Co-teaching literacy strategies for the inclusion of second-language learners: possibilities for professional development

Abstract

This narrative study, set in lower-secondary school (grades 7–9) in Finland, explores a co-teaching initiative aimed at developing inclusive education for second-language students by integrating literacy strategies in content-area teaching. The co-teaching was realized by three experienced content-area teachers working in dyads with a resource teacher with competence in literacy strategies. Co-teaching among experienced teachers at higher grade-levels and with the purpose of including second-language students is a relatively under-researched field. Data was collected through seven teacher–researcher conversations over two school-years. Word images were constructed from the field texts to represent four narrative themes, depicting the teachers’ narration of their knowledge construction and role formation: daring to share, extended expertise, fine-tuning roles and increased well-being. Results suggest benefits such as well-being through shared responsibilities along with increased expertise to support all students’ learning of both language and content in the co-taught classroom and beyond. Trust, being open to change and daring to share one’s knowledge landscapes and classroom space were seen as crucial in the negotiation of educational priorities and teacher roles. Co-teaching is suggested to be a viable option for the professional development of inclusive practices in the form of integration of literacy strategies across the curriculum.

Keywords:

• Co-teaching
• literacy strategies
• professional development
• second-language learners
• inclusion

1. Introduction

This study explores a co-teaching initiative aimed at developing and enabling inclusive education for second-language (L2) students by integrating literacy strategies in content-area classrooms. The co-teaching was realized through collaborations between three content-area teachers working in dyads with a resource teacher with competence in literacy strategies.

The integration of literacy strategies in content-area classrooms, also known as Language Across the Curriculum (LAC), originated in the needs of L2 students in Anglo-Saxon educational settings in the 1970s. However, LAC has gradually been integrated in core curricula world-wide as a competence to imbue all instruction, in particular in light of weakened PISA results in reading comprehension. LAC involves the contextualization of educational content to support students’ learning and the simultaneous development of reading and writing strategies pertaining to specific content areas (Cummins & Early, 2015; Shanahan & Shanahan, 2008). However, it still remains an under-developed area in teacher education and among practicing teachers at large.

The focus of this study is on the informants’ narrations of their evolving partnerships and competence building for LAC through a process of merging their individual professional knowledge landscapes, as this knowledge is embodied, relational and lived out in a particular time and place (Clandinin & Connelly, 1996). Thus, the study explores co-teaching as a tool for professional development. Co-teaching, short for co-operative teaching, is ‘two or more professionals delivering substantive instruction to a diverse or blended group of students in a single physical space’ (Cook & Friend, 2017, p. 2). Other terms used are collaborative teaching and tandem teaching. The models vary in complexity and degree of joint responsibility, from the model where one teaches and the other assists, to team teaching where whole group lessons are delivered interdependently by both educators.

Although Cook and Friend (2017) note collaboration between different categories of educators (see also Krammer et al., 2018), it is often explicitly defined as ‘the collaboration between general and special education teachers for all of the teaching responsibilities of all of the students assigned to a classroom’ (Dove & Honigsfeld, 2012; Pratt, 2014). By the end of the 1980s, it had become the preferred method of educating special needs students in general education settings at least in North American contexts, in the wake of reforms instigated to support disadvantaged student populations. In the early 2000s, co-operations between mainstream teachers and L2 teachers began to emerge and increase (e.g. Dove & Honigsfeld, 2014; Saloviita & Takala, 2010).

Although introduced into Finnish national curricula already in the 1970s as a partnership between special education and class teachers, co-teaching has not so far been frequently employed, particularly not above primary-school level (Saloviita & Takala, 2010; Sundqvist et al., 2021). However, in the last decade it has gained interest due to the prominence given to increased inclusion opportunities for special needs students. In the most recent National Core Curriculum for Basic Education (Finnish National Agency for Education, 2016) cross-curricular collaboration is given increased emphasis.

2. Theoretical background

Co-teaching has mainly been researched from two perspectives: co-teaching for the purpose of inclusive education; co-teaching as teacher learning (Rytivaara & Kershner, 2012; Rytivaara et al., 2019). Co-teaching for inclusive education has primarily focused on inclusion of special needs students into the mainstream classroom through different models of co-teaching and their implementation processes. Studies show benefits such as increased teacher resources per student, as well as added and combined teacher expertise concerning both content knowledge and instructional strategies for special needs (Jackson et al., 2017). Resources for co-teaching have often been allocated only for specific core-subjects, e.g. mathematics and language arts in the case of Swedish-medium schools in Finland (Sundqvist et al., 2021).

