“Who am I?” How Female Care Leavers Construct and Make Sense of Their Identity

Identity formation may be more complex for those who have been in foster care in the face of childhood abuse, difficult relationships, unstable environments, and multiple care contexts but this does not imply there is anything pathological about it. Given the higher levels of mental health difficulties in looked after children and the known role identity has in mental health, whether as a risk or a protective factor, it seems clinically significant to investigate what factors help construct or hinder the formation of identity for those who have been in care. Interpretative Phenomenological Analysis was used to analyze semistructured interviews of eight female care leavers about the understanding of their identity development. Three superordinate themes emerged which encapsulated participants’ identity development. These included Construction of Identity—How I Became Me, Understanding of Identity—Who am I, and Experience of Identity—How My Identity Plays Out. Participants’ construction of identity can be understood in the context of early adverse environments and developmental trauma. This construction of self, in turn, mediates how participants understand and experience their identity. Findings were discussed in relation to previous research, and limitations were outlined. Implications for future research included giving fuller consideration to the role of developmental trauma in identity formation. Clinical implications encourage understanding of looked after children and care leavers in the context of developmental trauma, rather than focusing on symptoms of various diagnoses.


Identity Formation
There is no consensus in the definition of identity (Bosma, Grassfsma, Grotevant, & de Lavita, 1994). Common to all definitions, importance is placed on the sameness and continuity of identity overtime (Bosma et al., 1994). The Framework for the Assessment of Children in Need and their Families (Department of Health, 2000) defines identity as the child's growing sense of self as a separate and valued person. It includes the child's view of self and abilities, self-image and self-esteem, and having a positive sense of individuality. Race, religion, age, gender, sexuality and disability may all contribute to this. Feelings of belonging and acceptance by family, peer group and wider society, including other culture groups. (p. 19) This definition is utilized within the current study. Erikson (1959) outlined a life span model of human development. Although identity formation is considered to be a lifelong progression, its initial development is thought to be a primary task of adolescence (Blos, 1970;Erikson, 1950;Marcia, 1980). With adulthood and independence on the horizon, adolescents attempt to discover and stabilize a sense of personal identity incorporating their past, present, and future. It has been suggested that the process of identity formation may be different for males and females (Gilligan, Ward, Taylor, & Bardige, 1988). Gilligan (1992) argued that female identity is built around belongingness and affiliation while male identity is centered on personal experiences that demonstrate aspects of achievement and autonomy. Preston-Shoot, and Wigley (2011) stressed that identity development in LAC is a somewhat subordinated area in terms of theoretical literature, with limited empirical research looking at young people's own sense of self (McLeod, 2007). In other broader literature, identity is conceptualized as a central part of human development and critical to mental health and well-being across the life span (Oyserman & Markus, 1990;Ruvolo & Markus, 1992) yet few researchers have gained an understanding about identity development for those in foster care.

Identity Formation for Looked After Children (LAC)
Within his model Erikson highlighted the contextual nature of identity development, arguing that an individual cannot be understood in isolation from their social context. In his model of psychosocial development, Erikson (1968) also emphasized the role of earlier experience, especially within the family, to subsequent phases of development. In addition, attachment theory (e.g., Bowlby, 1969Bowlby, , 1988 suggests that a loving and consistent caregiver provides children with a safe environment in which to explore and develop. Alongside this, Smith and Logan (2004) suggested that knowledge about one's personal and familial background is important in identity formation to integrate one's history into a continuing sense of self. Maintaining relationships with family and respecting cultural background is known to be important for LAC. Winter and Cohen (2005) illustrated the difficulties young people with no knowledge of their personal history can face in relation to their identity and wellbeing and the sense of loss that this brings. This underpins the National Institute for Health and Care Excellence (NICE; 2013) guidelines regarding the importance of life story work. Possible difficulties with identity formation for LAC may arise in the face of childhood abuse, difficult relationships, an unstable environment, and multiple care contexts. Identity formation, consequently, may be more complex for those who have been in foster care but this does not imply there is anything pathological about it. McMurray et al. (2011) interviewed adolescents (12-to 16-year-olds) in foster care and found young people's identity is shaped by their relationships, that identity can act as a protective factor and can be delayed or put on hold. Within this study, social workers argued that young people "may not have had the opportunity to start developing a coherent sense of self because they were still in the midst of very difficult situations" (p. 216). Young people themselves also recognized that a lack of independence and control while waiting for the outcome of decisions regarding their lives resulted in their future identity being put on hold. Kools (1997) interviewed 15-to 19-year-olds in foster care and found that the foster care status had a detrimental impact on identity. The stereotypical view of children in foster care contributed to the devaluation of the adolescents' sense of self. Despite this research, there remains a significant gap in our understanding of the process of identity formation in the context of growing up in foster care. Erikson (1968) highlighted that confusion regarding one's identity can result in low or unstable self-esteem and feelings of insecurity. According to the Diagnostic and Statistical Manual of Mental Disorders (5th ed.; DSM-5; American Psychiatric Association, 2013), negative or unstable selfesteem is a central feature in the diagnostic criteria of a large number of mental health difficulties. In contrast, positive self-esteem has been shown to contribute to better mental health and well-being (Mann, Hosman, Schaalma, & de Vries, 2004).

