Integrated Language and Science & Technology Instruction: A Cognitive Task Analysis of the Required Teacher Expertise

ABSTRACT

Integrated language arts and science & technology (ILS&T) instruction can make learning more meaningful and improve student learning outcomes for both subjects. Although the literature has stressed that such integrated instruction presents pedagogical challenges for elementary school teachers, little is yet known about the specific characteristics of the required pedagogical repertoire. To gain a better understanding of the teacher knowledge and skills required for ILS&T instruction, a cognitive task analysis was conducted. This analysis resulted in a hierarchy consisting of the various required constituent teacher skills for ILS&T instruction. The cognitive task analysis also revealed what knowledge teachers rely on and identified the factors influencing the complexity of ILS&T instruction. The findings presented here contribute to our knowledge about what ILS&T instruction requires from teachers and can serve as a basis for the development of substantiated, 4C/ID-based teacher professional development trajectories. Furthermore, it provides valuable insight into the pros and cons of a relatively new, systematic way of analyzing teacher expertise.

KEYWORDS:

• Cognitive task analysis
• integrated curriculum
• language arts
• science and technology
• teacher expertise

Integrating the elementary school curricula of science and technology (S&T) and language arts can make S&T education more appealing for teachers and increase the time spent on teaching S&T (Appleton, 2003; Martin et al., 2012). Moreover, it can improve students’ learning outcomes in language arts and S&T education (e.g., Guthrie et al., 2007; Vitale & Romance, 2011). Although in some countries, science and technology are taught separately, in the Netherlands S&T is taught as an integrated school subject. Teachers are guided by a set of 58 generic core objectives, which describe the overall goals and leave room for elaborations. Two of these objectives address students’ development of inquiry and design skills (e.g., objective 45: “Pupils learn to design, realize and evaluate solutions for technical problems”). The Dutch S&T objectives broadly correspond to the curriculum guidelines of the Next Generation Science Standards (NGSS), where students develop knowledge of the natural environment and technological artifacts, and develop a critical investigative attitude (see, NGSS Lead States, 2013). Scientific and engineering practices such as engaging in argument from evidence and communicating information include linguistic and epistemic S&T aspects (Erduran et al., 2015; NGSS Lead States, 2013), which makes integrated language arts and S&T instruction (ILS&T) logical and relevant.

ILS&T instruction requires teachers to develop competencies that are rarely taught in initial teacher training programs. A recent study conducted in the Netherlands by Smit et al. (2018) showed that designing and enacting language-oriented science lessons was profoundly challenging for primary school teachers. In their study, teachers learned to design and enact science lessons in which they promoted the development of oral language skills. Teachers formulated science and language goals and designed activities that promoted these goals. This is not common for teachers in the Netherlands, as school subjects (like language arts and science) are usually taught separately. The approach that was advocated in this study differs slightly from our definition of ILS&T instruction, as this encompasses the broader domain of S&T and promotes a balanced focus on language arts and S&T rather than emphasizing one subject. Because of the complexity of ILS&T instruction, various researchers have stressed the importance of professional development (PD) to equip teachers for ILS&T instruction (Bradbury, 2014; Carrejo & Reinhartz, 2012; Fishman et al., 2017). Although various relevant teacher competencies for teaching ILS&T have been identified in previous studies, such as scaffolding (see, van Driel et al., 2018) and interaction skills (see, Hackling et al., 2010), a detailed analysis of the full pedagogic repertoire is yet to be explored. Furthermore, previous studies do not elaborate on teachers’ reasoning and decision-making during ILS&T instruction, although this is crucial for determining how teachers can be prepared in PD programs. In this study, a cognitive task analysis (CTA) of ILS&T instruction was conducted to give more substantiated insight into the implicit and explicit knowledge and skills required of K-6 teachers. CTA is a systematic approach used to gain a better understanding of how complex tasks are performed by decomposing complex tasks into constituent skills (Clark et al., 2008). This often entails a combination of observations and interviews to elicit information about the process of teachers’ actions. The outcome of a CTA is often a skills hierarchy, which can be used for the instructional design of training tasks, such as in PD programs. This approach differs from other frequently used methods for analyzing teacher skills and knowledge, such as self-reports and surveys, and could provide valuable new insights. CTA offers a substantiated point of departure for the design of a PD program. Previous research demonstrated that CTA-based instructions are highly effective (Costa et al., 2022; Tofel-Grehl & Feldon, 2013). Moreover, through CTA, factors can be identified that add to the complexity of the task, which help sequence learning tasks in the design of a PD program.

This study aimed to answer the following research questions: (1) What constituent skills can be identified from an analysis of expert teacher performance of ILS&T instruction?; (2) What teacher knowledge can be identified from an analysis of expert teacher performance of ILS&T instruction?; and (3) What factors can be identified that influence the complexity of ILS&T instruction from an analysis of expert teacher performance?

The rationale for ILS&T instruction

First, language arts (i.e., reading, writing, oral language) and S&T share many (meta)cognitive and intellectual processes, such as predicting, and assessing the quality of arguments based on evidence (Bradbury, 2014; Casteel & Isom, 1994). Second, language helps students communicate about abstract S&T concepts, transform global ideas into coherent knowledge, and articulate ideas (Cervetti et al., 2006; Rivard & Straw, 2000). Third, S&T education engages students in complex scientific reasoning, for which everyday language no longer suffices. Rather, students need to develop a discipline-specific register to allow for more accurate descriptions of concepts and phenomena (Snow & Uccelli, 2009; Ward, 2013). Fourth, S&T education offers a relevant and authentic context to develop complex language skills. The distinctive uses of language in texts are referred to as genres, such as procedures or reports (Christie, 2017; Yore, 2004). Learning to recognize these distinctive uses of language enables students to better understand and interpret information (O’Hallaron et al., 2015).

Prerequisite teacher competencies for teaching ILS&T

There was no comprehensive overview of the entire required repertoire of skills that teachers need for ILS&T instruction. The literature suggests several fragmented competencies that teachers need during ILS&T instruction.