However, co-teaching for the inclusion of L2 students is a less researched field. Dove and Honigsfeld (2012) address the incessant growing demands put on teachers, including a sense of pressure from not being able to meet students’ differing learning needs, e.g. the needs of L2 learners in mainstream content-area classrooms. The growing demands contribute to increased rates of teacher attrition (Schleicher, 2018), and although attrition rates in schools in Finland are far below the ones in e.g. the U.S. or the U.K., the need for support throughout the teacher career is often highlighted (Niemi, 2015). Collaborative practices such as co-teaching can bring important emotional support and even reduce the workload, besides constituting job-embedded professional development (Alcalà Arxé et al., 2020; Dove & Honigsfeld, 2012). Among the benefits suggested for students is the reduction of possible stigma connected to pullout arrangements (Cook & Friend, 2017; Dove & Honigsfeld, 2012), and more linguistic input and support for L2 students’ learning (Alcalà Arxé et al., 2020; Creese, 2005; Davison, 2006; Dove & Honigsfeld, 2012). Cook and Friend (2017) note that even well-organized pullout instruction results in fragmentation and time-consuming relocation between rooms. They also highlight increased individualization opportunities through co-teaching, including supporting gifted students at their level. How increased support for L2 learning can be accomplished through co-teaching can be exemplified by Alcalà Arxé et al. (2020), who report on co-teaching contributing to an inclusive classroom and interactional competence in the context of English as L2 (Content and Language Integrated Learning) among predominantly Catalan speakers. In their study, teachers co-operated around making conversational adjustments in order to ensure understanding, achieve participation and help learners’ extend and improve their utterances. They suggest that the use of such interactional competence strategies provided students with more opportunities to produce good quality output in L2.

Studies of co-teaching as teacher learning have more typically focused on unequal partnerships, e.g. pre-service teachers working along in-service teachers, or other types of collaboration where responsibility is only partly shared, e.g. special education teachers supporting specific students in the classroom (Saloviita & Takala, 2010). As Rytivaara et al. point out, such endeavors are different from co-teaching where the professional knowledge landscapes of two experienced teachers meet. Studies also most typically involve infrequent collaborations within relatively brief time-frames. For professional development to be sustainable, i.e. resulting in enactment, the literature stresses long-term processes to enable the creation of professional learning communities (PLCs) collaborating towards shared goals (Gore et al., 2017; Hord, 2004). Overall, the importance of initially establishing partnerships and negotiating roles is stressed in the co-teaching literature (Rytivaara & Kershner, 2012; Rytivaara et al., 2019). This entails mutual identification and articulation of educational goals and beliefs about teaching, including negotiations of responsibilities and agreement on acceptable student behavior to avoid giving contradictory signals to students (cf. Cook & Friend, 2017). As a further example of the complexity involved on a micro-level, a study by Pesonen et al. (2020) suggests that primary school teachers’ sense of belonging in co-teaching teams is constructed considering a balance between three dimensions: teachers’ work practices (e.g. shared responsibilities and commitment), mutual relationships (e.g. knowing each other, supportive atmosphere), and individual characteristics (e.g. individual strengths, similar pedagogical thinking and practical knowledge).

To prepare for what the implementation of co-teaching involves, the general suggestion in the co-teaching literature is to include some type of supportive instruction in order to avoid discontentment or even failed processes. For example, Dove and Honigsfeld (2014) investigated enabling factors for co-teaching around the inclusion of L2 students at elementary school level and found professional development workshops, coaching sessions, team observations and feedback during the yearlong process to be of importance. In Brendle’s (2017) study, the teachers were already familiar with different co-teaching models, but still expressed the need to learn more about how to handle changing teacher roles in particular. The consequence was unequal responsibility patterns, notably seeing the special education teachers in assisting roles, as well as an unnecessarily low degree of collaboration throughout. As an example pertaining to the focus of the current study, Creese (2005) notes that teachers contributing literacy aspects risk unequal status if such aspects are less prioritized as educational goals (cf. Davison, 2006). Thus, professional development could contribute to support the handling of different challenges commonly connected to co-teaching.

The gradual evolution of partnerships in co-teaching is depicted in different models. Among the most frequently cited are Davison (2006) and Pratt (2014). Although local processes may differ, the models suggest certain stages to be particularly challenging and thus help raise awareness of issues that might be mitigated or even avoided.

Davison’s (2006) model includes five stages: 1. Pseudocompliance: a stage of passive resistance (in enforced processes), not yet co-teaching; 2. Compliance: willing participation, although with minimal effort and limited insights, not yet true co-teaching; 3. Accommodation: teachers experimenting with models and ways of enacting collaborative teaching, still in need of external support; 4. Convergence: teachers start engaging, sharing knowledge and experiences; 5. Creative co-construction: co-teaching preferred to other ways of working, interdependence through shared responsibilities, leaning on and learning from each other’s expertise. At this stage, partnerships are signified by trust, entailing constructive professional interactions. Conflicting views are seen as inevitable and even as opportunities to evolve.

Pratt’s (2014) model constitutes three phases. The first is the start of the process, comparative to Davison’s first two stages. The second phase, the symbiosis spin, is the start of proper collaboration through partnership building as teachers reflect on their mutual task and negotiate their respective roles. This phase share similarities with Davison’s stages 3 and 4. Finally, in the third phase, the teachers are in symbiosis, working as an integrated and interdependent team, benefitting from each other’s differences and expert knowledge. This mirrors the fifth stage of creative co-construction in Davison’s model. In Pratt’s study involving secondary school teachers, there were significant differences between the teams concerning when they had achieved the third phase: some within a few months of co-­teaching, while others were still working towards the final phase in their second year. Although studies suggest that specific instruction might be beneficial and even necessary in many contexts, the concrete adaptation of co-teaching may benefit from being more organic and develop according to locally relevant needs rather than strict implementations of prescribed models (cf. Rytivaara & Kershner, 2012).

