The hidden professionals? An interview study of higher education-based teacher educators´ professional identity

This article investigates teacher educators’ self-understanding by asking how they explain their professional identities as teacher educators, based on socialisation and further professional development. Teacher educators facilitate learning from the initial teacher education phase to in-service teachers’ further professional development. The data consists of thirteen qualitative semi-structured interviews and two focus-group interviews with teacher educators from two universities in Norway. Using Bryman’s four-stage approach of analysis, 15 categories were re-organised into three main categories: (1) Recruitment and socialisation, (2) Professional identity, and (3) Professional development. The findings illustrate that teacher educators have different understandings of being a professional. For some, their identity is rooted in the discipline in which they were educated. However, others have built identities as teacher educators supplementary to their primary careers. This knowledge has implications for how universities can and should support teacher educators in the development of their identities. This can apply to educational strategies, institutional leadership and the role of teacher educators themselves. Their further professional development can be facilitated by a clearer understanding and appreciation of this profession reflected in educational policy.


Introduction
The aim of this article is to investigate teacher educators' self-understanding in relation to their experience of teacher-education practice, and how they understand their identity as teacher educators working in higher education. The background of this study is a growing interest in the quality of teacher educators' work in an effort to strengthen the professional profile of all teachers. The pressure of changes and reforms in recent decades have resulted in changes for teachers working at different levels, including teacher educators (European Commission, 2014, p. 10). In Europe, changes in teacher education include evolving centrallydetermined goals as well as a shifting balance among curricular components, required activities and outcomes (Caena, 2014, p. 118). Therefore, questions about teacher educators' self-understanding and identity as professionals are worth investigating further in relation to this evolving context. Teacher educators are defined as 'all those who facilitate the formal learning of student teachers and teachers, whether at the level of initial teacher education or continuing professional development' (European Commission, 2014, p. 1). The professional identities of teachers and the value attributed to the function of teacher educators are themes that have been addressed by Ulvik and Smith (2016), Angelo (2016) and Hoydalsvik (2017), among others.
Teacher educators participate in a complex exercise that can be interpreted in relation to three elements: (1) Fields of practice, (2) The roles of researchers, and (3) Expected academic standards (Elstad, 2010). Most research concerning teacher education is directed towards student teachers or educational programmes, while fewer studies have teacher educators as their objects of research (Ulvik & Smith, 2016). This article highlights research about teacher educators and their learning and teaching experiences. The research question is: • How do teacher educators explain their professional identities based on socialisation into the role and their further professional development?
In this article, the data demonstrates that teacher educators as an occupational group express differing understandings of their professional identity, and relatively many of them still hold an identity rooted in the discipline they were educated for before becoming teacher educators.