Research indicates elementary teachers’ need to develop lesson planning skills for ILS&T instruction. Often, teachers need to (re-)develop instructional materials (Casteel & Isom, 1994; Stoddart et al., 2002) because it is challenging to find suitable material such as informational texts (Ford, 2004; Pearson et al., 2010). Lesson planning skills include formulating learning goals and (re-)designing instructional material to meet these goals (Allen et al., 2013). Studies have shown that during ILS&T instruction, teachers often tend to formulate learning goals that are activity-driven (e.g., “students will gain experience in carrying out hands-on science activities,” rather than “students develop inquiry skills,” as well as oral language and vocabulary (Fisher & Frey, 2014; Haug & Ødegaard, 2014).

Through the employment of interaction skills, teachers can promote effective classroom discourse and guide students’ conceptual understanding (Hackling et al., 2010; Haug & Ødegaard, 2014). Authentic, open-ended questions facilitate student learning (Chin, 2006; Oliveira, 2010). Offering appropriate cognitive feedback, pressing students to elaborate on their reasoning and relating student contributions to each other, redirects students toward learning goals and promotes students’ language use during S&T activities (Chin, 2006).

Language activities in S&T instruction such as argumentation practices, place high demands on students’ cognitive, metacognitive and social abilities (Cheuk, 2016). Scholars have therefore stressed the importance of scaffolding, especially for guiding discourse and argumentation in the context of S&T (Haug & Ødegaard, 2014; Kawalkar & Vijapurkar, 2013; van Driel et al., 2018). Scaffolding is a temporary support offered by a teacher and includes diagnosing what students need, determining and providing appropriate support, and a gradual shift in responsibility to the student as they become more skilled (van de Pol et al., 2010).

During instruction, teachers must monitor students’ progress as they work toward dual learning goals, which can be challenging (Choi et al., 2015). Cope et al. (2013) stated that often “the assessable object cannot be produced within the framework of test time” (p. 426). Researchers have argued that teachers should monitor students’ learning process through diagnostic, formative and summative assessments (Cope et al., 2013; Hackling et al., 2007).

Regarding teachers’ prerequisite knowledge, it has been argued that teachers need domain-specific knowledge of S&T and language arts, and knowledge of the appropriate instructional and pedagogical strategies to achieve subject matter specific learning goals (Glen & Dotger, 2013; Vitale & Romance, 2008). The concept of Pedagogical Content Knowledge (PCK) was introduced by Shulman (1986) and elaborated in various conceptual models (see Fernandez, 2014). Commonly, teachers’ PCK includes curricular knowledge, knowledge of instructional strategies, knowledge about students’ understanding, and knowledge of assessment (Ball et al., 2008; Magnusson et al., 1999). Researchers also underlined teachers’ knowledge of the interrelatedness of language arts and S&T processes (Casteel & Isom, 1994; O. Lee et al., 2013). Teachers should view language arts and S&T as synergistic domains, and recognize the language forms and functions appropriate for S&T topics (Stoddart et al., 2002). Finally, scholars addressed the need for teachers’ understanding of the nature of science, and the linguistic registers applied by scientists (Duschl & Osborne, 2002; Glen & Dotger, 2013).

In sum, the literature offered several suggestions for fragmented competencies that teachers need in the context of ILS&T instruction. However, it is imperative to acquire a holistic understanding of the entire pedagogic repertoire. From the perspective of whole-task learning, fragmented training of isolated competencies should be avoided (Frerejean et al., 2023). Further analysis was needed to elicit all the required competencies that teachers need to coordinate while teaching ILS&T instruction.

Cognitive task analysis

There is a need for a comprehensive analysis of the whole pedagogic repertoire required for ILS&T instruction. CTA offers a suitable methodology that allows researchers to gain insight into the cognitive and procedural information that experts use while performing a complex task. CTA examines how experts engage in goal setting and decision making, which are crucial aspects of complex tasks that require the integration of controlled and automated knowledge and skills (Clark et al., 2008; Schraagen et al., 2000). A combination of observations and interviews is used to obtain complete information about the process of teachers’ actions, because experts perform (parts of) the task in an automated way. By observing multiple experts, CTA aims to discern consistent patterns in their actions and cognitive processes. It is considered unlikely that multiple experienced professionals will leave out the same information consistently. Therefore, by analyzing their individual task performance, a comprehensive picture is expected to emerge of the cognitive components of a task (J. Y. Lee & Reigeluth, 2003). CTA results in an overview of the skills and knowledge required to perform a complex task, often in the form of a skills hierarchy, which offers a sound starting point for designing training programs (van Merriënboer & Kirschner, 2017). Moreover, CTA reveals which factors influence the complexity of the task, which can help sequence learning tasks from simple to complex (van Merriënboer & Kirschner, 2017).

Materials and methods

Research context

This CTA study was conducted in the Netherlands. Due to the focus on inquiry and design skills in the Dutch core objectives, all teachers in this study adopted an inquiry-based (Pedaste et al., 2015) or design-based pedagogy (Gómez Puente et al., 2011). Another important consideration is that the Dutch educational system prioritizes mathematics and language arts. Schools are held accountable for their performance in these subjects, which is not the case for S&T.

