Introduction

For some time after the introduction of Content and Language Integrated Learning (CLIL) in European schools, it was hoped that students would naturally improve their L2 skills through increased input alone (compare Dalton-Puffer, 2007), which frequently resulted in programmes where a regular L1 lesson was simply translated into the CLIL language. In Austria, such programmes were called “English as a medium of instruction”Footnote 1-programmes (Abuja & Heindler, 1993) where the foreign language was primarily seen as only a vehicle of learning and not related to the learning goals of the lesson per se. However, it soon became clear that for students to reap the full benefits of the CLIL approach, the language aspect would need to be addressed more explicitly and in connection with the content goals of the lesson (Llinares et al., 2012; Moore & Lorenzo, 2015; Cammarata & Cavanaugh, 2018). Ways in which this integration of content and language goals could be conceptualised have since become a primary concern in CLIL research, with the recent focus on literacies (Coyle & Meyer, 2021; COST, 2022) promising especially useful insights into how to effectively integrate content and language goals in CLIL practice.

A small number of earlier studies (Bauer-Marschallinger, 2022; Mathieu, 2021; Young, 2015) have shown that paying attention to how the integration of content and language goals is implemented in the classroom can lead to improvements in how language goals are made explicit in CLIL practice. The study presented in this paper pursues similar goals: the main aim here is to contribute to improving CLIL practice in the unique and arguably rather challenging context of technical colleges in Austria by providing novice CLIL teachers with a lesson planning tool tailored to the needs and demands of technical college CLIL. Whilst the study presented in this paper is part of a larger project with multiple research questions, the results presented here focus on ways in which subject literacy can be operationalised to make the connections between content and language explicit to novice CLIL teachers.

Literature Review

This study is set in technical colleges in Austria, a vocational upper secondary school type where English-language CLIL programmes tend to focus on the specialised technical content subjects unique to these schools (Austrian Ministry of Education, 2011). These technical content subjects, like recycling technology or control engineering, are often highly specialised and advanced, meaning that cognitive demand in these subjects is high both in terms of content and the specialised language used to verbalise this content. Whilst this is already the case for regular L1 lessons, these language demands often remain invisible to technical college teachers until CLIL and the addition of English as a foreign language shine a spotlight on them.

From a linguistics perspective, it might seem that these language demands are not as far removed from content teaching as technical content subject teachers might think: Schleppegrell (2004) stresses that any kind of “[s]chooling is primarily a linguistic process” (p. 1) as learning always occurs through language. From this point of view, content learning is characterised by working with the academic text types that constitute the subject, which not only involve understanding the register and genres of the field, but also grammar and vocabulary. The principal challenge of schooling is then to understand and apply “the genres of schooling and the purposes for which they are useful” (Schleppegrell, 2004, p. 83), which allows students to participate in academic and professional discourse in future employment. It is therefore unsurprising that the importance of CLIL teachers understanding genre has been emphasised in current research (Cammarata & Cavanaugh 2018; He & Lin, 2020). Yet whilst genre is a tangible theoretical concept that could indeed guide the development of content teacher language awareness, content teaching in Austria follows a largely oral tradition (compare Dalton-Puffer, 2007), which limits the transferability of research that is predominantly conducted on written educational genres. It therefore stands to reason that genre theory might be too far removed from the practice of technical content subject teachers and would thus exceed the limits of what they can realistically work with in the context of technical colleges.

Dalton-Puffer’s (2013) cognitive discourse functions (CDFs) present a more general approach towards the integration of content and language teaching methodologies. CDFs are the patterns created by the recurrent requirements of dealing with content knowledge, for the purpose of learning it, representing it, and communicating about it (Dalton-Puffer, 2015). By creating a natural intersection between the didactic traditions of both content and foreign language teaching, CDFs help close the gap between traditional and language-aware content teaching. The CDF construct has seven types of functions: categorise, define, describe, evaluate, explain, explore, and report (Dalton-Puffer, 2013), all of which can serve as a reference point when designing language-aware content tasks for a CLIL lesson. Yet whilst the CDF construct has proven to be successful in making language learning goals accessible to non-language teachers in other CLIL studies (compare Bauer-Marschallinger, 2022; Hasenberger, 2018), the teachers in this study were unconvinced of the suitability of the CDF construct for their context. They felt that CDFs could not fully capture the range of tasks covered in their technical content subjects and advocated for the use of a theoretical construct that was closer to their current practices.

