Five ways of being a schoolteacher

This exploratory study consists of a description of the five schoolteachers’ identities that showed five different configurations of teacher’s I-positions. Thirty-one participants were interviewed via a written survey, and data were analysed using qualitative and quantitative procedures. The findings revealed thirty I-positions held by the teachers, classified into nine broader institutional positions. Additionally, five broad categories of “ways of being” a school teacher were identified and labelled as follows: (1) teachers engaged mainly in instruction and collaboration with colleagues; (2) teachers engaged mainly in instruction and building a relationship with families; (3) teachers engaged mainly in instruction, improving educational practice, and collaboration with colleagues; (4) teachers engaged mainly in education and instruction, improving educational practice, collaborating with colleagues, and building a relationship with families; and (5) teachers engaged mainly in instruction, improving educational practice, collaboration with colleagues, building a relationship with families, and collaboration in the management of the school.

In the last fifteen years, a considerable number of studies have attempted to define the concept of teachers' identities. Douwe Beijaard, Paulien Meijer and Nico Verloop (2004) sought to address the absence of a clear definition of teaching identity by setting out four features of this notion. First, identity is a continuous process. Therefore, it is more dynamic than stable and is continuously evolving. Second, it is both individual and collective, as the notion of identity suggests a 'kind of person' within a particular context. Third, a teacher's professional identity includes sub-identities. These sub-identities may be more or less central to the general identity, and there must be a balance among them to avoid conflicts. Fourth, professional identity includes the active search for professional development and learning as part of teachers' pursuit of their professional objectives.
Other efforts to sketch the concept have been made by researchers such as Catherine Beauchamp and Lynn Thomas (2009) who have argued that teachers' identities are dynamic and that they shift over time under the influence of a range of factors. Some of these factors, such as emotion (Rodgers & Scott, 2008;Van Veen & Sleegers, 2006), are internal to a given individual, while others are external, as is the case of a person's work and life experiences in particular contexts (Flores & Day, 2006;Sachs, 2005). Elsewhere, Claudia Lenuta Rus, Anca Reluca Tom, Oana Luiza Rebega and Livia Apostol (2013) maintained that teachers' identities involve a combination of competing interactions between personal, professional and situational factors.
According to Judyth Sachs (2005), the teacher's professional identity form the cornerstone of the teaching profession. This identity represents the foundation upon which the teacher constructs his/her ideas on how to be, how to act, and how to understand his/her work. The resulting identities are shaped via a process of negotiation between professional experiences and the meaning attached to these experiences.
Our exploratory study consists of a description of five kinds of identities that school teachers assume in schools. Each kind of identity is composed of a specific configuration of teachers' I-positions, and each one represents a specific "way of being" a teacher in a school. More profound knowledge of some relevant individual differences among school teachers will be useful to educational policymakers, teacher trainers and leaders of educational change.