Participants: teachers and domain experts

Two groups of experts were consulted during the CTA. Although the definition and criteria for expertise vary across contexts, we adhere to the operationalization of expertise by Clark et al. (2008), who state that experts should have extensive experience with the task at hand, enabling them to consistently and quickly succeed at variations of the task. The first group of experts included teachers who had above-average experience in teaching ILS&T. Since ILS&T instruction is relatively new for teachers, few teachers who are highly skilled, and the participants of this study are considered pioneers. Nine teachers from regular elementary schools were recruited based on recommendations from a network of researchers, teacher educators and network organizations for S&T learning. Teachers’ expertise was checked in an unstructured interview with the first author, which focused on their experience with ILS&T instruction. The average age of the teachers was 43 years (SD = 7.94). They had an average teaching experience of 19.2 years (SD = 9.74). Most teachers had participated in PD programs focusing on S&T instruction. Two teachers (Christie and Alice) had participated in a PD program on the topic of ILS&T instruction (see, Smit et al., 2018). The contextual characteristics of the observed lessons are presented in Table 1. The second group of experts included experts in the domains of language arts and S&T, who were also selected based on recommendations from the authors’ network. This group of experts consisted of two teacher educators and four researchers, whose expertise included pedagogy of S&T and language arts and developing and implementing pre-service and in-service PD courses on ILS&T education. Informed written consent from the participants, and ethical approval from the ethics committee at the University of Twente was obtained before the start of the study.

Table 1. Contextual characteristics of teachers’ ILS&T lessons.

Name*	Grade(s) taught**	Number of students	Student characteristics	Learning goals
1. Christie	4	25	Low-SES, low language proficiency	Experience sound as a vibration. Understand the difference between high and low sounds and loud and soft sounds through experimentation. Develop domain-specific vocabulary.
2. Alice	Kinder-garten	23	Low-SES	Understand how mirrors work and what a reflection is. Develop oral language skills (conversations).
3. Sophia	Kinder-garten	16	Bilingual, low and high SES	Design an insect hotel. Develop domain-specific vocabulary. Develop oral language skills (conversations).
4. Janet	4, 5, 6	10	Bilingual, low language proficiency	Understand reasoning pattern pro-con. Create a design that solves a problem. Construct sentences using signal words and appropriate grammar. Develop domain-specific vocabulary.
5. Elisabeth	5, 6	10	Multi-lingual	Present design ideas. Understand the function and structure of the narrative genre.
6. Nicole	1	14	Low and high SES, bilingual	Experience what air pressure is and how it works through experimentation. Develop domain-specific vocabulary.
7. Mandy	6	22	No special characteristics	Understand what communication and noise are. Develop oral language skills (conversations, formulating conclusion). Develop reading comprehension skills.
8. Nadine	5, 6	16	No special characteristics	Understand how a bicycle chain transmits force through experimentation. Develop domain-specific vocabulary.
9. Kayleigh	2, 3, 4	44	No special characteristics	Understand the concept of condensation through experimentation. Distinguishing nouns, adjectives, and verbs. Develop reading comprehension skills.

*Names are pseudonyms.

**Grade equivalent in US school system.

Data collection

The data collection procedure for the CTA was based on the five steps from Clark et al. (2008). The CTA process consists of several consecutive steps, although the order of execution may vary depending on the context of the task under study (Clark et al., 2008). The five steps and the corresponding research activities of this study are presented in Table 2. The third and fourth step are described in more detail below.

Table 2. Implementation of CTA procedures.

CTA procedure (Clark et al., 2008)	Research activity
1. Collect preliminary knowledge	A literature study was carried out to review the characteristics and teaching demands of ILS&T instruction.
2. Identify knowledge representations	In this step, a particular method is chosen which determines how the types of skills and knowledge required to perform the complex task will be organized and represented (in step 5). Based on the Four-Component Instructional Design (4C/ID) model by van Merriënboer and Kirschner (2017), the knowledge representations are formatted in (a) a skills hierarchy that describes the prerequisite relation of the overall learning goal (i.e., teaching ILS&T) to more specific constituent skills, (b) the underlying body of knowledge and (c) the factors influencing task complexity.
3. Apply focused knowledge elicitation methods	Experienced teachers were observed while teaching an ILS&T lesson, followed by semi-structured interviews using stimulated recall.
4. Analyze and verify the data acquired	The data from the observations and interviews were qualitatively analyzed. Two expert meetings were organized with (a) teachers and (b) domain experts, to verify and expand the initial data.
5. Format the results for the intended application	In a follow-up study, the results will be used to design a 4C/ID-based PD program for in-service teachers.

Apply focused knowledge elicitation methods

Experienced teachers were observed while teaching an ILS&T lesson. Teachers were asked to deliver an inquiry- or design-based lesson that included language arts and S&T instruction. All lesson observations were video recorded. After the observation, a semi-structured interview took place using the stimulated-recall method (O’Brien, 1993). In this retrospective interview technique, videotaped fragments of a person’s actions are played back to encourage them to recall what they were doing and thinking. Several “cues” or “stimuli” are provided by the researcher to trigger the verbalization of thoughts. The researcher selected a series of 3 to 10 fragments (i.e., cues) to be reviewed with the teacher. The researcher selected fragments that specifically required teachers to employ their skills for teaching ILS&T. To this end, fragments were selected where the decision-making process of teachers’ behavior was unclear and remained implicit. For example, it can be observed that teachers offer student support in the form of scaffolding, but during a lesson observation it remains implicit why a teacher decided to offer this type of support at this time and to these students. Cognitive processes such as these were elicited through questions such as, “What are you doing here?” and “Why did you do that?.” This way, the authors were able to uncover the cognitive and procedural information that the experienced teachers used, for instance, for goal setting and decision making. Because experts often carry out tasks in a partially automated way, this protocol enabled the authors to elicit information about the ways in which these teachers perform the task (i.e., ILS&T instruction) that usually remains implicit. The interview protocol included questions about parts of the ILS&T lesson that could not be observed, such as the lesson preparation and follow-up activities, to obtain a more comprehensive impression of the whole task. Finally, teachers were asked to pinpoint factors that increased the complexity of ILS&T instruction. All interviews were recorded and transcribed. The quotes in this study were translated from Dutch.