The theoretical construct most entrenched in technical college teaching practices is Bloom et al.’s taxonomy (1956), which is already familiar to technical content subject teachers from their teacher training course. In their seminal work on educational objectives, Bloom et al. detailed how to formulate learning objectives that allow for outcome-oriented teaching and the designing of appropriate testing and assessment measures. They achieved this through proposing a model of the cognitive domain consisting of six categories: knowledge, comprehension, application, analysis, synthesis, and evaluation. These categories form a cumulative hierarchy, “based on the idea that a particular simple behaviour may become integrated with other equally simple behaviours to form a more complex behaviour” (Bloom et al., 1956, p. 18). This construct has proven highly influential in education, although it is Anderson and Krathwohl’s revised taxonomy (2001) and their categorisation of remember, understand, apply, analyse, evaluate, and create that is most widely used today. Yet whilst Bloom’s taxonomy can certainly help teachers when it comes to addressing the cognitive demands of a task and is used as such in technical college test design, it does not take the corresponding language demands into account.

The ways in which the cognitive demands of a task can be verbalised become easier to pinpoint when we turn to the concept of literacies, which can broadly be defined as the social practices embedded in the use of “literacy skills and knowledge, for socially constructed purposes, within specific sociocultural contexts” (Green, 2020, p. 13). Whilst the current interest in the concept of literacies has led to manifold definitions of subject, disciplinary, and pluriliteracies, the shared idea behind many of these constructs is that progress in a discipline means gaining both the procedural knowledge associated with the field and the academic and specific language skills needed to address subject-specific tasks. This idea is congruent with the language demands of technical content subject CLIL, as students need to solve tasks related to the discipline and work with advanced specialised texts in a foreign language. They also need to develop multimodal proficiency, as technical disciplines represent content not only through language but also through different representation modes like graphs, diagrams, and mathematical formulae (compare Airey, 2009). Coyle and Meyer’s (2021) pluriliteracies teaching for deeper learning (PTDL) model not only emphasises this demand for the development of subject literacies and plurimodal skills but also takes into account the need for learners “to be able to successfully and adequately communicate knowledge across cultures and languages” (Coyle & Meyer, 2021, p. 41). This conceptualisation of pluriliterate language use seems well-suited to support the primary goal of CLIL programmes at technical colleges, which is to prepare students to participate in technical disciplines on a global scale. Therefore, the PTDL model might usefully provide some guidance in addressing problems of practice arising from the implementation of technical content subjects through CLIL. However, the PTDL model comprises a complex integration of different skills and understanding that in its entirety might prove non-transparent to technical content subject teachers who may have a less developed awareness of the role language plays in their subjects. For the purposes of this study, the focus will therefore be on the plurilingual element of the PTDL model, which is arguably more accessible to novice CLIL teachers.

The Study

Context and Research Goals

The present study was carried out at two different technical colleges in the provinces of Lower Austria (School A) and Upper Austria (School B) during the 2021/2022 academic year. Technical colleges offer 5 years of vocational upper secondary education, starting at grade 9 (student age: 14/15) and ending with partly-standardised leaving examinations at grade 13 (student age: 18/19) (OeAD, 2022). Students receive a combination of general education and specialised training in different vocations, which allows graduates both to pursue tertiary education and to practice a number of professions that are legally protected under commercial code in Austria and other EU member states (Austrian Federal Ministry of Education, 2018). If they fulfil certain criteria, graduates of technical colleges are also entitled to bear the qualification title of “Ingenieurin/Ingenieur” (“engineer”), which has been equated to the education level of a bachelor’s degree in a 2017 national regulation (Federal Chancellery of the Republic of Austria, 2017). For this reason and due to the high skill level students generally graduate with, technical colleges enjoy great prestige in Austria.