Thus, co-teaching appears to be as much relational as it is dependent on knowledge of teaching models and methods (cf. Pesonen et al., 2020). Rytivaara et al. (2019) discuss how more open-ended experimentation appears to be an effective means of learning for experienced teachers in particular. However, Rytivaara et al. also suggest that experienced educators may have more fixed and even rigid knowledge landscapes, necessitating more effort to achieve well-functioning partnerships. The articulation of largely tacit knowledge requires effort from everyone, but concerning representatives from different knowledge disciplines, the initial challenges may be even greater. Still, as long as the process continues, the learning possibilities created from conflicting views should not be underestimated (Hollingsworth & Dybdahl, 2007; cf. Davison, 2006). Once the basis for the co-operation is set and the partners can focus on more substantial questions, the ensuing process of efficiently learning from each other’s perspectives can be relatively quicker for experienced teachers, as well as rewarding in terms of bringing renewed inspiration and energy (Cook & Friend, 2017). Gourvennec et al. (2022, p. 95-96) conclude:

(…) if the two collaborating teachers truly engage in co-teaching, give each other meaningful roles, and embrace the potential to learn from each other’s competence, then ‘one plus one teacher’ may equal more than the competence of ‘two teachers’ and result in a situation with much growth and synergy. By contrast, when that is not the case, one teacher may withdraw while the other feels burdened, and the metaphorical sum may be less than ‘two teachers’.

3. Methodology and methods

3.1. Methodology, aim and research questions

The methodological basis for this small-scale study is narrative research. Focus is on the relatively uncharted territory of co-teaching as professional development between experienced educators developing increased preparedness for LAC for the inclusion of L2 learners. This process entails the formation of teacher roles in the construction of a shared knowledge landscape (Clandinin & Connelly, 1996; Rytivaara & Kershner, 2012; Rytivaara et al., 2019). The ‘narrative turn’ (Riessman, 2008) of the last four decades has brought an increase in communities where experiences are shared during different stages of teachers’ professional career, e.g. peer mentoring. Furthermore, the use of teacher narratives as a research methodology in relation to teachers’ practices has increased in the last two decades, partly as a consequence of the opening for participants’ voices through the increased validity afforded to qualitative research (Burns & Bell, 2011). The process of gathering narrative data can simultaneously serve to support ongoing professional transformative processes through the meta-awareness in play when explicitly reflecting and co-constructing experiences (Edwards, 2017; Hollingsworth & Dybdahl, 2007).

Clandinin and Connelly’s (2000) development of the narrative approach builds on Dewey on human experiences as central narrative elements. The world is seen as experienced, understood and created by individuals or groups of individuals in interaction with their surroundings, and stories seen as a way of interpreting and communicating this perceived reality, not as reproducing reality (Riessman, 2008). The temporal aspect is also fundamental, shown through the continuity created as different experiences build on previous ones.

The study involves three content-area teachers co-teaching in dyads with the same resource teacher in a lower secondary school (grades 7–9) in Finland (N = 4). The exploration into the competence building and evolving partnerships of the co-teachers was guided by the following research questions:

1. What are the co-teachers’ experiences of their (individual and shared) knowledge construction?

2. What are the co-teachers’ experiences of the formation of their roles during co-teaching?

3.2. Setting and data gathering procedure

The co-teaching endeavor was initiated in 2017–2018, involving the resource teacher and two voluntary content-area teachers. During the initial year, the main task of the resource teacher was to bring LAC aspects into the mainstream classroom. The dyads co-taught between one and three weekly lessons (per six-week content-area course with three weekly lessons), depending on what the schedule of the resource teacher allowed in terms of mutual slots. This entailed four different student groups in the four content areas taught by the two content-area teachers. However, a more long-term vision of the school was to explore how co-teaching could contribute systematic support for the transition and inclusion of L2 students into mainstream education. Thus, from the second year onward, the co-teaching resources were scheduled to focus on one mixed group of 18 students as they started grade 7. Consequently, the co-teaching was expanded to involve more literacy-based content areas and the teachers of those content areas.

As a local of the area of the school, I learned about the ongoing co-teaching endeavor and contacted the principal to ask whether it would be possible to include their work in a study. After consent from the principal, the resource teacher and three content-area teachers volunteered to participate: the two content-area teachers initially participating as well as the group’s homeroom teacher who taught two of their content areas. In total, these three content-area teachers partaking in the study were responsible for five co-taught content areas: biology, geography, physics, chemistry and health education. The participants share previous experience of between 10 and 15 years in the same school.

The study follows the general ethical standards approved by the Finnish Advisory Board on Research Integrity (2019/2023) as required by the host university. Before data gathering, written consent was obtained from the informants, and they were assured in writing of their anonymity as well as their right to withdraw from the study at any point. Data was then gathered through teacher–researcher conversations (Hollingsworth & Dybdahl, 2007).