IDENTITIES
School teachers' identities will be conceptualised here via a self-positioning process whereby teachers express their professional identities in terms of the set of purposes or aims that they pursue in their current professional teaching activities (Vanassche & Kelchtermans, 2014). Three interrelated dimensions will be used to characterise these identities. The study will measure each participant's teacher's I-position, teacher's position, and configuration of the teacher's I-positions.
The term teacher's I-position comes from the dialogical self theory (Akkerman & Meijer, 2011;Meijers & Hermans, 2018). A teacher's I-position is a particular manifestation of the teacher's identity, and it shows a specific way of pursuing and achieving a particular educational aim. Each teacher's I-position is influenced by the subjective, personal and individual teacher's assets (made up of an individual's knowledge, conceptions, beliefs, skills, attitudes and other factors). This concept also includes a delimited set of tasks that serve as a means of achieving each educational purpose in real institutional scenarios.
The term teacher's position refers to a particular professional teacher's function carried out at a particular educational institution. Because we define the term teacher's position as synonymous of teacher's status function (Searle, 2010), the features of any teacher's position are not entirely determined by an educational administration or established by the legal standards governing his or her role. A teacher's position must be both personally assumed by the teacher and accepted and recognised by the institution and the educational community. A teacher's position will be performed by a teacher when he is carrying out his or her professional activity to achieve different purposes, according to his or her status in the institution .
Although there are some similarities between the terms teacher's roles and teacher's positions, they are defined differently in educational research. Data on teachers' roles tend to be collected via instruments that gather data on teacher's perceptions of the actions or responsibilities they carry out in practice and the set of tasks that teachers perform (Valli & Buese, 2007). In contrast, data on teacher's positions are gathered via tools that ask about the educational functions that they carry out in their institution and the goals or intentions that they want to achieve in their schools (Akkerman & Meijer, 2011;Searle, 2010).
Finally, we define the term configuration of teacher's I-positions as the set of interrelated positions and I-positions, which may characterise the identity of a teacher or a set of teachers with strong identity similarities. According to Hubert Hermans and Thorsten Gieser (2012), human identity is a theoretical construct consisting of different I-positions. Each I-position is expressed by an individual voice. In the field of education and teaching, Sanne Akkerman and Paulien Meijer (2011) have defined the teacher's dialogical self similarly, as "composed of multiple I-positions in the landscape of the human mind" (p. 311). A teacher's personal space of negotiation, populated by his or her Ipositions, allows the teacher to understand and explain him-or herself and the world.
The configuration of positions and I-positions of a particular teacher includes both the interrelated positions assumed by the teacher and the set of Ipositions included in each position. Each of these I-positions does not work in isolation. They cooperate and form groups of I-positions. A given teacher's dominant group of I-positions is called the teacher's core I-positions. The particular way in which a teacher's core I-positions combine determines how this teacher expresses his or her professional identity (Hermans & Hermans-Konopka, 2010) in a particular institutional scenario. Changes in teachers' core I-positions may occur when they have boundary experiences (Assen, Koops, Meijers, Otting, & Poell, 2018), or when they analyse the range of tensions that appear in teaching practice (Alsup, 2006). This study is part of a larger research project entitled "The identity of teachers and student-teachers". We report here the preliminary results of this project which show the professional identity of schoolteachers who were not principals in their schools at that time. In particular, we focus on the following two research questions: 1) What different I-positions held by teachers emerge from listening to and gathering together teachers' voices?; 2) Are there differences among the teachers' I-positions, and what clusters of kinds of teachers emerge in light of these differences?

METHOD
We adopted here the inquiry method called "analysis of multivoicedness" (Aveling, Gillespie, & Cornish, 2015). The method is used to analyse qualitative data and is informed by the tradition of dialogism. According to this tradition, the self is viewed as multivoiced because it can speak from a multiplicity of different I-positions, each of them with its distinct voice. Furthermore, a person-oriented approach was taken to analyse the qualitative data findings  (Bergman & Trost, 2006), making it possible to identify clusters of teachers with common I-positions.

Participants
The participants were 31 schoolteachers who responded to the survey, all of them working in the public education system at schools in Catalonia (Spain). All the participants were enrolled in a master's program devoted to increasing the quality of their educational practice. None of the teachers was serving as a school principal at the time of the study. After the participants had been informed of the aims of the study and invited to participate, 31 teachers voluntarily responded to the survey (36% of the total possible sample). Table 1 summarises their general characteristics.
As shown in Table 1, 77.4% of the participants were female, 61.3% of the participants were between 40 and 49 years of age, and all of them had more than ten years of teaching experience. Approximately one-quarter of the teachers had a master's degree (32.2%), while 76% of them had received more than 400 hours of in-service training courses. More than half of the teachers (64.5%) had specialised in primary education, and rest had specialised in pre-school education (35.5%).

