference tests for each pathway between total strengths and each youth outcome to examine whether they were significantly different across groups. There were no significant differences across the clinical and comparison group for any of the relations between parent relationships and youth outcomes (all ps > .05; Table 6).

Well-Being?
To examine whether character strengths may act as a moderator in the relation between parent relationship and youth well-being, a series of hierarchical multiple regression analyses were conducted according to the steps outlined by Keith (2006). Prior to these analyses, parent relationship and strengths measures were centered (Aiken & West, 1991).
In the first set of analyses, maternal relationship and youth total strengths were entered into the first block of three regression equations predicting youth depressive symptoms, youth anxiety symptoms and youth life satisfaction. The product term of maternal relationship by youth total strengths was added in the second block of each regression equation. For each analysis, across both the clinical and comparison groups, the addition of the product term did not result in a statistically significant increase in the variance accounted for in youth outcomes (Table 7; all analyses p > .05).
In the second set of analyses, paternal relationship and youth total strengths were entered into the first block of three regression equations predicting youth depressive symptoms, youth anxiety symptoms and youth life satisfaction. The product term of paternal relationship by youth total strengths was added in the second block of each regression equation. Again, for each analysis, across both the clinical and comparison groups, the addition of the product term did not result in a statistically significant increase in the variance accounted for in youth outcomes (Table 8; all analyses p > .05).

Discussion
The central objective of this study was to examine parenting and youth personal strengths as factors related to a holistic conceptualization of well-being for first year students with mental illness transitioning to university. As predicted, and consistent with previous research in other populations of university-aged youth (e.g., Crede & Niehorster, 2012;Gustems-Carnicer & Calderon, 2016;Henton, Lamke, Murphy & Haynes, 1980;Hirsch, Conner & Duberstein, 2007;Kenny & Donaldson, 1991;Koydemir & Sun-Selışık, 2015;Lounsbury, Fisher, Levy & Welsh, 2009;Small et al., 2011;Stander, Diedericks, Mostert & de Beer, 2015;Wintre & Yaffe, 2000), for students beginning university with a mental illness, both relationship with parents and youths' personal character strengths were important in predicting level of youth depressive and anxiety symptoms, and life satisfaction. Specifically, most indicators of a positive relationship with parents (i.e., higher levels of trust and communication; lower levels of alienation) were associated fewer symptoms of anxiety and depression and higher life satisfaction, and this pattern was similar to that seen in the comparison sample of students. Overall, results underscore the importance of continued positive and supportive relationships with parents in youths' adjustment and overall well-being even as they gain increasing independence during the transition to university.
In terms of youth character strengths and virtues, a number of virtues were associated with youth outcomes, which may suggest that youth who endorse higher levels of strengths across any of the virtues are likely to experience lower levels of depression and anxiety symptoms and higher levels of life satisfaction, and this pattern was consistent with that seen in the comparison sample. Further, the clinical sample endorsed lower levels of strengths overall and lower levels of many virtues, which may suggest that students experiencing mental illness may be less able to reflect on, acknowledge or use their personal strengths, which may have implications for continued lower levels of well-being.
For students beginning university with a mental illness, the character virtues of Humanity, Courage, Temperance and Transcendence and the overall level of strengths endorsed emerged as the strongest predictors of well-being. The virtue of Humanity is comprised of the strengths of Love, Social Intelligence, and Kindness. Given that these strengths focus on one's ability to create positive connections with others  and the importance of relationships, support and a sense of connection and belonging to one's well-being, it is not surprising that students who possess these strengths evidence higher levels of positive functioning. Courage is composed of Bravery, Perseverance, Honesty and Zest. Students who are able to maintain and call upon these types of strengths, particularly in the face of challenge (associated with mental illness or university life in general), may be best able to rise to face challenges and persevere to overcome them, while maintaining a positive sense of well-being ). Temperance includes the strengths of Forgiveness, Humility, Prudence, and Self-Regulation and the current study suggests that these strengths are particularly relevant to the mental health and well-being of students with mental illness (versus the comparison sample). These strengths may allow these students to let go of past negative experiences, to maintain positive interpersonal connections through the forgiveness of transgressions and to minimize the stress, guilt or shame associated with unresolved conflict (Toussaint & Webb, 2005), engage in appropriate self-monitoring of symptoms and exercise effective coping (Logan, Kilmer, & Marlatt, 2010). Finally, the virtue of Transcendence is comprised of the strengths of Appreciation of Beauty and Excellence, Gratitude, Hope, Humor, and Spirituality and these strengths are associated with the ability to connect with something greater than one's self and provide life with meaning . Students able to draw on these types of strengths may be best able to maintain a positive and optimistic view of life and to see beauty, goodness and meaning in the world, even in the face of challenges and difficulty.
These results continue to support the important role identifying, enhancing and capitalizing on the strengths individuals possess can play on overall adjustment and wellbeing across a diverse range of populations, including those with mental illness. Often the intervention focus among youth with mental illness is on monitoring and ameliorating symptoms of psychopathology, greater attention paid to youth strengths may have novel and important implications for well-being in the university context. While particular strengths may emerge as consistently strongly associated with indicators of well-being, it is noteworthy that a variety of virtues, as well as endorsing overall higher levels of strengths, were implicated in the experience of depressive symptoms, anxiety and life satisfaction, which suggests a diverse range of opportunities to enhance student positive functioning.
Contrary to hypotheses, the well-being of students in the clinical group did not appear to be impacted to any greater or lesser degree by parent relationships or personal character strengths than the well-being of students in the comparison group. This highlights the importance of supportive parental relationships and character strengths for all students as they transition to university. For university support services, these results suggest that creating a school climate that focuses on students' strengths and encourages appropriate levels of parental involvement, has the potential to impact on the students at greatest risk, while also enhancing the well-being and positive functioning of the student population in general. Additionally, though differences did not emerge in the present study, the potential detrimental consequences of not facilitating these protective factors may be more pronounced for students with mental illness in ways that were not examined in the present study, or may have an impact over time (e.g., higher risk of suicide, alcohol use, drop out, etc.). Thus, while this study supports the importance of these protective factors as similar for all students, further research is needed to better understand how a lack of these protective factors may manifest differently over time for students beginning university at higher-risk during to pre-existing mental illness.
Finally, this study examined a potential mechanistic relation among the parenting and strengths factors for predicting student well-being. Given that external protective factors, such as parent-child relationships, may not be within the control of students (or university support services) in the same way as internal capacities, character strengths were examined as a potential moderator in the relation between current parent relationships and youth outcomes, to determine whether personal strengths have the potential to reduce the impact of a negative parent-child relationship and facilitate positive well-being. Contrary to expectations, character strengths were not supported in serving as moderating mechanisms. While this may suggest that parent-child relationship and person character strengths operate as independent predictors of student well-being, it is likely that these factors are associated in ways that were not examined in the present study. For example, the current parent-child relationship has been found to better predict university student functioning (Wintre & Yaffe, 2000), and thus was examined here.
Some limitations of the present study require note. First, shared method variance should be taken into consideration, given that study variables were based on youth report alone. To address this, future research using other sources of information (e.g., parent and youth report of relationship, observed or objective indicators of well-being) would be beneficial. Similarly, the clinical and comparison groups were recruited on the basis of their self-report of whether they had received a mental illness diagnosis prior to beginning university. It is possible that students in the clinical group no longer met criteria for a diagnosable mental illness, or for students in the comparison group to have developed mental health issues since starting university. Particularly in light of the findings of many similar patterns across these groups, further research including more thorough consideration of psychopathology is needed to replicate these results. Despite this concern, significant group differences did emerge on all aspects of the mental health screening, symptoms of depression, anxiety and life satisfaction, potentially validating the distinction between groups. Examining pre-existing mental illness prior to university is novel and a strength of the current study, as this population may be distinct from students who develop mental health difficulties in response to a difficult university transition in terms of age of diagnosis, severity, length of illness and predisposing and precipitating factors related to the illness. Further research is needed to examine the potentially unique experiences and needs of students with pre-existing mental illness, and those who develop issues during university. This sample also included a mix of reported diagnoses, which may have impacted the results. However, 89% of the sample reported anxiety disorders, depressive disorders or comorbid anxiety and depression. Additionally, the diversity evidenced in the clinical sample is likely representative of the range of mental health needs of students attending university.
Additionally, data for this study were collected from students between November and March within their first year of study, as this was theorized to be the time of most significant period of adjustment and transition and would have allowed students a period of managing various academic tasks and challenges (e.g., midterms, assignments). Yet other aspects of the transition may not have been captured using this methodology.
Further research using a longitudinal design, and ideally following students from immediately prior to attending university and throughout their first year would provide a more comprehensive understanding of the dynamic challenges and experiences associated with this significant and complex life event, as well as how protective and risk factors influence functioning over time.
As indicated above, a technical administration error resulted in items 25, 43, 46 and 68 of the VIA Survey not being administered to participants. The strengths impacted included Curiosity, Love of Learning (under the virtue Wisdom and Knowledge), Spirituality and Gratitude (under the virtue Transcendence). Scores for these strengths were prorated for each participant using the other two items from each subscale. While this may have impacted the psychometric strength of these individual subscales, patterns were examined at the virtues level. These virtues were each comprised of five 3-item strength scales, making the missing items well within the 25% cut-off for prorating scores (representing 13% of the items for each virtue).
In conclusion, the results here suggest that for students beginning university with a mental illness, maintaining a positive and supportive relationship with parents and drawing on personal character strengths are important factors for promoting increased levels of well-being during a challenging time of transition. Creating a positive strengthfocused university climate that acknowledges the importance of appropriate levels of parental support in balance with student autonomy is likely to enhance the functioning of all students, including those at risk due to mental illness. Strategies that elicit parental support are relatively easy to implement, and enhancing students' strengths through academic programing is becoming increasingly common in post-secondary institutions.
Both present promising avenues for effectively addressing increasing mental health needs, and preventing further increases in rates of student mental health difficulties on campus.