Clinical Relevance of Identity Formation
LAC are much more likely to have experienced risk factors, such as abuse, neglect, family dysfunction, poverty, and parental mental health, for a range of mental health problems (Richardson & Lelliott, 2003). Research suggests the prevalence of mental health and behavioral problems among LAC are significantly higher than children in the general population (McCann, James, Wilson, & Dunn, 1996). Outcomes for those who have experienced multiple placements tend to be worse than those who experience stability within foster placements (Courtney & Barth, 1996) and those in kinship care as this increases children's sense of belonging through continued family identity (Selwyn, Farmer, Meakings, & Vaisey, 2013). Reoccurring messages of rejection, a lack of stability or a sense of belonging can negatively affect selfesteem (Yancey, 1992). Given the higher levels of mental health difficulties in LAC and the known role that identity and self-esteem have in mental health, whether as a risk or a protective factor, it seems clinically significant to investigate what factors help construct or hinder the formation of identity in LAC.

Rationale for Research
Of the limited research in this area, most studies have interviewed individuals who are currently within foster care, aged 12 to 19 years (Kools, 1997;McMurray et al., 2011). This is a similar age range to research on "typical" identity formation. It has, however, been repeatedly demonstrated that achievement of identity has seldom been reached by the end of adolescence (Waterman, 1982) and that identity formation continues into the twenties (Arnett, 2000). The current study, therefore, aims to look specifically at those who are slightly older, 19-to 25-year-olds, in the hope that they will have had a chance to reflect on their experiences and have gone through the process of identity formation for longer. Alongside this, as Kools (1997) found that foster care had a negative impact on identity, it was deemed important to identify how young people who are no longer in the care system view themselves and the meanings they ascribe to their experience of identity formation.
Interpretative Phenomenological Analysis (IPA) requires homogeneity within the sample. Given that within mainstream literature females are thought to strive to incorporate their identity in terms of affiliation (Gilligan, 1992), and the known attachment and relationship difficulties common within LAC (Golding, 2008), the current study focused on female experience of identity only. This study hopes to further the knowledge of how young women understand their identity and its formation within the context of having previously been in foster care.
Research Question: How do female care leavers make sense of their identity development?

Method
Design IPA was used to explore how individuals reflect on their experiences and the meaning people make from these lived experiences (Smith, Flowers, & Larkin, 2009). IPA uses a dynamic and interactive process in which both participants and researcher play integral roles in the analytic account produced (Smith et al., 2009). It was for these reasons that IPA was chosen to explore how young women who have previously been in the care system make sense of their experiences in relation to the question "Who am I?" and the understanding they give to how they constructed this sense of themselves. This study utilized in-depth semistructured interviews.

Participants
During the recruitment process, eight young women were identified and expressed interest but did not turn up to the initial meeting. Two others did not meet inclusion criteria and one changed her mind during the initial meeting. Given difficulties recruiting, the inclusion and exclusion criteria were altered during the course of recruitment. The initial criteria included the following: females, 19 to 25 years, significant period in care; multiple placements (3+), in the care system until independence; and fluent in English. Exclusion criteria included the following: adopted children, brief period in care (e.g., days/weeks); unaccompanied asylum seekers; kinship care; learning disability. During later stages of recruitment, unaccompanied asylum seekers were included and multiple placements were redefined as multiple homes, for example, having lived with different people prior to care and/or more than one placement in care.
Following IPA guidelines, a small sample was recruited (Smith et al., 2009). Although including participants from diverse ethnic backgrounds and unaccompanied asylum seekers contradicted the recommendation for homogeneity, the research question focused on identity formation in relation to significant instability and a substantial period in care. It was this similar experience and gender that provided homogeneity. Participants were eight young women (Table 1.). All experienced significant developmental trauma and were subject to a full (local authority) care-order. All participants experienced multiple homes prior to and/or during care, for example, passed around family members, friends, strangers, residential placements, and foster placements.

Recruitment
Participants were recruited through a charity for care leavers. Staff identified potential participants and the research was discussed with them. If the young person expressed an interest and demonstrated clear understanding what participation would involve, consent to be contacted was gained. Initial contact between researcher and young person took place in person. The research was explained and participants were given the space to ask questions. Confidentiality and its limits were discussed. The potential for distressing material to arise and participants' freedom to pause or stop at any time was also stressed. When the young person demonstrated significant knowledge of the study to make an informed decision, the consent form was completed.

Ethical Approval
Ethical approval was obtained from Canterbury Christ Church University's Ethics Committee. Risk of harm was considered and monitored throughout the interview process and appropriate safeguarding put in place. The study was conducted according to the British Psychological Society (2009) code of ethics and conduct.

Interviews
In developing the semistructured interview schedule, current literature was considered and consultations were undertaken with two psychologists experienced in working with LAC and a social worker in a leaving care team. A pilot interview was conducted with a care leaver. Following feedback and reflection, the questions were amended. Information derived from these sources was used to ensure questions would elicit information relevant to the aims of study.
Interviews took place in a private room within the charity. The interview schedule served as a guide, while the interviewer focused on actively listening to participants' accounts and following their lead. Interviews lasted between 64 and 97 min, guided by how long the participant wanted to talk about their experiences. At the end, participants were provided with a space to debrief. All participants reported they found it a positive experience some reporting that it was insightful, supportive, and therapeutic. Participants expressed their happiness at being given the chance to have a voice. No participants wanted to talk further with a clinician to discuss any issues following the interview. Participants received a £10 voucher as a way of thanks. Interviews were digitally recorded, transcribed, and anonymized for analysis. Interviews took place over a period of 6 months, and data were stored in line with ethical guidelines and data protection. The transcripts of the data are kept for 10 years.

