How Finnish experienced teachers learn to improve their interactions with pupils

(1)

How Finnish experienced teachers learn to improve their interactions with pupils

Lais Oliveira Leite

University of Eastern Finland Philosophical Faculty

School of Applied Educational Science and Teacher Education Master’s Degree Programme in Educational Sciences

Master’s thesis in Primary Education March 2018

(2)

ABSTRACT

Tiedekunta – Faculty Philosophical Faculty

Osasto – School

School of Applied Educational Science and Teacher Education Tekijät – Author Lais Oliveira Leite

Työn nimi – Title

How Finnish experienced teachers learn to improve their interactions with pupils Pääaine – Main subject

Teacher-student relationship

Työn laji – Level Päivämäärä – Date 11.03.18

Sivumäärä – Number of pages 85

Pro gradu -tutkielma X Sivuainetutkielma

Kandidaatin tutkielma Aineopintojen tutkielma Tiivistelmä – Abstract

This research aimed to investigate how Finnish experienced teachers reflect about their informal learning process to build positive interactions with pupils. Two questions were addressed: how teachers perceive that their professional agency support their learning process to build positive interactions with pupils and how they apprehend the general character of this learning process. To conceptualize teacher- pupils interaction, the “Teaching through interactions” model and its three main domains (emotional support, classroom organization, and instructional support) were used. The process of learning by teachers is theoretically based on the “Teacher professional agency”

construct and its three main elements (motivation to learn, efficacy beliefs to learn, and active learning strategies). This investigation was complemented with three teacher personality factors (work tool, lifelong learner, and oppositional factor to change) that emerged from the empirical evidence. They showed to be relevant to teacher-pupils interactions and teachers´ learning process. The research design adopted in this study was a confirmatory multiple-case analysis with a mixed method approach. A survey with 48 primary school teachers, who had at least 5 years of experience, was accomplished. It contained the scales Teacher professional agency and Boundary crossing. Descriptive statistics, reliability and non-parametric correlation tests were run over the collected data. The teachers with the highest scores of the scales and time availability were selected to participate in the in-depth semi-structure interviews. The participants´ reports were analyzed using theory guided and exploratory content analysis. For the first inquiry of this investigation, the findings revealed three groups of multidimensional factors: 1) professional agency factors that support teacher´s learning; 2) active learning strategies used by teachers to build positive interactions with pupils, and 3) personality factors that base teacher-pupils interactions. In the first group, the teachers´

motivation and efficacy beliefs in learning strongly supported their learning process regarding building better interactions with pupils. In the second one, the learning strategies showed a broad range of complexity. Reflecting about classroom situations and pupils´ feedback were very frequent. Boundary crossing appeared to happen across different settings and with different school actors. Learning by doing was the most relevant strategy, with emphasis on teachers´ developing sensitivity for pupils and building up positive climate. In the last group, although all these learning mechanisms were reported, teachers perceived that the fundamental base for their relationship with pupils is their personality, which guarantees the authenticity of the interactions. For the second inquiry, the teachers apprehended their learning process in two ways. First, they believe to still actively learn about classroom situations. Second, they perceive that their learning process does not have an impact of changing their interactions with pupils, but rather of general professional growth. The present study contributed to fill a gap in the literature and raise awareness about the importance of teachers’ informal learning in their everyday classroom situations, specifically with regard to build interactions with pupils.

Avainsanat – Keywords

Teacher´s professional agency; Teacher-student interaction; Teacher´s informal learning.

(3)

List of Tables

Table 1. Classroom Assessment Scoring System (CLASS) framework 13

Table 2. Survey participants profile by grade and gender 21

Table 3. Interviews’ general characteristics and participants´ profile 24

Table 4. Categories, themes and codes of the content analysis 26

Table 5. Means, standard deviations and intercorrelations among all subscales and teachers’ years of experience 30 Table 6. Interviews’ participants profile by classroom atmosphere and keyword 31

Table 7. Overview of codes´ frequency in the text corpus derived from the interviews 32

(6)

List of Figures

Figure 1. Hierarchical relationship between the codes derived from the “Teacher professional agency” concepts, “Teaching through interaction” model and “Teacher personality” factors 28 Figure 2. Hierarchical relationship between the codes, grouped in colors by the role

they play in the informal learning process of teachers 33

(7)

1 INTRODUCTION

Previous research has already showed that emotional and cognitive engagement of students in primary school are highly socially embedded, both being mediated by peers and teacher- student relationships, which in turn contribute to students perceived well-being (Pietarinen, Soini & Pyhältö, 2014).

In general, positive teacher-child relationship – the focus of the present work – can be characterized by respectful communication, supportiveness, sensitivity and responsiveness, focus and interest on teaching-learning content, and low level of disruptive behavior or conflicts (Jennings & Greenberg, 2009). This type of relationship has shown evidence of increasing academic achievement, self-esteem, active exploration of school environment, and engagement in peer interactions. On the other hand, good teacher-pupil relationship has also been reported to decrease social problems, such as disturbing behavior and aggression in classroom, besides peer rejection. Complementarily, it also compensates mal-adaptive family relationships (Hamre & Pianta, 2001).

As a consequence of positive teacher-student relationship, students become more secure, confident, and open themselves to participate in class and make mistakes (Jennings &

Greenberg, 2009). By experiencing teacher´s support, pupils feel to belong in the classroom, which reinforces their engagement in learning activities (Pietarinen et al., 2014). Thereafter, in a cognitive level, they recall new learned information easier (Titsworth, McKenna, Mazer

& Quinlan, 2013). On the other hand, when pupils experience negative emotions with teachers and the interaction is marked by conflict, they show behavioral and academic problems that can be tracked years later (Hamre & Pianta, 2001; 2005). It has been shown that they tend to block the lessons and refuse to engage in the learning process (Titsworth et al., 2013), as well as direct themselves to anti-social behaviors and a general disengagement to school (Jennings & Greenberg, 2009).

