General pedagogical knowledge, pedagogical adaptivity in written lesson plans, and instructional practice among preservice teachers

ABSTRACT

Lesson planning of teachers as a research field has received little attention in terms of modelling and measuring relevant competences. As an innovative measurement approach, we developed a standardised method for analysing written plans of demonstration lessons. Our focus is on the demand of pedagogical adaptivity, i.e., the ways in which lesson assignments fit with the cognitive level of learners so that they are guided into their zone of proximal development. This conceptualisation is operationalised by using several indicators (content analysis criteria) to reconstruct and quantify situation-specific planning perception, interpretation, and decision-making. We use the data from 172 preservice teachers in Germany with their first demonstration lesson during induction. Findings show their declarative general pedagogical knowledge of adaptivity (assessed via a standardised test) is a significant predictor for the situation-specific skill of pedagogical adaptivity in written lesson plans, and the latter effects preservice teachers’ self-reported instructional practice of teaching that lesson. Findings are discussed towards their implication for the design of teacher education.

KEYWORDS:

• General pedagogical knowledge
• instructional practice
• lesson planning
• pedagogical adaptivity
• teacher competence
• teacher education

Introduction

Teacher education in many countries worldwide aims at preparing highly qualified teachers and supporting preservice teachers’ development of professional competence (e.g. Cochran-Smith & Villegas, 2016; European Commission, 2013). Teacher knowledge as part of teachers’ professional competence has become an important research topic during the last decades (e.g. Baumert et al., 2010; Gitomer & Zisk, 2015; Tatto et al., 2012). Numerous empirical studies use paper-pencil tests to directly assess teacher knowledge domains as differentiated by Shulman (1987) into content knowledge (CK), pedagogical content knowledge (PCK), and general pedagogical knowledge (GPK). However, following current conceptualisations of professional competence, these knowledge tests have a limited scope (Shavelson, 2010), as they mainly focus on knowledge as a cognitive disposition (Klieme & Leutner, 2006). New perspectives on the modelling and measurement of competence (Blömeke et al., 2015), in contrast, emphasise the need for instruments that allow an investigation of teachers’ situational cognition, including the measurement of context-dependent, procedural teacher knowledge.

This need also becomes visible when considering the state of empirical research on planning competence of preservice teachers. Lesson planning is a challenging task teachers have to master in their daily work. Whereas declarative knowledge is supposed to provide a relevant basis, teacher planning is highly context-dependent (John, 2006), since teachers are required to address their planning to their relevant learning group. In particular, teacher expertise research has worked out that expert teachers, compared with novice teachers, highly align their planning decisions to the needs of their students (e.g. Berliner, 2004; Borko & Livingston, 1989; Hall & Smith, 2006; Westermann, 1991). This relates to teacher action such as assessing students’ state of knowledge, setting learning goals, considering and judging on appropriate teaching strategies to be implemented into the classroom of a particular learning group (Stigler & Miller, 2018). Paper-pencil tests comprising items on planning aspects in general might therefore fail to address the contextual nature and the complexity of the classroom situation teachers have to account for. From a perspective of modelling teacher competence, one might therefore argue that professional knowledge as directly assessed by paper-pencil tests is relevant, but also situation-specific skills have to be addressed in order to predict instructional practice (Blömeke et al., 2015; Darling-Hammond et al., 2013).

Against this background, the present study aims at accounting for such methodological concerns. To capture preservice teachers’ situational-specific skills as part of their planning competence, the study applies an innovative measurement approach, consisting of a standardised method for analysing written plans of demonstration lessons of preservice teachers during induction in Germany, i.e. during the practical phase of initial teacher education that preservice teachers enter after graduating from university. The approach highlights the demand of pedagogical adaptivity, i.e. the ways in which the assignments of the respective lesson fits with the cognitive level of the learning group so that learners are guided into their ‘zone of proximal development’ (Vygotsky, 1978, p. 84). Pedagogical adaptivity as a construct reflects empirical findings from teacher expertise research (e.g. Berliner, 2004; Borko & Livingston, 1989; Housner & Griffey, 1985; Kagan, 1992; Westermann, 1991) and refers to didactical concepts (Kansanen, 1995; Scholl, 2018), differentiated instruction (Dack, 2018; Tomlinson, 2014), the research on adaptive teaching (Parsons et al., 2018) as well as constructivist learning theories (Vygotsky, 1978). For its operationalisation, several indicators (content analysis criteria) serve to reconstruct and quantify situation-specific planning perception, interpretation, and decision-making (Blömeke et al., 2015). The study uses data from 172 preservice teachers in Germany with their first demonstration lesson during induction, self-reports of instructional practice of teaching that lesson, and additional paper-pencil general pedagogical knowledge test scores. We focus on two research questions:

1. Is preservice teachers’ declarative general pedagogical knowledge (GPK, assessed via a paper-pencil test) a significant predictor for the situational-specific skill of pedagogical adaptivity in written lesson plans?

2. Is the situation-specific skill of pedagogical adaptivity in written lesson plans reflected in preservice teachers’ self-reports of their instructional practice?

Literature survey

Lesson planning

Lesson planning is a crucial challenge for teachers. Teacher education, therefore, seeks to prepare future teachers as lesson planners. Although numerous course books exist aiming at providing theories or practical guides of lesson planning (e.g. John, 2006, for teacher education in the UK; Scholl, 2018, for teacher education in Germany), empirical research on lesson planning as a skill of preservice teachers is scarce (König et al., 2020, 2015).

Early studies in lesson planning investigated particular aspects, for example, which planning component a teacher pays attention to and the order in which a teacher works on such planning components (e.g. Clark & Peterson, 1986; Hill et al., 1983; Peterson et al., 1978; Taylor, 1970; Zahorik, 1975). For example, an in-service teacher survey conducted by Taylor (1970) provided evidence that during lesson planning teachers gave priority to student needs, learning content, goals, and methods. Teachers first started with thinking of the teaching context (comprising materials, resources), then with involving the specific student needs and learning dispositions, and finally thinking of curricular alignment. Another study by Hill et al. (1983) showed in a similar way that after selecting appropriate materials, teachers gave priority to planning decisions of how they can arrange these materials in the classroom so that their students use them as activities.

Lesson planning procedures of this kind can be described as a problem-solving process (e.g. Bromme, 1981; Yinger, 1977), highlighting the decisions teachers make on the basis of available information during pre-active teaching (Shavelson & Borko, 1979) and as part of their reflection on action (Parsons et al., 2018; Schön, 1983). Relevant skills can be considered as part of teacher competence and therefore should be an object of empirical investigation.

The research on the measurement of teacher competence has significantly increased over the last decade (Kaiser & König, 2019; König, 2014). Various research groups have developed standardised test instruments assessing teacher knowledge following the well-known classification by Shulman (1987), namely content knowledge (CK), pedagogical content knowledge (PCK), and general pedagogical knowledge (GPK). Moreover, recent research has started to assess teachers’ situation-specific skills in the area of professional vision and teacher noticing (Kaiser et al., 2015). However, to the best of our knowledge, situation-specific skills in the area of lesson planning has not been an object of investigation in the modelling and measuring of teacher competence.

Therefore, assumptions about processes of lesson planning seem hardly be supported by empirical evidence. For example, it remains an open question, whether teachers presumably make use of their specific knowledge in the subject area, subject-specific pedagogy, and general pedagogy and relate it to the specific planning situation that is predominantly determined by factors such as characteristics of the learning group, specific curricular goals, and the classroom context. At least some evidence exists that didactical models, being predominantly prescriptive rather than evidence based, are not necessarily applied by in-service teachers, although they are given high priority in initial teacher education programmes (John, 2006; Scholl, 2018).