Analyze and verify the data acquired

The preliminary knowledge representations and the observed actions and decisions derived from the interviews were elaborated into a preliminary skills hierarchy. A skills hierarchy refers to a schematization or structured overview of constituent skills required to complete a task (Clark et al., 2008). The preliminary interpretations were discussed with the participating teachers during a joint meeting. Four teachers took part in this expert meeting and five teachers provided written feedback. During the meeting, teachers were asked whether they recognized the actions and decision points as represented in the preliminary skills hierarchy and whether they agreed with their relevance for successful task performance. By asking teachers to elaborate and give examples, more detailed information about their thought processes was elicited. This way, teachers analyzed the systematic steps involved in their instruction, which helped to reach consensus on a revised skills hierarchy. Finally, factors influencing the complexity of ILS&T instruction were identified through the discussion of several authentic situations that appeal to their teaching skills. Those situations were ordered from relatively simple to complex during the expert meetings, which identified the complexity factors. A second expert meeting was held with the domain experts. First, the meeting served to verify the preliminary skills hierarchy. Second, the domain experts engaged in a discussion about the required level of competency of teachers. The discussions during both expert meetings were recorded. A summary of the main discussion points regarding the constituent skills (i.e., information that needed to be added or changed in the hierarchy according to the participants) was shared with all participants to check for accuracy and resonance (member check, see Creswell & Miller, 2000).

Data analysis

The literature study that was carried out in the first step of the CTA process (as described in Table 2) provided a starting point for the global structure of the knowledge representations, for which a skills hierarchy was chosen as the format in step two. Previous CTA studies of complex tasks for teachers (Meutstege et al., 2023; van Geel et al., 2019) indicate that teachers’ actions take place in four chronological phases; preparation of instructional unit (or lesson series), lesson preparation, lesson enactment, and evaluation. These four phases were empirically vindicated by our data. Parts of the transcribed interviews were linked to the corresponding phase, meaning that the researchers looked for remarks made by teachers about what they did during each of these phases.

Through inductive analysis, the authors then developed a set of codes that corresponded to the specific constituent skills that teachers used in each phase of instruction. The video recordings were used to illustrate teachers’ statements when the comments made by the teacher required visual illustration. For example, when a teacher talked about their guidance during an experiment, having a visual representation of what that experiment looked like was helpful. The interview transcripts were used to specify the constituent skills. The codes were refined with the other authors during several discussion rounds to reach consensus about the constituent skills and the knowledge components, which were then organized in a hierarchy. Finally, it was determined whether each skill could be broken down further, to give the most detailed description of what teachers are required to do.

The initial skills hierarchy was discussed with the teachers during the joint meeting for verification, revision, and refinement. The revised skills hierarchy was then presented to the domain experts during the second expert meeting for further verification. The domain experts mostly agreed with the initial findings but had more elaborate ideas about teachers’ actions and decision-making processes, which added detail to the descriptions of the constituent skills. Finally, the experts discussed the discrepancy between what teachers currently do, and what the ideal situation would be. The outcome of this discussion was used to identify complexity factors. These factors were grouped through open coding (Mortelmans, 2013) and reduced to four complexity factors. To illustrate, statements such as: “For many students, the language of instruction is not their first language” were coded as diversity in language proficiency. To verify these codes, it was checked whether all complexity factors that were mentioned by the experts could be grouped under these codes, which was the case.

Results

This study identified three components of teachers’ actions and reasoning during ILS&T instruction: the required skills, the required knowledge, and factors that influence the complexity of this task.

Teacher skills

Teachers demonstrated actions and reasoning during four phases of instruction: Preparation of instructional unit, Lesson preparation, Lesson enactment, and Evaluation. Figure 1 shows all constituent teacher skills in each phase. Horizontally, constituent skills have a prerequisite relationship, meaning that the skills on the far right are constituent or supportive of the skills to the left. Vertically, the constituent skills have a temporal relationship and can be performed simultaneously or successively, depending on the lesson and on teachers’ preferences (van Merriënboer & Kirschner, 2017). The findings are described in the following paragraphs.

Figure 1. Skills hierarchy for integrated language and science & technology instruction.

Figure 1 shows the skills hierarchy for integrated language and science & technology instruction. It includes the knowledge components as well as the constituent skills that were identified in the cognitive task analysis.

Preparation of instructional unit

In the phase of Preparation of instructional unit we identified one constituent skill: determining long-term goals (for a trimester, semester, or school year).

Determining long-term learning goals

Instructional units are often planned and carried out over the span of several lessons. Therefore, teachers (n = 5) indicated that long-term planning often starts by choosing a thematic unit, such as “sound” or “plastic soup,” which is used as a touchstone to determine long-term learning goals (e.g., inquiry skills, reasoning skills). For instance, Janet chose to elaborate the theme “sound” to “inquiry skills” and “formulating:” “Students should be able to formulate a hypothesis and think about how they can test it. I teach my students such inquiry skills across themes.” Thus, teachers should be able to develop long-term goals that contribute to language arts and S&T learning and recognize which thematic units fit well with inquiry or design learning. This requires an understanding of the scope and developmental sequence of the learning goals for S&T and language arts education.

Lesson preparation

The lesson preparation phase proved to call upon two constituent skills: determining short-term learning goals and deciding on the instructional approach.

Determining short-term learning goals

All teachers (n = 9) start by developing short-term learning goals that contribute to the long-term goals. Although the approach varies, a key pattern is that teachers reflect on relevant S&T concepts and processes within the thematic unit. Then, they decide what language skills can enhance students’ S&T learning or are relevant in the thematic unit. To illustrate, Janet stated: “I look at the S&T topic and think about what reasoning patterns, such as causal connections, fit well with that topic. Then, I decide what language skills are appropriate to address, such as signal words.” Alternatively, Elisabeth developed learning goals by analyzing the functions of language that are useful in S&T activities. She decided that when students pitched their design ideas, she should address the language functions of summarizing and persuading. Thus, teachers need to be able to reflect on the opportunities for language learning in the context of S&T education and choose learning goals that exploit the reciprocal relationship between language arts and S&T education.

Deciding on the instructional approach

All teachers (n = 9) determine the starting point for learning by reflecting on students’ prior knowledge regarding the learning goals.