Youth with Mental Illness
Decades of research have shown that caring for a family member with a mental illness can create a significant burden for families (e.g., Mitsonis et al., 2012;Nehra, Chakrabarti, Kulhara & Sharma, 2005;Reinhard, Gubman, Horwitz & Minsky, 1994;Scheirs & Bok, 2007), as these caregivers face a variety of stressors and demands, which can result in negative consequences in family relationships, as well as increase vulnerability to physical and psychological problems in the caregiver (Fadden, Bebbington, & Kuipers, 1987;Mitsonis et al., 2012;Ricard, Bonin & Ezer, 1998;Scheirs & Bok, 2007). While the focus of research has been primarily on the burden of caregiving, recently there has been a shift to consider the resiliency and rewards that may also be associated with the caregiving role. Promoting resilience and well-being is an important consideration beyond preventing or minimizing negative outcomes (e.g. Keyes et al., 2012;Seligman & Csikszentmihalyi, 2000), and resiliency factors may make important contributions in offsetting caregiver perceptions and experiences of burden.
Further, incorporating both negative and positive experiences provides a more holistic view of the parenting experiences, which is essential for a comprehensive understanding of not only how to minimize potential distress, but also how to help parents flourish and experience increased levels of well-being and positive mental health in spite of the challenges they may face (Keyes et al., 2012;Wood & Tarrier, 2010;Seligman & Csikszentmihalyi, 2000). For parents raising a child with mental illness, beyond their own experience of burden versus well-being, is the potential for these to significantly impact the parent-child relationship (Athay, 2012;Gerkensmeyer et al., 2011;Pickett, Cook, Cohler & Solomon, 1997). As outlined previously, these key relationships have been wellestablished as prominent factors in the promotion of well-being in youth (e.g., Armsden & Greenberg, 1987;Flouri, Midouhas, Joshi & Tzavidis, 2015;Masten et al., 1999;McLeod, Weisz & Wood, 2007;Pettit, Laird, Dodge, Bates & Criss, 2001;Wood, McLeod, Sigman, Hwang & Chu, 2003). Understanding factors that contribute to positive relationships between youth living with mental illness, and their potentially burdened parent-caregiver may have the potential to significantly impact the well-being of the family system, in the face of the challenges they may be experiencing. This study focuses on a specific challenge that these families face, their youth's transition to university. This is particularly relevant given the increasing number of youth with mental illness choosing to pursue post-secondary education (Hunt & Eisenberg, 2010;Iarovici, 2014). Beyond this being a significant, and often difficult transition for youth (Iarovici, 2014;Wintre & Yaffe, 2000), parents must also adjust to their child's increased independence and autonomy, which may be particularly challenging for parents who have supported their youth through significant mental illness, and whose youth may particularly struggle to manage these new challenges as a result (Davis & Vander Stoep, 1997).
Thus, examining the experiences of parent caregivers from a resilience perspective is important for gaining a broader understanding of their experiences as caregivers, as well as for informing how to best promote resilient processes that may contribute to both family and youth well-being and positive functioning. The focus of this study is to consider the experiences of parents of youth diagnosed with a range of mental illnesses as their youth transitions to university, with a particular interest in caregivers experience of positive adaption to the caregiving role (i.e., lower levels of burden and higher levels of reward) and, how this, in turn, contributes to their relationship with their child and their child's functioning during this important time of transition. Additionally, to inform how parents may best adjust to their changing role as caregivers to a more autonomous youth to provide appropriate and adaptive levels support, this study seeks to understand the type of caregiving support youth perceive to need from their parents as they begin university.

Caregiver Burden
It has been well-established that caring for a family member with mental illness can create considerable burden for caregivers and lead to a host of negative health and well-being outcomes (e.g., Mitsonis et al., 2012;Ricard, Bonin & Ezer, 1998;Scheirs & Bok, 2007). In contrast to mental health professionals, family members who face the challenge of caring for a relative with a mental illness typically have no formal training, are "on call" at all times, are often without options for relief or respite, and are emotionally connected to the individual they are caring for (Wing, 1982). This can result in subjective burden, which is described as the caregiver's personal emotional response and individual appraisal of burden (Heru, 2000;Ricard, Bonin & Ezer, 1998), and objective burden, which is the observable or concrete forms of burden, such as caregiving tasks, disruptions of everyday life, and impact on health and finances (Fadden, Bebbington, & Kuipers, 1987;Heru, 2000;Scheirs & Bok, 2007).
Traditionally, studies in the area of caregiver burden focused almost exclusively on families of adult patients with schizophrenia (see Fadden, Bebbington, & Kuipers, 1987 for review of some early studies), and more recently, bipolar and personality disorders (Chakrabarti et al., 1992;Chakrabarti & Gill, 2002;Magliano et al., 1998;Mitsonis et al., 2012;Nehra et al., 2005;Scheirs & Bok, 2007), likely because the chronic nature and severity of these disorders place an obvious strain on caregivers.
However, youth mental illness more broadly considered has become an increasingly acknowledged concern, with approximately 10 -20% of children suffering from a diagnosable mental illness (Waddell & Shepherd, 2002;World Health Organization, 2012). As such, research on caregiver burden has begun to shift to consider caregivers to younger populations as well (e.g., Cadman et al., 2012;Gerkensmeyer et al., 2011;Manor-Binyamini, 2012;Rosenbaum-Asarnow & Horton, 1990;Vaughan, Feinn, Bernard, Brereton & Kaufman, 2012). This has also expanded the range of mental health concerns considered to those more typically seen in clinical samples of youth, such as Autism Spectrum disorders, ADHD (Cadman et al., 2012), eating disorders (Coomber & King, 2012;González et al., 2012;Zabala et al., 2009), Conduct Disorder (Manor-Binyamini, 2012 and mixed samples of emotional and behavioral concerns (Athay, 2012; Davis et al., 2011;Gerkensmeyer et al., 2011;Vaughan et al., 2012). The overall pattern that has emerged within this area of research suggests that, across a range of diagnoses, patient ages and other characteristics, a significant majority of families experience moderate to severe levels of burden as a result of caring for a family member with mental illness (Bauer, Koepke, Sterzinger & Spiessl, 2012;Cadman et al., 2012;Chakrabarti et al., 1992;Chen & Lukens, 2011;Fadden, Bebbington, & Kuipers, 1987;Nehra et al., 2005;Pickett, Cook, Cohler & Solomon, 1997;Reinares et al., 2006;Rosenbaum-Asarnow & Horton, 1990;Scheirs & Bok, 2007;Winefield & Harvey, 1993;Zabala, Macdonald & Treasure, 2009). While this caregiver research has expanded considerably in the breadth of its focus, experiences of caregivers of youth transitioning to university have yet to be considered. This is a unique developmental period, particularly for youth already experiencing mental illness, given that the process of becoming more independent and autonomous may be particularly challenging in the context of their symptoms and other difficulties (Davis & Vander Stoep, 1997). This suggests that the role for parents of these youth may be different from both typically developing youth, and from previous stages of development in which youth are more dependent.
The research outlined above suggests that providing care for youth suffering from a range of mental illnesses, not only those with the most severe diagnoses, can create significant strain on families with a wide impact on functioning, health and well-being.
While research has sought to reveal some factors that may result in higher burden in some families versus others, results are often conflicting. Extant research clearly highlights the need to better understand the experiences within these families to elucidate how to best support caregivers and youth with mental illness to reduce the level of burden that is experienced and contribute to more positive functioning and overall well-being.