Background
Identity theory can be used to study informants' understandings of their professional roles and identities. G. H. Mead (1934) emphasised the importance of language and social interaction in the development of the self in his foundational work. He believed that human subjects form perceptions of themselves through seeing themselves through others. Mead defined self as arising through interaction and specifies that this is made possible through the identical reactions of the self and others (Mead, 1934, p. 317). Mead attached the development of identity to society's attitudes and values and believed that people can only become individual selves in relation to this. In addition, it is assumed that individuals are involved in certain cooperative activities within groups. The self consists of two phases, 'me' and 'I' , where one reflects the attitudes of the generalised others, and the other responds to them (Mead, 1934, pp. 174-175). We have constant access to the self, and the self is an object to itself (Mead, 1934, p. 136). Individuals experience themselves indirectly, through the position of other members of a given social group or based on the current attitudes of the entire community. We see ourselves as others see us by adopting their attitudes towards ourselves; Mead asks: 'How can an individual get outside himself (experientially) in such a way as to become an object to himself?' (Mead, 1934, p. 138). The span between the two parts of the self, 'me' and 'I' , ensures that individuality achieves expression, and therefore interaction and language are important in the development of the self. The process of constant construction of identity through community social interaction involves a continuous, objective analysis of one's own behaviour by taking the attitudes of the generalised others towards oneself, in a reflection on one's own point of view relative to others. Without this ability to reflect, humans would not possess self-consciousness, but only simple consciousness. No rational judgment can be made unless the self becomes an object of its own (Mead, 1934, p. 138). Mead's two-part definition of self, consisting of 'me' and 'I' , represents the collective and the subjective part of people and helps us retain our distinctive character and personality even though we consider ourselves to be others. No one has an awareness or identity that operates for itself, isolated from the social process it has emerged from and evolved within (Mead, 1934, p. 222). This is because the self both arises and develops in the social process (cooperative activities), through 'communication with significant others' (Mead, 1934, pp. 140, 165, 317). In this article, professional identity is not considered separate from personal identity; rather, the latter will influence who teachers and educators are as professionals (Hensvold, 2011). Following this logic, Akkerman and Meijer (2010, p. 308) define a teacher's identity as a dialogical self. In psychology, identity theory offers a more elaborate approach to teacher identity, conceived of as both unitary and multiple, both continuous and discontinuous, and both individual and social. The concept of identity is defined in various ways, with particular meanings of the concept specified differently in different scientific disciplines (Beijaard, Meijer & Verloop, 2003). First, positions about 'me myself ' in a professional role are produced in the process of qualification (Heggen, 2008, p. 324). Heggen (2008, p. 325) states that a professional identity is defined collectively, while also reflecting the construction of personal identity related to the exercise of a professional role. At the same time, professional identity is not limited to formal occupational requirements, but rather it fosters a more narrative identity or subjective self-esteem. Self-identity is fundamental, and professional identity must be based on it (Heggen, 2008, 324). In this sense, identity is not something one possesses, but something that develops during one's whole life (Beijaard, Meijer & Verloop, 2003). Livingston (2014) identifies certain implications of multiple, distinct identities, including the need for clarity about the role of definitions that people hold for themselves and the importance of embracing, rather than defining away, differences in teacher educators' identities.
As early as 2005, Murray and Male addressed the challenges new teacher educators face in establishing their professional identities. Their study analysed the tensions and conflicts arising for 28 teacher educators in their first three years of working in initial teacher-education courses in England (Murray & Male, 2005). The findings of this study showed that, despite their success in their previous teaching careers, the majority of informants took between two and three years to establish new professional identities as teacher educators. They faced challenges in two key areas: developing a pedagogy to teach teachers in institutions of higher education and becoming active in research. For the teacher educators to handle these challenges, significant adaptations were required affecting their previous identities as schoolteachers.
InFo-TED is an international network of researchers focusing on professional development (InFo-TED, 2019a). In their Conceptual Model of Teacher Educators ' Professional Development (InFo-TED, 2019b;Tack et al., 2015), identities are understood as influenced by the visions, boundaries, social changes, diversity and technological changes that affect the knowledge and practice of teacher educators. These components influence the teacher educators' perceptions of their personal visions, as well as their social and language competences. Moreover, these are influenced by the context of their current university and the teacher-education programme as a whole, as well as the local climate for networking and professional development. The national level, including national guidelines and frameworks, is also included in the model to contextualise the positions of teacher educators. Finally, the global level is recognised, including changes initiated by the European Commission and policy trends for teacher education programmes throughout the world.
The lower part of Figure 1 depicts the professional development throughout a teacher's career, which includes identity formation. Those just becoming teacher educators begin with a pre-initial phase. They then arrive at the initial phase before being subsequently defined as part of the induction phase. As understood in this model, new arrivals must go through an early-career phase before they can experience an identity as in-service teacher educators. The teacher educator has had to develop a self-understanding and an understanding of their work to experience themselves as being in-service. The final, lifelong phase is understood as being maintained throughout the rest of an educator's career.
Some studies highlighted the need to understand teacher educators' career-long professional development and identity in greater detail. Livingston (2014) argued that a more integrated and collaborative approach to teacher education is needed, based on a more thorough understanding of those who take up the role of teacher educator across their careers. She observed that, in many cases, participants in teacher education do not recognise themselves as -nor are recognised by those they work with asteacher educators. Drawing on an empirical study carried out with mentors in schools in Scotland, it is suggested that these teacher educators may be 'unrecognised' and remain 'hidden professionals' because of the identities they construct for themselves, including the values and priorities that they or others attach to their roles. Ping, Schellings and Beijaard (2018) have performed a systematic review of what, how and why teacher educators learn. They analysed 75 research articles and observed that research on teacher educators' professional learning appears to be a growing field of interest, but remains fragmented in focus. Among other observations, they concluded that teacher educators generally experience a need for continuous learning so they can perform their work. Another study by Tack, Valcke, Rots, Struyven and Vanderlinde (2018) collected evidence from 611 teacher educators in Belgium and used the concept of 'hidden professionals' . Their results confirmed the lack of attention to teacher educators' professional development in the Flemish system. Their analysis pointed to a structural need for participation in professional development activities more closely linked to teacher educators' own practices, organised as long-term sustainable professional-development trajectories and incorporated in formally-recognised systems (Tack et al., 2018). In relation to the teacher educators' identities, this study raised new questions about the connection between the interest-based groups and the reflections of the educators themselves.