Data Analysis
As recommend in IPA (Smith and Osborn, 2008), repeated reading of each transcript allowed an immersion in and familiarity with the data. Exploratory comments of descriptive, linguistic, and conceptual features of the data were recorded. Emergent themes were reviewed in light of the original transcript, and quotes were identified for each theme to ensure that all themes encapsulated direct participant experience. A list of all themes made on the transcript was printed out, and similar themes were clustered together and given a new title to represent a subtheme. This process was completed for all interviews individually. Themes were subsequently compared across all participants. Associated themes were grouped together under superordinate themes. The researcher returned to the original transcripts to ensure goodness of fit between the analysis and data and to confirm that the quotes used to support themes adequately represented the breadth of participants' experience

Quality Assurance
Currently, there are no standard criteria for the evaluation of qualitative research. The data analysis was subjected to quality assurance guidelines taken from Mays and Pope (2000) and Yardley (2000). The need for the study was evidenced by the limited existing literature. Commitment and rigor was accounted for by following established IPA methodological guidelines (Smith et al., 2009). Inter-coder agreement (Yardley, 2008) was used through comparisons with a supervisor's analysis of four transcripts, to ensure that the themes were grounded in the data. "Transparency" was detailed through a clear description of the data collection and analysis and an audit trail to demonstrate theme development. Direct quotes from the data were highlighted to support themes to make the analysis transparent and easier to evaluate by others. To allow for reflexivity (Yardley, 2000), a bracketing interview (Ahern, 1999;Fischer, 2009) was conducted with the researcher prior to conducting this study which allowed any preconceptions or assumptions to be identified that may bias interpretation. A reflective research diary was also kept throughout.

Results
Although the journey for each participant was unique, three superordinate themes emerged from the data which reflected the process and outcome of identity development. The outcome of identity development highlighted two dimensions: (a) how individuals saw themselves and (b) how their identity played out practically day to day. Twelve subthemes were subsumed under these three superordinate themes (see Table 2 for examples).
The results from the current study demonstrate that past experiences shape how participants understand their identity and influences how their identity plays out in day-to-day life. There are also bidirectional effects within and between how participants understand their identity and the actual experience of their identity which in turn has an impact on the continued development of their identity.
To make sense of the results a model was created (see Figure 1).