Most teachers (n = 7) then select meaningful learning experiences that give an authentic purpose for achieving the short-term learning goals. In Nicole’s lesson, students learned about air pressure. She argued: “These experiments require students to understand reasoning patterns, such as cause-effect, and learn the appropriate language to express these patterns.” This step requires teachers to recognize which learning experiences provide an authentic purpose for language arts and S&T learning and leverage the reciprocal relationship between the subjects.

Most teachers (n = 6) stated that they subsequently examine what knowledge and/or skills are required for specific activities, so that they can determine the logical sequence of activities. They argued that sometimes students should first develop specific language skills to express certain ideas or observations. On the other hand, language production sometimes calls for a thorough understanding of S&T concepts. Nicole stated: “If you understand something clearly, you can put it into words, so students need some level of understanding of the concepts first.” This indicates that teachers should reflect on the learning goals and analyze the experiences and reasoning steps required for students to understand a phenomenon or problem.

Next, many teachers (n = 5) stated that they subsequently decide what subject matter knowledge requires explicit instruction. Nadine argued: “Some knowledge cannot be acquired through discovery, for example, the difference between adjectives and nouns.” This step requires teachers to examine which subject matter students cannot discover independently and thus requires instruction. Several teachers indicated that they create an authentic need for language production and comprehension before giving explicit instruction to stimulate students’ active thinking and increase their motivation to learn. For S&T subject matter, this implies that teachers must determine what insights they want students to gain independently, and what subject matter needs explaining. Regarding language arts subject matter, teachers must reflect on their learning goals and decide what to address in their instruction. Thus, this step requires teachers to reflect on the insights they want their students to gain and the means to achieve them.

Subsequently, all teachers (n = 9) said that they select (or design) suitable teaching materials, because ready-made materials are rarely available. Nicole commented: “Especially when dealing with abstract concepts, it’s helpful for students to carry out (hands-on) experiments.” This implies that selecting suitable materials requires teachers to consider which perceptions and actions certain materials afford and how materials help achieve the learning goals. Teachers should reflect on the purpose of materials, for example, to visualize abstract concepts or promote language use (“You saw what happened, explain it to me”). Teachers also agreed that the availability of materials is an important precondition. In sum, this step requires teachers to select material based on their purpose and suitability for the learning goals.

Finally, several teachers (n = 5) explained that they decide what supportive teaching strategies (i.e., planned support) will help students achieve the learning goals. ILS&T instruction addresses learning content that is often complex (e.g., problem solving skills, critical thinking skills, abstract and technical language) and is acquired through active and student-driven learning. The teacher plays a crucial role in offering guidance without giving too much direction. Nadine noted: “I prepare myself by thinking about what kind of questions I will ask, and what reactions I expect from the students.” This is often referred to as a hypothetical learning trajectory (“a prediction of how students” thinking and understanding will evolve during learning activities,’ see Simon, 1995, p. 136). This indicates the need for teachers to select supportive strategies that will help students gain the insights needed to achieve the learning goals, and to identify challenging learning content and potential preconceptions.

Enactment of the lesson

During the enactment of the lesson, we identified three constituent skills, related to introducing the lesson, offering meaningful learning experiences, and ending the lesson.

Introducing the lesson

Most teachers (n = 7) start their lesson by sharing the learning goals. Mandy said: “I want students to see the bigger picture so they can make connections.” By sharing the learning goals, teachers facilitate students’ understanding of the activities, teachers’ expectations, and the relationship of these activities with their prior knowledge.

Next, teachers (n = 6) direct students’ attention with an experiment, object, or phenomenon to engage students and arouse curiosity. The domain experts argued that to motivate students to acquire new knowledge and/or skills, teachers should be able to create a “cognitive conflict;” a situation that creates a need for students to acquire new knowledge or skills, to understand or explain a phenomenon. To illustrate, Kayleigh demonstrated that blowing air on a mirror will cause it to fog up. This way, she aimed to create such a cognitive conflict, prompting the students to wonder what is happening.

Almost all teachers (n = 8) then retrieve students’ prior knowledge, which provides insight into students’ existing cognitive schemes and preconceptions. To this end, Christie started off with a class discussion about different types of sound that the students could produce and referred to previous class activities where they produced sound to gauge what the students already knew about this topic and which vocabulary they already knew. Teachers mostly informally assessed students’ prior knowledge and language proficiency. This implies that teachers must be aware of the starting point for learning, and what prior knowledge students need to engage in the learning activities.

Offer meaningful learning experiences

After introducing the lesson, teachers engage students in learning activities. Four constituent skills were identified in this phase: employ interaction skills, provide scaffolding, explicate connections between language arts and S&T, and monitor and regulate students’ learning process. The sub-components of this constituent skill are non-linear, meaning that they do not necessarily occur in a particular order.

While students engage in meaningful learning experiences, all teachers (n = 9) demonstrated that they employ interaction skills to promote students’ cognitive engagement and language production. All teachers ask open-ended questions, as Alice explained: “I want students to feel a sense of ownership, so instead of telling them what to do, I ask questions: who can elaborate; does anyone have a different answer?” This indicates that teachers must understand the varying purposes of questions (e.g., predictive, explanatory). Teachers can direct questions toward the whole class (planned support) or toward individual students (differentiated support). Thus, this step requires teachers to recognize opportunities to employ interaction skills that will help students gain relevant insights and develop language skills.

All teachers (n = 9) explicated that they guide their students by providing scaffolding to help students navigate the challenges of guided experimentation and help them take ownership of their learning process. Furthermore, it enables teachers to offer tailored support to students with varying ability levels and prior knowledge. The specific purpose for which scaffolding was employed varied amongst teachers. Nicole stated that she uses scaffolding to help students understand the S&T subject matter: “I try to make the steps smaller by referring back to what students learned during previous lessons and building on that.” Elisabeth explained that she always uses language scaffolds in her lessons: “I scaffold their language by prompting the start of a sentence, or keywords to use. These hints help them structure their thoughts.” Several teachers provide feedback to scaffold students’ learning. For example, Christie stated: “I ask students to be more precise in their language production. It is important not to accept vague descriptions, but to keep pressing them to use the right language.” To provide scaffolding, teachers need to be aware of their learning goals and need to diagnose students’ current ability level, to be able to determine the appropriate scaffolding strategies. They should consider what misconceptions or difficulties may arise, and how task structure may enhance or reduce complexity, regarding subject matter for both language arts and S&T.