From teachers´ perspective, pleasant classroom climate reinforces enjoyment and self- efficacy feelings, increasing enthusiasm and commitment to the profession, thus creating a positive feedback loop that may prevent teacher drop out from schools (Dicke, Elling, Schmeck & Leutner, 2015). Moreover, previous research on teachers’ pedagogical well- being also showed that interacting with pupils consists of a core task of primary teachers’

work, which can happen in both empowering and stressful contexts, affecting teachers’ job satisfaction (Soini, Pyhältö & Pietarinen, 2010). It has been also reported that pupils take their teachers’ socio-emotional competences as a model to develop their own, proving the

(8)

necessity of teachers to have a deep comprehension of how relationship-building behaviors impact the learning and teaching processes (Waajid, Garner & Owen, 2013).

Although all these empirical-based evidence regarding the importance of positive teacher- student relationship, significant research about how experienced teachers learn to improve their relationship with pupils was not found. There is, indeed, recent work showing that different domains of teaching practice and teacher´s competences can be improved through professional development experiences, such as in-service programs, courses and workshops (Brown, Jones, LaRusso & Aber, 2010; Domitrovich et al., 2009). However, only few studies (Hoekstra, Beijaard, Brekelmans & Korthagen, 2007; Soini et al., 2010) have shown, from the teachers` lived narratives, their continuous informal professional development.

Hence, there is no expressive descriptive material reporting teachers’ thoughts, beliefs, the process of change or reinforcement in their skills, insights, and experiences that provided them with the necessary conditions to be able to build positive relationships with their pupils.

Therefore, this research attempted to fill this gap in the literature by investigating how Finnish experienced primary teachers reflect about their informal learning process (Jarvis, 2004) to build positive interactions with pupils throughout their career. To support conceptually the interaction process, the “Teaching through Interactions” model (Hamre et al., 2013) and its three main domains of emotional support, classroom organization, and instructional support are used to understand different aspects of teacher-student interactions.

This model was chosen because of its emphasis on the relational aspect of teaching and learning processes. As reference to the concept of learning by teachers, the “Teacher Professional Agency” construct (Soini, Pietarinen & Pyhältö, 2016) and its three main elements of motivation to learn, efficacy beliefs of learning, and active learning strategies are used. These concepts accordingly consider the informal nature of teachers’ learning process through their professional experiences, as long as they have an active attitude towards this goal.

Finally, it is important to consider that the term "teacher-student relationship" is a broad concept in the literature that can represent diverse and complex behavioral, socio-emotional, and cognitive processes beyond the scope of this study. Hence, the present research uses one principal term to capture this complex process – “teacher-pupils interactions” – and account mainly (but not exclusively) the behavioral aspect of it. However, the reader will find both terms “relationship” and “interaction”, as well as “students” and “pupils” being used with the same meaning along this work, because of common use in the literature.

(9)

2 SHAPING TEACHER-STUDENT INTERACTIONS

2.1 Recent paradigms of teacher-student relationship and classroom management

For many teachers, cultivating a positive teacher-student relationship and, consequently, a pleasant classroom environment are still somewhat uncertain and it is not clear how to establish it, although many studies have already been undertaken since the beginning of the 20^th century under the heading of classroom management. (Brophy, 2006.)

The history of the research on classroom management starts with initial treatises made by common sense and aphorisms, such as “don’t smile before Christmas”. An illustrative example is the study developed by Bagley, which – although the offensive language approaching children as “savages that needed to be civilized” – already contained principles of management still recognized nowadays, such as trying to keep pupils focused on activities or avoiding punishment and competitive prizes, because they increase jealousy and frustration. (Brophy, 2006.)

In the following decades, research became empirical and started to focus on influence techniques, leadership style and group climate, investigating the consequences of punishments or praises on children’s behaviors. Worthy of mentioning is the work by Lewin and colleagues that classified three leadership styles: authoritarian, democratic and laissez- faire. According to this study, pupils had more time engaged on task and were more responsive and productive under the head of democratic teachers. Already from that period, the general conclusions oriented educators to favor positive-reward techniques rather than punishment-oriented ones. (Brophy, 2006.)

Then, educational and psychological researchers started to develop specific programs about classroom management. A major part of this initiative came from behavioral researchers, who developed experiments based on contingencies and reinforcement principles, mainly related to shape students’ behaviors. With time, the behavioral research changed to compound a package of procedures that should increase or decrease specific student sbehaviors. (Brophy, 2006.)

At the same time, an ecological approach in classroom management was fostered during the 1970’s, when USA government supported and funded studies on “process-product” theory, which assumed that teacher’s behaviors (“process”) affect student learning outcomes (“product”). This program tried to figure out what were the common management attitudes

(10)

and decision making process of teachers whose students were engaged with the studies and had better academic performance. (Weinstein, 1999.)

From this period on, research trends tried to demystify the idea that impacts derived from negative individual and family backgrounds could not be changed by school structural characteristics and organizational climate (Hultin et al., 2018). Currently, many researchers are engaged with the field, bringing traditional qualitative studies to the arena together with more recent quantitative methodologies. Hamre and Pianta (2001), for instance, contributed with this research trajectory emphasizing the classic Bowby’s attachment theory, according to which relationships with caregivers marked by trust and caring promote the development of internalized working models in children, who in turn feel secure to explore the world and take risks to learn.

Important research approaches have been developed in the first two decades of the XXI century (Hultin et al., 2018). Worthy to be mentioned, the Chilean educator Juan Casassus (2009) found in his research that the factor, which most influences and facilitates learning regardless of methodology, is the existence of a pleasant emotional climate in the classroom – directly related to the quality and the intensity of the interactions between student-student and teacher-student. The author found that schools, where students relate well with peers and teachers, present better academic rates than schools, where students have more relationship problems.

The research-based Collaborative for Academic, Social and Emotional Learning (CASEL) developed in 2008 five major emotional, cognitive and behavioral concepts that compound social and emotional competences: self-awareness, social awareness, responsible decision making, self-management, and relationship management. It is recommended that teachers and students learn these competences, with the support of school, throughout school period (Jennings & Greenberg, 2009). This progress in theoretical conceptualizations has boosted new research approaches in the field and facilitates Teacher Education regarding this topic.