Caregiver Resilience
Resilient families are those who evidence a greater ability to adapt and effectively face stress and adversity. Given the burden that can be created by parents' role in caring for a child with mental illness, examining the processes by which families are able to most effectively adapt and cope in this context is important for promoting positive adjustment and functioning, as well as informing clinical practice for intervention planning that meets the needs not only of clients with mental illness, but also of the family members supporting and caring for them (Greeff, Vansteenwegen & Ide, 2006;Nehra, Chakrabarti, Kulhara & Sharma, 2005).
In the context of post-secondary education, understanding how parents may adapt effectively to best support their youth with mental illness as they transition to university can inform how universities may effectively consider and incorporate family support into broader student support services. Currently, there are conflicting views regarding the level of parental involvement appropriate during late adolescence and emerging adulthood (Kenny & Donaldson, 1991) and utilizing the already established support systems of parents is often discounted (Agliata & Renk, 2008) for fear of parental overinvolvement at a time where youth autonomy "should" be promoted (Iarovici, 2014). This is particularly concerning for parents of youth entering university who have taken a significant caregiving role supporting their child through their mental illness. Lack of clarity regarding how to best manage their changing role as a caregiver is likely to associate with uncertainty and distress, particularly when their role is discounted altogether. The potential consequences of inadequate support for these youth are significant. Further research in this area is needed to establish best parenting practices for youth with mental illness at this stage of development, given that families are likely an important and positive source of support.
As previously suggested, positive caregiver functioning (i.e., well-being, life satisfaction) is not simply a function of low levels of burden. Research supports that different individuals can experience comparable levels of burden and varying levels of well-being (Chen & Lukens, 2011) and that burden may not be related to gratification experienced as a result of caregiving (Schwartz & Gidron, 2002). Thus, positive adaptation in caregivers appears to be a unique construct that requires research attention beyond the established literature on caregiver burden. Despite the potential theoretical and applied importance, few studies have adopted a resilience focus in examining parentcaregivers of youth with mental illness (Armstrong, Birnie-Lefcovitch & Ungar, 2005).
One resilience-related theme that has emerged within the caregiver burden literature is that despite the challenges, sacrifices and disruptions, many families are accepting of this burden and complain very little (see Fadden, Bebbington, & Kuipers, 1987). While some research has suggested that this may be the result of shame, guilt, denial (see Fadden, Bebbington, & Kuipers, 1987), or a hopelessness around lack of available services and support (e.g., Hoenig & Hamilton, 1966), more recent qualitative research has shown acceptance to be an adaptive coping strategy (Ganguly, Chadda & Singh, 2010). Some research has demonstrated that caregivers of individuals with mental illness are more likely to use positive, active and problem-oriented coping strategies, as opposed to negative strategies (Bauer, Koepke, Sterzinger & Spiessl, 2012;Chadda, Singh & Ganguly, 2007;Chakrabarti & Gill, 2002). In one study, use of meaning-or problem-focused strategies was associated with higher benefit finding and positive affect (Mackay & Pakenham, 2012). Faith has emerged as an important coping strategy for some caregivers, as well as maintaining a sense of hopefulness and optimism, and being able to use their own experience to help others in similar situations (Ganguly, Chadda & Singh, 2010;Mackay & Pakenham, 2012;Veltman, Cameron & Stewart, 2002). The personal strengths that caregivers are able to draw on to implement effective coping strategies may also be of considerable importance for understanding and promoting positive adaptation to the caregiving role and will be one focus of the current study. Some research has examined life satisfaction in caregivers of individuals with mental illness. One study found that caregiver life satisfaction fluctuates over the course of youth treatment and this change is associated with youth symptom severity (Athay, 2012). This study also demonstrated that caregiver life satisfaction is associated with several caregiver characteristics, including caregiver mental health history, marital status, age and relationship to youth (i.e., birth vs. non-birth parent; Athay, 2012). Other research has shown that while sibling and parent caregivers experience similar levels of subjective burden, siblings report significantly higher levels of emotional well-being (Chen & Lukens, 2011). Younger age, higher income, lower levels of family stress and more positive experiences with professionals were also associated with emotional wellbeing (Chen & Lukens, 2011). In that study, grief emerged as the most significant predictor of emotional well-being and depression symptoms, suggesting that addressing feelings of loss in caregivers as a result of mental illness is an important consideration for promoting positive adaptation to this role (Chen & Lukens, 2011). In research examining adult children with mental illness, the adult child's contributions to helping the caregiver, the caregiver's number of confidants, and support group membership have also been associated with personal gains experienced as a result of caregiving (Aschbrenner, Greenberg, Allen & Seltzer, 2010).
Caregivers' perceptions and particularly how they create meaning around their role may be of primary importance to well-being (Schwartz & Gidron, 2002). Lower levels of threat appraisals (i.e., caregiving events as threatening, limiting personal growth, uncontrollable) and higher levels of challenge appraisals (i.e., caregiving events as promoting personal growth, personal challenge, strengthening relationship, new skills) associated with caregiving have been found to related to better adjustment (Mackay & Pakenham, 2012). One study found that the effects of caregiver burden on two indicators of resilience, resourcefulness (i.e., cognitive-behavioral skills for managing adversity and continuing to perform daily activities at an optimal level) and sense of coherence (i.e., belief that the world is comprehensible, manageable, and meaningful), was mediated by the ability to think positively when faced with adversity (Zauszniewski, Bekhet & Suresky, 2009).
Other research has considered rewards associated with caring for a family member with mental illness. Studies have found that in addition to considerable burden, many caregivers also report rewards (Bauer, Koepke, Sterzinger & Spiessl, 2012; Heru & Ryan, 2004). In one qualitative study, factor analysis revealed that rewards reported could be termed as a primary "growth in character" factor, including themes such as selfesteem, inner strength, maturity, and life experience (Bauer, Koepke, Sterzinger & Spiessl, 2012). In a study employing a focus group design, themes emerged such as increased compassion, increased knowledge or understanding, improved relationship as a result of increased closeness because of the caregiving role (Ganguly, Chadda & Singh, 2010). Similar findings, such as experiencing gratification, love and pride within the caregiving role have emerged in other studies as well (Veltman, Cameron & Stewart, 2002). One longitudinal study found that across a one-year period, caregiver perceptions of rewards of caregiving were stable, though these were higher for caregivers of individuals with depression than for caregivers of individuals with bipolar disorder (Heru & Ryan, 2004). Rewards cited in this study were related to feeling needed, being able to help someone and seeing the process of recovery. Research highlights strong caring attitudes, sense of love, concern and compassion, and investment in their role as caregivers (Heru & Ryan, 2004;Veltman, Cameron & Stewart, 2002). Other rewards include satisfaction from fulfilling parental duty, learning about themselves and their personal strengths and limitations, and a deeper understanding of life priorities (Schwartz & Gidron, 2002).

Caregiver Resilience and Parent-Youth Relationships
Despite the established importance of positive relationships to resilient functioning in general, scant research has considered parent-youth relationships in the context of the caregiving experience, particularly in late adolescent samples. Within this limited area of research, better quality of relationship between adults with mental illness and their caregiver in terms of closeness, communication, similar life views and perception of how they "get along", associate with better caregiver adjustment (Mackay & Pakenham, 2012). One study noted that parents who provide more help to their child reported more burden, but also more personal gain. This was interpreted in light of a significant correlation with assistance provided by the adult child under care, and the authors suggested that it was the reciprocity in the parent-child relationship that contributed to these gains (Aschbrenner, Greenberg, Allen & Seltzer, 2010).
Beyond the potential impact on caregiver adjustment, the quality of the parentchild relationship is well-established to have important implications for youth functioning (e.g., Armsden & Greenberg, 1987;Flouri, Midouhas, Joshi & Tzavidis, 2015;Masten et al., 1999;McLeod, Weisz & Wood, 2007;Pettit, Laird, Dodge, Bates & Criss, 2001;Wood, McLeod, Sigman, Hwang & Chu, 2003) and for university-aged youth specifically, contributes to more positive student outcomes (Kenny & Donaldson, 1991;Wintre & Yaffe, 2000). Given that higher levels of burden have been associated with less positive appraisals by parents regarding their relationship with their child (Pickett, Cook, Cohler & Solomon, 1997), understanding how to decrease caregivers' experience of burden and maintain positive family relationships is of considerable importance.
Further, though considerable research has examined the impact of family dynamics on youth outcomes in general, few studies have specifically considered how caregiver burden, let alone sense of reward, is related to outcomes for the individual they are caring for. This is particularly relevant, as quality of parenting has been consistently found to relate to a child's positive adjustment and resilience in the face of adversity (Armsden & Greenberg, 1987;Armstrong, Birnie-Lefcovitch, & Ungar, 2005;Flouri, Midouhas, Joshi & Tzavidis, 2015;Masten et al., 1999), and higher levels of burden may be associated with less effective parenting practices, poorer family relationships, and overall, lower quality care (Athay, 2012;Gerkensmeyer et al., 2011). Indeed, researchers postulate that outcomes for children with mental illness highly depend on parent ability to adapt to the role of caregiving (Anuradha, 2004), and this is certainly the pattern seen in families dealing with physical child illnesses (Thompson & Gustafson, 1996). Finally, understanding the unique context of caregiving during university has implications for both intervention and prevention opportunities. Identifying malleable aspects within the caregiving context (e.g., relationship variables) associated with both caregiver and youth well-being can provide an important focus for intervention when working with youth with mental illness and their families. Understanding the role that family can play in supporting youth with mental illness at university is essential to help inform both parents and universities of best practices in balancing youth need for both support and autonomy during this important transition.

Personal Character Strengths and Caregiver Resilience
With a shift in focus to a more holistic view of caregiving acknowledging the full complexity of negative and positive experiences, it has been acknowledged that many caregivers evidence considerable strength and resilience as they adapt to the caregiving role. This suggests that understanding how caregivers may best draw on their personal strengths is important to promoting positive adaptation to the caregiving role.
Personal character strengths (Appendix A) are positive traits that individuals possess and which are evidenced in their thoughts, feelings, and behaviours (Park, Peterson & Seligman, 2004), which have been found to relate to a number of positive outcomes (Niemiec, 2013). These strengths and their overarching virtues have emerged as particularly useful constructs to assess positive personal characteristics. They have been supported as universal across cultures and belief systems (Niemiec, 2013;Park, Peterson & Seligman, 2006), are relatively stable, but can also be targeted and enhanced (Niemiec, 2013), and fostering strengths can be applicable and beneficial to individuals across the lifespan and in a variety of contexts and experiences. Given considerable research suggesting that using one's personal strengths can decrease psychopathology, promote positive functioning, improved quality of life and increase well-being (e.g., Biswas-Diener, Kashdan, & Minhas, 2011;Seligman, Steen, Park, Peterson, 2005;Wood et al., 2011), these virtues may be an important construct for understanding how caregivers may best respond to the caregiving role, and may be a potentially important focus for intervention with families caring for a youth with mental illness.
To date, limited research has examined how personal character strengths and virtues may relate to the caregiving experience, particularly in the context of mental illness. One study examined the impact of a support group for parent-caregivers of children with Cerebral Palsy in which parents were taught to identify and enhance personal character strengths in their child and in themselves (Fung et al., 2011). Parental stress and depression were found to decrease and hope increased immediately following four sessions of this strengths-focused group, and with the exception of depressive symptoms, which may have required more than 4 sessions to address, these gains were maintained at 1-month follow-up (Fung et al., 2011).
Aspects of parents' personalities have been found to predict child outcomes (e.g., Kochanska, Clark, & Goldman, 1997), and there have been a few studies examining how parents' personal character strengths may relate to youth functioning in typically functioning samples. Specifically, the strengths of Self-Regulation, Gratitude, Hope, Love, and Zest in parents have been associated with their child's life satisfaction (Park & Peterson, 2006a;2006b). One study found that parents' character strengths related to temperance (emotional/behavioral control and regulation), intellectual strengths (acquisition and use of knowledge) and interpersonal strengths (tending and befriending others) were positively related to their first grade children's social, emotional and behavioral adjustment to school (Shoshani & Aviv, 2012). No studies have examined strengths or virtues for parents of youth attending university or diagnosed with a mental illness during the important time of transition.