Against this background, the present study primarily aims at a rigorous measurement of preservice teachers’ skills in the area of lesson planning. However, due to the complexity of lesson planning as an object of investigation, only a particular, but highly relevant aspect will be examined: Our focus will be on pedagogical adaptivity, a construct we created (König et al., 2015) and that we define as the ways in which the assignments of the respective lesson matches the cognitive level of the learning group. This construct comprises several aspects, namely the way a teacher accounts for learning dispositions of students, the way learning tasks are planned, and how both learning dispositions of students and learning tasks are related to each other, so that learners are guided into their ‘zone of proximal development’ (Vygotsky, 1978, p. 84), which is defined as ‘the distance between the actual developmental level as determined by independent problem solving and the level of potential development as determined through problem solving under adult guidance or in collaboration with more capable peers’ (Vygotsky, 1978, p. 86).

Accounting for learning dispositions of students

The research on teacher expertise shows that expert teachers, contrasted with novice teachers, plan their lessons in a process-driven way (Stigler & Miller, 2018): Rather than using a stepwise procedure, expert teachers are capable to consider several planning elements simultaneously. Particularly, expert teachers rigorously relate the learning dispositions (e.g. domain-specific knowledge) of their students and the learning tasks chosen for the lesson to each other (Berliner, 2004; Borko & Livingston, 1989; Housner & Griffey, 1985; Kagan, 1992; Smith & Strahan, 2001; Westermann, 1991). Expert teachers are clearly aware of their students and are committed to involve student needs into their planning process. They perceive student learning dispositions as a key element of their teaching. They know how to interpret these dispositions, that is, they integrate diagnostic information retrieved from their students specifically into their lesson planning (Putnam, 1987). Their decision-making during the planning process can be characterised by a successful integration of conceptual and situation-specific knowledge (De Jong & Ferguson-Hessler, 1996), since important aspects of the planning situation are identified and are progressively merged with teaching and learning activities (Bromme, 1992; Cook, 1992; Ericsson, 1996; Leinhardt & Greeno, 1986; Schoenfeld, 1998).

Planning learning tasks

Selecting and creating learning tasks as part of student activities in the classroom can be considered as the core areas of lesson planning, since they enable teachers to integrate a range of further decisions (Bromme, 1981; Kang, 2017; Shavelson & Stern, 1981; Yinger, 1977; Zahorik, 1975). Learning tasks pool at least teachers’ decisions regarding the selection of content and the specification of objectives that are part of the lesson. The learning tasks chosen for a particular lesson reflect the objectives of that lesson. They refer to what students should learn, what knowledge students should acquire, or which competences students should elaborate. Classification systems or taxonomies support the analysis of learning tasks in relation to specific cognitive and motivational requirements (Anderson & Krathwohl, 2001; Commons et al., 1998; De Jong & Ferguson-Hessler, 1996; Johnson, 2003). For example, De Jong and Ferguson-Hessler (1996, p. 106) use a mechanics problem to illustrate differences in types of knowledge needed to solve that problem, e.g. situational knowledge (knowing about the problem’s situation) vs. strategic knowledge (knowing how to organise the information given). At the same time, complex learning tasks can cover a range of difficulty levels in different dimensions, allowing an alignment to existing student dispositions and needs in a differentiated way (Dack, 2018). For example, Hsieh et al. (2011, p. 287) provide an example on differentiating the following task in primary-level mathematics (‘A machine uses 2.4 litres of fuel for every 30 hours of operation. How many litres of fuel will the machine use in 100 hours if it continues to use fuel at the same rate?’), and they describe the challenge that teachers face when required to create a different problem of the same type that is easier for students to solve (e.g. ‘A machine uses 3 litres of fuel for every 30 hours of operation. […]’). Therefore, learning tasks serve teachers to implement teaching strategies of differentiated instruction (Tomlinson, 2014, 2015), that is, learning tasks are considered to support teachers’ proactively anticipating and responding to diverse needs of learners. Learning tasks support teachers when they account for the existing knowledge of learners and guide learners into their ‘zone of proximal development’ (Vygotsky, 1978, p. 84). Learning tasks can therefore be regarded as an important instrument of adaptive teaching (Corno, 2008; Corno & Snow, 1986; Parsons et al., 2018), which may generally be defined as effective instruction through which teachers manage to ‘adjust their teaching according to the social, linguistic, cultural, and instructional needs of their students’ (Parsons et al., 2018, p. 205). The way how a teacher deals with learning tasks during lesson planning might provide insight into his or her pedagogical adaptivity.

In the present study, we specifically look at those learning tasks students are required to work on during the lesson’s main activity phase. These tasks represent the work order that the teacher instructs to his or her students and are expected to trigger student information processing and cognitive activation of students (Neubrand et al., 2013). Cognitive activation relates to learning processes that allow in-depth conceptual understanding of the learning content (Baumert et al., 2010), usually following the idea that learners are actively involved in the knowledge construction process as suggested by constructivist learning theories (e.g. Piaget, 1964; Vygotsky, 1978). Students usually work on such tasks individually, in pairs, or sometimes in groups. Usually these tasks can be clearly identified in written lesson plans (König et al., 2015), not least as they emerge from the relevant lesson material (e.g. a worksheet or a number of differentiated worksheets) that guides student work.

Lesson planning as part of teacher competence

Due to little investigation into the field, empirical evidence on how teachers plan their lessons is fairly limited (Bromme, 1981; Jacobs et al., 2008). A number of surveys or qualitative studies have been conducted and provide relevant descriptive scientific knowledge. However, in order to proliferate explanatory knowledge, also approaches are necessary that directly assess teacher skills in the area of lesson planning. An exception can be found in the research work underlying the performance-based assessment for teacher candidates in the US, edTPA (Sato, 2014). Like its predecessor, the Performance Assessment for California Teachers (PACT), preservice teachers are required to complete several components related to planning lessons, teaching, assessing students, and reflecting on teaching, where they are asked to submit an outline for several lessons they are going to teach. The performance ratings are based on coding schemes with a 4-point continuum (Pecheone & Chung, 2006, p. 25). For the task ‘planning’, five guiding questions are used by the raters who have to score the quality of the instructional design (Pecheone & Chung, 2007, p. 27). These are related to how the students later taught by the preservice teacher have access to the curriculum, how the curriculum is addressed in a coherent and balanced way, how the students’ interest and needs are reflected and addressed, and how well learning goals, instruction, and assessments are aligned. Research work based on PACT provides important insights into a teacher performance assessment that is very close to typical tasks teachers have to master. In a more recent analysis on predictive validity, Darling-Hammond et al. (2013) showed the PACT overall scale as well as subscales such as planning can significantly predict student achievement. However, since information on the scaling procedure is limited (Pecheone & Chung, 2007), we conclude that research on measuring and modelling lesson planning as part of teacher professional competence can still be regarded as a research desideratum.