Overall, researchers state that nowadays there is a shift of paradigm regarding teacher- student relationship and classroom management models (Weinstein, 1999), illustrated by modern Socio-Emotional Learning Programs that target both students and teachers (Weissberg, Durlak, Domitrovich & Gullotta, 2017). These programs emphasize pro-social and cooperative behaviors among teachers and students, encourage supportive relationships with assertive limit-setting guidance, and aim at the prevention of misbehaviors through the daily construction of positive learning environments (Hamre & Pianta, 2005; Jennings &

Greenberg, 2009).

(11)

2.2 Beyond transferring knowledge: teacher task-perception of socio-emotional guidance

Teachers should have a holistic comprehension of their differential roles regarding pupils, interacting with them adequately depending on each particular situation. This understanding highlights the importance of their professional identity (Taylor, 2017). Teaching is not only about transferring knowledge, but rather about relating with students through different actions and realizing how these interactions affect students’ general growth and learning (Hamre et al., 2013).

Taking into account the previous evidence regarding the importance of a positive socio- emotional development for children to success in school, and given that teachers are key figures in children's daily lives, it's becoming more patent that schools should put emphasis on the quality of teacher-student interactions, and on the primary role of teachers for the socio-emotional support of pupils within these interactions (Jacobs & Struyf, 2015; Lam &

Hui, 2010). This recognition demands clarification, (re)organization, planning and (re)arranging of the roles of school personnel. In other words, socio-emotional work cannot be seen as a solely specialist (e.g. psychologist, counselor) task anymore, but rather a collective responsibility, with an active and fundamental involvement of teachers to integrate it in the daily school curriculum and within their routine relationships (Hornby & Atkinson, 2003).

It is not in the scope of the present study to discuss all the elements that must be involved in a whole-school approach capable of effectively promote a holistic socio-emotional development of pupils. However, a study by Jacobs and Struyf (2015) stressed out the fundamental role of an integrated and supportive network characterized by cooperation and communication among teachers; teacher training programs (professionalization); school structures and procedures; well-defined and shared school policy; and, in an indirect way, school principal support and school climate.

Focusing the attention on the teacher's role, it has been highlighted in the literature that teachers’ professional identity and beliefs have strong impacts on their teaching practice, commitment, resilience, effectiveness and engagement in new activities (Taylor, 2017;

Edwards, 2016; Jacobs & Struyf, 2015). Taking it into account, there seems to be an ambiguous understanding regarding the teacher’s role in providing socio-emotional support

(12)

for students (Lam & Hui, 2010). While some may believe there is a tacit assumption that it is, indeed, a teacher’s task to work with the socio-emotional competences of pupils – beyond transferring content-based knowledge, studies have already reported reluctances by teachers regarding this role. For instance, because some teachers believe their role concerns solely the pupils’ academic achievement (Lohrmann, Forman, Martin & Palmieri, 2008), getting involved with student guidance and counselling is outside their scope of practice (Hornby &

Atkinson, 2003) and territory of experts (Finney, 2006). Consequently, they tend to get less involved with student socio-emotional development (Lam & Hui, 2010).

Furthermore, teachers may resist to adopt a socio-emotional approach in their daily school activities, because it might contrast to what they learned and shaped as role and professional identity during their previous teacher education (Lam & Hui, 2010; Lohrmann et al., 2008).

Hence, in order for them to provide socio-emotional support to pupils, they might need to change not only their self-image, but also their global teaching repertoire: move from a generalized view of students towards students’ individuality, as well as from a teacher- centered approach to a more student-centered one (Rice, 2002). All of this demands great energy and time investment.

Moreover, it’s been already shown that reduced time available, heavy workload due to paperwork, too much emphasis on academic achievement, and high levels of stress are strong factors that compete and reduce teachers involvement in relating and interacting in a rich socio-emotional approach with students – even if they believe that socio-emotional support is part of an integral education (Lam & Hui, 2010). These limits of teachers’ possibilities lead them to deal only with at-risk students (Jacobs & Struyf, 2015).

Finally, Lam and Hui (2010) contributed for the field with one of the few qualitative research that showed evidence that those teachers who perceive themselves professionally as role- models and caregivers are more likely to get involved with developmental interactive and supportive tasks. This involvement was shown to be accomplished through different approaches, such as spending time with students outside classroom, helping them with career choices, identifying students with (mental) health issues and following cases of students that needed expert help (Lam & Hui, 2010).

Overall, positive attitude, beliefs, and task perception toward this role are fundamental in order to teachers effectively approach socio-emotional development with the pupils inside and outside the classroom (Jacobs & Struyf, 2015). Because of the additional responsibilities and workload into the hands of teachers, these factors present strong evidence for the need of a supportive network from the whole school (Lam & Hui, 2010; Jacobs & Struyf, 2015).

(13)

2.3 Teachers learn by interacting

As described previously, different theoretical frameworks approach the relevance, the impacts and the structural domains of teacher-student interactions, including and emphasizing various elements that play a role in this phenomenon. Yet, there is no complete model that can comprehend all the complexity of it.

Taking these limitations into account, it is still crucial to select a model that is able to organize teacher-student interactions and teacher learning into theoretical research-based elements, in order to further understand their components and interrelations. In the next sub- chapter, the theoretical framework to approach the phenomenon under investigation is presented. It consists of two main elements: the “Teaching through Interaction” model (Hamre et al., 2013), and the “Teacher Professional Agency” model (Soini et al., 2016).

The first model was chosen because it emphasizes the interactional and relational aspect of teaching and learning, which means, it is not about teacher or student behaviors isolated, rather how they relate to each other and how they build together positive interactions. This emphasis is fundamental, even if this study takes only the empirical teacher’s perspective into consideration. The second model accounts the fact that teachers informally learn and improve their teaching competences throughout their professional experiences – as long as they have an active role towards this development.

2.3.1 Teaching through interaction model

The “Teaching through interaction” model highlights, as a fundamental task of the teacher´s daily work, the interactions between teachers and students. It consists of a “multilevel latent structure for organizing teacher-student interactions” (Hamre et al., 2013, p.463) designed with three broad domains: Emotional Support, Classroom Organization and Instructional Support. Each domain consists of several dimensions that interact with each other and describe more specifically the interactions between teachers and students. Hamre et al.

(2012) considered dimensions of daily classroom interactions that have showed evidence for promoting children`s social and academic (language and literacy) development.