The Current Study
While limited, the research related to caregiver resilience outlined above demonstrates that many caregivers of individuals with mental illness evidence remarkable strength in spite of the considerable challenges they face. Based on the literature reviewed above, there are several aspects of the caregiving context as it relates to caregiver and youth well-being that may be important to examine. Research going beyond the consideration of caregiver burden to include positive aspects of caregiving is currently in its relative infancy. Examining the broader caregiving experience, including a range of both negative and positive experiences is essential for a holistic understanding of the caregiving context, as positive experiences and personal resources are found to uniquely contribute to risk and well-being (Wood & Tarrier, 2010) and may also help to buffer potentially distressed caregivers from developing psychopathology or other difficulties (Seligman & Csikszentmihalyi, 2000;Wood & Tarrier, 2010). By supporting caregivers in drawing on their personal strengths to promote positive adaptation in the face of caregiving challenges, a more comprehensive understanding of caregiving may contribute to promoting resiliency during this important family and developmental transition.
Further, given that personal strengths can be cultivated and enhanced (e.g., Biswas-Diener, Kashdan, & Minhas, 2011) to decrease psychopathology and increase well-being (e.g., Seligman, Steen, Park, Peterson, 2005;Wood et al., 2011), character strengths may be a potentially important focus for intervention with families caring for a youth with mental illness.
Identifying malleable aspects within the caregiving context (e.g., relationship variables) associated with well-being can provide an important focus for intervention when working with youth with mental illness and their families, yet few studies have specifically focused on the parent-youth relationship in the context of the caregiving experience, particularly in late adolescent or young adult samples. Further, no studies to my knowledge have incorporated reports of both the caregiver and youth to capture their unique experiences and perspectives. In university-aged youth, who rely less on caregivers and take on more of a role in their own mental health, understanding the youth's experience of their caregiver, and how they feel they may best be supported may be particularly important to promoting well-being for the family as a whole. Additionally, given the conflicting views around the appropriate involvement of parents as their youth transition to university (Agliata & Renk, 2008;Iarovici, 2014), and the importance of parental relationships in buffering the stress of this transition (e.g., Crede & Niehorster, 2012;Henton, Lamke, Murphy & Haynes, 1980;Kenny & Donaldson, 1991;Small et al., 2011;Wintre & Yaffe, 2000), examining the perspectives of both youth and parents regarding parental support may help to contribute to reconciling the current discrepant views.
Thus, the current study seeks to contribute to the literature on the caregiving experience in a number of novel ways. The overarching focus of this study is to examine some of the factors and processes (i.e., parent-child relationship; personal character virtues) that may be associated with positive adaptation (higher levels of reward, lower levels of burden) of parents to the role of caring for a first-year university aged child with mental illness and how this associates with youth functioning during this unique time of stress and transition. The consideration of youth functioning and the parent-child relationship in relation to caregiver burden and reward is novel, particularly in a youth sample, as is the consideration of personal character virtues. The use of both parent and youth report of the parent-child relationship, as well as regarding parental support during university is also a novel approach.
Towards these aims, Study 2 explores the experiences of parent-caregivers and their first-year university aged youth with mental illness, using a multi-method approach, incorporating both parent and youth perspective, as well as both quantitative and qualitative approaches to data analysis. Specifically, this study seeks to address the following research questions using quantitative methods: (1) How does satisfaction with the parent-child relationship relate to caregivers' experience of reward and burden in response to the caregiving role? (2) Which parent character virtues relate most strongly to caregiver reward and burden? (3) How does the caregiver's experience of reward and burden relate to their youth's experience of depressive symptoms, anxiety and life satisfaction? (4) Do parent character virtues relate to youth depressive symptoms, anxiety and life satisfaction? Based on previous research, it was hypothesized that higher relationship satisfaction, as reported by both youth and parents, would be associated with higher levels of caregiver reward and lower levels of caregiver burden. Given no previous research has examined the specific virtues associated with caregiver reward and burden, an exploratory analysis was conducted. It was hypothesized that higher levels of caregivers total strengths overall would be associated with higher levels of reward and lower levels of burden. It was hypothesized that higher levels of caregiver burden would be associated with greater youth endorsement of psychopathology and lower youth lifesatisfaction. In contrast, higher levels of caregiving rewards and character strengths would be associated with less youth psychopathology, and higher youth life satisfaction.
Following this, given limited research has specifically considered the supports that may be most beneficial to youth with mental illness as they transition to university, the study seeks to give voice to the perspectives of both youth and their parents using qualitative methods to explore the question, "How Can Parents Best Support Their Child with Mental Illness at University?" This question was examined via written responses to a structured interview of open-ended questions using thematic analysis (Braun & Clark, 2006) to identify common themes reported.

Participants
Participants were university students recruited for a larger study through the University of Guelph Introductory Psychology course participant pool. Potential participants were screened for those who reported a diagnosed mental illness prior to attending university and those who did not identify as having a mental illness prior to attending university. A total of 438 university students completed the larger study. 245 participants identified as having a mental illness prior to attending university and, of those, 91 consented for researchers to contact their parents to also participate in the study.
Of those parents, 23 participated in the research study (20 mothers, 3 fathers) along with their child (21 females, 1 male, 1 participant who indicated "Other" gender). Parents were aged 45 -57 (M = 49.79, SD = 3.49) and all identified as Caucasian (n = 23). The majority of parents were married (n = 18), while two were divorced, two were in a relationship and one was single. Most parents had completed college or university (n = 13) or graduate (n = 6) or professional (n = 1) degrees. The youth who participated in the study were aged 17 -23 (M = 18.64, SD = 1.33) and all identified as Caucasian (n = 23).
The majority of students indicated that they were single (n = 18) or in a relationship (n = 5). Nearly all participants lived with their parents prior to attending university (n = 21) and the vast majority were no longer living with parents, either in residence (n = 18) or with roommates (n = 4).
Many participants reported multiple diagnoses (n = 15). All but one participant reported visiting a mental health professional (n = 22) and most had sought services in the past 6 months (n = 14) or in the past year (n = 18). Many reported taking medication to treat a mental health issue (n = 18) and some had been hospitalized (n = 6). The vast majority reported being depressed or anxious for a period of at least 2 weeks (n = 22) and that this interfered with their functioning (n = 20).
Regarding parent mental health, many reported a past or current mental illness diagnosis (n = 7), which were primarily related to depression (n = 3) or comorbid depression and anxiety (n = 4). Many had themselves visited mental health professional (n = 19), some had taken medication to treat a mental health related issue (n = 9) and one parent had been hospitalized.

Measures
Mental Illness Screen. The Mental Health Questionnaire (Lumley & McArthur, 2016) is a nine-item questionnaire examining mental health history that was created by the researchers to screen potential participants in the University of Guelph Introductory Psychology course participant pool to identify those who reported a diagnosed mental illness. This questionnaire examines self-reported recent history (ranging from the past month to the past year) of contact with mental health professionals, use of medication to treat mental illness symptoms, and diagnosed mental illness. Eight items are answered in a "yes" or "no" format, and one question asks to indicate frequency of contact with a mental health professional.

Caregiver Burden.
Objective and subjective burden associated with caring for a child with mental illness were assessed using the 19-item Burden Assessment Scale (Reinhardt et al., 1994). Participants rate the extent to which they have experienced each item related to caring for a child with a mental illness within the past six months on a four-point scale ranging from Not at All to A Lot (or Not Applicable; e.g., To what extent have you had to cut down on leisure time; Felt trapped by your caregiving role). The measure yields a total score and scores for both objective (the observable behavioral effects of caregiving) and subjective burden (feelings, attitudes, and emotions expressed about caregiving), by calculating the sum of the ratings of the items comprising those scales. Higher scores on this measure indicate higher levels of burden. The Burden Assessment Scale evidences adequate internal consistency, content and construct validity, reliability and sensitivity to change over time (Reinhard et al., 1994). In the present study the Cronbach's alphas were .89 for total burden, .91 for objective burden, and .74 for subjective burden. Caregiver Rewards. The caregivers' perceived rewards associated with caregiving were assessed with ten questions adapted from Green's (2007) measure of Perceived Benefits of mothering a child with a disability and based on the authors' suggestions for additional items, to apply to caring for a youth with mental illness (Appendix B). Caregivers were asked to indicate the degree to which they agreed with each statement on a five-point scale ranging from Strongly Disagree to Strongly Agree (e.g., The pride and joy my child feels with each accomplishment makes all the work I do worth the effort). Total scores were calculated by taking the sum of the ratings on all items, with higher scores indicating higher levels of reward. In the present study the Cronbach's alpha for the Burden Assessment Scale was .78.