The study
Teacher educators from two universities in Norway were asked, among other issues, about their own professional development. In addition, the study examines how different teacher educators, with a range of working experience, understand and express their professional identities. Thirteen research interviews and two focus-group interviews were undertaken with participants in the categories of campus-based teacher educators and kindergarten-based teacher educators. Kvale and Brinkman (2015, p. 42) state that the qualitative interview is effective in understanding aspects of informants' daily lives from their own perspectives. The interviews were conducted faceto-face; the duration of the sessions was between 40-75 minutes and included such questions as: 'When you study or teach professional theory as a teacher educator, in what ways do you think about yourself as being a part of a profession?' , 'How did you become socialised into the "profession" of teacher educator?' , 'How did you qualify for this position?' , 'Do you have space to be yourself and develop yourself in this function?' , and finally, 'Could you say something about your autonomy as a teacher educator?' .

Context and selection
Interviews were chosen as the data-collection instrument for the present study because of their focus on the informants' perceptions, understandings and experiences. The institutions, referred to as university A and university B, were chosen because they were based in different parts of the country, had essentially the same size student populations and were both unknown to me before conducting the interviews. I knew none of these teacher educators beforehand. All the informants work as teacher educators in the bachelor programme, Early Childhood Teacher Education. I decided in advance to challenge all teachers on one specific teaching subject (Children's development, play and learning). These teachers were contacted at both universities through the programme leaders. Everyone, thirteen from this sample, said they were willing to participate in the study. Before the interviews took place, all participants signed consent forms and were treated with fictitious names to maintain anonymity (Creswell, 2012).
Thirteen semi-structured research interviews were completed with participants who were either campus-based or kindergarten-based teacher educators. The pseudonyms given to the informants began with the letter A for those from university A and with B for those from university B. (For details, see Table 1). The informants for this study vary in their current positions and roles, their educational and disciplinary backgrounds, and their teaching experience in schools or in kindergartens. In Norway, a teacher educator should ideally hold a Ph.D. degree, but this is not always the case. One common practice has been for former kindergarten 1 or school teachers to enter teacher education after gaining some work experience.
1. In Norway, all children beyond the age of 1 are entitled to a place at a kindergarten, a pedagogical institution in which the classes must be led by kindergarten teachers from the Early Childhood Teacher Education Programmes (mandated to work with children from 0-6 years of age).