The instrument developed to observe and assess these interactions, the Classroom Assessment Scoring System (CLASS). consists of 13 constructs grouped in three broad domains, as described below (Hamre et al., 2013). It has demonstrated both criterion and

(14)

predictive validity and provided evidence for its use in different body of students, grade levels (Brown, 2010) and across cultures (Pakarinen, 2017a; 2017b).

The Emotional Support domain considers the overall emotional tone of the classroom interactions: either positive, warm and respectful, or negative, stressful and marked by conflicts between teachers and students. This domain also considers how much teacher is sensitive and aware of pupils´ needs, motivations and perspectives, creating student-driven activities. It observes how teachers promote a learning environment where pupils feel safe to explore with autonomy. The Classroom Organization domain approaches the ability of teachers to prevent and redirect misbehavior by presenting clear expectations. The efficacy of teachers in providing routines and learning instructions is also covered. This domain equally considers the quality and diversity of the daily materials offered for children in order to maximize learning according to each child particularities. Finally, the Instructional Support domain considers the quality of teacher`s instructions regarding the stimulation of higher-order thinking in students. Likewise, the degree to which teacher`s feedback promotes formative evaluation to expand understanding and teacher`s speech formulation boosts differential uses of the language by pupils. Table 1 presents each broad domain, dimensions and respective descriptions.

Table 1. Classroom Assessment Scoring System (CLASS) Framework (Hamre et al., 2013, p.465)

Domain Dimension Description Emotional

Support

Positive Climate Reflects the overall emotional tone of the classroom and the connection between teachers and students.

Negative Climate

Reflects the overall level of expressed negativity in the classroom between teachers and students (e.g., anger, aggression, irritability).

Teacher

Sensitivity Encompasses teacher´s responsiveness to students´

needs and awareness of students´ level of academic and emotional functioning.

Regard for Students

Perspectives

The degree to which the teacher´s interactions with students and classroom activities place an emphasis on students´ interests, motivations and points of view, rather than being very teacher-driven.

Overcontrol Assesses the extent to which the classroom is rigidly structured or regimented at the expense of children´s interests and/or needs.

Classroom Organization

Behavior

Management Encompasses teacher´s ability to use effective methods to prevent and redirect misbehavior by presenting clear behavioral expectations and minimizing time spent on behavioral issues.

(15)

Productivity Considers how well teachers manage instructional time and routines so that students have the maximum number of opportunities to learn.

Instructional Learning Formats

The degree to which the teachers maximize students´

engagement and ability to learn by providing interesting activities, instruction, centers and materials.

Classroom Chaos

The degree to which teachers ineffectively manage children in the classroom so that disruption and chaos predominate.

Instructional Support

Concept

Development The degree to which instructional discussions and activities promote students´ higher-order thinking skills versus focus on rote and fact-based learning.

Quality of

Feedback Considers teacher´s provision of feedback focused on expanding learning and understanding (formative evaluation), not correctness or the end product (summative evaluation).

Language

Modelling The quality and amount of teachers´ use of language- stimulation and language-facilitation techniques during individual, small-group, and large-group interactions with children.

Richness of Instructional Methods

The extent to which teachers use a variety of strategies to promote children´s thinking and understanding of material at a deeper and more complex level.

Hamre et al. (2013) consider relevant organizing these competences in three broad domains, because it can help guide further research and teacher education in developing specific teaching skills. Moreover, studies have shown evidence that professional development experiences, such as courses (Hamre et al., 2012), school-based programs (Brown et al., 2010) and coaching interventions (Pianta, Mashburn, Downer, Hamre & Justice, 2008) with in-service teachers can improve teaching competences when these domains are intentionally and specifically targeted.

The CLASS scale is currently being used as crucial tool for researchers all around the world to observe and comprehend elements of effective teaching practices. In Finland, for instance, it is worth mentioning the First Steps Study, conducted as collaboration between the University of Jyväskylä, the University of Turku and the University of Eastern Finland. This research group has been developing longitudinal studies from 2006 until 2019, in which learning, motivation, well-being, among other educational factors are examined from preschool to secondary education at both school and home environments. Due to the study´s large data and frequent data points, it has gained worldwide attention (The First Steps study, 2006.)

(16)

Taking into account the contributions of the “Teaching through Interactions” model, its domains and dimensions are used as base for the coding process of the interviews that are developed in the second phase of the current study.

2.3.2 Teacher’s professional agency

Taking into account the importance of teachers with strong capacity to build positive relationships with pupils, it is still recurrent the dissatisfaction of teachers, both experienced and newly qualified ones, about the lack in their professional trainings of structured programs that approach classroom management and socio-emotional behaviors (Dicke et al., 2015; Waajid, 2013; Jennings & Greenberg, 2009; Jacobs & Struyf, 2015).

Complementarily, learning does not happen only in formal and structured programs. New research points out that, even though they are not organized structures, everyday classroom situations (Hoekstra et al., 2007; Korthagen, 2010) and work related occasions (Lohman, 2006; Kwakman, 2003) demand adaptive and flexible pedagogical practices, consisting of rich situated, embodied learning environments for teachers as informal learning settings.

In other words, students’ needs and interests, conflict situations, learning difficulties, bullying, loss of motivation, misbehavior, asking for help from colleagues etc. call teachers to reflect and improve their teaching methods. Different strategies were already reported to be used by teachers, such as experimenting new teaching methods, seeking students´

feedback, inhibiting or strengthening beliefs, among others (Hoekstra et al., 2007;

Kwakman, 2003).

An important contribution from Hoekstra et al. (2007) is that informal learning in classroom settings does not occur necessarily with goal-oriented purposes and full awareness from the teacher. On the contrary, many times informal learning happens in a reactive (not planned) or implicit/tacit (unaware) forms. Moreover, cognitive, affective, motivational and behavioral aspects must be taken into account when it comes to study learning from teacher´s perspectives.