Teacher–researcher conversations allow researchers to be responsive to unfolding stories, while aiming for a more equal and collaborative researcher–informant relationship (Hollingsworth & Dybdahl, 2007; Clandinin & Connelly, 1996). Rogan and de Kock (2005) report on a narrative study where the change from a structured inteview to a more flexible conversational style, where the informants were encouraged to contribute narrations in their own preferred manner rather than being restricted by the fact finding structure set by the researcher, enabled informants to participate more comfortably. Although researchers cannot step out of their role, Rogan and de Kock suggest that a more conversational style where researchers provide input to some extent can enhance motivation for informants to share their experiences. In this case, I opted to comment on specific experiences that the teachers had related, e.g. that a specific way of working corresponds to a specific model in the co-teaching literature. As it became clear that my intention was not to silently evaluate their effort, they were also open with their process of trail and error. Thus, I suggest that this method of data gathering served to add to the rigor of the study (cf. Hollingsworth & Dybdahl, 2007). Furthermore, the group conversations allowed the colleagues opportunity for reflections on and beyond their co-teaching process to learn from it even further (Edwards, 2017; Sarasa, 2015; Schwarz, 2001). By being aware of what experiences to share as well as observing researcher input in the ensuing analysis, the teacher–researcher conversations provided access to rich data sets, which outweighed potential drawbacks.

The seven teacher–researcher conversations were conducted 2018–2020 and comprise close to 340 minutes of recorded interaction. They were conducted with the informants in different constellations in accordance with Table 1. Unless I opted for specific informants, these constellations varied according to how teachers made themselves available for the times of my visits, and there was nothing in the conversations to suggest that this availability was to avoid taking part with specific colleagues. The approach to gather data both from individuals and in pairs and groups contributed to the validity by mitigating against specific individuals influencing other informants, a disadvantage often connected to data collection in groups. An advantage with group conversations is that they allow for possibilities for increased dynamics and in-depth discussions, as other participants’ input can evoke memories and add further perspectives (cf. Keene et al., 2016).

Table 1. Timeline, constellations and length of teacher-researcher conversations.

Nov 2018	Resource t. + Biology/Geography + Physics/Chemistry	77 min
March 2019	Resource t. + Physics/Chemistry	39 min
March 2019	Health education	41 min
March 2019	Biology/Geography	30 min
Nov 2019	Resource t. + Biology/Geography + Physics/Chemistry	51 min
April 2020	Resource t. + Biology/Geography	60 min
April 2020	Health education	42 min

I visited the school for all data gathering sessions except the two final ones, which were conducted remotely and audio-visually recorded through Zoom. The sessions at the school were audio-recorded and took place in an available classroom. During the initial session, to guide the conversations towards areas of interest for the study (cf. Hsieh & Shannon, 2018; Rogan & de Kock, 2005), the teachers were provided with a set of topics on slips of paper. They could address these topics in their preferred order: why co-teaching/how the process started; previous experiences of co-teaching; positive experiences; challenges at different stages; ‘good advice’/prerequisites for co-teaching. The topics were open enough to allow for addendums as well as flexibility concerning the weight they wanted to put on the topics. Each following conversation was further guided by the experiences gained from the on-going process. They were primarily teacher-led in that the teachers were encouraged to bring forth both positive and challenging issues they were experiencing at each point in time. However, for each session I had also listened through and/or read transcripts from the previous ones to be able to explore specific issues further if need be.

3.3. Data analysis

The recordings were transcribed verbatim in the weeks following each session. After transcription, the totality of field texts went through an analysis of narratives (Polkinghorne, 1995). In this process, I read the field texts repeatedly to enable initial coding highlighting aspects pertinent to the research questions. This process went on until the codes could be forged into over-arching themes emerging as resonant threads across the texts. While the themes were inductively derived, I recognize my analysis as at least partly guided and informed by previous research towards similar findings (cf. Vaismoradi et al., 2013). Throughout this thematic analysis, interpretations become an issue of probability and coherence (Riessman, 2008): despite the recognition of reality as constructed, researchers must critically examine their interpretations in relation to quality criteria and previous theoretical perspectives.

The thematic analysis was followed by a narrative analysis (Polkinghorne, 1995), which refers to producing explanatory stories, i.e. coherent narratives of differing form and length, as the research text. Inspired by Rytivaara et al. (2019), following Clandinin and Connelly (1996) and Connelly and Clandidin (1999), I chose word images as the research text, an unconventional representational form which challenges the traditional form of academic writing. Word images constitute a distillation constructed from field texts, as researchers take words and phrases and shape, or re-story, these into word images representing the findings. Word images are sometimes called found poems (Patrick, 2016), as researchers can be said to compose a poem based on their analysis. The images can be chronological or non-linear, based for example on the main themes of the narrative story.

Through a recursive interpretative process of re-storying while frequently checking back to confirm the context in the field texts, verbatim utterances were rearranged to explain and illustrate the over-arching themes. As the intention was not to perform a comparative analysis between individual teachers or partnerships, the co-teaching experiences were explored as co-constructed over and throughout the data sets, letting the teachers’ interacting voices form confluent word images (cf. Aveling et al., 2015). Thus, although based entirely on participant utterances found in the field texts, the creation of word images is, as the analytical process related above suggests, a creative interpretive process (Clandinin & Connelly, 1996). Table 2 provides an example of transcribed field text and corresponding lines in a word image connected to the over-arching theme Daring to share.