Construction of Identity-How I Became Me
This superordinate theme encapsulates the influences that have contributed to the participants' sense of self. . . " (Freya) " . . . When I went back to my mum I felt I was really happy and stuff then I realised after seeing my mum's behaviour this is the worse decision I ever made . . . " (Leah) Protection of identity " . . . We went to the hospital and I had black eyes (laughs) and mum (laughs) was like trying to cover it up with foundation and like, put hair in front of my face . . ." (Chloe) "I don't really think anyone knows me" " . . . sometimes I think erm like will I ever be able to form long term relationships with people because I've never been able to do that . . . but other than that it's fine." (Amara) Positive influences and turning points "It's like I broke that cycle when, that's when I found myself actually when I was pregnant . . . " (Freya) " . . . She was a nice lady, she looked after us and she just said to us I was in the same position . . . " " . . . talking to my foster carer cos she had been through experience, the same things we were and she was like it's ok, it will get easier . . ." (Nyala) Understanding of identity-Who am I?
I am an outsider "I was again excluded and I felt very much on the outside of that social circle so regardless of where you're put, you (sighs) your needs are never truly met . . ." (Chloe) "Being by myself and knew kind of thing and not having the baggage of my parents cos your whole life is just baggage. You're like baggage until you reach your age when everyone says you're an adult and nobody really care for your well-being . . ." (Dana) I am a survivormy active construction "I've come a long way; I've never given up. I'm brave and yeah." (Nyala) " . . . I'm writing who I want to be, so I think there's a need for strength for that, I needed strength to let go of a lot of things that had happened and build up a framework and then each day write down what do I want." (Ebele) I am bad, undeserving and unlovable " . . . I just give people the sort of love and affection that I don't get and what I wish I could sort of get back but it never seems to work that way, like I seem to give a lot more than I receive . . ." (Chloe) " . . . Sometimes I feel like it's because she had me why she had to go through that and sometimes she makes me feel like that . . ." (Dana) I help and care for others " . . . I'm volunteering at a mental health hospital erm like befriending people with mental health issues . . . I always wanted to like take care of people like that." (Amara) "I was an ambassador at university, erm that inspires people, that's so happy erm the Leah that would help anyone, the Leah that everyone that goes to for advice . . ." No sense of secure base. This refers to a lack of security in relationships and home instability. Instability characterized many participants' experience both before and after going in to care, frequently moving house and caregivers, receiving inconsistent care, and having no sense of belonging. The experience of instability and the powerlessness surrounding this was highlighted and ultimately affected participants' internal working models of themselves and others: It's like I stay there and then it's like woah ok they just move me, I can't trust no one, like I literally don't trust no-one. (Leah,p31,(635)(636)(637) Participants spoke about receiving unsafe caregiving before going into care, leaving them believing that adults don't keep you safe: . . . before I came into care cos he was an adult, he was someone that said he would take care of me, not even to me but just to my mum and I thought that I was safe . . . (Dana,p47) Experiences of unsafe and uncaring caregiving in foster care frequently mirrored participants' earlier family experiences: Erm, then we got moved to this carer, who was just so abusive and it was verbally like she'd say things cos she knew about our files and stuff, "You're going to end up like your mum . . . (Keisha,p8,(148)(149)(150) Ambivalence re birth family/ambivalence re foster care. Participants demonstrated simultaneous conflicting reactions, beliefs, or feelings toward their birth families and/or foster care. Participants both knowingly and unknowingly spoke about having mixed feelings regarding their birth families. Some participants spoke of unconditional love and loyalty toward their birth mother, which was accompanied by resistance, blame, and disappointment: . . . seeing my mum it was like, are you really my mum? Like it gave me mixed emotions cos I love her but it's difficult because I don't want to show her too much affection for her to just drop it like she did before . . . (Freya,p5,(91)(92)(93)(94)(95) I remember feeling so emotional about that, like seeing her make an effort even though she she'd, of course it doesn't make up for all the years of abuse . . . (Chloe, p72, 1481-1483) . . . she wasn't a bad woman . . . (p72, 1485) . . . from a very young age I remember looking at my mother and deciding that I was going to be everything that she wasn't . . . (p29, 575-578) Alongside this many participants also felt ambivalent regarding their experience in care.
. . . I'm so grateful for that (being given a childhood) cos otherwise I wouldn't have had it even living with my nan I wouldn't have had the experiences I've had being in care, but care's still horrible (laughs). (Freya,p29,(573)(574)(575)(576) Care was stressful, the beginning the middle and the end, everything was stressful because at the beginning you're like shit you just realised you're not going to see your family for ages . . . (Dana,p23,(396)(397)(398)(399) Going into care was the best thing that could have happened to me." (p28, 491-492) This ambivalence regarding both birth family and foster care demonstrates how conflicted and torn many of the participants felt and continue to feel.
Protection of identity. All participants described backgrounds of adversity including physical, emotional, and sexual abuse; neglect, domestic violence, loss, parental mental health, or substance misuse; and poverty and war in their birth country. Participants developed coping strategies as a way of surviving these adverse environments and recounting these experiences to protect themselves and their identity. One coping strategy was the use of humor to minimize recounts of traumatic memories: . . . my arm was broken and I had these black eyes and I took my top off and the nurse just burst into tears and ran out and then my mum looked at me like you bitch, you're going to get it (laughs) when you get home (laughs) . . . (Chloe,p7,(137)(138)(139)(140)(141) Participants frequently disconnected from any emotion related to experiences. When asked what is was like for Dana to have had multiple changes in home and caregivers, she replied, A positive reframe of experiences allowed participants to feel as if they had gained something from these traumatic experiences: I wouldn't take back anything that's happened to me I think that I've actually got a massive advantage of the majority of people . . . I've been through shit times and I've come out the other end of it. (Chloe,p46,(945)(946)(947) Escapism enabled participants to survive the difficult situations they found themselves in. These strategies provide distraction and temporary relief from an unpleasant reality: . . .I've always found something fun out of the situation. Like we could be walking around for ages with my mum . . . and I'd be like to my sister let's play a game . . . but just trying to find the fun in every situation because everything's so negative anyway . . . (Keisha, p32, 619-624).
Positive influences and turning points. All but one participant talked about positive influences in their lives including people, relationships or events. Positive influences seem to have helped people in a number of ways, one being encouraging self-confidence and self-belief: . . . My tutor (college) was like "you're so capable of going to uni" and I was like "no I'm not," I'd lost my confidence and then she brought me up again. (Leah,p36,(727)(728)(729) A number of participants highlighted turning points in their lives where significant change occurred for them. For several, the loss of key figures in their life was a major turning point: It was such a great loss (death of grandparents) I can't explain it . . . (Keisha, p51, 1015) . . . it just made me realise that I can't keep being how I was like and just thinking the world's against me and defeated and stuff that I need to be something they could be proud of . . . (p52, 1018-1020) For one participant becoming pregnant and being allowed to keep her child was a major turning point, providing her with a purpose, self-belief, and an identity outside of her care status: I had to have all these assessments and I passed all of them, which was more like it lifted me up even more to think I can do it and I can break this cycle . . . like my family goes in the care system back from four generations ago . . (Freya,p20,(384)(385)(386)(387) Cos I was never a reliable person but it, that all changed as well when I was pregnant, I was reliable, on time, organised just everything got into shape when I was pregnant. (p24, 486-488)