Kwakman (2003) unfolded from her literature review and empirical studies three learning principles that reflect teachers´ learning in the workplace: teachers learn by participating in real context-based activities within the school they work in; by collaborating with colleagues and taking part in social-oriented activities that boost individual reflections; and by accomplishing activities that aim to develop themselves professionally. Parallel to that, King and Newman (2000) stress three main individual activities that most support teacher

(17)

professional development: reading (keeping oneself updated with new insights and further professional developments), experimenting (intentional effort to try something new) and reflecting (recognizing problems and unsatisfied routines and thinking about them in order to change it). Other studies support these principles. According to Henning, Rice, Dani, Weade and McKeny (2017) and Soini et al. (2010), particularly solving challenging situations with pupils was considered one of the most important learning resources for teachers’ professional development. Moreover, self-reflecting about teaching practices must come with collaboration with students and colleagues to be effective (Hamre et al., 2012).

Whether teachers take these informal learning settings and situations as opportunities for their teaching improvement and further professionalization depend both on teacher´s characteristics (if they are akin to learn actively from what happens there) and the support of the school environment. Both factors cannot be taken for granted (Soini et al., 2016).

Regarding the support of the school environment, Jacobs and Struyf (2015) showed that cooperation within school personnel, participation of teachers in creating the school vision, school´s climate, and principal´s support are fundamental for active participation of teachers in professionalization activities. Lohman (2006) also points out that lack of time and resources at work, as well as lack of productive interactions with work colleagues are factors that limit teachers from engaging in informal learning activities at school. Kwakman´s (2003) study in Netherlands also reported a general lack of infrastructure in the schools that could promote this workplace as a potential learning environment for teachers.

Related to teachers' characteristics, factors that favor teachers to engage in learning at work are interest in the profession, love for learning, initiative and commitment to professional development (Lohman, 2006). A recent study by Perera, Granziera and McIveen (2018) reported that teachers with the "resilient" or "well-adjusted" personality profile from the Big- Five Dimension scale have the highest scores of self-efficacy in teaching, job satisfaction and – most important for the present research – work engagement. This profile is characterized by low levels of neuroticism and high levels of conscientiousness, agreeableness, and openness. In other words, teachers who would be more willing to learn and develop themselves professionally are those who are strongly conscious (have self- discipline and order); presented high openness scores (are curious and tend to vary their practices); and are consistently agreeable (tend to be tender-minded, altruistic, and straightforward). Other studies show that teachers with good socio-emotional competences and self- and co-regulative strategies are more open to learn and adapt their teaching styles to students' needs (Jennings & Greenberg, 2009).

(18)

Pyhältö et al. (2015) consider that teacher´s active learning at school must be understood with a holistic and integrative concept, such as the teacher’s professional agency construct, which means the capacity of managing the own process of learning and constructing the own context of learning at individual level. Though, social interactions are equally relevant and teachers must look after for both colleagues and pupils as learning resources, as well as act as a support for them (Edwards, 2005). According to Pyhältö et al. (2015) and Soini, et al.

(2016), this complex behavior is highly relational, context-dependent and includes three interrelated elements: motivation to learn, efficacy beliefs about learning and active learning strategies.

These elements must be simultaneously present for one to presume that a teacher has professional agency. The teacher should identify him/herself as an active learner capable of intentional learning by experimenting and reflecting about it (efficacy beliefs). It also requires motivation to accomplish self-improvement, self-oriented learning and face the challenges of it (motivation). Finally, knowledge is not enough, one has to have the proper skills and strategies to change previous behavior or develop new ideas (learning strategies) (Pyhältö et al., 2015).

A fundamental aspect related to professional agency and collective knowledge building in the school (Soini et al., 2016) is that one cannot assume that innovative pedagogical practices are automatically shared with others and move spontaneously from particular classrooms to the world (Henning et al. 2013). The fact that teachers learn from classroom situations and from discussions with colleagues doesn't mean that it necessarily happens simultaneously.

According to Akkerman and Bakker (2011), this simultaneous process requires boundary crossing (BC) skills, which means the capacity of a teacher to transit in different professional contexts and identify learning opportunities, extracting from them possible solutions for correlated problems. This consists of an interactional and communicative process and leads to transformative practices in a large level (Akkerman & Bakker; 2011, Soini et al., 2016).

All seems to lead to the conclusion that teacher professional agency is deeply social embedded and demands reflection upon the interactions with students and school community in order for teachers to adapt roles and performances that can reach the needs of these relationships (Pyhältö et al., 2015). Moreover, both willing to learn with pupils and other teachers are in the “heart of meaningful teacher learning” (Soini et al., 2016, p.391).

However, as mentioned previously, only few studies have directly addressed the informal learning process of how teachers learn to improve their interactions with pupils and how they see the impacts of such learning development.

(19)

2.4 Research questions

This research aims to progress the field of teacher continuous professional development and fill a gap in the literature regarding teacher informal learning about relationship with pupils.

In other words, it tries to understand how experienced teachers reflect about their learning process in informal contexts in order to continuously develop their performances, when it comes to build positive interactions with their pupils.

In a few words, my research questions are:

1) How do Finnish experienced teachers perceive that their professional agency supports their learning process to build positive interactions with pupils?

1.1. How do they reflect about their motivations to learn?

1.2. What do they think about their efficacy beliefs to learn?

1.3. How do they portray their active strategies that most support their learning process?

These questions try to capture, from the narratives of Finnish experienced teachers, how they perceive that their professional agency in terms of motivations to learn, efficacy beliefs to learn, and strategies to learn help them to improve their interactions with pupils during their everyday teaching practices. The “Teacher Professional Agency” constructs (Soini et al., 2016) take into account the informal character of teachers’ learning process and are fundamental factors that maintain it.

The fact that teachers try to remember their informal learning process along their career means that some of this informal learning can remain unconscious. Hence, this question tries to capture the most significant types of factors and learning events that marked their informal learning process regarding how to improve interactions with pupils.

2. How do teachers apprehend the general character of their learning process related to build positive interactions with pupils?

This question looks into the general impression the teachers have about their learning process all the way through their career and how it impacts their everyday relationship with children.

It approaches both their holistic perceptions of the influences of their learning over the relationships with pupils, as well as the intensity of their learning process throughout the teachers´ career.

(20)

3 METHODOLOGY

In the next sections, the research design and methodological choices of the present study, followed by measurements, data collection, participants, and analysis processes are described and explained in further details.