Table 2. Example of transcribed field text and corresponding lines in word image.

Field text	Lines 1–4 for theme Daring to share
First we discussed what can co-teachers do, what different tasks could be brought in? And that was of course literacy aspects for second-language students. And literacy aspects for everybody. We’re going to add to what can be achieved, not take away.	1 what can co-teachers do? 2 literacy aspects for second-language students 3 literacy aspects for everybody 4 we’re going to add, not take away

Vaismoradi et al. (2013) address challenges regarding how to evaluate and achieve rigour in qualitative studies. Among these is the subjectivity inherent in latent interpretations. A key question to consider is whether an informant has reason to highlight a certain message or whether other data point to the same direction. To strengthen the validity of the study, a draft of the findings was shared with the participants. No objections were raised regarding the accuracy of interpretations. Vaismoradi et al. conclude that the quality of a study depends on whether enough insights can be gained from findings to increase understanding and inform practice.

4. Results

In the analysis of narratives, four over-arching themes were derived from the teachers’ (individual and shared) knowledge construction and role formation during co-teaching: daring to share, extended expertise, fine-tuning roles and increased well-being.

Below, each theme is first opened up and contextualized before being represented through word images. Throughout, the shared co-teaching experience unfolds as following the chronology of the story regardless of when and by whom events had been related in real time. The images do not distinguish individual voices, except utterances by the resource teacher that are bolded to clarify that they reflect a different role than the content-area perspective.

The numbers in parentheses signify corresponding lines in the ensuing word images for reference. As the word images constitute brief distillations of verbatim utterances from the narrations, the findings can be better grasped by first reading each description in their entirety, then the corresponding word images, before referring to specific lines. The word images have been translated into English. When deemed necessary for understanding, explanatory words have been added in square brackets. Underlining signifies speaker emphasis.

4.1. Daring to share

The first theme communicates the need to share – both individual knowledge landscapes and classroom space – to enable successful co-teaching. Every co-teacher brought their expectations to the process. Starting from negotiations of educational priorities and what they each were to contribute, they found a shared vision and common goals (1–10).

The initial negotiations between trusted colleagues were swift as all participants were open to co-teaching, even eager for inspiration after having worked alone in their classrooms for many years (11–13). One of the content-area teachers expressed difficulty concerning the ability to let go of complete control; however, these challenges were not attributed to the partnership (14). Repeatedly, they all came back to the need to trust the process of mutual adjustment and adaptation: being ready to hand over responsibility and trust as well as being prepared to fail (15–16). Mutual trust was also a prerequisite for allowing the co-teaching partner to ‘step in’ during instruction, e.g. to ask for more in-depth explanations or clarifications. This was done with the two-fold aim of supporting the students’ learning as well as modelling how to ask questions and take part in the discussion of lesson topics. Rather than becoming embarrassed or offended, the teacher being ‘questioned’ needed to find alternative ways to express the instructional content. It also took some courage to be the one to step in to ask for clarifications (17–20).

1. what can co-teachers do?

2. literacy aspects for second-language students

3. literacy aspects for everybody

4. we’re going to add, not take away

5. teach them how to read texts, explain words

6. what texts work for close reading

7. if I can’t understand it, there’s something wrong with the text

8. physics and chemistry are very exciting

9. challenging to understand

10. super interesting to use for close reading

11. I feel rather lonely in my subjects

12. several years of going through the motions

13. my least favorite courses become more motivating

14. a bit hard to let go of all control

15. a need to give up the belief in one’s own way being the best way

16. maybe none of us thought a failure really was a failure

17. when you react to something I said:

18. ‘did you understand what [N.N.] said now?’

19. ‘okay how can I say this differently?’

20. provide a model for how to discuss content

4.2. Extended expertise

As none of the participants had previous experiences of co-teaching, they started out by slowly exploring how to collaborate in different student groups mainly decided according to the schedule of the resource teacher (1–2). The creative experimentation resulted in viable experiences, but sometimes also prompted a change in direction and new priorities for the instructional goals (3–6). For example, the teachers argued that allocating resources for last minute remedial teaching with the purpose of aiding individual students to pass courses was less sustainable than teaching literacy strategies in whole-class contexts (7–9). A major change took place from the second year onward when the inclusive practices in a specific class starting grade 7 became the focus for the co-teaching (10–13).

The narration of their experiences to an outsider during the teacher–researcher conversations made the process more visible to the co-teachers themselves, making them realize how much they had accomplished and learned (14). They noticed that the co-teaching had resulted in added benefits beyond doubling the number of staff in the classroom. Literacy integration into content-area teaching was seen as enabling more successful learning for more students, not as a reduction of time spent on content-area content (15–17). For the content-area teachers, it enabled competence building particularly related to how to integrate literacy strategies into content-area teaching as well as additions to their toolkit in the form of ready-made literacy materials to be used in non-co-taught classrooms. The resource teacher learned more about how to adapt literacy strategies to specific content-areas (18–24). In addition, working with the same student group in several content areas enabled helping students to gain some cross-curricular insights (25–27).