Theoretical framework

Modelling of planning competence

Planning a lesson is dependent on the context in general (John, 2006; Mutton et al., 2011). This is reflected by the present study, since with the construct of pedagogical adaptivity emphasis is given to the situation of planning in which a teacher is required to account for the characteristics of his or her learning group. As outlined in the literature survey, when planning a lesson, expert teachers master to perceive and interpret student dispositions in order to make decisions for planning their lessons, especially when they select and create learning tasks. We consider that learning tasks should be adapted to the cognitive level of students following the ‘zone of proximal development’ (Vygotsky, 1978, p. 84). This may become apparent when a teacher makes use of strategies of differentiated instruction (e.g. Dack, 2018; Tomlinson, 2014) and plans a differentiated set of tasks, which may involve, for example, more challenging tasks for strong students and tasks with auxiliary material for weak students (Tomlinson, 2014, p. 90). Using the construct of pedagogical adaptivity, we will define and investigate a situation-specific teacher skill that is relevant to master such requirements.

Moreover, we consider such a skill as part of teacher professional competence. To highlight the situated perspective of lesson planning as a professional teacher activity, we build our framework on the model of ‘competence as a continuum’ by Blömeke et al. (2015) who account for the complex interaction of personal, situational, and social characteristics. According to this general model, competences can be described along a continuum of personal dispositions, such as teacher professional knowledge, situation-specific cognitive skills, such as perception, interpretation, and decision-making, and the observed teacher performance in the classroom. Following this model, we also consider that pedagogical adaptivity as a construct of situation-specific cognitive skill has to be investigated in combination with other measures. Teacher professional knowledge as investigated as cognitive disposition by previous studies should be a relevant antecedent of such a situation-specific skill, whereas situation-specific skills are to be seen more proximal to actual performance in class. Figure 1 illustrates this idea. It serves as an heuristic to locate the constructs of the present study (set in square brackets) in an overall model.

Figure 1. Heuristic of modelling lesson planning as part of teacher competence (following Blömeke et al., 2015, p. 7) with constructs of the present study set in square brackets

Preservice teachers’ professional knowledge needed for lesson planning will be investigated using a standardised test measuring their general pedagogical knowledge (GPK) developed by König et al. (2011) in the context of IEA’s Teacher Education and Development Study: Mathematics (TEDS-M) (see, for sample items, Table A1 in the appendix). This test relates to generic dimensions of teaching quality and, therefore, measures knowledge allowing teachers to deal with heterogeneous learning groups in the classroom (‘adaptivity’), to prepare, structure and evaluate lessons (‘structuring’), to motivate and support students as well as manage the classroom (‘motivation/classroom management’), and to assess students (‘assessment’). GPK of adaptivity, which presumably is of particular interest in the present study, comprises knowledge of differentiated instruction (see first and second sample item in Table A1) as well as knowledge of the use of a wide range of teaching methods (see third sample item in Table A1). In the test, both bodies of knowledge are covered after they had been identified during test development as relevant categories to describe GPK of adaptivity (König & Blömeke, 2009; König et al., 2011). However, a joint subscale only measures GPK of adaptivity in the test, that is, a more detailed empirical differentiation into the two bodies of knowledge is not intended. Teachers need such knowledge to deal with heterogeneous learning groups in the classroom (e.g. Brophy, 1999; Dack, 2018; Gräsel et al., 2017; Tomlinson, 2014). However, such GPK of adaptivity is measured in a general way. The test addresses basic teacher knowledge, as it requires teachers to recall or apply basic terminology (e.g. the differentiation into external and internal differentiation) or general concepts (e.g. teaching methods). However, the test does not specifically account for the contextual nature of teaching a particular learning group.

By contrast, pedagogical adaptivity as a skill is investigated using the data of authentic lesson plans that de facto were enacted as demonstration lessons. The nature of pedagogical adaptivity as a situation-specific skill, therefore, is different from GPK of adaptivity as measured using the paper-pencil approach. Instructional practice as an indicator of classroom performance is captured using self-reports of preservice teachers they were asked to provide for the specific lesson after performing that lesson. This is a valuable source of data, since teachers are aware of strengths and weaknesses of their instructional practice which should be reflected in their ratings of questionnaire items. Compared with student ratings that are frequently used, teacher ratings may also be more valid regarding the evaluation of instructional concepts that are behind the visible teaching process in the classroom (Wagner et al., 2013). However, since teacher ratings may be “biased by self-serving strategies or teaching ideals“ (Kunter & Baumert, 2006, p. 231) leading to a gap between the expressed and the actual practices (e.g. Holzberger et al., 2013), empirical results should be interpreted with caution. In the methods section, we provide further information on the constructs.

Lesson planning as part of teacher education

Teacher education programmes intend future teachers to learn how to plan lessons. Corresponding opportunities to learn are provided by teacher education institutions in many countries (Clift & Brady, 2005; Schmidt et al., 2011). While courses in the academic setting often primarily aim at the acquisition of theoretical knowledge, in-school opportunities to learn give future teachers the chance to connect their knowledge to practical situations in the classroom. Lesson planning might be a particularly complex challenge for novice teachers, as they are required to link their professional knowledge to the concrete learning group they are going to teach (Hall & Smith, 2006; John, 2006). However, the question arises, whether future teachers manage to link their theoretical knowledge to practical problems of lesson planning. For example, a survey of preservice teachers on their learning opportunities in Germany showed that only practical learning opportunities helped them to develop their planning competence, whereas learning opportunities in the subject, subject-specific and general pedagogy were not relevant (König et al., 2017a). This finding mirrors the gap between theory and practice as it has been discussed in teacher education reforms for decades in many countries worldwide (e.g. Flores, 2016; Zeichner, 2010). An analysis of how preservice teachers during induction relate their lesson planning on previously acquired theoretical knowledge might therefore be a valuable contribution to the teacher education theory-practice discourse.

Context of the study

The German teacher education system has a consecutive structure. Its programmes are spread over two separate phases, a theoretical at university and a practical at small teacher training institutions operated by state governments (König & Blömeke, 2013). Preservice teachers finish their first phase at university with a master of education nowadays, requiring coursework that emphasises the acquisition of theoretical knowledge in the teaching subjects, subject-specific as well as general pedagogy. By contrast, most practical learning opportunities are then provided in the 1.5-year second phase. This phase serves as induction for the preservice teachers who then work part-time at schools and attend courses in general pedagogy and subject-related pedagogy. They are assessed by their teacher educators and mentored by one or two teachers at school. Lesson performance is usually based on a written lesson plan comprising detailed information about a large number of planning aspects such as objectives, teaching methods, the learning group, activities, time schedule, and embedding the lesson into the larger teaching unit. For this, preservice teachers are required to have observed or even taught the learning group in advance, so that they are familiar with the students and had the opportunity to learn about the students’ prior knowledge and motivation. Preservice teachers are required to give demonstration lessons and to submit the relevant written plan at regular intervals over the duration of the second phase. This phase ends with a state examination consisting of a practical part including at least two lessons performed in two different subjects.

The present study

The structural divide into a theoretical and practical initial teacher education phase in Germany offers an excellent opportunity to analyse how preservice teachers at the transition into teaching practice relate their teacher knowledge to challenges of lesson planning and how they reflect about instructional practice. We therefore assessed preservice teachers’ GPK at the beginning of their induction and required them to submit a first demonstration lesson plan with a rating of their instructional practice. With the present study, we address the following research questions:

• 1. Is preservice teachers’ declarative general pedagogical knowledge (GPK, assessed via a paper-pencil test) a significant predictor for the situation-specific skill of pedagogical adaptivity in written lesson plans?