Understanding Identity-Who am I?
This superordinate theme encapsulates how participants make sense of who they are and how they present themselves.
I am an outsider. This subtheme represents the self in relation to others. All participants spoke of feeling like an outsider, feeling different to others, and that something was wrong with them, that they were not "normal." Having no sense of a secure base and ambivalence regarding birth family and foster care left participants having never experienced a sense of belonging either within their birth family or within the care system: . . . there's actually quite a lot of racism towards me (laughs) funnily enough my mother is mixed race, my brother is black so I was the white kid in a black family and she actually sort of hated me for it . . . (Chloe,p5,(89)(90)(91)(92) This lack of belonging resulted in participants feeling like an outsider whoever they are with. Participants spoke of not having a normal "family" and the impact that this had, compounded by being in foster care, which further sets them apart from their friends: I think if I wasn't in the care system, I had a normal family I'd have been like all my other friends who've graduated by now . . ." (Keisha, p60, 1182(Keisha, p60, -1183 I am a survivor-My active construction. Given the adverse environments participants grew up in they had to actively construct an identity in order to survive both dangerous and abusive environments and living with constant instability and inconsistency. The strategies used in protection of identity, positive influences, and turning points enabled participants to see themselves as a survivor. Participants had to create a self to survive the different environments they found themselves in: . . . maybe I'm just adaptable . . . (Leah, p34, 692) . . . like I call it adaptable . . (Chloe,p34,698) Participants spoke of how their experiences shaped them and they found strength in enduring difficult experiences: It shapes me to be who I am, I think I'm really strong in comparison to a lot of people I know . . . (Keisha,p37,(733)(734) All participants were immensely resilient and resourceful. Some participants spoke of actively attempting to try and provide themselves with a better future: . . . I was on a mission . . . for a better life. I've been taken out from a situation that was really bad and I'm here . . . but I thought you know what I'm going to give my best in education and I'm going to try and make something good out of myself. (Ebele,p25,(499)(500)(501)(502)(503)(504) I am bad, undeserving, and unlovable. This theme relates to an internalizing sense of self understanding. All participants spoke of feeling bad, undeserving, and unlovable. As highlighted in the model above (Figure 1), this theme has bidirectional links with feeling like an outsider. Participants' experiences of multiple homes and caregivers (no sense of a secure base), for many, both before going in to care and while in the care system, resulted in feelings of rejection and abandonment and had an impact on internal working models of themselves: . . . it feeds my need to prove to myself that . . . like no-one likes me, no-one loves me, no-one cares about me . . . (Chloe,p39,(802)(803)(804) . . . thing is right now I feel like such a bad person after what happen with my ex . . . he didn't want to be with me . . . I feel like am, I am a bad person . . . (Leah, p55, 1131(Leah, p55, -1136 Some participants believed that the reason they ended up in care was because they were unwanted and unloved by their parents. This ultimately led participants to believe it was their fault they were in care and that they are clearly unlovable: . . . you're in care cos you're mum doesn't want you, I've had that said to me hundreds of time . . . (Freya,p13,(258)(259)(260) Many participants experienced repeated rejection within the care system, the devaluing of self and disregard by others that they previously encountered within their birth families, reinforcing the view of themselves as bad, undeserving, and unlovable: . . . I could never have someone in my house and treat them like you're not wanted or I'm, you're just here because I'm getting paid for you, I don't want you. I think being in the care system theres already enough reason to feel isolated and rejected . . . (Keisha,p50,(990)(991)(992)(993)(994)(995) I help and care for others. In relation to not having their own needs met throughout childhood (no sense of a secure base) and having to meet the needs of others, for example, parents or younger siblings, all participants spoke about helping and caring for others in their current life whether this was birth family, friends, young people in care, and other vulnerable individuals. Reflecting on their experiences, some participants were able to identify why helping others was a part of who they are: I think it's given me a very mumly nature because I try to be a mum to everyone. (Leah,p39,(790)(791) Helping others also provided participants with a sense of personal value and achievement: . . . I like to help people and I, it just makes me happy to see that you can actually help someone . . . (Nyala,p41,(828)(829)(830) Some participants spoke about being indispensable to others, which provides participants with a sense of worth and the feeling of being needed by others: I seem to be there to help her a lot like my sister says stuff to me like I'm her rock, and you know she don't know where she would be without me . . . (Keisha,p40,(788)(789) Other participants had aspirations of having a career within a helping profession and some were currently volunteering with charities: . . .I'm really in to charity and volunteering . . . I'm about to go to India for three months . . . to raise livelihoods in rural villages and raise awareness of vector-borne diseases . . . (Chloe,p3,(43)(44)(45)(46)(47)(48)(49)(50)