Some teachers (n = 4) stated that they explicate the connections between language arts and S&T concepts and processes by discussing how the learning goals and activities are connected and which cross-disciplinary knowledge and skills were developed. This helps students understand the added value of integrated instruction, because certain language skills are needed to engage effectively in S&T activities, and the context of S&T can improve their language skills. That way, students are more engaged and motivated to learn. To illustrate, when trying to explain why certain objects float and other sink during an experiment, students may refer to the “heaviness” of objects although this term does not suffice. This experiment urges students to learn the term “density” to accurately describe the phenomenon. In this regard, Janet indicated: “After experimentation I asked students to express what they saw, using the signal words they just learned to express a causal relation.” She then explained to the students that these signal words will help them communicate more accurately. In sum, this step requires teachers to be able to recognize the reciprocal relationship between language arts and S&T instruction and to recognize opportunities to convey connections between the two subjects.

All teachers (n = 9) stated that they monitor and regulate students’ learning processes throughout their lesson and adjust their instruction if necessary. For example, Elisabeth stated: “Students use a checklist to self-assess their skills.” The domain experts argued that currently, there is no urgency for Dutch teachers to formally assess students’ S&T achievement, because there is no external framework or system that monitors these learning outcomes. Notwithstanding this, most teachers expressed that they informally monitored students’ progress toward the learning goals during interactions. Christie described: “During the experiment I deliberately checked in with each group to ask whether they heard the difference between the sounds (high and low) and could point out what the medium was.” This step requires teachers to apply appropriate monitoring strategies to gain insight into students’ emerging understanding and skills. This implies that teachers need to be constantly aware of the dual learning goals and to regularly check for students’ understanding, either at the whole-class or the individual level.

Ending the lesson

Teachers end their lessons in various ways. Janet explained that during the final phase of the inquiry or design cycle, she includes a summative assessment: “I always include a language product assignment. This is a good way to check whether the students can apply the new academic vocabulary.” In most cases, the ending of the lesson included a recap of the learning goals and a preview of the next lesson.

Evaluation

After the lesson ends, teachers evaluate the lesson on its process and product. A distinction can be made between short- and long-term evaluation.

Short-term

Teachers (n = 6) expressed that they evaluate the lesson based on the insights students have gained. Teachers reflect on whether the short-term learning goals were achieved by all students, and whether the next lesson needs to be adjusted. At the end of Mandy’s lesson, students had to formulate their conclusion in one sentence, which they hung up on the wall. During the interview, she said: “Some sentences were a bit poor, so we will leave them up for next time.” She planned to revisit these formulations during the next lesson. Thus, teachers need to be aware of what relevant insights students have gained, what difficulties students may have experienced during the lesson, and what learning content requires additional attention during the subsequent lesson.

Long-term

Some teachers (n = 6) also evaluate whether the lesson needs to be improved for future implementation, and if so, they adjust the lesson plan. Sophia explained that she did not have sufficient time to address the concept of functional requirements of a design. She indicated that next time, she would allow more instructional time for this. Teachers also reflect on their own professional skills. Elisabeth expressed: “I can see there is a lot of potential in language, science and technology, but I am still at the beginning and there is loads more to be done.” Teachers thus need to be able to critically reflect on their teaching, and to judge whether the instructional activities and strategies effectively contributed to the accomplishment of the short-term as well as the long-term learning goals.

Teacher knowledge for ILS&T instruction

The findings indicate that teachers rely on three types of knowledge during ILS&T instruction that enable them to execute the constituent skills: subject matter knowledge, Pedagogical Content Knowledge (PCK), and knowledge of subject integration. Below, we demonstrate how teachers used these kinds of knowledge to inform their decision making.

Subject matter knowledge

Teachers and domain experts emphasized the need for teachers’ subject-matter knowledge of language arts and S&T. Mandy stated: “You need to have a deep understanding of the scope and developmental sequence of the learning goals of language arts and S&T, especially if you need to design or adapt learning materials.” This knowledge component requires an understanding of S&T phenomena, concepts and technical systems, and of the organization of learning activities in the inquiry and design cycle. Teachers need to understand the linguistic skills appropriate to the various steps of the inquiry and design cycle, such as precise formulation of a research question or causal reasoning, and which S&T contexts offer meaningful opportunities for developing language skills. Teachers made implicit references to their subject matter knowledge throughout all phases of instruction. For example, while preparing their lesson, teachers relied on their knowledge of S&T topics (e.g., “sound”) and of students’ prior knowledge and verbal skills to determine the learning goals and to determine the starting point for learning.

Pedagogical content knowledge

Teachers’ PCK enabled them to devise effective routes from the students’ starting point to the desired learning outcome (see, Fernandez, 2014; Shulman, 1986), although during the interviews they mostly referred to PCK implicitly. In consideration of the various components of teacher PCK, teachers mostly referred to their knowledge of instructional strategies for teaching ILS&T. For example, Janet described how she relied on her PCK during the lesson preparation, to determine how to decide on the instructional approach: “I try to determine which reasoning steps are involved in the S&T and language arts subject matter and decide what type of instruction and support the students need.” This requires teachers to recognize opportunities to employ interaction and scaffolding skills and understand the purpose of these strategies. Some teachers also referenced their knowledge of students’ understanding of specific topics. Teachers identify challenging actions or reasonings, and the required support. Finally, several teachers commented on their knowledge of the S&T and language arts curriculum. They used this knowledge to select suitable materials.