1. when we started none of us knew anything

2. good thing we started with the teachers

3. ‘well let’s try this’

4. so many times we have said ‘let’s try!’

5. ‘well that didn’t really work’

6. but most of the time it worked out fine

7. last-minute focus on students at risk of failing

8. of course we helped them pass

9. not the best value for money

10. now our thought was ‘what group will benefit the most?’

11. now my main task is not only close reading

12. the advantage is that the group is one and the same now

13. second-language students integrated early

14. notice we have actually done a great deal

15. we reach new levels together

16. all students benefit

17. a bit different content focus

18. a bigger toolkit

19. what I can use

20. what I can do

21. eye-opener for what words are difficult

22. you have started explaining other words than terms and concepts

23. I had thought of some of the issues before

24. that the textbooks disappoint me

25. best case scenario I’m in the know:

26. ‘you’ve already done unit conversion in physics’

27. and I understand the chemistry content since I’m there

4.3. Fine-tuning roles

The theme of continuous fine-tuning of roles concerned each content teacher in relation to the resource teacher, and for the resource teacher in relation to all individual colleagues, which took a great deal of effort (1–2). For the resource teacher, this process was also in relation to the expectations of the students of what a teacher, or the teacher’s assistant, usually does in the content classroom (3–5). However, throughout the process, the goal was to enable student access to both teachers as much as possible despite their differing roles (6–7).

Specific models of co-teaching were more frequently favored in some content-areas, but there was also variation within content-areas depending on the aims and content of lessons. After more frequent and thorough co-planning initially, they were soon able to build on their shared experience to achieve interdependent team teaching. Also when making use of the one teach–one assist model, they built on each teacher’s expert knowledge, with shared responsibility e.g. for continuous assessment or for posing questions to each other during lessons for the benefit of all students. Thus, although with different roles, they became more equal as they were able to find solutions that could bring forward their respective expertise and adapt as they saw the need, not always having to plan whole lessons together in detail. Still, in making use of many models of co-teaching and developing and utilizing them for specific purposes, they also frequently intensified co-planning as needed (8–16).

While the co-teachers saw the experiences from the first year useful to build on during the later stages, the resource teacher in particular recognized a possible challenge in that previous experiences could create expectations that new co-partners might find difficultly in meeting, e.g. not automatically taking assisting role (17–22).

1. takes more energy than I thought

2. has to be somewhat of a chameleon

3. need to rethink one’s role

4. not afraid to show the students I don’t know the content area

5. my role is to provide other content

6. for each new group last year it took a while to accept my role

7. the [L2] students also ask for your help

8. we’re already comfortable in our roles

9. often teach in a dialogic manner

10. we’ve tried out some things that we know work

11. what I can and can’t do and what my resources are

12. notice individual development

13. formative assessment

14. sometimes I ask:’I did not get that, what about you?’

15. we often say, ‘we do what we usually do:

16. I plan literacy tasks if you plan for the subject content’

17. now when I go in with a new teacher

18. for better or worse I have experience

19. not only with different content areas [but] also with my role

20. I have some expectations, too

21. not a para-educator at the back of the class or zooming out

22. the other teacher may have completely different expectations

4.4. Increased well-being

The extended expertise contributed to the co-teachers’ well-being, particularly through their increased ability to differentiate according to individual student needs. This included being able to challenge and support students with high grades to develop beyond their level (1–6). Overall, shared responsibilities contributed to an increased sense of security. For example, an extra pair of eyes enabled more frequent directed observations for information that otherwise easily goes unnoticed, not only for assessment purposes, but also to be able to address a variety of student-related matters such as disciplinary issues when needed (7–8). Compared to the initial year, co-teaching complete courses provided a more comprehensive view that reduced stress as both planning and execution of instruction were facilitated (9–12).

In the spring of 2020, the start of nearly two months of remote teaching due to COVID-19, resulted in a need for re-orientation (13). Although planning became less smooth (14–16), the co-teachers were able to benefit from each other’s support also on-line and build on their previous experiences. After some initial confusion, they continued delivering instruction according to familiar co-teaching models applied to the remote setting. This brought much needed stability to the situation for both teachers and students, and the already familiar literacy assignments could support the more independent work connected to texts and textbooks that was a consequence of the remote teaching (17–20).

During this period, a new co-teaching partner came in – new as a colleague and in relation to the students. This meant further role negotiations for the resource teacher. In the unfamiliar remote context, the initial stage of building trust as well as finding a shared vision and common educational goals became more challenging (21–22, 24). However, the new co-teacher was open to letting the experienced resource teacher take the lead. Thus, the resource teacher and familiar ways of co-teaching became a connection to normalcy for the students meeting a new teacher on-line (23).

The conversations end with the resource teacher pondering unresolved issues around roles and educational goals: the inquiry into their practice continues (25–26).

1. we’ve learned how to lean on each other’s strengths

2. students with high grades really developed

3. they had to rethink the subject

4. maybe fail to answer if they can’t find it in the book

5. students with lower grades can excel

6. more used to struggling with a challenge

7. we share responsibility

8. you observe other issues more relations in different groups

9. we want all lessons and plan the whole course

10. so instruction won’t wobble

11. takes more planning for single lessons

12. less flow

13. at first we thought ‘now what do we do?’