Experience of Identity-How My Identity Plays Out
This superordinate theme encapsulates how participants' identity plays out practically on a day-to-day basis.
Doing it alone. In a context of relational trauma, rejection, and abandonment, as a way of protecting their identity and through feelings of ambivalence regarding birth family/foster care, participants became increasingly self-reliant, ultimately leading to increased isolation from others and reinforcing negative self-beliefs. Doing it alone also links bidirectionally to being a survivor becoming self-reliant is in part how they survived and is also a way of taking back control. Their experience, however, was of loneliness and solitude: I feel really isolated from everyone, like not that people have isolated themselves from me it's just what I've done because living by myself and just trying to live . . . that's why I'm probably lonely quite a lot . . . (Freya,p1, Participants spoke about relationships with others being difficult and not having longevity, ultimately leaving them feeling alone: . . . like me and people just grow apart so, yeah. (Amara,p3,41) Based on past experiences of people and relationships, participants spoke about being increasingly self-reliant as others cannot be relied upon or trusted.
. . . I've had to be my own mum, my own, my own person and it pained me at times but every day I say to myself I'm going to get this, and I'm going to get this done because then I know I did it, I, I helped myself . . . (Ebele,p40,(825)(826)(827)(828) Participants also frequently isolate themselves as a way of protection: . . . I didn't realise I was doing it but I took my chair and I sat outside the circle . . . I sort of already socially exclude myself because I expect to be excluded . . . (Chloe, p49, 1005(Chloe, p49, -1012 Taking back control. Similar to doing it alone, taking back control was developed as a way of protecting themselves in the face of frequent feelings of powerlessness. Uncertainty and danger in the past meant participants needed to take back control of their own lives as others cannot be trusted to do the right things for them. Control was used to reduce anxiety related to an intolerance of uncertainty: . . . I like things done in a certain way . . . and if it's not done in a certain way I tend to let it get to me and then I worry about it . . . (Nyala,33,(626)(627)(628)(629) For some participants, control was about being prepared and limiting potential dangers in the future: Analyse everything and I always do some weird things where I think about what I would do in certain scenarios like I'm trying to prepare myself for a life or different kind of scenarios . . . .so I know what's coming. (Dana,p35,(624)(625)(626)(627)(628)(629)(630)(631)(632) Another way this control was displayed was in relation to participants' emotions. Participants frequently felt unable to display emotions as this would demonstrate weakness and vulnerability. If they displayed these emotions, people's responses would no longer be within their control; therefore, it is safer to protect themselves by keeping emotions hidden: I don't like being emotional with people . . . people see a weak part of you . . . when you're emotional, cos people can then try and tap in to that . . . (Dana, p57-58, 1031(Dana, p57-58, -1044 The mosaic/fragmented self. All participants displayed this mosaic/fragmented sense of self. Participants spoke of having to constantly adapt to different environments and having to change themselves to fit in. To combat feeling like an outsider and being different to others, participants spoke of how they become a reflection of those they are around as a way of managing social situations: you're constantly being moved your sense of self changes because you're just sort attaching yourself on to whoever you're living with . . . you're there trying on loads of different personalities . . . you try so hard to fit yourself in to these sort of social groups or even sort of boxes . . . (Chloe, p69, 1422(Chloe, p69, -1430 . . . I try to be . . . what's that thing I don't know, malleable like I can be different around different people. (Amara,p27,(475)(476)(477) The ambivalence participants feel in relation to their birth family and foster care, alongside having to be a different person depending on who they are with and where they are is likely to have led to a conflicted sense of self. Most participants frequently made contradictory statements about the sort of person are they: For all participants there was a sense of an external and an internal identity or a two-part self: . . . people say I come across very confident but . . . it's a facade you know, it's a front I'm not, I'm dying inside (laughs) but like you've obviously got to put on this front to people and that's something that I've become very, very equipped with and very on point I'd say it's my act. (Chloe, p49-50, 1013(Chloe, p49-50, -1017 Many participants spoke about the sense of confusion and loss this two-part self-created for them: I forget that it's the other side of me that everyone sees that they don't see at home erm and it's really people love that Leah but they never really see the . . . that's why I'm lost I'm lost in-between the two. (Leah,p16,(311)(312)(313)(314) There were some aspects of themselves that induced feelings of shame. To protect themselves from these feelings participants kept these aspects hidden from others: . . . nobody knows I'm in care unless I tell them so if I don't tell them they don't know . . .hiding it from people I feel strong and capable . . . (Freya,p26,(516)(517)(518)(519)(520)(521) Psychological impact. This theme describes the impact past experiences have had psychologically on participants, in relation to mental health and well-being. All participants spoke about difficulties with mental health, some more severe and enduring than others. Participants spoke of ways in which they tried to manage the psychological impact, one method being self-medicating: I used to drink a lot, I would to drink myself to sleep, I would cry myself to sleep because when you've been through abuse when you've been through moving around and no family the emptiness crave in you. (Ebele,27,(553)(554)(555)(556) Another way of managing the psychological impact was through self-harm: . . . that sort of led me to sort of like self-harming and sort of other mental issues . . . (Chloe,p11,(224)(225)(226) Participants were able to reflect on how their experiences had an impact on them psychologically: . . . we were talking about why I have mental health issues but I wanted to say because of my mum, because of my dad, because of what happened . . . (Leah,p9,(170)(171)(172) Most participants were able to relate the impact their history had on present day low self-esteem: . . . I've got low self-esteem because I don't feel confident and I don't wanna talk to anyone (Amara, p19, 337)