Parent-Child Relationship.
To assess satisfaction with the parent-child relationship, parent and youth each completed a modified version of the Relationship Assessment Scale (Hendrick, 1988), adapted to be appropriate to non-romantic relationships (RAS-G; Renshaw, McKnight, Caska & Blais, 2010). The RAS-G is a seven-item self-report scale, with each item scored on a five-point scale ranging from Not Satisfied to Very Satisfied (e.g., How good is your relationship with your caregiver compared to most?). The overall score was calculated by taking the mean of all items, with higher scores reflecting higher relationship satisfaction. The original measure evidences good internal consistency and convergent validity in multiple samples (e.g., Hendrick, 1988, Hendrick, Dicke & Hendrick, 1998 and the modified version has demonstrated good internal consistency, item reliabilities, test-retest reliability, and factorial validity across various close relationships (Renshaw et al., 2010). In the present study the Cronbach's alpha for the youth report was .89 and was .85 for the parent report.
Caregiver Personal Character Strengths. Caregivers' personal character strengths were assessed using the shortened version of the VIA Inventory of Strengths (VIA-72;  Temperance was below acceptable reliability at .47, and therefore this virtue was excluded from analyses. Due to a technical error, items 25, 43, 46 and 68 were not administered to participants. For each participant, scores for these items were prorated using the other two items from each of their respective subscales. Youth Functioning: All youth functioning variables were assessed using the youth's self-report. Youth depressive symptoms were assessed using the Beck Depression Inventory-II (BDI-II; Beck, Steer & Brown, 1996). The BDI-II is a widely used 21-item self-report scale designed for individuals 13 years of age and older to assess the number and severity of depression symptoms over the preceding two weeks. Each item contains four statements scored on a four-point scale (e.g., 0-I do not feel like a failure to 3: I feel that I am a total failure as a person) and scores were calculated by summing all items, with higher scores indicating higher severity of depression symptoms.
The BDI evidences high internal consistency, test-retest reliability, and construct validity (Beck et al., 1996). In the present study the Cronbach's alpha for the BDI was .92. Youth anxiety symptoms were assessed using the Beck Anxiety Inventory (BAI; Beck & Steer, 1993). The BAI is a widely used 21-item self-report scale designed for individuals 17 years of age and older to assess the number and severity of anxiety symptoms over the preceding week. Each item is scored on a four-point scale assessing how bothered the individual has been by each symptom, ranging from Not At All to Severely (e.g., Nervous; Numbness or tingling). Scores were calculated by summing all items with higher scores indicating higher severity of anxiety symptoms. The BAI evidences good internal consistency and test-retest reliability (Beck, Epstein, Brown & Steer, 1988). In the present study the Cronbach's alpha for the BAI was .93. Youth life satisfaction was assessed using the Brief Multidimensional Students' Life Satisfaction Scale (BMSLSS).
The BMSLSS is a six-item self-report scale that asks participants to indicate their current

Procedure
Youth participants were recruited through the University of Guelph Introductory Psychology course participant pool. All students in this participant pool were first screened regarding their mental health history, using the Mental Health Screen measure described above, and those meeting study criteria were automatically given access to the study through the participant pool system and invited to participate online. Information regarding the study was provided online and those interested in participating were directed to an online consent form outlining the study's purpose, procedures and potential risks. Consent was assumed if participants choose to continue with the survey.
The online youth survey took approximately 1 hour to complete and included the Mental Health Screen, RAS-G, BDI, BAI, BMSLSS, and written interview questions, among other measures as these data were a part of a larger study. Following completion of the survey, youth were provided with an additional consent form for them to complete if they agreed to consent to having their parent contacted to participate in the study.
Information regarding the parent study and youth confidentiality was provided. Consent was assumed if participants choose to provide their parent's name and contact information. An electronic debriefing form was provided and students were automatically granted 1.5 course credits for participating.
When youth consented to parent participation, parents were contacted at the email address that was provided by their child with an information package containing information about the study. Parents willing to participate in the study accessed the study measures online through a unique link contained within the e-mail that generated an identification code identical to that of their child, in order to match parent-child dyads for analyses. Similar to the procedure described above, parents were directed to an online consent form outlining the study's purpose, procedures and potential risks. Consent was assumed if parents choose to continue with the survey.
The online parent survey took approximately 1 hour to complete and included the Mental Health Screen, Burden Assessment Scale, Caregiver Rewards measure, VIA-72, RAS-G, and written interview questions, among other measures as these data were a part of a larger study. After completion of the survey, an electronic debriefing form was provided and parents were sent an electronic gift card for $10 by email.

Qualitative Analysis
To gain a sense of what kinds of support youth identified as being most important from their parents, and the ways in which parents' were providing support, both youth and parents were asked to provide written responses to open-ended questions that were included with the quantitative measures online. After data were collected, analysis followed a thematic analysis approach, as outlined by Braun and Clark (2006). Analysis began with youth responses only, and upon completion, the process was repeated with parent data.
First, to become familiarized with the data, the researcher read through participant written responses to each question twice and, during the second reading, began thinking about commonalities that appeared to be present across responses, and these ideas were noted. In the second phase of analysis, initial codes were manually generated. Each response was re-read and each specific element of the response that was relevant to the question(s) asked were identified and given a code (e.g., "care packages"; "feel loved"), as were the overarching ideas communicated within the response (e.g., "talked about balance between support and independence"). A comprehensive list of all codes was generated. The next phase of analysis involved sorting this list of codes into potential themes by placing codes that were similar or appeared to be related together. These potential themes were further refined by examining whether data in each category appeared to be meaningfully related, whether each quote fit within the category and whether categories were too broad (and couple be separated into smaller categories) or too narrow (and could be combined into a broader theme). Following this refinement, the responses of all participants were re-read by the researcher and a second researcher to examine whether the potential themes appeared to capture the ideas expressed by the participants or whether other potential themes could be identified. Themes were then named, described and subthemes were identified (e.g., theme of "instrumental support" with subthemes of "financial support" and "problem-solving") with quotes that exemplified each subtheme. Themes are described in depth below.

Quantitative Data Screening
To manage missing data, when less than 25% of items on a subscale were missing, the missing items were prorated using the mean of that subscale. Data were examined for outliers using scatterplots in SPSS. No outliers were observed for caregiver reward, caregiver burden (objective and subjective), caregiver reported virtues and total strengths, youth or parent reported relationship satisfaction, or youth reported depressive or anxiety symptoms or life satisfaction. Data were examined for skew and kurtosis. All variables above were below an absolute value of 1.5 for both skew and kurtosis. Multicollinearity was examined by regressing the independent variables (parent and youth relationship satisfaction, virtues, total strengths, caregiver reward and burden) on each of the dependent variables (caregiver reward and burden, youth depressive symptoms, anxiety symptoms and life satisfaction). In all analyses, tolerance was greater than 0.1 and VIF was less than 3.0, indicating no issues with multicollinearity.

Descriptive Characteristics of Sample
Descriptive statistics and inter-correlations of study variables are presented in Table 9 and Table 10. Given the limited sample size, the effects of gender and age in relation study variables were not examined.

Burden?
Relations among parent-child relationship satisfaction and caregiver reward and caregiver burden were explored by examining the bivariate correlations (Table 9).
Though caregiver reward, total burden and objective burden were not related to parent or child report of parent-child relationship satisfaction (all ps > .05), parents' report of parent-child relationship satisfaction was significantly negatively associated with subjective caregiver burden, r = -.46, p = .04, 95% CI [-.76, -.06].

Do Parent Character Virtues Relate to Caregiver Reward and Burden?
Relations among caregivers' character virtues and caregivers' experience of reward and burden were explored by examining the bivariate correlations between these variables (Table 10). In these analyses, the virtue of Wisdom and Knowledge emerged as negatively related to caregiver experience of reward, r = -.49, p = .02, 95% CI [-.74, -.08].

Does Caregiver Reward and Burden Relate to Youth Functioning?
Relations among caregiver reward, caregiver burden and youth report of their own functioning were explored by examining the bivariate correlations (Table 9)

Do Caregiver Virtues Relate to Youth Functioning?
Relations among caregiver virtues and strengths and youth report of their own functioning were explored by examining the bivariate correlations (Table 10) 1 Given some previous research suggesting that the relation between strengths and virtues and outcomes may be non-linear, the possibility of a quadratic relation was explored using hierarchical regression. While most analyses were non-significant, three did emerge as significant (Wisdom and Knowledge in predicting Caregiver Reward and Humanity and Transcendence predicting Subjective Burden), suggesting this is worth exploring further in a sample of sufficient size and power.

How Can Parents Best Support Their Child with Mental Illness at University?
To better understand the ways parents may best support their child at university, students and their parents were each asked a number of questions about the ways that they either provide or receive support. Their responses were examined to identify commonalities in their responses. The goal was to gain a more comprehensive understanding of what youth with mental illness need from their parents while attending university, and potentially to help inform parents, and universities regarding how these needs may be best met.