We consider that preservice teachers’ GPK is a relevant resource for their lesson planning. However, not all GPK content areas might be required in a similar way when preservice teachers have to solve the lesson planning problem of pedagogical adaptivity as defined by our study. We thus specifically hypothesise that GPK of adaptivity is positively associated with adaptive lesson planning (H1a), since GPK of adaptivity comprises teacher knowledge of differentiated instruction as well as knowledge of the use of a wide range of teaching methods (see, the item examples in Table A1 in the appendix). General pedagogical knowledge of structure, classroom management/motivation, or assessment might be relevant as well, but less strongly related to our relatively narrow definition of pedagogical adaptivity and therefore only weakly associated with adaptive lesson planning (H1b).

• 2. Is the situation-specific skill of pedagogical adaptivity in written lesson plans reflected in preservice teachers’ self-reports of their instructional practice?

Although teacher planning and the actual performance in class will hardly correspond perfectly to each other, we assume pedagogical adaptivity in written lesson plans being positively associated with instructional practice in the area of cognitive activation and constructive support during that lesson (Praetorius et al., 2018). That means the teacher supports student learning to a higher degree due to a better fit between learning pre-requisites of students and activities assigned to students. Differentiation of tasks for the different learning needs of students is a typical element of constructive support (Fauth et al., 2014). Designed to support students’ ‘zone of proximal development’ (Vygotsky, 1978), tasks should activate student knowledge in the particular domain of a school subject. We, therefore, hypothesise that pedagogical adaptivity predicts cognitive activation and constructive support as dimensions of instructional practice (H2a). In contrast to this, we assume that classroom management can not necessarily be predicted as well, due to the rather narrow definition of pedagogical adaptivity we use when analysing written lesson plans (H2b).

Methodology

Sample

The data derives from the PlanvoLL-D project, an empirical research study conducted in Germany in 2016 in order to investigate the planning competence of German language preservice teachers. For this, preservice teachers who had entered induction (i.e. their second phase of teacher education) in spring 2016 were sampled in two federal states, North Rhine-Westphalia and Berlin. North Rhine-Westphalia is a region not only with the largest number of students and in-service teachers in Germany, but also covers between 20% and 25% of all students and in-service teachers in Germany (KMK, 2015). The sample consists of preservice teachers attending a teacher education programme that would qualify them as lower secondary teachers only (Haupt-/Real-/Gesamtschule) or as lower and upper secondary teachers (Gymnasium/Gesamtschule). In Berlin, the corresponding teacher education programme was focused on, a comprehensive teacher education programme that would qualify preservice teachers as lower and upper secondary teachers (Integrierte Sekundarschule/Gymnasium). In both federal states, these teacher education programmes are the only programmes that qualify German language teachers for secondary schools. However, only those are certified to teach at grammar schools (Gymnasium) who have the qualification to teach German in both lower and upper secondary level. That means preservice lower secondary teachers in North Rhine-Westphalia do not qualify to teach at the Gymnasium later on, although the Gymnasium starts with grade 5. Lower secondary teachers teach at the Hauptschule or the Realschule as the other two school types of the tripartite system or at the Gesamtschule (comprehensive school).

A random sample of training units was drawn for the lower and upper secondary teacher education programme in North Rhine-Westphalia, whereas, due to smaller populations, a census was applied for the other two programmes. Participation rate on the level of training units was good (92% in North Rhine-Westphalia) or at least acceptable (70% in Berlin). Within these training units, all preservice teachers were included into the survey. Participation rate on the individual level was good (91% in Berlin) or at least acceptable (68% in North Rhine-Westphalia). The sample consists of 378 preservice teachers.

Research assistants of the project team administered a paper-pencil questionnaire that the preservice teachers were asked to complete under observation. This questionnaire included the standardised test to assess preservice teachers’ GPK. After the survey, preservice teachers were asked to submit a copy of the written plan of their first demonstration lesson and to complete a short questionnaire related to the execution of that lesson. With 172 plans and questionnaires submitted that finally could be linked with the previous survey data, participation rate was moderate (46%), but a drop-out analysis did not show sample bias. In the following, we will focus on this subsample of 172 preservice teachers. They were, on average, about 29 years old at the entrance into induction (M = 28.9, SD = 4.9). The majority was female (more than 85%). Sixty preservice teachers (35%) are from Berlin and qualify as lower and upper secondary teachers, whereas 112 (65%) are from North Rhine-Westphalia, with 73 (42%) qualifying as lower and upper secondary teachers and 39 (23%) qualifying as lower secondary teachers.

Measures

In the following, we describe the measures we focus on in the present study: First, the preservice teachers’ general pedagogical knowledge (GPK) was assessed using a paper-pencil test. Second, they were asked to submit a written plan of their first demonstration lesson. These written plans were used to capture pedagogical adaptivity. Finally, after teaching that lesson, they had to complete a short questionnaire, capturing their self-report of instructional practice related to that specific lesson (see, for further details on instruments and methodology, the PlanvoLL-D project overview outlined in König et al., 2020, p. 64).

General pedagogical knowledge

As outlined previously, preservice teachers’ GPK was measured using a test that was developed in the context of IEA’s Teacher Education and Development Study: Mathematics (TEDS-M), which provided evidence for curricular validity as well as reliability of the test (König et al., 2011). The test concept can be characterised in general as follows: It approaches teachers’ GPK to be structured in a task-based way and explicitly not according to the formal structure of general pedagogy as an academic discipline; it accepts to regard instruction as the core activity of teachers, so the central demands placed on teachers are clearly related to students’ cognitive learning; and the operationalisation of teachers’ GPK refers to the extensive research on instruction and effective teaching. Item examples are provided in Table A1 in the appendix. Due to time constraints in the present study, we used a shorter version with 40 items only (36 being dichotomous, 4 partial-credit) that was developed on the basis of in-depth scaling analysis of TEDS-M test data (for a more detailed description of the test see, e.g. König, 2014; König & Blömeke, 2010; König et al., 2011). For the open-response items, the TEDS-M coding rubrics were used. The codes were intended to be low-inferent. Two raters coded the answers of 27% of all questionnaires independently of one another. Cohen’s Kappa was computed showing very good consensus (κ =.90). Overall test reliability is acceptable (Cronbachs Alpha = .71), and even for three of the subscales reliability is in an acceptable range (adaptivity: .51, structure: .58, assessment: .68) with the exception of classroom management/motivation (below .30), which cannot be used as a subscale in the present study.

Pedagogical adaptivity in written lesson plans

In order to investigate pedagogical adaptivity of preservice teachers in their written lesson plans as a situation-specific skill, the written plans of demonstration lessons were analysed and indicators created on the basis of an existing coding system developed in a previous study (König et al., 2015). The coding system differentiates indicators into four components (see, Table A2 in the appendix): On a descriptive level, written plans are analysed whether the learning group is described (component 1) and whether descriptive information is given for the learning task that primarily governs students’ activities during the lesson (component 2). It is important that teachers account for cognitive and motivational preconditions as the learning dispositions of students (component 1). In case a teacher plans strategies of differentiated instruction, it is important to analyse plans whether learning tasks have different cognitive levels which allow the differentiation of students in the lesson. On an analytical level, plans are analysed whether the descriptions given by a preservice teacher for his or her specific learning group and the learning task or tasks planned are logically and pedagogically consistent (component 3). Here we make the following distinction: A teacher may relate one learning task to a general level of the whole learning group (i.e. the whole class) he or she plans the lesson for. A teacher may also plan more than one task for the lesson, since he or she relates different tasks to individual students or different group of students which we then denote as student differentiation. This application of the given descriptions to the specific situation comprises the examination whether the learning task or tasks is or are adapted to the cognitive level of students following the ‘zone of proximal development’ (Vygotsky, 1978, p. 84). Therefore it is necessary that the lesson plan contains an outline that shows how the task (or even a differentiated set of tasks) given to the students connects with what the students (or groups of students) have learned so far, e.g. in a preceding lesson of the unit that contextualises the plan of the demonstration lesson. The connection between tasks and prior knowledge of students needs to be addressed by the preservice teacher and he or she should relate this connection to the situation of the particular lesson. Finally, plans are analysed whether such adaptive teaching is linked to other important elements of planning such as the connection to learning goals (component 4).