Analysis
After the interviews were transcribed, the data was analysed using Bryman's four stages of qualitative data analysis (Bryman, 2008): reading and re-reading, coding and recoding, categorising and re-categorising before constructing a hierarchical matrix of codes. The first two stages of the analysis therefore followed an inductive approach using a data-driven method. Stages three and four were more deductive stages following a theory-driven approach. During these iterative processes, I observed three main categories: (1) recruitment and socialisation, (2) professional identity, and (3) professional development, which will be presented and discussed sequentially. The rationale for these particular categories is that they emerged through the four phases of the analytical work as the most important. They represent terms that contain a multitude of data with one or more unifying aspect. A check-back was conducted to increase the trustworthiness of the study by sharing the results of the analysis with the participants. The study's findings will now be presented, and the interpretations of these findings will then be discussed in relation to some of the factors in InFo-TED's model ( Figure 1).

Findings
This section will describe the study's findings according to the three categories mentioned above; the first category to be discussed is recruitment to the position and the role as teacher educator. This includes the first phases in the profession of teacher educators and will be followed by the other categories: professional identity and professional development.

Recruitment and socialisation
The category of recruitment to the position of teacher educator contains the experiences of teacher educators in the sample concerning their first encounter with, and their entrance into, recruitment for a position in the university programme. This category also includes their socialisation into the faculty and their experience of meeting the expectations and content expected of the role. In this sample, four of the thirteen participants have worked for five years or less as teacher educators at the university level. Meanwhile, as many as seven of the informants have 15 years or more experience as teacher educators. Only one of the thirteen is in the middle phase, as Beate, with eight-years experience as a teacher educator, is defined as in the early-career phase. Table 2, inspired by InFo-TED's six phases (InFo-TED, 2019b), displays the data in a new way. The sample offers insight into a group of teacher educators belonging to various phases, defined as pre-initial, initial, induction, earlycareer, in-service and lifelong phases (For more details, see Table 2). The division into these six phases is influenced by the results found by Murray and Male (2005), who observed that employees needed between two and three years to establish new professional identities. In such cases, the informants were placed in the initial phase. In addition, Murray and Male (2005) showed that teacher educators often need two to three years before shifting their identities to align with their new roles. In the present study, one teacher educator in the initial phase is Anita from university A. During this period, she has struggled to find her own approach to handling work-related pressures while also being socialised into her teacher-educator team: About her pre-initial phase, we understand that Anita worked too hard, and has learned through reflecting on how to survive as an employee, not to be too engaged or too busy to listen to others. She does not yet have a permanent position, but now that she has found her way into the profession, she wants to remain. Only a few of the informants planned to become teacher educators years before they began in this role. It seems that an element of randomness prevails in entry to this profession: Many were recruited into their university positions in a number of ways. This aligns with the next interview transcript, where Bert reflects on the randomness of the reception and assistance given in the first semester. Bert says, ' And that's how it happened' . Although it was well-received and welcomed, only the practical preparations for teaching were facilitated: I have received good guidance all the way. I was very well received by both the programme leader and those who had this subject before me. And the profession? This is how I began learning: "Today you will have your first class, and it will last for 3 hours, this and that are the topics, and in that classroom (pointing to a door) are your students. " Then I was there. And that's how it happened.
(Bert: Physical Education, 5th year) Bert told something about his first foray into the profession. He came from a military background and had some teaching experience, but teacher education was a new domain that nobody explained to him. He had to spend time finding his place, discovering the goals of the education programme and so on. However, there is also a suggestion in the transcript of a tradition in the university: those who taught the same topic the previous year felt a responsibility to support the new employee. New instructors could inherit expertise, supporting documents and learning material from those who were in the same position the previous semester. Perhaps it was this culture Bert refers to as 'good guidance'? Meanwhile, Anita experienced a different process of recruitment. She began as a teacher educator by working together with a 'master' in the profession for several years: During the thirteen interviews, this was the only example given of co-teaching or working closely with a master in the field when entering the profession; the example itself took place more than thirty years ago and therefore belongs to another time and tradition.