Youth Perspective on Parental Support During University
First, to examine youths' perception regarding how their parents may best support them in university, youth were asked, "What would be most helpful to you from your mother and/or father to support you while attending university?"; "What do you need most from your mother and/or father while attending university?"; and "How has your mother and/or father been supporting you while attending university?" Most students provided responses about what they felt they needed from their parents, with only two of the twenty-two participants indicating that they did not know what support would be helpful.
A number of themes around what students wanted from their parents emerged from student responses, including: satisfaction with current support, instrumental support, emotional connection, trust in the youth, acceptance in the face of struggle, and balance of support and autonomy (Figure 3). These themes will be further illustrated below using excerpts taken from student responses.
Youth Satisfaction with Current Support. In examining responses, many students indicated that they felt their parents were already providing the support they needed or would be the most helpful. The following excerpts from the students' reports highlight their feelings of satisfaction with their parents' current level of support: "Exactly what she's doing." "What they're currently doing." "Just to keep doing what they are doing…" Further, when youth were specifically asked, "Is there anything different you wish your mother/father could do to support you?", the majority indicated that there was nothing different they wanted than what their parents were already providing: "No, they were wonderful to me this year." "No they are doing the best they can and are doing a good job." "Nope they are doing everything they possibly can" "no, if there was, I could bring it up and they would do their best to support me." These findings suggest that within this population, youth were generally satisfied with the level of support their parents were providing for them. Their reports indicated that these youth typically felt that their needs were being met by their parents. Their reports indicated generally positive feelings, that their parents were trying hard to meet their needs, and that their parents would be receptive if they had any further needs.
Instrumental Support. Some students indicated that they were looking for instrumental or tangible forms of support from their parents. The minority of students who highlighted instrumental support spoke specifically of financial support. Though few students identified financial support as what they needed most from their parents, a number of students did indicate that their parents were currently providing this type of support: "money" "Help me financially" "Visiting me often and paying my credit card bill." "They have been supporting me mentally, financially, and in so many other ways... They have lent me money when I am broke." "Financially and nutritionally." Youth reports indicate that for some students parents were helpful by contributing financially, either by paying bills or providing loans, while they attend university.
Among students who spoke of instrumental forms of support, the majority sought personal or academic advice from their parents. Many students spoke of their parents as being someone to talk to. Students indicated turning to their parents for help with solving specific problems, working through issues and seeking advice, and described their parents as being available to help them with this wide range of issues, as illustrated in the following quotes: "Every time I have a concern or issue regarding school they did absolutely everything within their means to help me out and give me the skills to deal with issues on my own more efficiently." "My mom helps me with the more personal issues that I am dealing with (housing, friendships, relationship advice) while my dad focuses on the academic stuff if we talk about it." "… to give me any possible advice considering they both graduated from University" "… and maybe essay critiques." "mother helps me with some homework if i don't understand it" "My mother has been supporting me a lot throughout University by offering for any clarification or help if I need it. She also helps me with managing my time so I don't get stressed out too easily. My father has been supporting me also quite a bit throughot University by giving me a lot of positive advice and feedback to keep motivated me to work hard but not over work myself." "They help me get through bad marks or issue with friends and other little things that seem small but still need to be addressed." Beyond seeking support with specific issues or problems, students also indicated turning to their parents when they just need someone to talk to: "She is always there if I need help or if I want someone to talk to." "In many ways -mostly letting me vent" "My mother talks to me everyday to see how my day was and lets me vent about the struggles of it." "We skype whenever I am feeling stressed or upset as well as just to talk. And visiting when I need to see them." These responses highlight that parents continue to play an important social support role for their children, even as they enter university. Many students appear to consider their parents to be an important and trusted advisor, as well as a confident to whom they can simply vent, when needed. Youth reports indicate that parents were helpful by imparting their own wisdom and skills as they help them navigate academic demands, social pressures, or deal with personal issues that arise within the university context. The parents were reported to help provide strategies around work completion (e.g., time management), to clarify aspects of course work, and to help more tangibly by reviewing work (e.g., proof reading). Parents were also reported to help with stress management, motivation, and to help keep youth feeling positive in the face of setbacks or stressors.
As will be explored below, while some students sought instrumental support, others identified needing various forms of emotional support or connection from their parents.
Emotional Connection. Many students expressed a desire to maintain an emotional connection with their parent while at university, even though they might be away, on their own, or more independent. Student reports reflected a desire to continue to feel an emotional bond with their parents in a number of ways, including feeling that their parents are there for them, thinking about them, and to feel that they are loved and cared for.
Many students described this emotional connection in wanting to feel a sense that their parents were still there for them and thinking about them as they attended university, as illustrated in the following quotes: "Just reassurance that they are there and everything will be OK." "I went through a lot of hardship in my first year, and they were there for me every step of the way. I love knowing that if I ever go through anything like that again, they will be there for me." "To know that even though there is distance between us, they are here." "To feel like they miss me." "She is always there if I need help or if I want someone to talk to." "Provided me with everything I need and are continuously there for me, even if they are hours away." "They are there when I need something from them." For a number of other students, staying emotionally connected to their parents was reflected in responses that mentioned love (or "love and support") as being what they need most from their parents: "Love, hugs and occasionally, money." "Something I need the most from my mother and father while attending University is love and support." "Knowing that they love me." "Their love is the most important thing I need as it helps me to know I can always count on my family to be there." While the above quotes speak to students feeling a sense of connection and support just from knowing that their parents care and will be there if they need them, other students described the more tangible and explicit ways their parents evidence this form of emotional support and strive to maintain an emotional connection.
"Just a quick text everyday about something random to know your loved and missed." "Skyping to talk as well visiting me when I feel homesick." "I wish i was able to see my mother more often to support me through stressful time." "Once a month I get a Skype call from each parent and a care package." "They often give me a text or a call just to check in. Since they are not able to support me all that much financially, they drop off a care package of food and other little supplies weekly, which is an enormous help, seeing as I have been working almost fll-time to put myself through school." "Frequent Skype calls/emails/text messages are exchanged between myself and them. I remain closer than ever with my mother. And I email my dad every once in a while for an update!" "Lots of facetime and phone calls, visiting me or bringing me home as much as possible and as needed by myself" These quotes demonstrate that for many students, continuing to feel emotionally connect to parents during university is an important aspect of feeling supported. Students reported that it is helpful for their parents to engage with them in ways that reassure them that parents continue to be there for them emotionally. For some students, parents maintained this emotional connection by being available when their youth reached out for support, by providing reassurance, or by letting their youth know that they are loved and missed. Many students highlighted that their parents demonstrated their connection, closeness and care by checking in with them, visiting or sending care-packages. As one student indicated, even contact about something unimportant or "random" provided support and comfort, knowing that their parents were thinking about them "The most helpful way they could support me is just to understand how busy I am and try to not nag me too much. I know there are things I have to get done, but I am busy all the time and I just want them to understand that and trust that I will get done wht I need to get done." "let me figure things out on my own" "… as well as expressing their confidence in my ability to succeed." As these students are becoming more independent and autonomous, it appears important for them to feel that their parents trust them to take on this responsibility and support the decisions that they make for themselves. Parents' expressions of trust seem linked to youth confidence, the opportunity to find their own paths and to figure things out on their own. In addition, students want their parents to understand that adding extra pressure is not helpful, because they are already aware of what needs to be done and the decisions that need to be made. For some students, it was not only important for them to feel supported in establishing their autonomy, but also to feel that that their parents believe in their ability to be successful.
Acceptance in the Face of Struggle. Some students described that what they needed most from their parents was for them to understand what they experience as they attend university, and beyond this understanding, for their parents to be accepting of them even in the face of struggles. These youth expressed a strong desire for their parents' unconditional love, respect and acceptance even if they potentially make mistakes, stumble, or even fail.
"To know that they are not disappointed in me even if I am struggling with my mental health" "I wish they could be able to talk to me without judgment and not just look at our relationship as needs based." "to be understanding of me getting marks lower than expected and not getting mad" "be more understanding, i have to lie about my marks" As the above quotes illustrate, youth acknowledged that things may not always go as well as anticipated as they adjust to the new and unique challenges of attending university, while also managing mental illness. It appears important to youth to not feel that their parents are disappointed, angry or judging them, as they cope with these challenges. These youth reported wanting their parents to understand the various demands that they are juggling and to see and acknowledge that they are doing the best they can to successfully manage this.
Balance of Support and Autonomy. Across the themes described above, a few students specifically noted the need for balance in the support provided by their parents, as illustrated below: "If they continue doing exactly what they are doing, no more and no less. They are still there for me, but they are allowing me to become an adult and to learn how to take care of myself."

"I think it would be helpful for them to see that I am succeeding on my own and that while their support is beneficial, they need to realize that I have matured even more than I already had in high school and I don't need them to tell me what I am doing wrng or how I should be going about living my life."
"I think the most helpful thing my mother and father could do to support me while attending University is support my decisions and if they're concerned about me to tell me" "Every time I have a concern or issue regarding school they did absolutely everything within their means to help me out and give me the skills to deal with issues on my own more efficiently." These quotes highlight that while students desire, and appreciate support from their parents, many are also welcoming the opportunity, and eager to take on more independence and autonomy. These students acknowledge that they want their parents to allow them to become their own adult and support their decisions, while stepping in with appropriate levels of support, when needed. Students were able to recognize the importance of learning how to manage their responsibilities and how to take care of themselves, and a desire for their parents to be able to see them as a maturing and responsible adult.