Coding examples

The lesson plan coding was done using a content analysis method based on deductively formed categories (Mayring, 2014). Table A3 in the appendix provides examples from the lesson plan coding that show how indicators of the components 2, 3 and 4 were applied. These examples derive from different lesson plans. In each case, the criterion described by the single indicator was fulfilled in the lesson plan by the individual preservice teacher. Examples in Table A3 clearly show that our lesson plan coding is linked to content (e.g. using rhetorical devices of advertising, listening to a recorded scene and taking notes of the characters’ feelings). Appropriate lesson plans contain information about specific content to be taught and learned in a particular lesson (Rusznyak & Walton, 2011). When coding information according to the coding rubrics, such content was inextricably taken into consideration and determined the code assigned, however, not as a subject-specific code. Therefore, the lesson plan coding is not genuinely content-related, but more located on an aggregated level, i.e. a pedagogical level. If a preservice teacher was not dealing with content in his or her lesson plan, coding would not be possible due to missing information that would be needed for judging whether categories of pedagogical adaptivity were fulfilled or not fulfilled. So the coding actually presumes preservice teachers’ reasoning in the area of pedagogical content (see right column in Table A3), but the indicators themselves then abstract from the detailed content (see left column in Table A3). That means, the construct of pedagogical adaptivity has the potential to be investigated in different subjects, which was done in a previous study where the coding system was developed (König et al., 2015). Methodologically, our approach is similar to PACT, where one generic rating scheme is applied to lesson plans of different subjects (Darling-Hammond et al., 2013).

Using the items created for the descriptive components 1 and 2, this allows us to reconstruct preservice teachers’ perception and interpretation of the learning group during the process of planning the lesson. When applying the items created for the analytical components, preservice teachers’ decision-making comes to the foreground. We therefore consider this approach as suitable to model situation-specific skills in the area of adaptive lesson planning (König et al., 2017b).

Scaling and reliability

Written lesson plans were first coded according to this coding system and the existing coding manual (König et al., 2015). As in their lesson plans about two-third of the preservice teachers outlined students with special needs they accounted for regarding the description of differences in the learning group, we decided to separately code item 11, 22, and 41 for this specific planning situation. That is, the 11 items outlined in Table A2 were extended by three more optional items, since not all preservice teachers necessarily had to deal with special needs students in their specific learning group. All items were coded as 1 if the relevant criterion as described in the coding manual was matched. The code 0 was given if the criterion was explicitly not matched. If the written plans did not contain sufficient information about the particular analysis criterion, a 9 was given for missing value. Interrater reliability was good (κ > 0.73). IRT-scaling analysis was carried out using the software Conquest (Wu et al., 1997). This approach allows us to define optional items as missing if they do not correspond to the specific planning situation. Item 42 was excluded due to negative discrimination. Our scaling analysis therefore comprises 13 items making up a reliable scale (EAP reliability 0.68). Weighted mean square of each item is in an acceptable range. Discrimination index is, on average, .47, which is good.

Self-reports of instructional practice

When submitting a copy of the written plan of their first demonstration lesson, preservice teachers were also asked to complete a short questionnaire related to the execution of the lesson, comprising 22 items describing their instructional practice delivered to students during that specific lesson. These items are conceptually related to the three basic dimensions of instructional quality (effective classroom management, cognitive activation, constructive support; Praetorius et al., 2018). Items were derived from existing instruments and modified for the purpose of surveying teachers’ self-reports (Depaepe & König, 2018). Each basic dimension is measured by two subscales. To measure classroom management, the subscales ‘preventing disorder’ (four items, e.g. ‘I always knew exactly what happened in the classroom.’, α = .67) and ‘providing structure’ (four items, e.g. ‘I frequently told the students what they had to remember.’, α = .74) were used. Cognitive activation was measured using the subscales ‘cognitive demanding tasks’ (three items, e.g. ‘I asked the students questions they had really to think of.’, Cronbach’s α = .46) and ‘stimulating students’ cognitive independence’ (three items, e.g. ‘When working on challenging tasks, I allowed students to apply their own strategies.’, α = .51). Constructive support was measured using the subscales ‘encouraging students’ (four items, e.g. ‘I showed an interest in every student’s learning.’, α = .72) and ‘differentiated instruction’ (four items, e.g. ‘The single students often had different tasks.’, α = .57). Taking into account that the ratings are highly situation-specific and only a limited number of items could be provided in the short questionnaire due to time constraints, reliability for each scale was still in an acceptable range.

Control variables

As preservice teachers in Germany use individual formats to write their lesson plans, the length will be controlled for using a dichotomous variable (with 0 = short version up to 4 pages and 1 = 5 pages or more). Besides age and gender of preservice teachers, we control for teaching type using dummy coding: type 1 serves as a first dichotomous variable indicating the lower secondary teaching type (coded as 1), type 2 serves as a second dichotomous variable indicating the lower and upper secondary teaching type in North Rhine-Westphalia, and preservice teachers in Berlin serve as reference group.

Analyses

The following intercorrelation and path analyses are carried out using the software package Mplus (Muthén & Muthén, 1998–2015). Missing data are handled with the full information maximum likelihood option implemented in Mplus. Standardised coefficients are reported.

Results

Descriptive results and correlational analyses

Table 1 provides descriptive statistics for all study variables (except control variables). The pedagogical adaptivity measure is extracted from IRT-scaling analysis, thus following the logit scale metric. As can be seen, preservice teachers generally performed on the GPK test with an average score of 27.5 out of the maximum possible score of 44. Self-reports of instructional practice show that preservice teachers’ evaluation had a generally positive tendency, since four scales were rated above the theoretical scale mean (M > 2.5). However, the aspect of ‘differentiated instruction’ shows the lowest mean (M = 2.34) with the largest standard deviation (SD = .66).