Yet whilst the technical training students receive is generally considered excellent, the intensive curriculum at technical colleges leaves little space for foreign language education: throughout the majority of grades, only two weekly contact hours are allocated to foreign language teaching, with English often being the only foreign language on offer (Austrian Federal Ministry of Education, 2015). This has left graduates unprepared for the foreign language demands of a globalised labour market and excluded them from the academic discourse in technical disciplines, which is largely conducted in English. To address these shortcomings, English-language CLIL was made mandatory at all technical colleges in 2011 (Smit & Finker, 2016). The main incentive for this curricular change was the option of increasing students’ exposure to English without further overloading their already full schedules; instead of trying to fit in more designated EFL lessons, the curriculum now stipulates that students in grades 11 and 12 (student age: 16–18) ought to receive 72 CLIL lessons per school year, whilst students should receive 36 mandatory CLIL lessons in grade 13 (student age: 18/19) (Smit & Finker, 2016). The downside of this curricular change is that schools are obligated to fill these CLIL slots, regardless of whether enough teaching staff feel competent teaching CLIL (compare Wolf, 2016); whilst CLIL programmes at technical colleges are more regulated than CLIL teaching elsewhere in Austria, many teachers would wish for more support from educational authorities in addressing this perceived or actual lack in skills (Smit & Finker, 2016). This leads to a host of issues at technical colleges, especially amongst the staff most pressured to participate in a school’s CLIL efforts, since the Austrian Ministry of Education recommends focusing on specialised technical subjects rather than on general academic subjects (Austrian Federal Ministry of Education, 2011). The former subjects are typically taught by engineers or technology experts without a teacher education degree (Wolf, 2016). Many of these teachers start their CLIL journey without having received formal training in the targeted medium of instruction or in language teaching methodology (Hüttner et al., 2013). This often results in a perceived or actual lack of target language proficiency (Wolf, 2016) and lower levels of general language awareness, making the integration of content and language goals in CLIL lessons even more challenging. Whilst these issues are now increasingly addressed in Austrian CLIL studies (compare Bauer-Marschallinger, 2022), there is a decisive lack of research into practical, design-based solutions to the challenges CLIL teachers at technical colleges are faced with. The primary goal of this study is to close this gap through the development of a lesson planning tool for CLIL lessons that guides novice CLIL teachers through the process of integrating content and language goals, thus improving CLIL practice at technical colleges.

Methods

The study was carried out within a design-based research (DBR) approach. DBR is a relatively recent, mixed-methods approach in educational research with the express “intent of producing new theories, artefacts, and practices that account for and potentially impact learning and teaching in naturalistic settings” (Barab & Squire, 2004, p. 2), thus effectively closing the gap between theory and practice. It is similar to action research, but pursues additional goals: where the focus in action research is on addressing local needs (McKenney & Reeves, 2018), DBR takes this one step further by formulating new, overarching theories from the knowledge gained in solving the local problem (Barab & Squire, 2004). DBR is also conducted by co-design teams consisting of local practitioners and educational researchers rather than solely by the teachers (Anderson & Shattuk, 2012), thus circumventing issues like researchers’ bias and time constraints whilst also combining strengths like in-depth knowledge of the institutional culture with theoretical expertise. This type of partnership makes DBR an especially suitable methodology for this study, which seeks to provide theoretical guidance to new and inexperienced CLIL teachers whilst drawing on their expertise as content teachers (compare McKenney & Reeves, 2018).

Several different models for conducting design-based research have been proposed (e.g. van den Akker, 1999; Anderson & Shattuk, 2012; Mintrop, 2016), all describing an iterative design and evaluation process in slight variations. McKenney and Reeves’ (2018) generic model for conducting design research in education connects these core characteristics of DBR to instructional design and curriculum development theory, allowing for the flexible, holistic approach to design research needed for this study. The model consists of (1) an analysis and exploration phase, (2) a design and construction phase, and (3) an evaluation and reflection phase which outline the general stages of a DBR project but may also be revisited across a series of research cycles (compare McKenney & Reeves, 2018). As the research cycles progress, the theoretical understanding of the subject grows and the intervention design matures and is repeatedly implemented (McKenney & Reeves, 2018). This project followed the same, non-linear research process whilst unfolding over the course of three consecutive project stages that were adapted from McKenney and Reeves (2018) to better reflect the structure and timeline of this study. Table 1 illustrates this research design.

