Theory on Teaching Effectiveness at Meta, General and Partial Level

Abstract

This chapter focuses on theories in the context of educational effectiveness research. Three levels of theory are distinguished: (1) The meta-theoretical level uses a multi-level framework of measurable facets of educational quality to provide a structure into studying teaching and its effects on student learning. (2) The general theory of teaching effectiveness focuses on major substantive dimensions of pro-active, interactive, and retro-active strategies as well as “structured independence”. (3) Partial theories are more specific in being closely tied to empirical outcomes, with “direct teaching” and tentative explanatory mechanisms in the realm of providing emotional support in classroom teaching being two examples. The three levels are considered to discuss the value of theories for research on teaching and its effects on student learning.

1 Introduction

In this chapter I would like to reflect on the role of theory for research on teaching effectiveness. The focus on effectiveness reflects the preference to address teaching as instrumental to learning and student outcomes. Not only is this more interesting from a scientific perspective than focusing on teaching in general, as it sets the stage for causal theories about “what works, but it also has a natural connection with the “applied” question of educational quality. The structure of the paper is based on a sequence of three ways of theorizing about teaching effectiveness. A first contribution of “theory” to teaching effectiveness is the presentation of an underlying structure for empirical research and educational discourse. This is done by positioning teaching effectiveness in a multi-level framework of educational effectiveness and measurable facets of educational quality (Sect. 3). In a second step, major substantive dimensions of the basic framework are discussed: the elementary parts of teaching, pro-active and retro-active regulation in teaching, “structured independence”, and classroom management, which includes classroom ecology and climate. These substantive dimensions are considered as building blocks for a general theory of teaching effectiveness (Sect. 4). Next, in a third step, more specific partial theories, closely tied to empirical outcomes, will be addressed by providing two illustrative examples: the theoretical underpinnings of “direct teaching” and tentative explanatory mechanisms in the realm of providing emotional support in classroom teaching (Sect. 5).

In the discussion section the status and usefulness of theory in an empiricist field like teaching effectiveness research will be assessed (Sect. 6).

2 Terminology

I will discuss theory about teaching, as an empiricist theory framed as teaching effectiveness. An empiricist theory, according to Chambers (1992), is one in which generalizations about observable variables are related to one another in ways that accord with observation (cited by Gage, 2009 p. 22). According to Snow (1973, p. 78) “its simplest form, a theory is a symbolic construction designed to bring generalizable facts (or laws) into systematic connection”. The latter element, “systematic connection” reflects the aspiration that generalizations are connected to conjectures that explain the “how” and why” of the empirical findings. The logical structure of scientific explanations is the so called “covering law model”, which means that research-based, empirical generalizations can be subsumed under a more general law or principle, which could be part of a broader and already established theory (Hempel & Oppenheim, 1948; Von Wright, 1971).

In the rational empiricist research tradition theoretical conjectures and empirical findings are intertwined in a cyclic process of inductive and deductive phases (de Groot, 1961). In the case of teaching effectiveness this process is still largely dominated by inductive interpretations based on empirical generalizations. So, when discussing theory in the field of teaching effectiveness there is likely to be more emphasis on “bottom up” theory formation, than on research that is guided “from above” and driven by existing theory.

Meta-theory is defined by Snow (1973, 79) as “a theory concerned with the development, investigation or description of theory itself”. As a first example he mentions the Stimulus-Response versus the Stimulus-Organism-Response model in psychology, also indicated with the term “paradigm”. Other examples of meta-theories are methodologies and epistemological positions, like the distinction between nomothetic and ideographic approaches. In this chapter I will use three distinct interpretations of meta- theory with respect to teaching effectiveness:

In the sense of a logical structure of causal conditions to realize intended educational outcomes, based on a model from systems theory, which distinguishes context, input, process, and outcome indicators (CIPO). I will usually refer to the distinction of educational effectiveness within this framework with the term “educational effectiveness research paradigm” (Sect. 3).

In the sense of the epistemological premises of the scientific method on which educational effectiveness research depends.

In the sense of the nature of substantive theory (ontology). In normal science is has an interpretation in the sense of defining characteristics of a substantive theory.

Substantive theories on teaching effectiveness are to be defined in close connection to the state of the art of empirical research, as manifest, for example, from meta-analyses of research studies. Substantive theory will be discussed at two levels of abstraction and empirical specificity:

• As contribution to a general theory of teaching, in the sense of a “conceptual map” which distinguishes a limited set of substantive dimensions through which teaching affects student learning and student outcomes; (Sect. 4). These substantive dimensions correspond to the level of abstraction in what Gage (2009, p. 123) refers to as “sub-theories” defined for main components of teaching and Praetorius et al. (2020) describe as basic dimensions of teaching in their TBD (Three Basic Dimensions) model.