14. a lot less discussions in class

15. co-planning more challenging

16. less opportunities for quick and efficient check-offs

17. it feels safer that we can co-operate

18. we do lean on what we’ve already learned

19. students read texts and construct mind-maps

20. summarize [content] as usual

21. works best between’ us’ with most experience together

22. new content-area teacher dares to hand over control

23. trusts me to include activities the students are used to

24. but no time to develop an’ us’ in the classroom

25. ‘what is my role?’

26. ‘what other competences should the students learn now?’

5. Discussion and conclusion

This narrative study explores a multiannual co-teaching initiative with the purpose of developing and enabling successful inclusive education for second-language (L2) students by including literacy strategies across the curriculum. The four co-teachers were experienced educators and long-time colleagues; however, the resource teacher was the only one with more extended competence in literacy strategies. Co-teaching was new for all of them. In the analysis, four interconnected themes were drawn from the co-teachers narrations: daring to share, extended expertise, fine-tuning roles and increased well-being.

Rytivaara et al. (2019) stress the importance for collaborative partnerships to be established and initial negotiations around classroom management, educational goals and teacher roles to be resolved as far as possible before teachers can efficiently implement co-teaching. In this process, relational aspects come to play, most importantly trust, to be able to share one’s knowledge landscape and classroom space (Cook & Friend, 2017; Dove & Honigsfeld, 2012, 2014; Pesonen et al., 2020; Rytivaara & Kershner, 2012; Rytivaara et al., 2019). The co-teachers stressed the need to give up the belief in one’s own way being the best way, being ready to trust and hand over responsibility as well as trusting the process.

Starting from the initial negotiations of educational goals, the search for and establishment of roles was prominent throughout. Connected to this was the central aspect of the construction of a shared knowledge landscape (Clandinin & Connelly, 1996). Cross-curricular collaboration when different paradigms meet is not without its challenges, and the articulation of largely tacit knowledge can be demanding. Rytivaara et al. (2019) suggest that experienced educators in general may have more fixed and even rigid knowledge landscapes, necessitating more effort to achieve well-functioning partnerships. Here, however, with a basis of trust already in place, the initial stage of the process was smooth. The co-operation can largely be defined by open-ended experimentation (Rytivaara et al., 2019) and competence building over time through constructive professional interactions. Thus, they were able to move directly into and quickly through the stages of accommodation and convergence in Davison’s (2006) model, and the second phase of symbiosis spin in Pratt’s (2014) model, before transitioning into the stage of creative co-construction. The majority of time, they seemed to be working as integrated and interdependent teams benefitting from each other’s expert knowledge in partnerships signified by trust (cf. Pesonen et al., 2020). The most complex task befell the resource teacher, who not only had to establish co-teaching relationships with all the co-teaching partners, but also navigate students’ expectations of teachers and assistants in the content-area classrooms. A new teaching-partner coming in during a period of remote teaching was addressed by the resource teacher as more challenging, as the necessary relation had not yet been built. Thus, this study adheres to those suggesting the importance of building collegial trust which allows participants to step out of their comfort zone and move forward beyond the initial stage(s) of the evolving partnership, as depicted in Davison (2006) and Pratt (2014). This suggests benefits of focusing resources for long-term collaborations, allowing both teachers and students the safety of not having to re-negotiate and rebuild working relationships more often than necessary.

The goal and vision of inclusive education guided the teachers’ work of utilizing their respective expert knowledge and forming their roles in the adaptation of suitable models of co-teaching. Learning about different models of co-teaching and how to implement them is often stressed in the literature as a prerequisite for a successful process (Brendle, 2017; Dove & Honigsfeld, 2012, 2010, 2014). In this study, however, most of the established models of co-teaching developed organically from the start without theoretical underpinnings supporting the process. This included team teaching, often described as the most challenging model due to its interdependent style of execution in the classroom as well as the shared responsibilities all around (Cook & Friend, 2017). Team teaching can be seen as comparable to the advanced stages of creative co-construction in Davison (2006) and the interdependent symbiotic phase in Pratt (2014). As team teaching allowed for efficient integration of LAC, it continued to be used on a regular basis. Also the model of one teach–one assist was purposefully used, e.g. for observing as part of formative assessment or posing clarifying questions to each other to aid understanding and model classroom discussions (Alcalà Arxé et al., 2020; Cook & Friend, 2017); thus, not putting one co-teacher in an inferior position in the way this model has often been used (Brendle, 2017; Creese, 2005). The way the co-teaching models were implemented contributed to establishing a collective responsibility (Pesonen et al., 2020), communicating to students that the teachers’ different areas of expertise were equally important and not seen as intrusions into each other’s domains (cf. Dove & Honigsfeld, 2014). Thus, the process was not dependent on the initial mastery of teaching models, rather highlighting co-teaching as an exploration based on trust and the ability to communicate around roles and responsibilities in a changing knowledge landscape.