Table 1 Research design
Full size table

The final evaluation and reflection stage (compare McKenney & Reeves, 2018) is still ongoing at the time of writing. Whilst a second refined version of the lesson planning tool is presented in this paper, the more thorough analysis and evaluation currently underway might precipitate further changes and a final refinement of the design. A corresponding distilment of effective design principles will then allow for an overarching theorising on CLIL lesson planning tools and their potential role in CLIL teacher professional development.

Participants

As is typical for DBR projects, this study was characterised by a close collaboration and working partnership with the practitioners involved. An experienced technical content subject CLIL teacher as well as an EFL teacher and CLIL teacher educator at school B provided input during the needs analysis phases and at the preparation and development stage, and acted as consultants whenever necessary during field work. The main research partners, however, were Clemens (school A) and Werner (school B), the two teachers who piloted and co-constructed the lesson planning tool throughout the 2021/2022 academic year. These teachers were looking to incorporate CLIL methodology into their content subject teaching for the first time during that academic year and agreed to form a close collaborative partnership in order to close the gap between theory and practice in the area of CLIL lesson planning. This partnership was devised as the intersection between two different disciplines: whilst the practitioners provided technical expertise, their pedagogical content knowledge as content subject teachers, and their insights into the research context, I contributed knowledge of foreign language teaching methodology and CLIL theory gained through my experience as an EFL teacher and CLIL researcher. The outcome of this study in the form of a CLIL lesson planning tool for technical content subject teachers at technical colleges hinged on the success of this collaboration, where these two different skill sets were brought together. The practitioners did therefore not fill the role of research subjects but contributed to the design process as active participants and research partners (compare McKenney & Reeves, 2018). Table 2 gives an overview of the partner teachers’ professional backgrounds.

Table 2 Partner teachers
Full size table

During the field work phase, both teachers were enrolled in a 60 ECTS in-service qualification course at a college of education, which is a pre-requisite for lateral entrants into the teaching field wanting to teach at a vocational upper secondary school in Austria. For both of them, this represented their only formal teacher training.

Results and Discussion

Whilst the collaborative setup of the study was crucial for its success, the research partnership with Clemens and Werner organically evolved throughout the study. Since the consideration of the language dimension of their content teaching had never played a substantial role in their lesson planning prior to the study, meeting expectations in that regard was their biggest initial concern. In the early feedback sessions, both Clemens and Werner approached their CLIL teaching from their perspective as content subject teachers and sought my support in my role as former EFL teacher who could help identify potential language hurdles posed by the content they had prepared. As the study progressed and their understanding of CLIL methodology grew, we moved from collaborators with separate areas of expertise to partners considering all aspects of planning a CLIL lesson together. By the second research cycle, both Clemens and Werner were more confident in their role as CLIL teachers and needed less input concerning their lesson planning; we redirected our efforts towards refining the lesson planning tool for optimal use by other novice CLIL teachers of technical content subjects, a process in which the unique expertise of both teachers was invaluable. These changes to the lesson planning tool present the primary output of this study.

CLIL Lesson Planning Tool

This CLIL lesson planning tool draws on both Tedick and Lyster’s (2020) instructional design for immersion and dual language classrooms and on Meyer’s (2010) CLIL-pyramid unit template, which have been adapted for the specific demands of CLIL teaching at technical colleges in Austria. Tedick and Lyster’s (2020) lesson planning tool is a comprehensive resource that helps the teachers involved in immersion and dual language programmes, who are primarily language teachers, consider diverse student needs, and incorporate secondary learning objectives like cultural and social goals into their lesson planning. However, expecting the same level of diversification from technical content subject teachers who were not specifically trained and had limited guidance available could lead to frustration amongst those teachers. To avoid overburdening these teachers, significant adaptations to Tedick and Lyster’s (2020) lesson planning tool were necessary.