Knowledge of subject integration

To effectively combine two subjects, teachers should recognize the reciprocal relationship and common processes and patterns. To illustrate, Nicole stated: “There are many steps involved in the inquiry and design cycle, and many different corresponding language skills. If you can incorporate all these language skills, you could potentially replace your textbook, but then you need to be aware of these connections.” Understanding subject integration makes teachers aware of the rationale for ILS&T instruction and how integration enhances student learning. Teachers adopted different approaches for ILS&T, which is in part due to different views on subject integration. Most teachers’ views match with theories of content and language integrated learning and content-based instruction (Snow, 2010; Stoller, 2008), which view language as a general means of communication and therefore emphasize oral language and vocabulary. Elisabeth’s on the other hand, drew from theories of genre pedagogy (see Hyland, 2007), which uses genres (e.g., reports or narrative) to elicit the distinctive purposes and features of language. Teachers use their understanding of subject integration to reflect on opportunities for language arts learning in S&T instruction and develop learning goals. This helps teachers recognize which learning experiences offer an authentic purpose for language arts and S&T learning. Thus, teachers’ conception of subject integration determines their approach for finding meaningful connections between language arts and S&T subject matter.

Factors influencing the complexity of language arts and S&T instruction

ILS&T instruction is more complex in some situations than in others. By comparing teachers’ experiences, four complexity factors were identified: diversity of students’ language proficiency, phase of the inquiry or design cycle, facilitation, and degree of collaboration. In this regard, ILS&T instruction is least complex with a homogenous group of students in terms of language proficiency, during the initial phases of the inquiry or design cycle, with much facilitation by the school (leadership) and a high degree of collaboration with other teachers.

Diversity in language proficiency

Several teachers worked in schools in multicultural neighborhoods, which results in varying language proficiencies among their students. For example, during the interviews Nicole stated: “The composition of my group is very diverse, with many different home languages, which requires a lot of classroom differentiation. That is a challenge with this group, because sometimes there is a language barrier. I therefore have to think carefully about how to get students to talk more.” The expert meeting revealed that other teachers had relatively homogeneous student groups. During the joint meetings, the experts agreed that ILS&T instruction is more complex when teaching heterogeneous student groups with diverse language proficiencies, as it requires more instructional differentiation.

Phase of the inquiry or design cycle

Most often, the inquiry or design cycle is implemented over the course of several lessons. Most teachers expressed that they find it easier to provide instruction for ILS&T learning when their lesson addresses the earlier phases of the inquiry or design cycle (e.g., orientation phase) because the instructional goals are still exploratory (e.g., observing and discussing a teacher-led experiment). Teachers argued that during later phases of the inquiry or design cycle (e.g., conclusion phase), students often write scientific reports or engage in argumentation, which is more complex and requires more teacher guidance. Christie stated: “During the orientation phase, I mostly stimulate students to explore the topic. After a few lessons, I focus on whether the students are achieving the learning goals.” The domain experts pointed out the fundamental openness of inquiry- and design-based learning, indicating that it is complex for teachers to anticipate students’ diverging directions.

Facilitation

Generally, there is little educational material for ILS&T instruction, so teachers expressed that they need time to prepare their lessons, as well as materials and spaces for hands-on activities. To illustrate, Nicole stated: “You need to prepare the lessons and gather materials, set everything up and clean up afterward, this takes up a lot of time.” During the expert meeting, some teachers expressed that they were provided with additional time to prepare ILS&T lessons. After this discussion, all teachers agreed that the degree to which the school facilitates these activities makes teaching ILS&T either more or less complex.

Degree of collaboration

During the expert meeting, some teachers expressed that they collaborated with peer teachers, while other teachers did not. For example, Mandy expressed: “We no longer use a textbook for our reading comprehension lessons. Teachers from grades 4 through 6 worked together to restructure the scopes and developmental sequences of our curriculum thematically.” Therefore, Mandy can refer to this overview when preparing her ILS&T instruction, as it shows how the scopes and developmental sequences of language arts and S&T are connected. During the expert meeting, teachers agreed that preparing ILS&T lessons with no support from or exchange with other teachers is more complex.

Discussion

This study aimed to obtain insight into the actions and reasoning of primary school teachers while teaching ILS&T, through an in-depth CTA. Our analysis resulted in a skills hierarchy consisting of several constituent skills, as well as an overview of the underlying required knowledge and of the factors that reduce or increase the complexity of the teacher task.

The findings contribute to our knowledge of the pedagogical demands of ILS&T instruction. In reviewing the literature, four prerequisite skills were identified: lesson planning skills, interaction skills, scaffolding skills and monitoring and assessment skills. These skills were empirically vindicated by our study, which showed that they are part of a larger skill set related to four chronological phases: preparation of the instructional unit, lesson preparation, lesson enactment and evaluation. The skills hierarchy showed that there is a prerequisite relation between the constituent skills and the overall learning goal of ILS&T instruction.

This study showed that a major component of teachers’ lesson (unit) preparation refers to the formulation of learning goals, especially if instructional materials are scarce. In line with previous studies (Fisher & Frey, 2014; Haug & Ødegaard, 2014), teachers mainly included activity-oriented goals (e.g., perform experiments), oral language skills and vocabulary. This may be because some teachers did not yet recognize language as a discipline-specific register in S&T, but rather as a general means of communication. Another possible explanation is that teachers are hesitant to focus on other language skills, because they find it challenging to find suitable material (Ford, 2004; Pearson et al., 2010).