• As partial theories of teaching, to be defined in close connection to the state of the art of empirical research. Partial theories have a place in the main dimensions of the conceptual map, described in Sect. 4. Of these I will give just two illustrations: theoretical conjectures with respect to “direct teaching” and “social emotional support in teaching” (Sect. 5).

In the sequel basic terms like theory, model, paradigm, (conceptual) framework and conception will be used in accordance with the Introduction chapter of this volume (Praetorius & Charalambous, 2023, Table 2, p. 10/11). The distinctions between these terms are somewhat fuzzy, for example when it comes to distinguishing models and theories. Still there is an ascending order in grades of formal development and sophistication when going from conception, via framework and model, to theory. This would imply that the more ambitious concepts include less ambitious ones, in the sense that, for example, all frameworks are also conceptions, but that not all conceptions are frameworks. Given this global understanding I have applied these terms in a liberal way, where I would sometimes indicate a certain configuration as a conception, a framework, or a model.

3 Meta-Theory. A Systems Model on the Functioning of Education as the Underlying Structure of an Educational Effectiveness Research Paradigm and Its Practical Implications

In this section an underlying structure for empirical research and educational discourse is presented. This is done by positioning teaching effectiveness in a multi-level framework of educational effectiveness and measurable facets of educational quality. As indicated in Sect. 2 , such a general underlying structure is seen as a meta-theory.

3.1 A Model Representing the Educational Effectiveness Research Paradigm

The elementary design of educational effectiveness research is the association of hypothetical effectiveness enhancing conditions and output measures, mostly student achievement. The basic model from systems theory, shown in Fig. 4.1, is helpful in clarifying this design. The major task of educational effectiveness research is to reveal the impact of relevant input characteristics on output and to “break open” the black box in order to show which process or throughput factors “work”, next to the impact of contextual conditions.

Fig. 4.1

A basic systems model on the functioning of education

The model, shown in Fig. 4.1, sometimes indicated with the acronym CIPO (which stands for Context, Input, Process, Output) can be used at different levels of aggregation, the level of a national educational system, the school level, the level of the instructional setting, often indicated as the classroom level and the individual student level. The levels are nested, in the sense that schools function within an educational system at national level, classrooms function within schools and students within classrooms. This is illustrated in Fig. 4.2.

Fig. 4.2

Integrated multi-level model of education; solid arrows represent managed control actions, the dotted arrows from one system level to the next represent across level influences; feedback-loops (not shown in the diagram) are assumed to run from outcomes at each level to the box containing ecology and active policies at each object level and from lower to higher levels

Measured outputs, like (aggregated) achievement test data, make up the effect side of the model. Next, malleable or “managed conditions” (inputs and processes) hypothetically associated with outcomes form the core in the multi-level representation. The term “antecedents” is used to indicate previously given contextual conditions. These antecedents may represent the larger environment and culture, higher level policies as well as existing characteristics of teachers and students. At school and classroom level, the term ecology refers to partly controllable composition effects and their interaction with malleable variables (e.g. the interaction between classroom SES composition and a “good” relational climate at school (Luyten et al., 2005). A more detailed description of the model is given in Scheerens (2016).

Given the multi-level framework a bottom-up logic could be used for designing an overall structure where lower-level processes (starting from student learning) are “boosted” by higher level conditions. For example, when it is established that “control strategies” in students’ learning are effective (Artelt et al., 2003), structured teaching could be seen as a strategy that particularly supports weaker students.

The degree of higher- level control versus lower-level autonomy is an issue of central importance at all levels. At system level it is about effective patterns of functional decentralization, which means that, perhaps dependent on the larger context, certain patterns of centralization in some functional domain (e.g., the curriculum) and decentralization in another domain (e.g., financial management) work best. At school level it is about the degree of participative decision making, or “distributed leadership”, and at classroom level it refers to the balance between strongly structured didactic approaches and more open teaching and learning situations that are expected to invite self-regulated learning. Structure versus independence is a red line that dominates policy and research agendas in education. Critics of educational effectiveness thinking sometimes have a tendency to depict it as a “closed” mechanistic perspective, neglecting, for example the professional autonomy of teachers. Here, on the contrary, autonomy is built into the system.

Further elaboration of the framework is provided in Scheerens et al. (2011, 2016). Multi-disciplinary applications from economics, sociology and psychology are discussed in Scheerens and Bosker (1997). The framework can be seen as a structure for multi-disciplinary educational research.