Data collection
Data were collected using a survey named School Teacher Identity, the result of an adaptation of two instruments created by Doug Hamman, Kevin Gosselin, Jacqueline Romano and Rommel Bunuan (2010) and Rudy Vandamme (2014), but we focused on collecting data on the range of schoolteachers' voices that emerge from each I-position. The survey took the form of a written, structured interview with two sections. Written communication tends to be both more complex and more explicit than oral communication and provides time for participant reflection (Keats, 2009).
The first section included items that gathered participants' personal information, academic background and professional experience. In the second section, we collected data concerning I-positions. The open-ended questions required each participant to describe seven to ten I-positions that schoolteacher regularly performs in school given his or her status as a teacher. The beginning of this section provided a clear definition of the term teacher's position (a specific function carried out by a teacher at his or her school, defined by a teaching objective). Then, we asked participants to provide detailed information on three questions related to each I-position: a) the name of the position; b) the associated purpose of the function, that is, what the teacher wants to achieve in implementing this position; and c) a typical teaching task that is linked to this position and that exemplifies how he or she acts in order to achieve this purpose. Participants were required to answer items b) and c) using a minimum of 50 words for each.
One of the authors was responsible for the data collection procedure. He was the master's degree coordinator at the time of the data collection process, but he did not act as a teacher trainer within the program. Acting as a researcher, he requested the teachers' participation in the study as an independent task not related to their studies. The data collection procedure was carried out via email in May-June 2017. The email included the link to access the survey, designed using Google Forms and stored on Google Drive. The entire process of completing the survey usually took approximately 45-60 minutes. Participants were given two weeks to complete the survey and send it back. After the allotted time, two reminder emails were sent to participants who had not yet responded to the survey.

Data analysis
The software MAXQDA2018 (https://www.maxqda.com/) was used to carry out the complete data analysis and categorisation process based on multivoicedness (Aveling et al., 2015). We created 31 documents, one for every participant. Each document included individual socio-professional data and all the text written by that participant.
A deductive-inductive content categorisation analysis was carried out following the principles of grounded theory (Bryant & Charmaz, 2007). The thematic unit was applied as the unit of analysis to maintain the meaning of each written textual fragment used by participants to describe every I-position. Each thematic unit included the name given to of the I-position in question, the description of the associated personal purpose and the example of how the teacher would typically act to achieve this purpose.
The process of categorising each thematic unit began with identifying the voice of each I-position held by the teachers and assigning a proper name to each thematic unit, all in light of the three pieces of information provided. The iterative process of categorising data consisted of two steps. First, we categorised each thematic unit by assigning a name that describes a specific participant's institutional position at his or her school. Then, we constructed a second-order categorisation which consisted of re-categorising each thematic unit, assigning another name to of them. This new name was related to the particular manner in which the participant performs this institutional position. Following this process, each thematic unit was given two labels, one corresponding to the position itself, and the other corresponding to the individual's way of acting in this position. The combination of these two labels served to name each I-position.
The description of each thematic unit matched one single I-position in most cases (over 95%). In order to accurately reflect the meanings of each thematic unit voiced by teachers, we split some thematic units into two different Ipositions, and when we identified two redundant thematic units voiced by a single participant, we counted just one I-position.
A total of 278 I-positions were identified, with an average of 8.97 per participant. The length of the written text for each I-position ranged from 60 to 110 words. Table 2 summarises the categorisation. To assess the degree of interrater agreement, two independent analysts reviewed a random sample of 20% of all thematic units. The Cohen's kappa values were 0.80 and 0.89.
In order to determine the set of I-positions that schoolteachers adopt in teaching at school, a descriptive statistical analysis of the frequencies (N) and percentages (%) of the prevalence of the thematic units was conducted. Table 2 shows the frequencies (N) and percentages (%) of the total number of teachers in each I-position. To addressing the research question 2, a hierarchical cluster analysis (HCA) was used to classify participants using Ward's method. The data used were the total number of I-positions that each participant mentioned within each of the nine teacher positions. The final number of clusters was selected based on the mean of a silhouette plot (see Figure 1), the predictive validity of the clustering variables, and the interpretability of the cluster solutions.