Professional identity
In their interviews, participants offered explanations of their individual paths of socialisation into the profession and the importance of the culture within the programme to build collegial identities among teacher educators in Early Childhood Teacher Education. In institution A, for example, it is evident that programme leaders and instructors have built a discourse of 'Who are we? What are we in the process of becoming?' , seeing the identities of teacher educators as works in progress. Audrey, who had yearned for the same profession that her student teachers are going to become, gave an example: I never say to the student teachers: 'No, I'm not a kindergarten teacher' . I never say that. I present myself as a kindergarten teacher educator, similar to the others.
(Audrey, Physical Education, 16th year) Teacher educators have clearly worked through various aspects of their self-understanding. However, this awareness is not necessarily visible to others, though the evidence gathered in this study suggests that a strategic and robust process has been developed in one university in particular. At university A, teacher educators have worked out how to present themselves, who they ought to be in front of the student teachers, and who they are as a profession and as professionals. However, even here, some of the teacher educators retain their first career as the most conscious and readily-expressed part of their identities: As Arne reflected on his identity during the interview, he presented these observations as new thoughts in spite of being a part of a university A, which promotes a culture supporting the formation of identities as teacher educators. On the other hand, Bodil in university B has extensive experience in the university's teacher-education programmes, after a first career as a kindergarten teacher; she expressed her experiences in this way: The 'kindergarten teacher' role is the role and profession that her student teachers are qualifying for; the classes and education she offers for this profession seem to influence her understanding of her own profession as a teacher educator. Autonomy concerns the ways in which the informants experienced a space to be themselves and develop themselves as teacher educators. This theme emerged in different responses, including this one from Audrey: Yes, I really think so (that I have this space). We are obviously guided by the framework plans and study plans. I am given 'elbow room' in the teacher-educator team. (Audrey, Physical Education, 16th year) The kindergarten-based teacher educators, Agate and Bill, both explained their previous and current roles as a contrast over time. They felt more like teacher educators and part of a larger team of university educators with their earlier personal contract structure, which has been replaced by new contractual forms. The working contracts now connect the program to each specific kindergarten. With this type of contract, they feel that they have fewer rights and are less connected to their universities. Bill explained it as follows: I feel that way too (that I am a teacher educator), but just then there was a feeling that I belonged here. There were quite a few others who were there, and we met.
Yes, I really enjoyed that period. (Bill,25 th year) The external frames, here the national guidelines for contracts and the forms of association with colleagues, are important factors for Bill's understanding of his identity as a teacher educator. It seems that the framework plans function as frame-setters too, within which to develop. The importance of being heard and having a voice in the debate and development of the programme is something Berit commented on: I know that I have a large space [to express myself] and that I'm being heard, that I have great influence on the teams of teacher educators, both in terms of content and literature choices. (Berit,Pedagogy,5th year) What Berit says here represents various comments showing how the teacher educators experienced the freedom to choose their own methods. In addition, it is clear that they have arenas for discussion that work well, where the roles of teacher educators can be discussed and negotiated. Berit suggests that the framework for teacher education is not seen to be too restrictive or too strictly defined outside the teams of educators themselves. This can, of course, be interpreted as a strength. However, the downside might be seen if this space of action is large because the profession itself is invisible or hidden. Few are even aware of how this group of teachers goes about identifying and solving its tasks.
So, being a teacher educator, I have been one almost my entire working life. (Bodvar,Religion and Ethics,22nd year) Regarding these educators' experiences of autonomy in research, this was expressed as a complicated role to perceive, especially in the early years of teaching. However, to be given time and resources to develop academic texts together as a research team makes a difference: We are going to Krakow for a week, writing and sharing texts, and have designated time for doing research. (Bodil,Pedagogy,16th year) To produce articles for national and international publications is one of the main expectations of the Norwegian teacher educators and is understood as part of their professional development.