The pilot version of the lesson planning tool consisted of three main resources: a unit planning template, a lesson planning template, and a scaffold for writing language objectives. The lesson planning template proved to be the most important resource in Clemens’ and Werner’s lesson planning process and will therefore be the focus of discussion moving forward. The template is divided in different subsections which largely stayed the same throughout the design process, despite numerous modifications and adaptations to the lesson planning tool at large (see below). The first section is dedicated to general information about the lesson topic, the larger module the lesson belongs to, as well as the time frame and student number. The latter is especially important at technical colleges, where teaching often has to be adapted to large class sizes of up to 36 students. The next section, “Learning objectives,” features subsections for content and language learning objectives as well as for the related language that students will need to use; the latter section was divided in “content-obligatory” and “content-compatible” language in the pilot version of the tool, but this distinction was ultimately removed. In the first research cycle, this was followed by a separate section for the “Materials and media” used throughout the lesson, which was later complemented by additional sections for “Subject-specific modes” and “Scaffolding.” The final two sections are dedicated to the “Lesson sequence” and to “Evidence and assessment,” allowing the teachers to detail the individual activities making up the CLIL lesson as well as their plans regarding formal assessment procedures. The former section initially featured a subsection for “Scaffolding” on the level of individual activities, which was later restructured to enclose the entire lesson. The updated version of the lesson planning template is illustrated in Fig. 1 and demonstrates all of the changes made to the pilot version over the course of two research cycles.

Fig. 1
figure 1

Lesson design template

Full size image

In the first research cycle, the tool also featured an input resource on cognitive discourse functions (CDFs), the theoretical construct that also informs Tedick and Lyster’s (2020) lesson planning tool. However, both partner teachers reported dissatisfaction with using the CDF construct during the second round of interviews. Clemens admitted that “the terms of the CDFs are […] still not tangible for [him]” (Clemens, mid-stage interviewFootnote 2) and suggested that the CDF construct might work better in a humanities discipline like literature studies, whilst a construct like Bloom et al.’s taxonomy (1956) made more sense in the context of his teaching. Werner pointed out that he was already successfully incorporating Bloom’s taxonomy into his testing design and would find it logical to apply the same reasoning to planning his CLIL lessons; he also admitted to finding the distinctions between CDFs and Bloom’s taxonomy hard to pinpoint and confusing. It is important to note that at this point in the study, the teachers were only slowly developing an awareness of the role language played in their teaching. This might explain their need to take recourse in Bloom’s taxonomy, a construct they were already incorporating into their teaching and which, on a cognitive level, appears similar to the CDF construct given that the latter demands deeper understanding of literacies. We therefore decided to follow their suggestions, putting the use of the CDF construct on hold and instead incorporating Bloom’s taxonomy directly into the lesson planning template. The updated Anderson and Krathwohl (2001) version was used to make this design change, but the hierarchical element from the original 1956 taxonomy was requested by the teachers as guidance. Following a suggestion from Clemens, we also added a section where the teachers could define the aim behind making this specific lesson a CLIL lesson, as the levels of language support needed would differ according to whether content previously covered in the L1 was repeated or expanded upon, or whether entirely new content was introduced in English. Both the “CLIL lesson aim”-section as well as the “Bloom’s taxonomy”-field next to every separate activity helped the teachers determine how difficult an activity or set of activities would be on a cognitive level; if they concluded this level of difficulty to be high, they would seek to provide additional language support, for example in the form of extended vocabulary boxes or translations to German. Clemens occasionally simplified authentic texts he deemed above the language level of his students. As it had emerged from the findings that the content covered was typically even more interconnected than initially assumed, we also redesigned the “Scaffolding”-section to focus on measures that could be used throughout the lesson rather than on the level of individual activities. Finally, we removed further distinctions between content-obligatory and content content-compatible language items from the “Related language”-column, since the teachers involved in the study felt like making these distinctions was beyond their competence as non-language teachers. These design changes seemed to work well in the second research cycle, with both teachers remarking on how much easier and more natural it felt to plan CLIL lessons with the updated lesson planning tool.