Teachers’ need for interaction and scaffolding skills during the lesson enactment phase of ILS&T instruction has been widely addressed in the literature, as these are important mechanisms through which teachers support students’ learning process (Mercer et al., 2004; Wellington & Osborne, 2001). Our study revealed that during ILS&T instruction, teachers explicated the connections between language arts and S&T to foster students’ awareness of the reciprocal relation between the two subjects. This constituent skill has not yet received much attention in the scientific literature, although some studies introduced strategies to build on the connections between different types of knowledge, learning events, and texts. Pappas et al. (2003) introduced the concept of intertextuality to refer to the connections between various types of text, including diagrams, figures, and recounting of previous events. Similarly, Haug and Ødegaard (2014) referred to link-making strategies designed by Scott et al. (2011), which include supporting knowledge building by integrating every day and scientific concepts, promoting continuity by recounting events from prior lessons and linking positive engagement with the subject matter. Although these strategies were introduced to promote scientific knowledge, that purpose can be extended to link-building between language arts and S&T.

In contrast to other studies (e.g., Choi et al., 2015), the teachers in this study did not perceive monitoring and assessment as a challenging aspect of ILS&T instruction. Most teachers were confident of their ability to monitor students’ learning process informally. This may be attributed to the fact that currently Dutch official agencies are doing little external inspecting of students’ S&T achievement. In this study, we found that teachers used their insight into students’ learning process to evaluate instruction after the lesson had ended, to determine whether they need to adapt the next lesson and/or instructional unit. In countries with an agreed-upon curriculum and/or national assessment system for S&T, teachers can rely on more input, but this will not change the position of monitoring as a key skill in the hierarchy.

The findings from the CTA confirmed and elaborated on the knowledge demands from the existing literature: subject matter knowledge, PCK and knowledge of subject integration. Particularly, subject matter knowledge informed teachers’ decision-making during all phases of instruction. The findings show that teachers relied on PCK during ILS&T instruction, regarding their knowledge of instructional strategies, students’ understanding and the curriculum. PCK literature commonly also cites knowledge of assessment and teachers’ orientation toward teaching (science) (Ball et al., 2008; Magnusson et al., 1999). These types of PCK were not mentioned by teachers, although it would seem apparent that they are also important for ILS&T instruction. More research is needed to fully grasp the scope and content of PCK for ILS&T instruction. Knowledge of the nature of science did not emerge from our analysis. Previous studies indicated a lack of knowledge about the nature of science among teachers, which may be caused by their lack of experience with the normative discourse practices of the scientific community, as they have no experience with conducting authentic research (Duschl & Osborne, 2002). Without this knowledge, teachers may hold limited views of the linguistic registers applied by scientists and engineers, and therefore may replicate superficial discourse practices in their classroom (Duschl & Osborne, 2002; Glen & Dotger, 2013). Although it did not emerge from our analysis, a basic understanding of the nature of science, and the corresponding linguistic registers, may in fact be very beneficial for teachers in the context of ILS&T instruction.

The complexity factors that were identified in this study were not previously addressed in the literature, although some of the factors are commonly understood as influencing the complexity of teaching in general (e.g., facilitation and degree of collaboration). Complexity factors were identified at the lesson (phase of inquiry or design cycle), the classroom (diversity in language proficiency) and the school (facilitation, degree of collaboration) level.

Implications for teacher professional development

The findings provide insights that can serve as the basis for the design of a PD program in which teachers learn how to prepare and teach ILS&T instruction. The descriptions of the constituent skills provide a good basis for formulating concrete teacher performance objectives that describe the criteria for successful task performance. The complexity factors can help sequence learning tasks in the PD program (van Merriënboer & Kirschner, 2017). Furthermore, the findings provide a starting point for the development of an observation instrument to rate teacher performance with respect to ILS&T teaching based on the performance objectives.

The value of CTA for analyzing teachers’ actions and reasoning

This study contributes to our understanding how CTA can be applied to analyze teacher expertise. The added value of the CTA method for PD is that it elicits the thought- and decision-making process of teachers while performing classroom activities. CTA is still relatively uncommon in the context of teacher PD. We know only of one other study in which this has been done (van Geel et al., 2019), and like those authors, we found that a CTA is a valuable method for analyzing the components of complex tasks for teachers.

The authors found it challenging to identify advanced experts in ILS&T instruction due to a lack of an unambiguous, easily applicable, definition. Moreover, ILS&T instruction is still a relatively new phenomenon? in the Netherlands, which explains why there are currently relatively few experts. Therefore, the authors relied on the recommendations of others, and thereafter tried to verify the level of expertise of the teachers that were involved in the CTA. Based on the recommendations from the network of the authors and the verification interview with the first author, the teachers in this study were more competent than their peers, but the term expert should be interpreted with care. Some teacher actions were not deliberate, and at times teachers were unable to explicitly articulate why they made certain decisions. It seemed that teachers at times insufficiently recognized and distinguished between underlying models, theories and approaches, and were not yet fully aware of all the possibilities and affordances of ILS&T instruction. The findings may be somewhat limited with respect to the cognitive and procedural information for some of the constituent skills. Furthermore, while not all teachers proved to be high-level experts, a general pattern of the required skills and knowledge emerged after the observations and analyzing additional teachers did not change that picture. As such, this study adds to our understanding of the pedagogical demands of teaching ILS&T.

Finally, it should be noted that this study reflects characteristics of the Dutch school system and the knowledge and skills of Dutch teachers with respect to ILS&T instruction, which may be different from other countries. It will therefore be interesting to verify our findings in other countries, and with teachers with more experience with ILS&T instruction.

This paper describes a teacher skills hierarchy of the prerequisite teacher competencies for ILS&T instruction that were identified through a CTA. Additionally, the required teacher knowledge and factors influencing the complexity of ILS&T instruction were discerned. The findings can offer a starting point for the design of a PD program focused on preparing and enacting ILS&T instruction. This study contributes to our understanding of the value of the CTA method as a systematic way of analyzing complex tasks.

Acknowledgments

We wish to express our appreciation to the teachers and the domain experts with whom we worked during this study.

Disclosure statement

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

Data availability statement

The datasets generated and analyzed during the current study are available in the Dans EASY repository: https://doi.org/10.17026/dans-zm3-xm2t.

Additional information

Funding

The work was supported by the TechYourFuture.