Parent Perspective on Parental Support During University
To gain a sense of parents' perspectives on the support they feel their children need while attending university, parents were also asked, "In what ways will you support your child while he/she is attending university?" A number of themes were noted in parents' reports, many of which were consistent with those reported by youth. The themes that emerged in the parents' reports of the support they are providing for their child at university include instrumental support and maintaining emotional connection.
Parents also specifically spoke about supporting their child in continuing to access mental health services, wanting to foster their youth's autonomy and a willingness to provide anything their child needs (Figure 4).
Instrumental Support. Similar to youth report, parents described providing financial support for their child. Providing transportation and food were another tangible supports that parents reported: "She has complete financial support and is welcome to live at home during university. She chooses to work but knows that she does not need to make money if her studies become too demanding to combine with a job." "… financial support and boxes of food she takes back with her after visiting…" "We help her financially as well as hold her accountable for her budget. We set up the house WITH her and I stocked the freezer with her favorites. We pick her up and drive her back when se asks." "I offer to talk to her about her assignmentsand to help her by reading her drafts and giving her feedback though she usually prefers to work this way with peers." These parents reported that providing instrumental support, such as financial assistance, food, and transportation, is one way they try to decrease the stress and burden of the university experience for their child. These quotes also reflect parents' willingness to allow, or actively increase, youth autonomy to make their own decisions about accepting instrumental support, while expressing a willingness to provide a "safety net" should it be needed.
Emotional Connection. Consistent with youth reports, maintaining an emotional connection was a theme that was described by the majority of parents. Parents described this in a number of ways that were consistent with those described by youth, including checking in with their youth, expressing that they are available to their child when needed, and ensuring that their youth feels loved.
Parents sought to maintain an emotional connection with their youth in a number of ways, including check-ins/communication, visits and care packages: "Answer texts, talk on Skype/phone, send care packages…" "She is still geographically close, and can call or come home anytime" "Routine texts, random notes and emails on topics of mutual interest." "… bringing her home when she needs a break or some perspective…" "I maintain communication with her always. I know her schedule and ask about her activities and experiences." "He comes home for thanksgiving, Xmas and reading week… I feel by having my son come home on these holidays, it breaks up the time away from home and helps him cope better.
"Keeping in regular contact (phone, text, skype and visits) to ensure that she is handling university." Beyond more the more tangible forms of demonstrating that parents continue to be connected to their youth, parents also described specifically seeking to ensure that their child knows that they are there for them and available if needed, as illustrated in the quotes below: "by being available when and how she needs me" "My son knows I am here for him as well as his father whenever he needs." "I try to be emotionally available if ever and whenever I am needed." Parents also described ensuring that their child knows they are loved and that they are thinking of them: "I tell her about home life here let her know she is still part of our family here" "I let her know that we love her and will face challenges together." Overall, these quotes demonstrate the parents strive to support their youth by maintaining an emotional connection in ways consistent with what many youth described.
Texts and phone calls are one way that parents described letting their child know that they are there for them and thinking about them. As some quotes illustrate, parents also used check-ins as opportunities to assess how their child was managing the demands of university and whether other supports might be needed. Parents described the importance ensuring their child knows they are loved and that parents are available when needed.
Mental Health Services. Parents specifically discussed wanting to ensure that their child was continuing to access mental health services when they needed to as in the following quotes: "By encouraging counseling when she needs it,…" "Daily contact, reminders of whre to seek assistance, encouragement to advocate as appropriate, visits and comfort, connections to Health office." "Continue with supporting her therapy." These quotes demonstrate that parents recognize the importance of mental health services in helping their child manage their mental illness and succeed in post-secondary education. Parents described themselves to support and encourage their youth's mental health service use, providing youth with the independence to take on increasing responsibility in managing their own mental health.
Balance of Support and Autonomy. Similar to youth, across the themes identified above, parents often explicitly acknowledged the importance of balancing the support they provide while also encouraging their child's autonomy as they transition towards adulthood and more independence.
"We are available for occasional visits, but try to provide her with as much independence as possible, so that she continues to build a sense of self-worth and ability" · "I support her by giving her advise when she asks and support when she needs it. I try to let her make her own way so she becomes self-sufficient." "We help her financially as well as hold her accountable for her budget. We set up the house WITH her and I stocked the freezer with her favorites…." These quotes highlight that parents appreciate the importance of providing parental support that also balances their child's ability to take on increasing responsibility, independence and autonomy. Parents described the importance of allowing their child to have the opportunity to build competency in these areas by doing things independently for themselves.
Anything Needed. Overall, many of these parents indicated more generally that they were willing to do anything that their child needed in order to support them through university: "In any way she needs. Emotionally, mentally, financially or otherwise." "In any way she needs me too!! The sky is the limit!!" "I am here for her whatever her needs may be" These quotes demonstrate that parents are willing to provide anything that their child feels that they need from them as they attend post-secondary education.

Quantitative Analysis
The central objective of the quantitative analyses was to understand the experience of caregivers of youth with mental illness transitioning to university.
Specifically, this study sought to examine parent-child relationships and parents' personal strengths and virtues as factors related to reward and burden associated with the caregiving role, as well as how this impacts youth functioning.
Partial support for the relation between parent-child relationship satisfaction and caregiver reward and burden was found. When parents reported higher levels of satisfaction with the parent-child relationship they reported lower levels of subjective caregiver burden, indicating a perceived positive relationship with their child corresponds with more positive feelings, attitudes, and emotions about caring for that child. Moderate to large effect sizes were obtained in some statistically insignificant correlations. In particular, for the relation between both parent and youth reported satisfaction and caregiver reward, as well as the relation between parent reported relationship satisfaction and overall and objective burden. Taken together, results do provide some support for previous research suggesting that relationship factors relate to a more positive experience of caregiving (Mackay & Pakenham, 2012) and suggest that maintaining or enhancing positive relationships between parents and youth may be an importance avenue to overall family well-being during the transition to university.
Regarding parent strengths, caregiver reward and youth functioning, contrary to expectation, higher levels of the parent virtue of Wisdom and Knowledge was associated with lower levels of caregiver rewards and higher levels of youth depressive symptoms.
Higher levels of Courage emerged as associated with lower levels of caregiver subjective burden at a trend level. This suggests that high levels of some virtues may be detrimental to well-being, while others are beneficial, in the context of caregiving. In particular, high levels of Wisdom and Knowledge may be related to less optimal functioning, while virtues such as Courage and Humanity may relate to more positive functioning.
The virtue Wisdom and Knowledge is comprised of cognitively based strengths that focus on acquiring knowledge. It may be that individuals who possess high levels of these strengths may place higher value on productivity and achievement, and this may make the stressors associated with raising a child with a mental illness to be perceived as particularly burdensome or out of line with the parents' goals and values. In contrast, the virtues of Courage and Humanity are comprised of emotional and interpersonally based strengths. Specifically, the virtue of Courage embodies the determination and ability to persevere towards one's goals in the face of challenge, which are qualities that may well equip parents to face the challenges associated with caregiving. Further, parents high in the virtue of Humanity are likely to place high value on relationships and to be nurturing by nature, which likely align well with their roles as parents.
These findings are consistent with resent research in the area of positive psychology that suggests that you can have too much of a good thing (i.e., too much value on Wisdom and Knowledge may come at the expense of other virtues, such as Humanity), and that moderate levels of virtues that are in balance with each other may be most optimal (e.g., see Pawelski, 2016). As an alternative, some research indicates that the relations between strengths, virtues and outcomes may be non-linear, suggesting that the optimal level may be mid-range and balanced strengths (e.g., see Gruman, Lumley & Goonzalez-Morales, 2017). In addition, individual strengths may be less informative then examining how patterns of different strengths work together to produce positive outcomes (Allan, 2015;Gruman et al., 2017). Although exploring more nuanced or nonlinear relations could not be justified given the present sample size, the present data do suggest quadratic associations may be present for some of the relations explored here (e.g., Wisdom and Knowledge in predicting Caregiver Reward and Humanity and Transcendence predicting Subjective Burden). Overall, results suggest that the relation between character virtues and caregiver functioning may be complex, and due to sample limitations could not be fully explored in the present study. Again exploring the magnitude and directions of these patterns is an important future research goal. Consistent with expectation, higher levels of caregiver reward and lower levels of caregiver burden were associated with better youth functioning. Specifically, higher reported rewards of caregiving were associated with higher levels of youth reported life satisfaction, while higher levels burden were related to depressive and anxiety symptoms and lower life satisfaction. This is a novel contribution to extant literature, and is consistent with research regarding physical child illnesses (Thompson & Gustafson, 1996). These findings support that youth outcomes are related to their parent's ability to adapt to the role of caring for a child with a mental illness (Anuradha, 2004). Further research with a larger sample is needed to explore the potential mechanisms through which caregiver reward and burden impact youth functioning (i.e., whether this is a direct effect or, potentially mediated by parenting practices, family relationships, or quality of care, for example; Athay, 2012; Gerkensmeyer et al., 2011). Alternatively, youth mental illness symptoms impact parent well-being in the form of how they are coping with caregiving role and it is likely that transactional analytic models would be important to investigate in future work. At present, these results support that interventions for first year students with mental illness should consider the needs of caregivers, and that this has important implications both for their student well-being and for the larger family system.
In summary, the results presented here underscore the importance of supporting caregivers as a part of promoting youth well-being during first year university.
Specifically, when caregivers experience less burden, and find parenting more rewarding, their children are more likely to exhibit positive functioning (i.e., higher life satisfaction, lower symptomology) during a major transition to first year university. This study provides partial support for maintaining or enhancing positive relationships between parents and youth as one avenue for achieving this overall well-being within the family.
The influence of caregiver strengths on caregiving and youth functioning was less clear, suggesting this relationship is more complex than could be explored in the present study.
Future research is essential to better understand how internal capacities (e.g., strengths, positive characteristics) and external factors (e.g., family relationships) may be best capitalized upon to support parents in their caregiving role, and in turn be able to support their children in achieving their own optimal functioning at this prominent developmental transition.