3.2 Scientific and Applied Use of the Framework

Although research on teaching is likely to take place in field settings, and in this sense applied, there are differences in degree to which the overriding interest is scientific as compared to supportive of practical or political decision-making. Making this difference more explicit is another example of meta-theoretical reflection, in line with Snow’s definition, cited in Sect. 2.

The context- input-process- outcome framework can accommodate in-depth description of teaching, as well as associations of teaching processes with all other components in the framework, like for example the influences of school level policies on teaching strategies. From a scientific perspective the causal association of characteristics of teaching processes with outcome measures is of interest in assessing the predictive validity of measures of teaching processes (like direct observation schedules) and to test theories about the way specific teaching approaches affect learning and learning outcomes. Measurement issues concerning various components of the framework and questions of causal attribution, like for example the issue of “value-added” outcome indicators, are part and parcel of both a scientific and applied use.

Turning to the applied use of the framework, the structure is fit to define core policy issues in education, which can be seen as measurable facets of educational quality (Scheerens, 2016; the C., I., P., and O associations refer to the components of Fig. 4.1):

• educational productivity can be highlighted by focusing on output variables at different aggregation levels, the well-known comparisons between mean performance levels in educational assessments studies between countries, such as TIMSS and PISA, are examples of comparisons of national systems on educational productivity (O-quality)

• educational effectiveness would seek to determine the “net” effect of malleable educational conditions, defined at different levels, on outputs, while controlling for relevant antecedent conditions at the level of individual participants (IxPxO-quality)

• educational equity is captured by examining disparities between resources and processes as well as the variation between students and schools in educational outputs; and the degree to which achievement levels and disparities hang together with specific antecedents of students, schools and school contexts, e.g. the reading performance of girls from cultural minority background and the average achievement levels of schools in rural areas; (IxPxO-quality|sub-populations)

• educational efficiency would address questions of input provision and effectiveness at the lowest possible costs (IxPxO-quality|costs)

• educational responsiveness represents adaptive or proactive outreach of the educational system towards the relevant environment to shape intended outcomes (e.g. by means of a representative survey of support for a revision of educational goals and curricula) (CxO)

It is important to note that the responsiveness perspective ‘transcends’ the instrumental effectiveness perspective by not only looking at the question of how to do thing right, but first considering the question of how to do the right things. In other words, the responsiveness perspective leads to a critical analysis of educational goals. Adaptation to contextual demands can be situated at system, school and classroom level. For example, defining the school curriculum as a means to adapt to national standards, as well as to the specific environmental local context.

4 Substantive Building Blocks for a General Theory on Teaching Effectiveness

In this section we are leaving the realm of meta-theory to which the conceptual framework or paradigm of teaching effectiveness belongs, and address building blocks of a general substantive theory on teaching effectiveness. Substantive theories on teaching effectiveness are to reflect the state of the art of empirical research, as manifest, for example, from meta-analyses of research studies. After the introduction Sect. (4.1) I will present a summary of indicators, supported by empirical research, and categorized by the section on the classroom level in Fig. 4.2. Next, in Sect. 4.3 I will explore the possibility to describe a general theory of teaching effectiveness as a “conceptual map”, which distinguishes a limited set of dimensions through which teaching affects student learning and student outcomes and explain in what way such dimensions could become “building blocks” for a general theory on teaching.

4.1 Introduction

Snow’s (1973, 87) work on distinguishing “grades of theory” and steps in the process of progressively developing higher levels of theory, provides a basis for speculating about the level of development of proclaimed theories of teaching, including dimensional models as indicated in the above. Starting out from “relatively simple summarization of empirical relationships without substantial inferences or deductive logic”, which characterize lower level theories, he sketches the trajectory to further developed higher levels of theory. Conceptions that credibly simplify the description and explanation of relationships among observed variables have higher theoretical status than categories that are just summary labels. Other criteria that should be met are that conceptions represent meaningful dimensions, manifest parsimony, and have explanatory potential (e.g. when linked to more established theory). As such conditions are met, lower-level theories evolve to middle level theories, characterized by “continuing interaction” between provisional theoretical concepts and data (Reezigt et al., 1999). (ibid, 87). Higher level theory is characterized by a formal logico-deductive structure, with derived hypotheses guiding empirical testing. As to the standing of theory formation on teaching effectiveness, earlier analyses pointed out that it is predominantly at the lower end of the continuum presented by Snow; with “formative hypotheses”, “elementism” (development of key-concepts and instruments), and “descriptive theories and taxonomies” as its major accomplishments and “eclecticism”, or connecting with eclectic fields of developed theories, as rare exceptions (Scheerens, 2016, 260/261). Gage‘s attempts to connect “structuring” as one of his “sub-theories of teaching” to “covering laws” is such an exception (Gage, 2009). The theoretical support for the effectiveness of explicit direct instruction on the basis of cognitive information processing theory, to be discussed in Sect. 5, is another (Kirschner et al., 2006).