Professional development
The length of time the participants have spent as teacher educators seems to influence the process by which they form professional identities, whether weak or strong ones. At the same time, personal awareness and a kind of local pressure to continue their professional development as teacher educators seems to have influenced the paths of the informants. Anita and Adele both specialise in pedagogy, Anita at an initial stage, having only two years as a teacher educator, while Adele, with 35 years' experience, is in the sixth and final, 'lifelong' category (InFo-TED, 2019b): Now I'm on a qualifying course for teacher educators. I think I will get a lot of growth opportunities to see myself in different ways through this study. (Anita,Pedagogy,2nd year) Universities in Norway offer no specific training for teacher educators. However, after hiring, they are offered one or two courses nearly every year to follow continuing education programmes for teacher educators. The intention of this practice is to strengthen the content they offer and the quality of their work. Anita tells us that she will be able to reflect and gain additional awareness through her course.
Yes, we have a good routine now and have been through a good round of awareness-raising. (Arne,Drama,15th year) To create supportive structures for further development, university B has gone through significant discussions and made important changes. Ambitious timetables have been constructed, even though they may not be fulfilled: Our teacher meetings are the most important activity for further professional development. To be honest, we have had too few of these this autumn, because I have been so busy (…) People are running around everywhere. (Brita,Pedagogy,3rd year) As is true of other teacher educators in the programme, Brita has plenty to do. Still, it seems she is aware of ways to strengthen her and her colleagues' professional development. The teacher meetings are understood to be most important for this purpose. As a team leader, she would like to ensure a more regular schedule, less dependent on participants´ busy schedules. The role is also expressed as a kind of lonely, selfmotivated work where you feel you are responsible for setting quality standards: You work relatively alone, and I would not like to have been 25 and come in here, because then I would have been out soon!. (Anita,Pedagogy,2nd year) In addition to working in various programmes that were described as 'hit or miss, ' teacher educators also need time for the research portion of their employment. Bodil has several plans for research projects: Yes, it is planned as a research project because we will write an article, but it is a project too -because it looks at and probes more deeply into these changes.
(Bodil, Pedagogy, 16th year) These findings point to the different ways teacher educators experience socialisation processes in their profession, and differences in the way they enter into their new roles and functions. To meet the expectations of their roles, new hires described working over a number of years to identify their role and grow to be prepared for their various functions. Participants explained their individual social paths into the profession and the importance of the culture of the programme to build collegial identities as well as their own professional identities. Teacher educators have clearly worked through various aspects of their self-understanding; while this awareness may not necessarily be visible to others, the evidence gathered in this study suggests that a strategic and robust process is needed for teacher educators to participate in further stages of professional development. In the following section, these findings will be discussed in greater detail.

Discussion
The findings presented above will now be discussed in relation to the research question and the theoretical approaches set out earlier. Inspired by the sections presented in Figure 1, the professional identities of teacher educators will be examined in relation to (1) recruitment and socialisation, (2) the four levels of explanatory factors and finally (3) identity, as plural dynamic aspects will influence teacher educators' professional identities.

Recruitment and socialisation
A relationship may be seen between the career phases discussed in InFo-TED's conceptual model (InFo-TED, 2019b) and the findings of this study. The model divides a career into pre-initial, induction, early-career, in-service and lifelong phases. Few of the informants in this study were found to be in middle categories. Those recruited into the universities over the past three years were interpreted as belonging to the 'initial phase' . Teacher educators with 15 or 16 years of experience were defined as belonging in the 'in-service phase' and those with more than 20 years of experience were defined as belonging to the 'lifelong' phase. Meanwhile, entry into the role of teacher educator is difficult for several reasons. To begin in this job brings with it a new field of practice, new roles as researchers and new academic expectations, along with substantial environmental pressures (Elstad, 2010). These contrasting expectations arise beginning on the first day, even though it can take between three and five years before an identity as a teacher educator is built (Murray & Male, 2005). This means that the first years are very critical for the recruitment process and determine whether the teacher educator will want to continue in the profession. If Anita, who came from university A and was in her second year as teacher educator, was to encounter too many new or hidden expectations and found herself recruited into a profession with lower status than she had previously as a school or kindergarten teacher, other incentives to continue would have to be strong. Although some teacher educators had been teachers in schools or kindergartens, their previous experiences were not entirely relevant, since now they are required to offer support for the development of adult student teachers. In addition, they have joined a profession that is hidden, nearly invisible outside the university.
The present study has shown that the recruitment and socialisation processes are unlike the way teacher educators are welcomed or helped into their new roles and functions. To meet the expectations of their roles, new employees describe hard work over years identifying the role and growing into preparedness for it. Anita offers an example when she reflects about the relatively lonely work involved in fulfilling the role of teaching in higher education. Perhaps it is easier to become engaged in cooperative activities (Mead, 1934, p. 164) in the research part of the function of teacher educators than the more typical teaching parts of their role? What Anita describes may be an example of this dilemma.
Recruitment plays a key role of entry into a profession, but the teacher educators in the present study were recruited into universities largely by coincidence. None of the informants said that they had worked over time to become a teacher educator, or that they had specifically entered into a qualifying process for it. Although this understanding may not be generalisable, it is still worth reflecting on. If an academic career in higher education is a career path that the actors enter gradually, this will influence their developing professional identity. Since this profession is difficult to spot, several random factors are required to 'recruit' new teacher educators, who require even more time to find their place with the autonomy that is required to make this profession their own.