Table 1. Descriptive statistics and intercorrelation of study variables (without control variables)

	1.	2.	3.	4.	5.	6.	7.	8.	9.	10.	11.
	GPK	GPK Ad.	GPK Str.	GPK Ass.	PA	IP CM 1	IP CM 2	IP CA 1	IP CA 2	IP CS 1	IP CS 2
1.
2.	.58***
3.	.69***	.18*
4.	.81***	.26***	.36***
5.	.02	.21**	–.06	–.05
6.	–.01	.05	.03	–.05	.07
7.	–.10	–.17*	.01	.01	–.04	.00
8.	.03	–.04	.16*	–.05	.06	.19*	–.02
9.	.11	.05	.06	.15*	.09	.09	.07	.30***
10.	–.02	.01	–.07	.02	.10	.29***	.06	.28***	.17*
11.	.02	.01	.05	.00	.17*	.11	.05	.21**	.23**	.32***
M	27.50	8.30	8.86	3.50	–.34	3.09	2.55	3.25	2.44	3.24	2.34
SD	5.23	1.88	2.16	2.57	1.11	.44	.61	.43	.59	.50	.66
Min.	13.00	4.00	2.00	0.00	–3.09	2.00	1.00	2.00	1.00	2.00	1.00
Max.	38.00	13.00	13.00	9.00	2.74	4.00	4.00	4.00	3.75	4.00	4.00

Abbreviations. GPK – general pedagogical knowledge overall score; GPK Ad. – GPK of adaptivity subscale; GPK Str. – GPK of structuring subscale; GPK Ass. – GPK of assessment subscale; PA – pedagogical adaptivity in written lesson plans; IP – instructional practice; CM 1 – dimension classroom management, subscale ‘preventing disorder’; CM 2 – dimension classroom management, subscale ‘providing structure’; CA 1 – dimension cognitive activation, subscale ‘cognitive demanding tasks’; CA 2 – dimension cognitive activation, subscale ‘stimulating students’ cognitive independence’; CS 1 – dimension constructive support, subscale ‘encouraging students’; CS 2 – dimension constructive support ‘differentiated instruction’.

*** p <.001; ** p <.01; * p <.05.

The correlations of all study variables show that GPK subscales on the one side and instructional practice subscales on the other side are intercorrelated, respectively. Between GPK and pedagogical adaptivity, only the correlation with the subscale GPK of adaptivity is significant (r = .21, p < .01). Between instructional practice and pedagogical adaptivity, only the subscale ‘differentiated instruction’ is significant (r = .17, p < .05). Due to reliability limitations of variables involved, also the disattenuated correlation was computed (r’ = .36 for GPK of adaptivity and pedagogical adaptivity and r’ = .27 for ‘differentiated instruction’ and pedagogical adaptivity), showing that true correlations are slightly higher.

Findings from regression analysis

Effects of GPK on pedagogical adaptivity

To investigate our first research question, a series of simple path models were conducted in Mplus (Table 2). Pedagogical adaptivity served as dependent, general pedagogical knowledge (GPK) served as independent variable. In the first model, GPK was integrated as an overall test score, followed by more differentiated models for GPK subscales. Several variables were controlled for (such as age, gender, GPA), but only the length of written plans and the differentiation into lower and upper secondary teaching type in North Rhine Westphalia (type 2) turned out to be significant. As findings show, only the subscale GPK of adaptivity is a significant predictor for pedagogical adaptivity (β = .20, p < .01).

Table 2. Findings from regression analyses on pedagogical adaptivity as dependent variable

	M1	M2	M3	M4
	β	β	β	β
GPK overall score	.06
GPK of adaptivity		.20**
GPK of structuring the lesson			.01
GPK of assessment				–.05
Teaching type 1	–.03	–.04	–.04	–.04
Teaching type 2	–.31***	–.28**	–.30**	–.29**
Age	−.09	–.09	–.10	–.10
Gender	.04	.06	.05	.05
Length of written plan	.30***	.31***	.30***	.31***
R^2	.24	.28	.24	.24

Abbreviations. GPK—general pedagogical knowledge; Teaching type 1—lower secondary teaching type in North Rhine-Westphalia; Teaching type 2—lower and upper secondary teaching type in North Rhine-Westphalia.

*** p <.001; ** p <.01; * p <.05.

Effects of pedagogical adaptivity on self-reported instructional practice

In order to examine our second research question, a path analysis model was specified in Mplus with pedagogical adaptivity as independent and the six aspects of self-reported instructional practice as dependent variables. As findings show, pedagogical adaptivity significantly predicts the instructional practice scale ‘differentiated instruction’ (β = .17, p < .05). However, it does not significantly predict the other instructional practice scales, neither in the area of classroom management (β = .07 for ‘preventing disorder’ and β = —.04 for ‘providing structure’), nor in the area of cognitive activation (β = .06 for ‘cognitive demanding tasks’ and β = .10 for ‘stimulating students’ cognitive independence’). Also, it does not significantly predict the other subscale in the area of constructive support (β = .10 for ‘encouraging students’). However, at least for the latter two subscales (‘stimulating students’ cognitive independence’ and ‘encouraging students’), predictors were positive and reached the lower level of small practical relevance (.10), showing at least a tendency of effect.

Overall path model

To summarise findings, we specified a path model (Figure 2) with the subscale of GPK of adaptivity as an antecendent and instructional practice (aspect of differentiated instruction) as outcome of pedagogical adaptivity in written lesson plans. This model shows a good fit (χ² = 1.35, df = 4, p = .853; CFI = 1.00; SRMR = .019). As previous analyses have demonstrated, GPK of adaptivity effects pedagogical adaptivity and the latter effects a teacher candidate rating of the instructional practice of the lesson regarding how he or she had provided differentiated instruction in class. In this model, 27% of the variance in pedagogical adaptivity can be explained by the GPK and control variables, 3% of the instructional practice aspect is explained.

Figure 2. Path model to examine pedagogical adaptivity in written lesson plans

*** p <.001; ** p <.01; * p <.05

Discussion and conclusion

Main research findings

The present study investigated the relationship between GPK, pedagogical adaptivity in written lesson plans, and instructional practice, using a sample of 172 preservice teachers at the beginning of their induction in Germany. Two research questions were addressed: First, we asked whether preservice teachers’ GPK could be considered a significant predictor for the situation-specific skill of pedagogical adaptivity in written lesson plans. Second, we examined whether pedagogical adaptivity in written lesson plans would be reflected in preservice teachers’ self-reports of their instructional practice.

Regarding the first research question, as we had hypothesised (H1a), specific declarative GPK of adaptivity (i.e. knowledge of differentiated instruction as well as knowledge of the use of a wide range of teaching methods) significantly predicted the situation-specific skill of pedagogical adaptivity. Furthermore, neither the GPK overall score nor other GPK subscales predicted pedagogical adaptivity in written lesson plans, which also corresponds to our hypothesis (H1b) to a certain extent. These findings show that pedagogical adaptivity in written lesson plans as a construct has a specific focus and that corresponding knowledge in the area of adaptivity can be regarded as a relevant knowledge base. Findings provide insights how preservice teachers make use of a selection of their theoretical GPK when solving a specific problem of lesson planning. As empirical investigation into lesson planning is very scarce, the finding that a particular situation-specific skill of lesson planning is correlated with the corresponding de-contextualised GPK gives rise to the assumption that preservice teachers’ lesson planning is not necessarily independent from their theoretical knowledge. Findings may also be interpreted in the way that they provide evidence for construct validity of the innovative approach measuring aspects of planning competence (Borsboom et al., 2004).

The findings related to the second research question showed that as hypothesised (H2a), pedagogical adaptivity significantly predicted one aspect of instructional practice measuring the way differentiated instruction took place during that lesson. Together with the aspect ‘encouraging students’, ‘differentiated instruction’ formed the operationalisation of the instructional quality dimension of constructive support (Fauth et al., 2014). As expected, pedagogical adaptivity could not significantly predict other subscales of instructional quality related to classroom management (H2b). However, against our expectations (H2a), predictors were not significant for the other subscale of constructive support (β = .10 for ‘encouraging students’) and for the subscales of cognitive activation, where only one subscale had a predictor at the low end of practical relevance (β = .10 for ‘stimulating students’ cognitive independence’). Therefore, only a tendency could be observed in the data.