By the end of the second research cycle, it had also become evident that both teachers were connecting the scaffolding measures they designed to the discipline-specific subject literacy their students had previously acquired in the subject. In recycling technology, the subject focus was on different recycling processes and the students had thus become very adept at working with and producing process diagrams; Clemens used this literacy feature to provide scaffolding measures bridging the differing cognitive and linguistic skill levels that were challenging his students, but still at the heart of the subject matter. Werner likewise recognised that he could use the mathematical formulae and the graphs that described the steps in solving a control engineering problem to help his students work with the research literature he was having them read, without further complicating the advanced technical language the students were presented with. The more they incorporated this approach into their lesson planning, the more smoothly they reported their CLIL lessons ran, an impression supported by the lesson observations. Two conclusions were drawn from observing this:

  1. 1.

    Scaffolding measures benefit from being multimodal, especially if the modes used represent how knowledge is typically constructed in the discipline.

  2. 2.

    The subject literacy the students have previously acquired in the discipline is therefore an essential part in advancing their learning in the L2 and has to be activated in order for them to make sense of authentic materials from their discipline in the L2.

This seemed to work well for both teachers, even though there are significant differences in the ways their respective CLIL subjects are structured, the subject-specific literacies were actioned, and how learning goals were defined: where recycling technology consists of several smaller, self-contained units, control engineering is primarily one extensive topic that is covered in depth. A similar approach has likewise been applied in yet another subject, namely multilingual mathematics education: in Prediger et al.’s (2016) instructional approach for meaning-making, the different languages and registers at the students’ disposal are being connected to symbolic and graphical representations typical for the field in order to help students construct knowledge and engage in meaning-making processes when learning mathematics through an L2. When reflecting upon these connections between different disciplines with Clemens and Werner, we found the similarities in how they had solved their scaffolding challenges despite the differences between their subjects too significant to disregard. We deliberated how we could operationalise their approach for teachers of other subjects and settled on incorporating literacy skills into the lesson planning tool. Whilst I drew on Coyle and Meyer’s (2021) conceptualisation of pluriliteracies to design this section, the teachers and I were cautious of overburdening novice CLIL teachers and chose to focus on the plurilingual and multimodal elements present in technical content subject CLIL teaching as an accessible pathway towards a deepened understanding of pluriliteracies. We decided against incorporating a new theoretical resource, similar to the original input on CDFs, into the lesson planning tool and realistically opted for a more implicit starting point where the concept of subject literacy is integrated with other steps in the CLIL planning process. We did this by inserting a section for “Subject-specific modes” into the lesson planning template, located between the sections for “Materials and media” and “Scaffolding.” This new section was incorporated to help guide CLIL teachers through the cognitive processes involved in designing teaching materials by first identifying the subject-specific modes and realisations of subject literacy involved in constructing the targeted content knowledge and then operationalising these modes as scaffolding measures to aid with the language aspect of a CLIL lesson. Whilst this approach only touches upon the complexity and potential of pluriliteracies and the PTDL model, it seemed vital to meet the teachers in their burgeoning understanding of pluriliteracies rather than alienating them with expectations they might be unable to meet at this stage in their CLIL teacher professional development. We therefore relied on combining the “Subject-specific modes”-section with Bloom’s taxonomy. Whilst the taxonomy helps the teachers establish the cognitive demand of their lesson and determine the level of language support needed to meet this demand, the emphasis on subject-specific modes and pluriliteracies aids them in developing appropriate scaffolding measures to provide this support. In this sense, the lesson planning tools also acts as a scaffolding measure for the teachers themselves: using the teachers’ implicit understanding of their own subject literacy to inform lesson planning and prompt a reflection on subject literacy practices, eventually guiding the teachers towards making these practices explicit in their teaching to assure learner progression. This third iteration of the lesson planning template is illustrated in Fig. 1.