Professional identity
The relationship between the individuals and their environment, language and social interactions (Mead, 1934) are important for the growth of their identities as teacher educators. In this section, the environment is defined by personal, local, national and global levels which influence educators' professional development (InFo-TED, 2019b).
The individual level is close to the teacher educators themselves. Understanding self-perception as any perception, feeling, belief or knowledge a person has about himself (Skaalvik & Skaalvik, 1988, p. 13), the 'personal level' in Figure 1, is important. The findings of the study suggest that the teacher educators interviewed showed different perceptions regarding continuity concerning their own roles and positions. According to Akkerman and Meijer (2010), professional identities can be both continuous and discontinuous. For the teacher educators in this study, their emerging professional identity seems to be influenced by personal perceptions. Teacher educators have developed, performed and adopted professional identities throughout their lives. Individuals experience themselves indirectly (Mead, 1934, p. 138). Anita, in her initial phase, understands her own socialisation as a step-by-step process. Finally, after her first year surviving the pressure as her sole objective, she was able to listen to the others and co-operate effectively with the other teacher educators. Moreover, Bodvar speaks of his identity as something firm; this identity is not something he simply possesses, but something he has developed in the course of his whole life (Beijaard et al., 2003). Teacher educators generally experience an imperative to learn how to fulfil their new functions (Ping et al., 2018).
Arne defined himself as being in the 'in-service phase' of his carrier and still retains his identity as a drama teacher more than as a teacher educator. In his 'personal level' , he has an unconscious understanding of his role. It seems like he has been far from having any significant conversations concerning these questions (Mead, 1934, p. 165).
At the local level, the data emerging from the interviews are mainly oriented around the professional teams, programmes and the institutions in which the participants teach. The findings of this study reveal clearly how these teacher educators need each other. At university A, they are more conscious after years of working about how to present themselves in front of the student teachers, and who they are as a profession. What Audrey stated about these aspects is close to Mead's theory of how the self arises through language and social interaction (Mead, 1934, p. 138).
At the same time, tensions between different levels also emerge in this context, such as the introduction of new frameworks or new requirements for teacher educators mandated at a national level. The national level is used as a frame of reference when the framework for the teacher-education programme is discussed, or when informants talk about cross-border cooperation in professional development or collaborative research projects. One example is when two kindergarten-based teacher educators discussed the 'good old days, ' when they had an individual connection and appointment to a university. Both of these participants longed for the prior national model where their role as teacher educators was represented not only by an academic affiliation, but was also a formal appointment with more visible and permanent positions. The leaders of this programme pointed to how important it was for them to be affiliated with a national network, conducting meetings and sharing experiences. Finally, especially in the 'in-service' and 'lifelong' phases of their careers, teacher educators worked in research networks with researchers from several national universities.
In this study, informants situated themselves in relation to the global level, as in Bodil's satisfied observation that she is part of a research group travelling to Krakow. As in this case, the global context is mentioned especially when participants discuss their research and the national guidelines influenced by global viewpoints. However, trends in teacher education and new discourses may also be observed to impact how teacher educators look at themselves and their practices. In a way, it is possible to divide influences on the identity of the teacher educators into these four levels -personal, local, national and global -all of which are important in understanding the entire, complex processes by which professional identities are formed.