That the path coefficients are relatively low and only the aspect of differentiated instruction could significantly be predicted might be interpreted from different perspectives. First, teacher planning hardly corresponds perfectly to their actual performance in class. Performance may be influenced by a teacher’s reflection in action (Parsons et al., 2018; Schön, 1983) or interactive teaching (Shavelson & Borko, 1979). Expert teachers might even be more flexible when implementing their plan, as for example, expert-novice differences show (Berliner, 2004). As a consequence, a higher quality level of instructional practice might even be reached through flexible implementation of lesson plans. An expert teacher may observe if the planned learning tasks are not appropriate as expected and, as part of reflection in action (Schön, 1983), modify tasks or task assignment in situ. Second, as already mentioned, self-reports of teachers regarding their instructional practice have their limitations, for example, they might be ‘biased by self-serving strategies or teaching ideals’ (Kunter & Baumert, 2006, p. 231) leading to a gap between the expressed and the actual practices (e.g. Holzberger et al., 2013). However, the correlation found in the present study is somehow in line with a previous study with a small sample of 22 preservice teachers whose written plans were analysed following the same procedure. Their pedagogical adaptivity could predict student ratings of differentiated instruction using the same items in a student questionnaire (König et al., 2015, p. 396): The regression coefficient in multi-level analysis was β = .37 (p < .05). So it seems that in the present study, the disattenuated correlation of r’ = .27 between pedagogical adaptivity and ‘differentiated instruction’ might be trustworthy. Finally, that cognitive activation as a dimension of instructional practice could not be predicted significantly may also be due to measurement problems. There seems to be room for improvement regarding validity. As Praetorius et al. (2018, p. 423) conclude from their review of the basic dimensions of instructional quality, measurement instruments have different operationalisations for cognitive activation and constructive support and ‘the relation between the three basic dimensions and adaptation [of instruction] has not been sufficiently clarified so far’. Future studies therefore should disentangle the relationships between adaptivity and the basic dimensions of instructional quality (Gräsel et al., 2017).

Theoretical and educational implications

The present study, although conducted in two federal states of Germany only, contributes to the international state of research on modelling and measuring teacher competence. Lesson planning is a crucial challenge in many education systems worldwide, and empirical research should focus on situation-specific skills that go beyond the measurement approach of paper-pencil knowledge tests. Situation-specific skills have been investigated so far in the context of professional vision and teacher noticing (Kaiser et al., 2015), focusing on classroom situations, frequently presented by video material. This, however, does not necessarily cover teacher pre-active decision-making, i.e. thoughtful decision-making that precedes the teaching process and that can not necessarily carried out during performance. Lesson planning represents the creative action and reflection of teachers in constructing learning opportunities for their students (Schön, 1983). This is why an investigation of corresponding skills as part of modelling and measuring teacher competence is so important. As the planning skills focused on in the present study relate to challenges that might occur in other contexts as well (e.g. Chizhik & Chizhik, 2018, for the US), principles on the conceptualisation of planning skills in the present study may be of interest in other cultural contexts and teacher-education systems as well.

Our approach further has relevance for the teacher education discourse. Whereas paper-pencil tests measuring teacher knowledge may be useful to clarify academic standards that should be reached by preservice teachers, instruments measuring teacher situation-specific skills may have the potential to support preservice teacher learning (Darling-Hammond et al., 2013). There is a critical discussion about the function of practice in teacher education and the orientation towards practice in teacher education learning environments (Ball & Forzani, 2009; König & Rothland, 2018). The present study findings suggest that knowledge preservice teachers have acquired during initial teacher education forms a basis for planning authentic situations during induction. As teacher expertise literature suggests, preservice teachers may profit from their knowledge during lesson planning if they can access a repertoire of teaching strategies from which they can select those that are suitable for teaching a particular learning group (Stigler & Miller, 2018).

Lesson planning has, at least in Germany, been dominated by theories or practical guides. For example, Scholl (2018, p. 18) identified more than 100 theories published since 1945 that did not build on each other and that have never been empirically examined. As a result, advancement of knowledge in the field is very limited, a problem that seemingly exists in other countries as well (e.g. John, 2006, for the UK). Against this background, the present study started with a different approach by suggesting pedagogical adaptivity as a teacher competence construct that reflects empirical findings from teacher expertise research and refers to didactical concepts (Kansanen, 1995; Scholl, 2018), differentiated instruction (Dack, 2018; Tomlinson, 2014), the research on adaptive teaching (Parsons et al., 2018) as well as constructivist learning theories (Vygotsky, 1978). Pedagogical adaptivity therefore is linked to existing research approaches and its operationalisation allows a clear definition. Such a definition as well as the indicators used (Table A2) may support discussions that aim at clarifying which criteria high-quality lesson plans should fulfill. At least in Germany, there is the need to increase the quality of assessment criteria for demonstration lessons (Kärner et al., 2019; Strietholt & Terhart, 2009). Research on lesson planning competence following the approach demonstrated in the present study can make a valuable contribution to such a reform issue in teacher education.

Limitations

As mentioned previously, pedagogical adaptivity as measured in the present study is only one aspect of lesson planning. The scope of the construct is therefore limited. Much further research is necessary to elaborate on the term ‘lesson planning competence’ and its meaning. For example, although the construct of pedagogical adaptivity presupposes preservice teachers to link their planning to content, the genuinely domain-specific challenges of pedagogical adaptivity have only indirectly been reflected in the lesson plan coding (Table A3). In order to start filling this research gap, the subject-specific potential of the learning tasks, being a relevant factor for cognitive activation of students (Neubrand et al., 2013), will be analysed in the project PlanvoLL-D, accounting for the domain-specific challenges of pedagogical adaptivity. However, one should keep in mind that such a content-specific analysis might be much more complex. The question arises whether in the written plans of demonstration lessons, the preservice teachers provide such very detailed content-specific information that have substantially added-value to our more generic lesson plan coding.

Written lesson plans have their limitations as well. Preservice teachers do not necessarily write down all their planning decisions (e.g. Borko & Livingston, 1989; McCutcheon, 1980). Particularly, preservice teachers being novices and therefore in contrast to expert teachers have difficulty in diagnosing student needs and prior knowledge (e.g. Putnam, 1987; Westermann, 1991), leaving open the reliability and validity of preservice teachers’ perception and description of the learning group. Future research should address an examination, for example, by comparing preservice teachers’ perception with student assessment data. Curricular requirements posed on preservice teachers also influence their written plans. Preservice teachers may seek to reach other goals such as fulfiling examination regulations or personal expectations of teacher educators (Calderhead, 1996). However, the sample of the present study consists of preservice teachers from two different federal states in Germany (where educational regulations are in charge of) and a large number of training units within these states with acceptable or even good response rates. Therefore, a certain generalisability should be possible and the bias of having coded irrelevant information from written plans that could be related only to a single training unit should presumably play a minor role.

Conclusion

Lesson planning is a relevant instrument that governs teaching and learning in the classroom. It allows teacher to pre-active decision-making and reflection on action. The present study has provided evidence that preservice teachers’ knowledge is relevant for pedagogical adaptivity as one particular aspect of lesson planning and that this kind of lesson planning investigated has some consequences for the way a preservice teacher deals with student heterogeneity in class. We need much further empirical research in this area, but the present study may provide a valuable suggestion for a possible approach to further investigate teacher competence in lesson planning.

Supplementary material

Supplemental data for this article can be accessed here.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by the Federal Ministry of Education and Research, Germany [Bundesministerium fur Bildung und Forschung, BMBF, grant number 01PK15014A, 01PK15014B, 01PK15014C].