RQ1:
What different I-positions held by teachers emerge from listening to and gathering together teachers' voices?
Findings show thirty types of teachers' I-positions, which reflect a range of contemporary ways of being a teacher at their schools. Below, we present a selected quote from a teacher for each of the thirty I-positions, broken down into their corresponding institutional positions. The seven I-positions included in the third institutional position have in common the teacher's purpose of supporting and assessing learning, and we have named them as follows: to support and assess learning by using preassessment instruments, by issuing learning assignments, by guiding the learning process, by motivating and encouraging student learning, by personalising learning, by using multiple methods of learning assessment, and by grading students.
03 To support and assess learning 031 By using preassessment instruments Taking into account a learning preassessment conducted of their students and lending support to the individual learning needs of their students. On some occasions, they can conduct an initial assessment of every student. To develop assessment mechanisms for learning. Learning assessment has to provide the teacher with information about the students' learning process. The teacher has to use rubrics and self-and co-assessment instruments and students need to know the learning assessment criteria.    Table 2 shows the number and percentage of teachers that assume each I-position within each institutional position.
RQ2: Are there differences among the teachers' I-positions, and what clusters of kinds of teachers emerge in light of these differences?
The results of the hierarchical cluster analysis are shown in Figure 1. This five-cluster solution divides participants into five groups: Cluster 1 with 7 participants, cluster 2 with 5 participants, cluster 3 with 9 participants, cluster 4 with 5 participants, and cluster 5 with 5 participants.
Cluster 1 was labelled Teachers mainly engaged in instruction and collaboration with colleagues. Figure 2 shows that most teachers assume at least one I-position from among the institutional positions involved with planning instruction (100%), supporting and assessing learning (71%), and collaborating with colleagues (71%).
Besides, a large percentage of the teachers assume the I-positions called to plan instruction by creating learning assignments (100%); to support and assess learning, both by issuing learning assignments (57%) and by using multiple methods in their learning assessment (57%); and to build a relationship with families by establishing communication and collaboration (57%). Also, a large number of teachers assume the I-positions to plan instruction by creating learning assignments (80%); to support and assess learning, by guiding the learning process (80%), by motivating and encouraging student learning (100%), by personalising learning (60%) and by using mul-  Cluster 3 was labelled Teachers mainly engaged in instruction, improve educational practice, and collaborate with colleagues. Figure 4 shows that most teachers assume at least one I-position from among the institutional positions related to planning instruction (100%), supporting and assessing learning (100%), improving educational practice (100%), and collaborating with colleagues (78%).

Cluster 2 was labelled Teachers mainly engaged in instruction and building
A large percentage of teachers also assume the I-positions called to plan instruction by creating learning assignments (100%), by creating instructional materials (89%) and by designing learning environments (67%); to support and assess learning by guiding the learning process (56%), by personalising learning (56%) and by using multiple methods of learning assessment (78%); to improve educational practice by reflecting on their educational practice (67%) and by attending training courses (100%), and to collaborate with colleagues by participating in the school's decision-making processes (56%). Cluster 4 was labelled Teachers mainly engaged in education and instruction, improving educational practice, collaborating with colleagues, and building a relationship with families. Figure 5 shows that most teachers assume at least one I-position from among the institutional positions related to educating children (80%), supporting and assessing learning (80%), improving educational practice (80%), collaborating with colleagues (100%), and building a relationship with families (100%).
There are also a large number of teachers who assume the I-positions called to plan instruction by creating learning assignments (100%); to support and assess learning by guiding the learning process (60%); to improve educational practice by reflecting on their educational practice (60%) and by attending training courses (80%); to collaborate with colleagues by building a community-shared vision of education (100%) and by participating in the school's decision-making processes (60%), to build a relationship with families by establishing communication and collaboration (100%), and to collaborate with external professionals by establishing shared actions (60%).
Cluster 5 was labelled Teachers mainly engaged in instruction, improving educational practice, collaborating with colleagues, building a relationship with Figure 6. I-positions included in teacher cluster five families, and collaborating in the management of the school. Figure 6 shows that most teachers assume at least one I-position from among the institutional positions related to planning instruction (100%), supporting and assessing learning (100%), improving educational practice (100%), collaborating with colleagues (80%), building a relationship with families (80%), and managing the school (80%).
Besides, several teachers assume the I-positions called to plan instruction by creating learning assignments (80%) and by creating instructional materials (60%); to support and assess learning by guiding the learning process (100%), by motivating and encouraging student learning (60%), by personalising learning (80%), and by using multiple methods of learning assessment (100%); to improve educational practice by reflecting on their educational practice (80%) and by attending training courses (100%); to collaborate with colleagues by building a community-shared vision of education (60%), and to manage the school by serving as a member of the school managerial team (60%).