3.1 Research design

The present research consisted of a mixed method approach of confirmatory multiple-case analysis. The purpose of accomplishing a mixed method of quantitative and qualitative approaches was to apprehend multiple ways of emerging as well as different possibilities of meanings and interpretations of the phenomenon under investigation (Creswell & Clark, 2018). Moreover, this is a confirmatory study because it based the research questions and further investigations mainly on theoretical assumptions deducted from the literature.

In a nutshell, an initial broad-scale survey followed by semi-structured depth interviews were developed. Therefore, general data regarding teacher professional agency was initially collected through the survey. The teachers who presented the highest scores of professional agency were selected, because of their engagement in continuous learning. The study was followed up with interviews of the most suitable individuals who completed the survey to bring in-depth information and insights that could expand the meanings of the survey results.

Finally, the main analytical framework of this study was deductive, theoretical guided content analysis. However, it was complemented by inductive analysis of relevant empirical evidence that emerged from the collected data.

In general, the research consisted of four main phases:

1) A broad-scale research with primary school teachers of the municipality of Joensuu, who volunteered to answer a survey with the Teacher's professional agency in the classroom scale and Boundary crossing scale (Soini et al., 2016).

2) Analysis of the frequencies distribution of the sample, reliability of scales, possible gender biases, correlations of the collected data, and further identification of the teachers with the highest scores of teacher professional agency. The teachers´ time availability to participate in the interviews was also considered.

3) Accomplishment of individual, semi-structured and face-to-face interviews with the previous selected teachers.

(21)

4) Development of theory guided Content Analysis (CA) over the text corpus, using as theoretical framework the “Teaching through Interaction” model (Hamre et al., 2013) and

“Teacher’s Professional Agency” construct (Soini et al., 2016) as initial references to frame the main categories and code values. Posteriorly, the analytical frame was complemented by Teacher´s Personality factors that showed empirical relevance for the analysis and interpretations of the findings.

3.2 Quantitative phase: measures, data collection and participants

As mentioned previously, the quantitative research procedure consisted of applying the Teacher's professional agency in the classroom and Boundary crossing scales (Soini et al., 2016). The former scale consists of a questionnaire that measures teacher professional agency, including motivation to learn, efficacy beliefs about learning, and learning strategies through two factors: collaborative environment and transformative practices (CLE – six items), and reflection in classroom (REF – four items). The Boundary crossing scale (BC – three items) measures teacher´s capacity to utilize ideas from other teachers in his/her own practices, as well as to share his/her own practices with others.

In order to collect as much data as possible, a study permit application (see Appendix 1) was sent to Joensuu municipality, asking permission to develop the research with all the schools of the city.

With the permission in hands, a pilot of the English version of the questionnaire was applied with seven Finnish teacher students of the School of Educational Sciences and Teacher Education, at University of Eastern Finland (UEF), Joensuu Campus. The pilot provided good suggestions regarding improving the questionnaire, such as: use of everyday vocabulary in the items, instead of academic/technical words, and provision of examples of what each item means to facilitate the comprehension of teachers. The general structure of the survey was also improved. For instance, the participants showed fields of the items to be marked that were not in the questionnaire.

When the final version in English was completed (see Appendix 2), the questionnaire was open to be answered starting from end of August 2017. The survey could be filled online through Monkey Survey website, or with paper-pencil materials, depending on the teachers’

preference. The whole survey had a mean time of 5 to 10 minutes to be answered. In the end of the questionnaire, participants were asked to give their contact information, in case they were available to participate in the further phases of the study.

(22)

It is important to mark that the participation in the research was voluntary and participants could cancel their involvement at any moment they wanted. Moreover, the fact that the teachers provided their contact information did not hurt the anonymity character of the research, because their identity was kept privy during the whole research process.

Initially, only few teachers answered the survey and some of them were giving as reason the fact that the questionnaire was in English. Hence, a Finnish version of the survey was developed with the assistance of a Finnish person graduated in English Language.

The Finnish version was piloted by seven other Finnish teacher students of the School of Educational Sciences and Teacher Education, at UEF, Joensuu Campus. The piloting gave important suggestions to improve the translated items for the Finnish version, such as content related improvements and structure of the phrases. Finally, the Finnish version (see Appendix 3) was available to be answered online and with paper-pencil materials starting from October 2017.

In total, the data collection of the survey started in the end of August 2017, taking five months to reach 48 teachers with both English and Finnish versions. The participants consisted of primary school teachers from the municipality of Joensuu, who worked with 1st to 5th grades – some of them teaching more than one grade. They could read and speak English fluently, had from 5 years to 39 years of teaching experience and are located from the stabilization phase to the final phase of teaching career (de Vries, Jansen & van de Grift, 2013). The distribution by gender featured 70.8% of participants being female, they had 19.9 years of experience on average (SD = 9.02) and the grade with the biggest portion of teachers working was 5^th grade, followed by teachers who worked in more than one grade (characteristic typical in Finland).

Table 2. Survey participants profile by grade and gender

Grade

Gender 1st grade 2nd grade 3rd grade 4th grade 5th grade More than one grade

Total

Female 5 6 8 1 6 8 34

Male 0 2 1 1 8 2 14

Total 5 8 9 2 14 10 48

It is important to mention that all the collected material was stored in a folder that only the researcher had access. The questionnaires were organized and coded according to each participant in order to recover any needed information.

(23)

3.3 Qualitative phase: rationale of interviewing and interviewees

Individual semi-structures interviews were accomplished, because the interactive relationship between interviewer and interviewee has potential to build new knowledge through the exchange of viewpoints and meanings about the investigated phenomenon.

Moreover, depth interviews have the advantage of providing details about personal and professional experiences, decisions and action sequences (Bion et al., 2011) – which were relevant to understand what and how the experiences of the teachers supported them to learn and develop professional competences.

One of the most relevant topics regarding interviews is the sample of respondents. The main purpose of qualitative research is not to count the opinions or to measure a trait in a population (as in quantitative research), but rather to explore the different positions and representations individuals of a specific population have regarding a phenomenon (Altheide et al., 2011; Bion et al., 2011). Hence, the main concern when selecting a sample is to comprehend individuals from significant milieus to the research’s topic. Not necessarily statistical socio-demographic samples will comprehend the most significant viewpoints of a phenomenon, but perhaps natural groups that play a significant role in the topic which is being studied is better. Besides, more interviews don’t mean necessarily a better understanding of the topic, since there’s a saturation point of the opinions and perspectives that can be reached from a sample (Altheide et al.; 2011, Bion et al., 2011).