The dynamics of professional development
In the search for an identity, teacher educators encounter a complex situation while bearing several different identities, which in turn are contextually and culturally conditioned. This situation reflects Mead's theory of identity and the 'I' and 'me' function as a framework for understanding. Teacher educators in this sample have high grades and self-esteem. Most likely it is only Anita, one of the most experienced participants in the sample, who experienced an organised system (co-teaching) over many years. She has had the opportunity to concentrate on seeing herself as others see her by adopting the attitudes towards herself as a process for recruiting into the profession (Mead, 1934, p. 138). Usually, these processes go on interactively; teaching teams, meetings and cooperation areas represent the two phases of 'me' and 'I' , reflecting the generalised other and allowing other individuals the possibility to respond to these activities (Mead, 1934, pp. 174-175).
The evidence presented here shows that identity can, to some extent, be both static and flexible over the same period. At the same time, the development of identities is far from being a linear process (Akkerman & Meijer, 2010). The data shows various forms and a wide variation of understandings of how to strengthen the identities of teacher educators at the local institutional level at both universities A and B. It seems that these differences have influenced the understanding and strength of identification of teacher educators at the individual level. An active and conscious relationship to their identities as teacher educators, in social groups and in teams, seems to support the emergence of new individual professional identities. The present study found a correlation between the length of time the teacher educators have been in their positions and the strength of their identities as teacher educators. At the same time, teacher educators have quite varied understandings of their professional identities (Lunenberg, Dengerink, & Korthagen, 2014;Smith, 2011). Some maintain an identity rooted in the discipline they were educated for, as a Drama teacher or a Physical Education teacher. At one of the institutions selected, the leadership of the programme has encouraged the formation by faculty members of new identities as teacher educators as a supplement to their first careers. Meanwhile, the possibility of wearing multiple identities -that several identities can coexist at the same time (Livingston, 2018) -has also been supported in various ways. The transformation of identity seems to depend on maintaining a link between the legitimisation of present activity and the reinterpretation of past biography (Berger & Luckmann, 1966). Murray and Male (2005) have affected this study's assumption that the employee's process can take three to five years before assuming an identity based on the functions of the new job. Here it is important for teacher educators to be offered opportunities to be themselves and to develop in their new function. This autonomy was described in several ways; overall, informants did not experience restrictive requirements for the role. Rather, this study found framework conditions that provided opportunities for new teacher educators.
The question arose whether teacher education is a hidden profession facing fewer expectations from the outside, aside from delivering research results. The present study also found that teacher educators feel that their profession is hidden; one reason for this might be that teachers in higher education are often largely unaware of their own construction of professional identities. Another finding is that some of the teacher educators appear to feel more invisible now than they had previously been as teachers in schools and kindergartens.

Implications and further research
The experiences and understandings presented by the informants in the present study are not easily transferable to other contexts; contextual factors would differ even more based on different countries' educational frameworks. Still, the study can be used to inform the ways universities recruit and support new teacher educators. In addition, it seems necessary for policy makers to attain a stronger understanding and appreciation of this profession. The identities formed by teacher educators depend on their implied status and the expectations of professionalism and quality within their practice. Research on teacher educators' professional learning to date has been fragmented in its focus (Ping et al., 2018). Beauchamp and Thomas (2008) state that we must try to incorporate what we know about the contexts and communities and their influence on the shaping of teacher identities into our teacher education programmes to prepare new teachers for continuing to build strong professional identities in positive ways. One vision for further international research would be to ask additional questions about teacher educators' experiences of professional development and their self-understanding of their identities. This study points to the importance of teacher educators as bearers of these key roles and responsibilities; it may be time for them to become more visible as a profession.