CONCLUSIONS
This study aimed to describe schoolteachers' identities, who were not principals in their schools at that time, by identifying five types of configurations of teachers' I-positions. To this end, we have developed a conceptual framework based on the theoretical construct of the teacher's position and I-position (Assen et al., 2018), and we have sought to define and describe teachers' identities by posing two research questions: 1) What different I-positions held by teachers emerge from listening to and gathering together teachers' voices?; 2) Are there differences among the teachers' I-positions, and what clusters of kinds of teachers emerge in light of these differences?
Regarding the first question, we identified thirty I-Positions held by the teachers in the study. Each of these teaching I-positions is the result of the process teachers engage in to give meaning to their professional activity (Linell, 2009). The voice that a given teacher uses to characterise each I-position is contextually constituted via the social interactions this teacher has with other teachers and with his or her social surroundings. This voice reflects an individual teacher's subjective way of thinking and acting concerning every institutional position (Hermans, 2015;Searle, 2010).
Regarding the second question, we identified five clusters of teachers, each of them characterised by the presence of certain dominant I-positions and the absence of other I-positions (Hermans & Hermans-Konopka, 2010). Each of these five "different ways one can be a teacher" (Vandamme, 2018) reflects a clear snapshot of a kind of a teaching self, "composed of multiple I-positions in the landscape of the human mind" (Akkerman & Meijer, 2011;p. 311). These different categories also consider the meaning of a teacher's identity, defined as "an ongoing process of negotiating and interrelating multiple I-positions in such a way that a more or less coherent and consistent sense of self is maintained throughout various participations and self-investments in one's (working) life" (Akkerman & Meijer, 2011, pp. 318-319).
This characterisation of teachers' I-Positions as components of their identities as teachers should be considered an alternative to the current ways of describing and classifying teachers' professional activities based on describing teachers' perceptions of the roles they perform in their daily activity (e.g., Valli & Buese, 2007). Whereas these previously existing classifications reflect teachers' perceptions of the tasks they perform, the use of the concept I-position allows us to identify a large number of real teachers' I-positions as they are manifested in their institutional positions at a particular time and place, as well as to identify the core I-positions of five clusters of teachers.
The study has three main limitations. First, findings do not show the dynamic and changing nature of the teachers' identity, and they may only be considered a photo finish of the teachers' identity at a given time of the teachers' professional development. Second, the sample is small, and the results of the study may have been affected by the particular characteristics of the participants. This last statement is particularly crucial in the case of trying to identify a teacher's existing I-positions. Third, we are aware that the data collected are based on individual written contributions, and as such, the study only reflects the list of teacher I-positions the participants have voluntarily reported.
This exploratory study should be considered the first step in this line of research, and, consequently, more research is needed in order to arrive at a more in-depth and detailed characterisation of teachers' identities. Three different paths for future research are worth highlighting. The first would consist of increasing the number and diversity of participants. The second would consist of using a new data collection tool which allows us to establish clear connections among the I-positions held by a teacher. An example of such a tool would be the elicitation cards technique (Monereo, 2019). Finally, the third would be to obtain data from daily teaching practice. This last possibility would allow us to use the triangulation technique to analyse data.
All these findings taken together could be useful in that they could encourage and facilitate further efforts by designers of teacher education programmes and by teachers themselves. Our findings could be of particular interest to designers of teacher education programmes who may wish to use new approaches to the planning of programmes and courses for pre-service and in-service training, based on the set of I-positions teachers will need to be able to adopt in real classrooms. Besides, our findings would also be useful for teachers who want to identify their predominant I-positions at a certain point in their professional careers.

Funding:
The author(s) received no specific funding for this work.
Disclosure statement: No potential conflict of interest is reported by the authors.
Statements on open data, ethics, and conflicts of interest: This research paper was developed following the ethical standards of the American Psychological Association (2010). Participants were informed in advance of the general aim of the research, its duration, and the procedure to collect, store, and analyse the information provided by them, and they read and signed an informed consent form. Following this notification, participants freely decided to answer the online survey. Data collected have been stored and managed by observing the law on data protection and the right to confidentiality.