Qualitative Discussion
The central focus on the qualitative analysis was to better understand what youth with mental illness need from their parents while attending university, and how parents may best support them. A consideration of both youth and parent perspectives on this topic was undertaken.
Youth indicated seeking a variety of types of support from their parents during first year of university. For some students, instrumental support, in the form of personal advice, academic help and, to a lesser extent, financial support, was identified as important. For many youth, emotional connection, as evidenced by feeling their parents are there for them and having a sense of being loved and supported, was a particularly salient theme. Youth described wanting to feel as though their parents were able to trust the youth to make their own decisions and to be successful at managing the new responsibilities associated with increased autonomy. Youth also spoke of wanting their parents to acknowledge their successes, including things that may appear to be small, but could be celebrated as accomplishments. Beyond this, youth wanted to be accepted unconditionally even when they face obstacles and challenges. Finally, some students explicitly commented on the developmentally appropriate need to balance parental support and youth autonomy. This suggests that these youth recognize the need to move toward independence, but also the important role their parents continue to play during this time of transition.
Overall, the themes within parent responses regarding the support they are providing to their youth as they attend university were in line with the important themes identified by their youth. This includes instrumental support, in the form of financial support, transportation and food, as well as maintaining emotional connection by checking-in, being available when needed, and ensuring their child knows they love them and are thinking of them. Uniquely, parents identified supporting their youth in continuing to access needed mental health services. Finally, in line with youth reports, parents identified the need to balance parental support and youth autonomy, again suggesting that many parents want to allow their youth to develop independence by providing support that is appropriate and adaptive.
The themes present in youth and parent responses are informative in a number of ways. Regarding advising parents how they may best support their youth, it is likely heartening to know that many of the things youth express needing are relatively simple for parents to provide. Maintaining a sense of connection, feeling loved and perceiving unconditional acceptance from parents appears to be primary for many youth. This is also in-line with extant research suggesting that a positive and supportive relationship with parents can serve as an important protective factor to promote youth well-being (Armsden & Greenberg, 1987;Flouri, Midouhas, Joshi & Tzavidis, 2015;Kenny & Donaldson, 1991;Masten et al., 1999;Wintre & Yaffe, 2000). While knowing how to provide positive support, particularly during times of heightened stress and difficulty, can be challenging for parents, it is certainly possible for parents to develop skills in order to achieve these aims (e.g., active listening, validation).
To post-secondary institutions, the present data speaks to many parents' willingness to do "anything needed" to support their child in first year university. First, it is important to recognize that even when parents are perceived as overinvolved or potentially providing support in ways that are perceived to be less adaptive, this is likely well-intentioned for the vast majority of parents. Second, parents are a potentially costeffective and easy to implement source of support with a vested interest in the success of their child (Small et al, 2011). As such, the potential of parental support could perhaps be better capitalized on. Further, in line with the perspective of many post-secondary institutions, both parents and youth speak to the importance of balancing this support with promoting autonomy. Yet parents may not know what the optimal balance is, or how to achieve it. In particular, for youth already diagnosed with a mental illness, it may be even more difficult for parents to promote independence, as parents may fear the consequences of providing insufficient support (e.g., higher risk of suicide, higher likelihood of risky behaviour and negative avoidant coping strategies). Additionally, parents may feel less able to "take a step back" when they feel as though their role as a caregiver is devalued or discounted altogether by post-secondary institutions (e.g., Agliata & Renk, 2008).
Post-secondary institutions may best serve students by acknowledging the value of parent support in promoting student success and well-being, while providing education to parents on how to best support youths' needs in a balanced way. Investing in this preventative education for parents may be especially effective if it reduces the burden on individual student support services (e.g., counselling). Further, reducing discrepancies in recommendations for how to best support youth (e.g., Kenny & Donaldson, 1991) and providing clarity and direction for parents, may help reduce the burden of caregiving during this stressful time of transition, and increase perceived parenting rewards, which in line with the results reported above, may also have a positive impact on youth adjustment and well-being.

Limitations
Similar limitations described above for Study 1 are also present in the current study, particularly related to study recruitment (e.g., on the basis of self-report) and limited diversity of the sample. A difficult to recruit clinical sample of youth and their parent is a strength of the current study and provides a rich glimpse of the caregiving experience, parent-child relationships, strengths and emotional functioning in a sample of families that might typically be seen in clinical practice, using a multi-method approach incorporating both parent and youth report. Yet, the present sample size clearly influenced power to detect significant patterns. The effect sizes and magnitude of several correlations suggest that further research in a larger sample is warranted. Since the focus of the study was on the experiences of caregiving (e.g., reward, burden) in the context of youth experiencing more chronic mental illness, students with a diagnosed mental illness prior to beginning university were recruited for the study, rather than students with mental illness or symptoms subsequent to starting university. However, the current sample size did not permit examination of course, severity or duration of illness, which are likely to influence level of caregiver burden (e.g., Chakrabarti, Kulhara, & Verma, 1992;Jungbauer et al., 2003;Mitsonis et al., 2012).
Only one parent, who was selected by their child, participated in the present study.
It is possible that the observed associations may have differed significantly if relationship, reward and burden measures were collected for non-participating parents. This is underscored by a significant difference in youth reported parent-child relationship satisfaction, with higher satisfaction with the relationship with the participating versus non-participating parent, t(20) = 8.87, p = .001. Additionally, the majority of parent participants were mothers and I was unable to examine whether patterns of results differed for mothers versus fathers. Yet, it is likely the parent that youth referred to participate in this study would be the parent from whom they would seek support, and who may be involved in their treatment. Similarly, only a portion of parents who were contacted elected to participate in the study. Thus, it is important to note that the support provided by these parents, and the views expressed by parents and youth reflect a specific subset of parents of youth with mental illness. Examining these themes within other samples of parents would be beneficial, particularly given that the most burdened parents, or those with strained parent child relationships may have been the less likely to participate in the current study.
The current study examined youth and parent perspectives of parental support using participants' typed responses to qualitative interview questions. While this allowed for the identification of some important themes, future research using an interview based approach would provide further depth in understanding these themes and would allow for a deeper understanding of the individual narratives of these families.
As indicated above, a technical administration error resulted in items 25, 43, 46 and 68 of the VIA Survey not being administered to participants. The strengths impacted included Curiosity, Love of Learning (under the virtue Wisdom and Knowledge), Spirituality and Gratitude (under the virtue Transcendence). Scores for these strengths were prorated for each participant using the other two items from each subscale. While this may have impacted the psychometric strength of these individual subscales, patterns were examined at the virtues level. These virtues were each comprised of five 3-item strength scales, making the missing items well within the 25% cut-off for prorating scores (representing 13% of the items for each virtue). In addition, due to low lower than acceptable reliability, the virtue of Temperance was excluded from analysis. While reliability in previous studies have been acceptable (e.g., Diener et al., 2010;Park, Peterson, & Seligman, 2006), this study is comprised of a very specific sample and it may be that the organization of character strengths virtues look different in this sample.

Conclusion
The transition to university is a significant transition for any youth and involves increased autonomy and less dependence on parents. For both youth and their parents, this is even more challenging with the added stress of managing mental illness and these youth are particularly vulnerable to further distress and a variety of negative outcomes.
Given that the number of students experiencing mental illness is increasing (Hunt & Eisenberg, 2010;Iarovici, 2014), understanding how to promote best outcomes for these youth is essential, and personal character strengths and relationships with parents have been identified as important avenues to explore. As such, this dissertation examined the experiences of first-year university students who reported having a diagnosed mental illness prior to attending university and the experiences of their caregivers.
Study 1 highlighted the importance of maintaining a positive and supportive relationship with parents and drawing on personal character strengths for promoting increased levels of well-being for all students, including those entering university with a mental illness. Study 2 provided support for the importance of supporting caregivers who are parenting youth with mental illness, because when caregivers experience less burden, and find parenting more rewarding, this associates with their student reporting less psychopathology and greater life satisfaction. Again, maintaining or enhancing positive relationships between parents and youth are implicated in supporting the well-being families as a whole. Finally, in terms of specific support that youth with mental illness need from their parents as they attend university, maintaining a sense of connection, feeling loved and perceiving unconditionally acceptance, trust and understanding from parents appears to be primary for many youth. These themes again speak to the continued importance of positive and supportive connections with parents, even as youth transition to more independent functioning.
Overall, the present study suggests that, while transitioning to university with a mental illness presents a number of unique challenges, strengths and parent-relationships are factors that appear to be beneficial to the well-being of all students. This is a promising finding for post-secondary institutions to consider, as it suggests that creating a positive strength-focused university climate acknowledging the importance of ideal levels of parental support is likely to enhance the functioning of all students, including those experiencing mental illness. As described above, initiatives and coursework focused on enhancing students' strengths and virtues are becoming increasingly common in colleges and universities and such initiatives are demonstrating positive results (Bowers & Lopez, 2010;Seoane, Tompkins, De Conciliis & Boysen, 2016;Stebleton, Soria & Albecker, 2012;Williamson, 2002). Currently, some universities have resources for families of their students available online, and may offer parent webinars and newsletters. Some universities also offer a parent and family orientation and a few facilitate connection among students' families. Many of these resources focus on practical considerations (e.g., paying tuition; resources on campus; calendars) and encouraging student autonomy.
Increasing the focus on supporting student well-being and providing education on and encouraging adaptive levels of parental involvement is likely to be beneficial.
Additionally, facilitating positive attitudes of campus faculty and staff toward parents of students and their involvement is essential for creating a climate of collaboration in the interest of student success. Taking a strength and family-focused approach is likely to be cost-effective and relatively easy to implement, given that it targets and enhances extant student and family capacities and resources. Further, this may allow for individual mental health services, which are expensive and limited, to be better allocated and utilized within a broader and more holistic strength and family-based approach to student well-being.
Longitudinal research will be essential to understanding the efficacy and impact of targeting personal character strengths and family relationships for this population of students, compared to targeting symptoms of psychopathology alone.
Future research examining the feasibility and impact of building strengths and how to best build capacity in parents to support their youth with mental illness through university is warranted and appears to hold promise for addressing the rising mental health needs in university students.       Note: All variables centered except for product *p < .05 **p < .01 ***p < .001

Instrumental Support
Emotional Connection Mental Health Services "… financial support and boxes of food she takes back with her after visiting…" "I try to be emotionally available if ever and whenever I am needed." "Continue with supporting her therapy." "I let her know that we love her and will face challenges together." "By encouraging counseling when she needs it,…"

Balance of Support and Autonomy
Anything Needed "We are available for occasional visits, but try to provide her with as much independence as possible, so that she continues to build a sense of self-worth and ability" "I support her by giving her advise when she asks and support when she needs it. I try to let her make her own way so she becomes self-sufficient." "In any way she needs. Emotionally, mentally, financially or otherwise." "In any way she needs me too!! The sky is the limit!!" "I am here for her whatever her needs may be"