Discussion

Both teachers were satisfied with their progress regarding the incorporation of language and subject-literacy specific learning goals into their CLIL teaching and reported feeling more secure in their CLIL teaching methodology skills. They also remarked positively on the role of the lesson planning tool, with Clemens detailing how using the lesson planning template was “becoming more of an intuitive process for [him]” (Clemens, mid-stage interview) and how he often referred to his planning when making ad-hoc decisions in class. Werner stated he would recommend the tool to his colleagues, but would probably feel the need to “help them in the beginning” (Werner, mid-stage interview). This observation illustrates the conception of the lesson planning tool as a measure that can support but not supplant guided CLIL teacher training. As such, it has proven to be a useful starting point when it comes to creating language-awareness in technical content subject teachers, which has in turn improved their ability to successfully integrate content and language goals in their CLIL teaching. The tool also managed to increase the visibility of subject literacy modes in technical disciplines and helped Clemens and Werner operationalise these modes to support their learners in reaching CLIL learning goals. However, it is important to note that the teachers have not fully embraced a pluriliteracies mindset yet and have just started developing a deeper understanding of the role language and literacies play in their discipline. It remains to be seen how this professional development is furthered through a prolonged engagement with the lesson planning tool.

The small sample size and qualitative nature of the data present further limitations to these findings. Whilst the lesson planning tool has been a valuable resource for CLIL lesson planning at technical colleges and in the subjects recycling technology and control engineering, it might not have the same effect in different CLIL contexts. Considering that the lesson planning tool was tailored to the exact needs of technical content subject CLIL and to those of the partner teachers, it seems likely that some adaptations will be necessary to optimise the tool for use in other contexts. The setup of the study also has some implications for the transferability of findings: whilst the close collaboration between myself and the partner teachers has proven invaluable in designing the lesson planning tool, my active presence in all stages of Clemens’ and Werner’s CLIL teaching is a factor that cannot be disregarded. Whether their progress was by merit of the lesson planning tool alone or partly also due to my involvement as EFL teacher, co-planner, and occasional CLIL coach is impossible to determine from the data. A follow-up study in which the lesson planning tool is integrated into regular CLIL teacher training could shed light on this. Finally, it should also be considered that both teachers chose to work with grade 13 classes consisting of students aged 18–19, who represent the most advanced group of learners in recycling technology and control engineering and are thus likely to have acquired a higher degree of subject literacy in these subjects than lower-level learners. It remains to be seen whether the promising results concerning the operationalisation of these subject literacy skills in CLIL teaching could be replicated with less advanced students.

Conclusion

Given the information presented above, it can be concluded that the study reached its main aim of setting novice CLIL teachers at technical colleges towards a path of integrating content and language goals in their CLIL teaching. However, the fact that neither of the teachers involved in this study had any experience with language teaching needs to be considered when drawing conclusions about their professional development over the course of this study. The teachers’ rejection of the CDF construct in the first research cycle could be interpreted as a symptom of their limited understanding of the role language plays in their discipline at the beginning of the study, which in itself is an important finding that has influenced the consecutive design process accordingly. In order to meet the teachers in their understanding about language and learning theory, subsequent versions of the lesson planning tool have connected Bloom et al.’s taxonomy (1956), a construct well entrenched in technical content subject teaching, with the concept of subject literacy, in particular the construct of pluriliteracies developed in Coyle and Meyer’s (2021) PTDL model. Unlike the CDF-based resource in the first iteration of the lesson planning tool, which sought to point out content and language intersections in all manners of disciplines and school-based genres, this approach allows the teachers to address the language needs of their technical content subjects more directly. Whilst the lesson planning tool has helped the two novice CLIL teachers involved in the study to implement a more content-and-language-integrated approach in practice, it is also a tool specifically tailored to the specific demands of technical content subject CLIL. Connecting Bloom’s taxonomy with the construct of subject literacies is the product of a close collaborative partnership with two novice CLIL teachers of technical content subjects and has only been optimised for use by CLIL teachers in this specific context, whilst the CDF construct has successfully informed CLIL designs in a wider variety of contexts (Bauer-Marschallinger, 2022; Hasenberger, 2018). This might limit the transferability of the lesson planning tool to other CLIL contexts. Further research is needed to explore these limitations and determine whether the lesson planning tool can produce equally favourable results in other CLIL contexts.