Discussion
In summary, participants' construction of identity can be understood in the context of early adverse environments and developmental trauma. Developmental trauma has been conceptualized as " . . . multiple, chronic or prolonged exposure to developmentally adverse interpersonal trauma in early life . . ." (Rahim, 2014, p. 549). This construction of self, in turn, mediates how participants understand and experience their identity. For many participants, survival was their main priority during childhood which influenced who they became as adults. This resonates with Erikson's theory regarding the contextual nature and role of earlier experience within the family, in identity development.
Participants' early environments were characterized by fear, unpredictability and instability. All participants experienced parental unresponsiveness or absence. Many also experienced repeated physical and emotional abuse from their parents and others. Frequent changes of primary caregiver were also common, never being with one person long enough to develop significant relationships. Within the attachment literature this is conceptualized as a lack of a secure base (Bowlby, 1988), the absence of warmth and security within early relationships is considered to have significant effects on later development (Waters, Crowell, Elliott, Corcoran, & Treboux, 2010). All of these changes reiterate the repeated experiences of loss participants encountered.
Participants' survival strategies were and remain as "doing it alone" and to "take back control." Participants spoke of providing their own safety as children and adolescents: it was, and remains, too frightening and dangerous to let others be in control. This need for control is driven by anxiety and fear (Howe, 2005) in relation to an insecure attachment (Golding, 2008). Control provides an environment that is more predictable. The results of "doing it alone" and "taking back control" support Kools's (1997) findings of considerable emphasis on autonomy and control. Participants in the current study highlighted difficulties within relationships, feeling vulnerable around others and recurring worries of rejection and abandonment. As a way of managing this, participants became increasingly selfreliant which, unknowingly, increased their isolation from others and reinforced the belief of having to be self-reliant.
Participants' needs were frequently unmet in the context they grew up in. Unlike the holding environment described by Winnicott (1960), in which parents hold in mind the needs of the child, participants were instead used to meeting the needs of their parents. Given that parents' needs take priority, children learn to survive by attuning to other people's needs and disregarding their own (Heller & LaPierre, 2012). Heller and LaPierre (2012) spoke of a shame-based identification in which individuals believe having their own needs is bad and they are not entitled to be fulfilled. Howe (2005) suggested that abused children often feel responsible for causing their carer's hostility, leading them to feel bad for upsetting their parents. Similarly, as Shengold (1979) contended, "The child needs his abusing parent. Since the parent cannot be bad, it is the child who must be the bad one" (p. 535). The current study corroborates these theories with participants feeling bad, undeserving, and unlovable. Alongside this Heller and LaPierre suggest that alongside a pride-based counter identification strategy, individuals become caretakers. Having others depend on them suppresses feelings of shame about having needs or the fear of rejection if they express their own needs. Caring and helping others may be an indirect way participants within the current study have of getting their own needs met.
Participants had a number of conscious and unconscious ways of coping with these adverse environments. During the interviews, participants frequently minimized their experiences, disconnected from the impact, and associated emotions and used positive reframes and splitting. This dismissing of past trauma provides protection through avoidance of negative affect that as children they learned not to express (Purnell, 2010). Attempts made by participants to manage difficult early experiences accorded with psychodynamic concepts of defensive mechanisms in response to early trauma (Freud, 1939). Defenses are theorized to become activated when anxiety feels too strong. Splitting is thought to be one of the earliest defenses, defending against feelings of both love and hate for the same object (Klein, 1935).
Given the inconsistent nature of their experiences, children exposed to developmental trauma have difficulty piecing together a coherent narrative (Byng-Hall, 1995;Rustin, 2005). The care system frequently exacerbates this incoherence through multiple moves, new environments, new attachment figures, and new ways of living (Conway, 2009). This disjointed narrative was a common feature for the current participants. In an unsafe environment, it is necessary to feel ambivalent about things to keep safe (Golding, 2008). The ambivalence regarding birth family and care is understandable given the intense need for and extreme fear of contact these individuals feel (Heller & LaPierre, 2012). As a result of volatile and threatening caregiving, participants may have developed a fragmented strategy for managing this stress (Main & Solomona, 1990). Cumulatively, these experiences potentially lead to the fragmented sense of self seen in the current study.
Both the social workers and young people interviewed by Kools (1997) described young people not presenting their "real" selves to others. This study extends those findings suggesting an internal and external identity: parts that are acceptable to display and parts that need to remain hidden. Similarly, Kools also found that young people presented different identities in different contexts. Social workers described this as a protective mechanism to avoid dealing with their history. The current study, however, suggests this was used as a survival strategy given the numerous environments they found themselves in. By becoming a "chameleon," participants were able to adapt to different environments and people. The lack of belonging within a family or a given place, along with constant instability and inconsistency, undoubtedly left participants feeling confused about who they are and where they belong. Presenting differently in different contexts is an adaptive way to manage this anxiety and allow participants to fit in to the different environments they find themselves in.
Unlike much of the current literature, this study highlights the role of developmental trauma in the construction, understanding, and experiences of identity for these young women. Kools (1997) drew on how past experiences impacted identity; however, this was not specifically related to trauma. In part, many of the ways participants construct, understand, and experience their identity is in relation to past survival strategies. This provides some support for Heller and LaPierre (2012) who suggest that the identity of abused children becomes based on the persistence of survival strategies that continue beyond when they are needed. In many ways, identity for young female care leavers could be considered as a successful adaptation to trauma and shame rather than being pathologized, as is frequently done so within the literature. Given that these are adaptations to early adverse environments, difficulties frequently arise when once useful survival strategies restrict individuals as adults when they are no longer in an environment that requires them (Heller & LaPierre, 2012).