Notes on contributors

Johannes König

Johannes König is a full professor of empirical school research at the University of Cologne, Germany. His current research includes school research, teacher education research, teacher competencies, teacher knowledge and international comparisons.

Albert Bremerich-Vos

Albert Bremerich-Vos is a full professor of linguistics and of teaching German language at the University of Duisburg-Essen, Germany. His current research includes PIRLS (progress in international reading literacy study) 2016, process-oriented assessment of reading, writing and orthography proficiency (grades 5 -7).

Christiane Buchholtz

Christiane Buchholtz is an academic councilor at the Technical University Berlin, Germany. Her current research includes teacher education research, esp. measuring teacher competencies and modification of teaching patterns.

Nina Glutsch

Nina Glutsch is a research assistant at the University of Cologne, Germany. Her current research interests include teacher education research, teacher motivations and competencies.

Appendix

Table A1. Item examples from the GPK test

GPK dimension	Item example	Correct solution/correct sample response [explanation]
Adaptivity (12 items)	Name three teaching methods that accomodate student differences that allow students some choice during the lesson. 1) 2) 3)	1) Weekly lesson plan 2) Station learning 3) table offering activities/work sheets that are selected by the students
	Which of the following learning objectives require measures of external or internal differentiation? Check one box in each row. [Response Categories: external differentiation, internal differentiation] A. Students’ learning processes should be individualised. B. Classrooms should be as homogeneous as possible. C. Students’ individual differences should be taken into consideration within the classroom. D. In elementary schools, students with a learning disability should be integrated into regular lessons.	External differentiation: B, Internal differentiation: A, C, D
	Imagine you are planning an exercise lesson for a first grade class on addition of the numbers until 20. Since the class is very heterogeneous considering their abilities, you want to offer three different types of exercises which correspond each with Bruners modes of representation: enactive, iconic, and symbolic. Assign the following examples to the three exercise types. Check one box in each row. [Response Categories: enactive, iconic, symbolic] A. Calculate: 8 + 12 = __ B. Arrange red and yellow gummy bears to mixed groups of the same size. C. Encircle white and black birds on a worksheet to groups of 20 birds.	Enactive: B Iconic: C Symbolic: A
Structuring (13 items)	Imagine you are helping a future teacher to evaluate her lesson because she has never done this before. To help her adequately analyse her lesson, what question would you ask? Formulate ten essential questions and write them down.	1) Do your students have prior knowledge about the subject? 2) What are your objectives? 3) Are the students working individually or in groups? … 10) Have your students gained the intended knowledge from the lesson?
Classroom management/motivation (9 items)	Which of the following cases represents an example of intrinsic motivation, and which represents an example of extrinsic motivation? Check one box in each row. [Response Categories: Intrinsic motivation, Extrinsic motivation] A student learns before a test in mathematics, because he/she … A. expects a reward for a good grade B. wants to avoid the consequences of a bad grade. C. is interested in problems of mathematics. D. does not want to disappoint his/her parents. E. wants to maintain his/her relative rank in the class.	Intrinsic motivation: C, Extrinsic motivation: A, B, D, E
Assessment (6 items)	Assessment methods are fair and accurate if three quality characteristics that describe educational assessments are fulfilled. What are these characteristics? A. Neutral, Reliable, Valid B. Objective, Reliable, Valid C. Objective, Reliable, True D. Neutral, Reliable, Valid	B

Table A2. Coding scheme for analysing pedagogical adaptivity in written plans of demonstration lessons (König et al., 2015)

Component	Indicators (Codes)
(1) Description of situation-specific factors	The teacher describes the learning group. (10)
	The teacher describes interindividual differences in cognitive preconditions of the learning group. (11)
	The teacher describes interindividual differences in motivational preconditions of the learning group. (12)
(2) Description of the learning task	The teacher describes the learning task. (20)
	The learning task explicitly comprises different cognitive levels (explicit instruction of student differentiation). (21)
	The differentiation of learning tasks outlined under 21 is supported with reasons. (22)
(3) Applying descriptions to the specific situation	The teacher describes the specific cognitive level of students (whole learning group) towards the learning task following the „zone of proximal development“. (30)
	The teacher describes the specific cognitive levels of students (student differentiation) towards the learning tasks following the „zone of proximal development“. (31)
(4) Connecting elements of planning	Learning task(s) and lesson objective(s) is/are connected. (40)
	Grouping students is connected with student and task(s) differentiation. (41)
	Check of results required by learning task(s). (42)

Table A3. Examples from the lesson plan coding

Example from the lesson plan coding (Component 2)
Indicator (Code)	Text from lesson plan
The learning task explicitly comprises different cognitive levels (explicit instruction of student differentiation). (21)	“Each student of a working group gets different task material (…). Using this assignment, the teacher can make sure that strong students work on more challenging aspects (closely related to the text) and weak students work on aspects (related to tenses) that are easier to master.”
The differentiation of learning tasks outlined under 21 is supported with reasons. (22)	“The method of partner puzzle was chosen, since it enables students to inform each other and supports weak students when checking results.”
Example from the lesson plan coding (Component 3)
Indicator (Code)	Text from lesson plan
The teacher describes the specific cognitive level of students (whole learning group) towards the learning task following the “zone of proximal development”. (30)	‘The lesson described is the ninth lesson of a unit. The unit consists of 14 lessons altogether. In the course of the unit, students have already worked with rhetorical devices of advertising language (…). In a preceding lesson, students created a print advert from given modules. In order to do this, they had to understand and make use of rhetorical devices of advertising such as hyperbole, anglicisms (…). As a next step, students are asked in this lesson to compose an advertising slogan for a common product following some given guidelines. The slogan is part of a self-designed advertising spot which should appeal to the respective target audience.’
The teacher describes the specific cognitive levels of students (student differentiation) towards the learning tasks following the “zone of proximal development”. (31)	“Then students are assigned to groups 1, 2, and 3, whereby each group closely investigates one citation. These citations can be found on the individual worksheets. The difficulty of identifying the meaning of the individual citation varies. Therefore, student grouping will be done as described above [i.e. every student is assigned to a table according to his or her reading and writing competence which can be minimum, medium or maximum standard related to reading and writing competence specified for the lesson’s objective]. Further differentiation will be done through scaffolding and different learning tasks. All tasks, however, require from students to select appropriate examples.”
Example from the lesson plan coding (Component 4)
Indicator (Code)	Text from lesson plan
Learning task(s) and lesson objective(s) is/are connected. (40)	“By resolution of our school’s German language department the stage play ‘Heart of a Boxer‘ (Herz eines Boxers) for young people is part of the 8th grade syllabus of our school. It corresponds with the task type 6: “Production-oriented writing”. Furthermore, it promotes the competencies outlined in the standard curriculum for German in lower secondary schools (Realschule) in North-Rhine Westphalia. Teaching the play follows the requirement “Dealing with literary texts” and in particular the aspect “Listening” as stated: “The students find key words and form sentences in order to demonstrate their understanding and the content of recited texts” (see Curriculum German for Lower Secondary Schools). The student assignment in the opening phase of this lesson is to listen to the recorded scene and take notes of the characters’ feelings.“
Grouping students is connected with student and task(s) differentiation. (41)	‘When grouping students, I payed attention to student heterogeneity regarding their reading competence and writing competence. I expect that strong students will help weak students.’