The purpose of developing co-teaching to benefit student inclusion and learning was enabled through collaborative inquiry, i.e. co-teachers repeatedly inquiring into their shared teaching practice (Hargreaves & O’Connor, 2018). Although the development of more inclusive practices for L2 students was at the center, it was readily recognized that all students should be able to benefit from LAC. As relevant literacy strategies cannot be successfully applied to content areas through a simple model of ‘one fits all’, this process necessitated and enabled increased insights into both generic and disciplinary literacy strategies (Fang & Coatoam, 2013; Shanahan & Shanahan, 2008). Here, it befell the resource teacher to develop further insights into a variety of disciplinary literacies to be able to support the process. According to the content-area teachers, the literacy integration did not detract from content coverage, rather it enabled content instruction to be delivered in a manner that overall helped improve students’ comprehension of the content. Thus, the outcome of the co-teaching was increased professional development for integrating literacy instruction for differentiated student needs both in the co-taught groups and beyond (cf. Davison, 2006; Dove & Honigsfeld, 2012), a competence beyond what could be reached by collaborating with someone with the same area of expertise (Hollingsworth & Dybdahl, 2007). This added professionalism, besides being invigorating in itself, brought about an increased sense of well-being compared to the feeling of having to stretch too thin in relation to student needs (Cook & Friend, 2017; Dove & Honigsfeld, 2012, 2010; Gourvennec et al., 2022).

From the second year onwards, the decision was made to continue avoiding pullout alternatives unless for specific purposes. With shared classrooms, students can become ‘our students’ rather than ‘my students’, which was further strengthened by the realization of the co-teachers’ different but equal partnerships. Hargreaves and O’Connor (2108) discuss this aspect as collective responsibility among their ten tenants of Collaborative Professionalism (cf. playing out a shared knowledge landscape). The teachers exemplified how this had worked by the way students saw all co-teachers in the classroom as their teachers. These results further strengthen the suggestion by Creese (2005), Davison (2006) and Rytivaara et al. (2019) that the co-teaching role of a L2 teacher or a special needs teacher can be broader, serving all the students in the classroom. Sharing responsibility, in particular regarding assessment, brought a sense of well-being, as did the different dynamic for interaction with another adult in class (Cook & Friend, 2017; Pesonen et al., 2020) as well as the mutual support during the period of remote teaching.

Scheduled or otherwise organized time for co-planning is stressed in the literature as another requirement for the successful implementation of co-teaching. Although there is no doubt teachers’ work benefits from being allocated enough time for planning, and more crucially so in the initial stages, Dove and Honigsfeld (2012) point out that scheduled co-planning is not a cure-all in the intricate process of co-teaching. In this study, the amount of time and effort required for co-planning varied depending on the phase of the process, the teachers’ familiarity with the content to be taught and student needs. The reasons for planning individually were pragmatic, to save time, not e.g. connected to cooperation difficulties, as discussed in Brendle (2017). When need be, such as for further developing and purposefully utilizing different models of co-teaching and their respective roles of expertise, co-planning was intensified. The content-area teacher who came in during the second year attested to a need for more time for co-planning compared to the others. The teacher in question openly discussed this need primarily as a personality trait (cf. Pesonen et al., 2020). Grounded in the trust he/she put in his/her co-teaching partner, his/her solution was to start taking a meta-perspective on his/her own reactions as a reminder of the necessity to let go of complete control and the need for detailed planning.

It is often stated that for successful implementation, co-teaching should be voluntary (Dove & Honingsfeld, 2012). This aspect cannot be addressed in this study, as all co-teachers taking part were eager to try out co-teaching as a way of learning both from and with each other for the benefit of their students. A limitation concerns the challenge to account for the importance of individual characteristics, e.g. the tenacity required by the resource teacher in establishing several co-teaching relationships (cf. Pesonen et al., 2020) alongside developing extended competence in disciplinary literacy, or the possible benefits for the co-teaching process from one of the participants gaining such knowledge and experience. In situations where partnerships are less established than in the current study, preparedness for less smooth co-operation is also warranted (Gourvennec et al., 2022). It further needs to be recognized that the diverging constellations in which the teacher–researcher conversations were conducted might have affected dynamics in conversations and, thus, possibly also the results.

Overall, it should be noted regarding limitations that results from small-scale studies from one context cannot be generalized or automatically transferred to other settings. Still, as suggested by Vaismoradi et al. (2013), smaller case-studies offer possibilities for other contexts to draw implications from, as well as adding to the total body of research in the field in dialogue with previous research. Although constituting a small sample, this study is suggested to offer some valuable perspectives through its longitudinal exploration into co-teaching between experienced teachers above primary-school level and with focus on L2 students, which to date have been researched to a limited extent. In particular, it highlights long-term co-teaching as a viable tool for professional development as different expertise meet to focus inclusive education, and, eventually, for building whole-school capacity, to teach literacy strategies across the curriculum. It further strengthens the suggestion by Rytivaara et al. (2019) that co-teaching is seen as a space that is actively created rather than as a fortunate coincidence (see also Pesonen et al., 2020). However, further inquiry is needed into the conditions and support strategies that enable such co-creation, particularly conditions pertaining to relational trust and necessary expertise.

Disclosure statement

No potential conflict of interest was reported by the author(s).