Limitations and Future Research
A number of limitations and biases may have been introduced by recruiting through a charity helping care leavers find employment. Participants were willing to engage with services and actively sought employment or were in employment and/or university. This excludes many care leavers who are not in the position to make contact with services or find employment: for example, in 2015, 40% of care leavers aged 20 to 21 years were not in education, employment, or training (Department for Education, 2015). Furthermore, the sample was self-selected. Nothing is known about those who were approached but declined the research. It is unknown to what extent these findings apply to those or to male care leavers. Findings from the current study may outline a gendered experience to identity development. For example, participants in the current study frequently help and care for others, which resonates with Gilligan's (1992) theory that female identity is in part constructed in terms of affiliation. As this study focused on females, it is unknown whether this is a gendered experience or whether male care leavers would also take this role. This would therefore be an important area for future research.
The narratives participants chose to share may have been shaped by a number of factors including sensitivity of the topic and myself as a White female in a mental health profession. Results are suggestive of an internal and external identity, the extent to which this played out in the interviewee-interviewer relationship is unknown. Some participants did make suggestions toward this " . . . I would never be able to tell you it if I'm in it, or let's say any problem I have right now you'd never hear about it . . ." (Dana, p57, 1032(Dana, p57, -1035. When asked how else Leah would describe herself she replied, "Erm . . . I'm trying not to be negative to be honest . . ." (Leah,p45,(915)(916).
Participants' experiences prior to and in foster care varied, for example, the level and type of abuse and trauma, the number of homes prior to care, and number of foster placements. Participants also differed in the age they entered care, the knowledge they had about their history, experience of therapy, and current contact with birth family and/or foster carers. In addition, as this interview was a snapshot of one time point, interpretations cannot be made about temporal aspects of experience. These factors would be helpful to consider in future research.
The impact of developmental trauma is documented within the literature with known effects on neurodevelopment, self-regulation, information processing, and relationships (Streeck-Fischer & van der Kolk, 2000). However, the impact of trauma specifically on identity is absent from the literature. The current study, therefore, may serve to broaden this knowledge base. Future research could give fuller consideration to the role of developmental trauma in identity formation.

Clinical Implications
All participants spoke of difficulties with current or historical mental health, some receiving diagnoses of depression, anxiety, and bipolar or personality disorder. Ackerman, Newton, McPherson, Jones, and Dykman (1998) acknowledged that most children and young people who have experienced developmental trauma receive diagnoses for things other than trauma. Ackerman et al. (1998) highlighted that treating the symptoms of these disorders does nothing in addressing the underlying distress caused by years of trauma. Many of the behaviors displayed by LAC and care leavers can be understood in the context of coping strategies and attempts to manage their distress as highlighted in the current study. These efforts to maintain their own safety are frequently considered to be maladaptive, dysfunctional, and oppositional (van der Kolk, 2005). Individuals in the current study already feel bad, different, and defective. Mental health services, therefore, may indirectly serve to enhance these self-beliefs with the diagnostic system focusing solely on deficits located within the individual.
LAC and care leavers' identity is frequently termed pathological in some way. Thomas and Holland (2009) found social worker descriptions of children's identity tended to be negative. The current research highlights the extent to which identity, for these individuals, is formed in the context of trauma and based largely on strategies for survival. These behaviors are frequently deemed in need of "treatment," with no consideration of the function these behaviors serve for the individual. Intervention should be based on the individual's needs and personal goals, not the goals of social services, foster carers, or other professionals, while taking account of their strengths. A key task for clinicians may be the provision of a containing, holding space, in which LAC and care leavers feel adequately secure. This may facilitate the lowering of defenses enabling the exploration and testing of current beliefs. In addition, a key task for clinicians will be to enable significant others to better understand a young person's identity and behavior.
The aim here is not to deny that LAC and care leavers have mental health difficulties. The literature consistently finds that children in care are more likely than not to have psychological difficulties (Tarren-Sweeney, 2008). Participants in the current study themselves highlighted the distress they face psychologically. It is essential to identify the difficulties and struggles that many LAC and those caring for them have to support them most appropriately. The current study highlights that LAC and care leavers need to be understood in the context of developmental trauma, rather than focusing on symptoms of various diagnoses. Mental health and psychological wellbeing within the LAC population is clearly a complex issue, and without specialist knowledge and understanding of developmental trauma and young people's survival strategies, those working with these individuals will struggle to understand and support them in the most appropriate way. Tarren-Sweeney (2010) highlighted that generic child clinical assessment methods, including psychometrics and formulations, are unsuitable for LAC, suggesting this in part highlights gaps in the knowledge base for this population. The current study may therefore serve to broaden this knowledge base and provide a way of formulating with this population. This will become increasingly important as children transition into adult mental health services where less knowledge of this area exists than in specialist LAC mental health services.

Reflexivity
Throughout this study, the researcher reflected on experiences as they arose during the research and in the context of previous experience both professionally and personally.
Prior to undertaking the research, the researcher engaged in a bracketing interview to "bracket" off any preexisting assumptions. During the interviews, transcribing and the analysis, the traumatic accounts expressed by participants evoked strong reactions in the researcher as she listened to the abuse and the intensely painful stories the participants shared. These experiences were recorded in a research diary and made use of supervision.

Conclusion
A search of the literature regarding identity development for LAC suggested a large gap in the research, especially for those who have since left care. The current findings suggest that for these participants, their construction of identity in part is based on survival strategies in the face of developmental trauma and adverse early life experiences. This construction of self, in turn, mediates how participants understand and experience their identity presently. The impact of developmental trauma is documented within the literature with known effects on neurodevelopment, self-regulation, information processing, and relationships (Streeck-Fischer & van der Kolk, 2000). This is, however, one of the first known studies to highlight the impact of trauma specifically on identity in this group of young people. These findings, therefore, require further investigation.

Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding
The author(s) received no financial support for the research and/or authorship of this article.