Another important aspect of interviewing is that it is time and energy consuming. That’s why previous survey was applied in order to better select the most suitable teachers who would take part in the interview. That is, the teachers who are more akin to learn from their everyday classrooms situations, their pupils, the teacher community and from other personal contexts too. Those teachers would be more able to express, in-depth, their experiences, opinions, values, motivations, insights etc. In sum, their learning process and consequential developments.

Considering the theoretical background of “Teaching through interaction” and “Teacher professional agency” models, the first draft of the interview guidelines was developed. The main idea of the initial phase of the interview was to warm up the teachers about the topic of research. This warm up phase had a predominant emotional tone and asked the teachers to link a color to the classroom atmosphere where he/she works. The colors provided were:

black, blue, brown, green, orange (dark yellow), purple, red and yellow (bright yellow),

(24)

following the study of Wexner (1954). The participant also had to write a keyword that represented the meaning of the chosen color.

After the warm up, the interview was oriented to explore how the teachers understand their interactions with pupils in a broad manner. Sequentially, the questions approached how the teachers used to interact with pupils in the beginning of their career compared to nowadays, looking for changes and improvements during his/her career. Detailed aspects of the theoretical framework were asked in order to understand how the professional agency factors play a role in the informal learning process of teachers. In the end of each interview, the interviewees were presented to one of the survey outcomes in order to gather their opinion about the matter (further explanation about the survey outcomes in the next chapter).

With the purpose to test the interview protocol, a pilot study was accomplished in a focal group of six teacher students of the School of Educational Sciences and Teacher Education, at UEF, Joensuu campus. From the pilot, the main take away lessons were: the need of a simultaneous Finnish-English translator for some teachers that might be uncomfortable with English, and the conclusion that individual interviews would be more able to gather detailed in-depth information than a focal-group approach.

Taking this into account, the final version of the interview was developed and would have been translated to a Finnish version, in case any participant would request a Finnish translation. However, all the participants of the interview sessions did not request the Finnish version or a personal translator. Hence, the written English version (see Appendix 4) was available to the participants during the interviews, divided in small pieces of papers with individual questions in order to facilitate the reading of the repondents.

To accomplish the interviews with the most suitable teachers, the criteria for the selection consisted of the highest scores of the TPA and BC scales and teachers’ availability to take part in the interview. Eight teachers were primarily chosen to be interviewed. The interview invitation was sent by email, in English and Finnish, and five teachers accepted the invitation. They could choose the time and place most convenient for them, as long as it could guarantee the needed privacy and conditions for the interview. Also, this sample of participants reached the saturation point of data collection, showing no more need for more interviews to be done.

In general, most interviews took place in a classroom of the schools where the teachers work.

In two occasions the interviews needed to be interrupted, because the teachers had visitors.

One participant decided to meet in the city library, because it was convenient for both interviewer and interviewee. All the participants chose to be interviewed after their workday,

(25)

in the afternoon, and all of them gave permission to the interviews to be audio recorded. The overall climate of all interviews was informal and relaxed. Some participants felt comfortable to narrate in details their teaching experiences and others were more objective and short in words. Below you can find the general profile of the participants of the qualitative phase as well as general information about the interviews.

Table 3. Interviews’ general characteristics and participants´ profile Gender Years of

experience

Teacher activities Grades Place of interview

Time of interview 1. Female 39 Teacher/Educator 1st-5th Classroom 14h – 15h30 2. Male 30 Teacher/Headmaster 3rd Classroom 12h – 13h30

3. Male 20 Teacher/ICT 4th Classroom 13h – 14h

4. Male 19 Teacher 2nd Classroom 13h – 14h30

5. Male 26 Special teacher 5th City library 15h30 – 16h30

3.4 Analysis of quantitative data

With the software IBM SPSS Statistics 19, descriptive statistics were run to check the normality of the data and frequencies distribution. Because the sample was below 50 participants, the suitable normality test was Shapiro-Wilk. It showed that the data was not normally distributed, what called for non-parametric tests. Cross tabulation analysis was done to verify any gender biases in the results. Non-parametric Spearman´s rho correlations were run between the subscales of TPA scale (CLE, REF), BC scale and years of experiences of the teachers. Cronbach’s alpha was also investigated to confirm the internal consistency of the scales and the reliability of the results.

The analysis of the quantitative data had fundamental relevance for the next phases of the research. The descriptive statistics gave a general view about the TPA and BC characteristics of the broad sample of participants. From this global picture, the individuals with the highest score of the scales were selected for the next moment of the study – also considering their availability. Hence, the analysis and results of the survey was the starting point for the interviews, working as a purpose sampling tool.

Finally, it is important to mention that some results from the survey analysis called attention and needed further investigation. Hence, the participants of the interview phase were asked to express their opinions about the survey findings in order to raise multiple view points and understandings about it.

(26)

3.5 Analysis of qualitative data

The analysis of the text corpus gathered with the interviews was made with the software Atlas.ti 8, using mainly as analytical method theory guided CA based on the previous theoretical framework. Complementarily, exploratory CA was also developed due to the relevance of new empirical evidence that emerged from the data. Both were developed in order to make holistic inferences about the informal learning experiences that the teachers accomplish to improve their interactions with pupils. The analysis also aimed to grasp the teachers´ general impression about the intensity of this learning process and its impacts on their relationship with pupils throughout their career.

To effectuate CA, both qualitative and quantitative processes took place: before any quantification descriptions (statistical treatment of code frequencies), contextual interpretation of the text corpus was developed. Semantic analysis over the whole material focused on connotation and denotation meanings of what was said by the participants (Andren et al., 2011.)

Then, a coding frame for the analysis was developed. This is important because, “While a text corpus is open to a multitude of possible questions, the CA interprets the text only in the light of the coding frame, which constitutes a theoretical selection that embodies the research purpose.” (Andren et al., 2011, p.6). In order to develop the coding frame, there are two kinds of approaches: deductive (from theory to practice) and inductive (from practice to theory). In this study, the deductive approach was mainly followed, because it is appropriate to test existing theory. Hence, the units of analysis of the coding frame were selected based on the research questions and the theoretical framework. However, the inductive method was also accomplished, because the text corpus derived from the teachers´ reports demanded exploratory analysis of new relevant empirical evidence.

According to Cho and Lee (2014), the process of data analysis follows three basic steps:

establishing categories and themes, selecting the units of analysis, and defining code values.

The two main theoretical models were used as global categories for the codes. Then, the themes and units of codes were compound by the three broad domains emotional support, classroom organizations, and instructional support of the “Teaching through Interaction”

model (Hamre et al. 2013), as well as the elements motivation to learn, efficacy beliefs, active learning strategies, and boundary crossing of the “Teacher Professional Agency”

model (Soini et al., 2016). Finally, the code values were looked for and observed according to the theoretical descriptions of these elements.

(27)

For analytical purposes, because the concept active learning strategy had a fundamental role in the interviews, this code was rooted into two sub-groups, apprehending the following theoretical factors: boundary crossing (Soini et al., 2016) as actually a collaborative learning strategy; and reading, experimenting, and reflecting (Kwakman, 2003) as individual learning strategies. Moreover, it was accentuated in the interviews that teachers most of the time

"learn by doing" (Kwakman, 2003) - further discussion in the next chapter. Consequently, the three domains of "Teaching through interaction" model were actually considered practices with the character of active learning strategies that teachers implement in their daily routines and learn from them. They have both individual and collective characteristics.

Posteriorly, an inductive approach was accomplished because of the relevance of new empirical evidence. It was observed from the teacher´s voicing the importance of a third category regarding the learning process of teachers, when it comes to building and improving interactions with pupils: the teacher´s personality. This factor appeared spontaneously in three main directions according to the teachers' reports: personality of teacher as a "natural"

work tool; personality of teacher as lifelong learner, a passionate person for learning throughout the whole life; and personality of teacher as the stable base of teacher-student interaction and consequently an oppositional factor to change these interactions. Hence, these three codes were added to the final coding frame, according to the following table:

Table 4. Categories, themes and codes of the content analysis Category Teacher Professional

Agency

Teaching Through Interaction (active learning strategies,

“learn by doing”)

Teacher Personality

Themes/

Codes

Motivation to learn Emotional support - Positive climate - Negative climate - Teacher sensitivity - Regard for students perspectives

- Overcontrol

Work tool

Efficacy beliefs to learn Classroom organization - Behavior management - Productivity - Instructional learning formats - Classroom chaos

Lifelong Learner

(28)

Active learning strategies:

- Reading (individual) - Reflecting (individual) - Experimenting

(individual)

- Boundary crossing (collective)

Instructional support:

- Concept development

- Quality of feedback - Language

modelling - Richness of

instructional methods

Oppositional factor to change

This coding frame was developed and evaluated along the whole analysis process by its internal coherence, simplicity and interconnection between the code values (Andren et al., 2011). Through these dimensions, CA was a valuable tool to find patterns, trends, turning points and even contradictions within the teachers’ reports about their learning process. The CA made possible to understand how the theoretical elements appeared in the participants´

discourses, as well as the intensity of their appearance in the collected material. Finally, it was intended to build “maps of experiences”, trying not only to identify isolated experiences, but a network between them, finding internal and external relationships (Andren et al., 2011).

Altogether, the context-oriented CA mapped from the teachers discourses how the elements of their professional agency and the characteristics of their personality supported their learning process to build positive interactions with pupils. Moreover, it revealed the global impression the teachers have about the influence that their learning have on their relationship with pupils. The hierarchical code structure derived from the theoretical framework and empirical material can be better visualized in the Figure 1 below.

(29)

Figure 1. Hierarchical relationship between the codes derived from the “Teacher professional agency” concepts, “Teaching through interaction” model and “Teacher personality” factors

In sum, the research design adopted in this study was a confirmatory multiple-case analysis with mixed method approach. A survey was accomplished to gather general data regarding teacher professional agency and to select the most suitable teachers to participate in the in- depth semi-structure interviews. The text corpus was analyzed using theory guided content analysis, based on the theoretical framework, but also taking into account the relevance of teacher's voicing and new concepts that came out from their participation.

(30)

4 FINDINGS AND DISCUSSION

The next sections present and discuss the findings of both quantitative and qualitative phases of this study in order to answer the research questions. To approach the first inquiry of this investigation, the survey results are presented to describe in details the sample of participants. Next, the statistical analysis made over the questionnaire’s data is showed.

Further, the teachers´ voicing gains a fundamental role in the results in order to understand how teachers´ professional agency and personality support their learning process to build positive interactions with pupils. Descriptive frequencies of the qualitative data are presented and posteriorly, the codes are analyzed regarding their relevance, meanings and interrelations.

To approach the second inquiry of this investigation, the second part of the chapter discusses how the teachers perceived the general character of their learning process regarding building positive relationship with pupils. First, the role of their personality is stressed and discussed in details. Then, the apprehension about the intensity of the teachers´ learning throughout their career is analyzed by comparing previous literature, the survey results and teachers´

reports.

4.1 How teachers´ professional agency and personality support teachers´ learning process to build positive interactions with pupils

The first quest of this study aimed to investigate how Finnish experienced teachers perceive that their professional agency in terms of motivations to learn, efficacy beliefs of learning and strategies to learn during their everyday teaching practices help them to improve their interactions with pupils. This investigation was complemented with two teacher personality factors (work tool and lifelong learner) that showed empirical relevance to support teacher- pupil interactions and affect teachers´ learning processes. In the following sections, the results of both quantitative and qualitative phases are presented in order to map how these elements interrelate to each other.

4.1.1 Survey findings: reliable sample and the intriguing correlation

The survey with the TPA and BC scales accomplished in the initial phase of this study gathered general data about the participants’ professional agency. The means, standard deviations and intercorrelations for all subscales and teachers´ years of experience are presented in Table 5.

How Finnish experienced teachers learn to improve their interactions with pupils