For the current presentation this state of affairs, with theory development progressing from lower to middle level of theory formation, is taken as an encouragement to working towards a parsimonious set of key dimensions, considered as building blocks for a general theory of teaching.

In summaries of the research literature on educational and teaching effectiveness authors have frequently subsumed operational variables that had received empirical support under a limited number of categories (Anderson, 1991; Creemers & Kyriakides, 2008; Klieme, 2012; Praetorius et al., 2020; Reezigt et al., 1999; Seidel & Steen, 2005; Scheerens, 2016; Stringfield & Slavin, 1992). Creemers and Kyriakides for example, distinguish the following “factors operating at student and classroom level”: orientation, structuring, modelling, application, questioning, assessment, management of time and the classroom as a learning environment. Klieme (2012) mentions “content exposure and structure”, “classroom management”, supportive classroom climate”, and “cognitive activation” as major rubrics under which 20 more specific variables are categorized. Gage (2009), distinguishes four “sub-theories of teaching”: a conception of the process of teaching (in which he basically contrasts traditional and constructivist-oriented teaching), a conception of the content of teaching (with curriculum and instructional alignment as the key factor), a conception of students’ capabilities and motivation, and a conception of classroom management. Chapters 4 and 6 of this volume show further developments in the dimensional models by Klieme and co-authors and Kyriakides and co-authors, respectively. My particular take on the tentative improvement of the current dimensional models is driven by (a) the interest in exploring to what extent dimensions could be conceptualized as continua, over and above colligations of discrete variables and (b) the challenge to find a parsimonious synthese of empirical variables from earlier work (Scheerens, 2016, p. 46). As a first step in trying to meet this latter challenge, I will start this section with a paragraph that present empirical indicators, embedded in one intersection of the model depicted in Fig. 4.2.

4.2 A Closer Look at Classroom Level Teaching

In my book “Educational effectiveness and ineffectiveness” (Scheerens, 2016) I used the model, illustrated in Fig. 4.2 to summarize research results. Apart from discussing the overall systemic properties of bringing together these levels in educational effectiveness, I summarized major outcomes in models of “learning”, “teaching”, effective “schooling” and “system level levers of educational effectiveness”. Summary results based on empirical research at each level were formulated as indicators of major input, contextual/ecological and malleable process conditions. I will use the summary results on teaching effectiveness as a starting point for the development of substantive building blocks for a general theory of teaching. The overview of teaching variables is shown in Fig. 4.3.

Fig. 4.3

Overview of indicators of effective teaching

The three columns in Fig. 4.3 show teacher characteristics, as a particular category of input indicators, indicators on classroom ecology and climate (seen as partly given and partly malleable) and teaching processes. In this chapter no further attention will be given to teacher characteristics, as the emphasis is on teaching processes and creating productive learning environments (the second and third column). The sub-categorization in the third column distinguishes pro-active, interactive and retro-active facets of teaching. I was prompted to use this particular categorization on the basis of contributions in the area of variations of the rational planning model (Scheerens et al., 2003). For completeness’s sake I would like to mention that in the overall multi-level structure (Fig. 4.2) teaching effectiveness is seen as embedded in school and system level influences; a perspective that will not be elaborated on in this chapter.

Further reflection on the underlying structure of the overview of indicators, in the sense of a limited set of meaningful dimensions, is the next step in discussing “building blocks” for a general theory on teaching. How exactly such dimensions are considered as a step in theory formation is indicated in the next sub-section.

4.3 Can Summary Categories of Variables Be Interpreted as Meaningful Dimensions for a General Theory on Teaching?

As it comes to the identification of core dimensions of teaching there is no need to reinvent the wheel. We can remain close to the contributions that are made in the literature, cited in the introductory part of this section. The main dimensions distinguished by Klieme and Gage are appealing in their coverage and parsimony. Klieme (2012) mentions “content exposure and structure”, “classroom management”, supportive classroom climate”, and “cognitive activation” as major rubrics under which 20 more specific variables are categorized. Gage (2009), distinguishes four “sub-theories of teaching”: a conception of the process of teaching (in which he basically contrasts, traditional and constructivist oriented teaching), a conception of the content of teaching (with curriculum and instructional alignment as the key factor), a conception of students’ capabilities and motivation, and a conception of classroom management. Closely following Gage (2009), and building on the structure of the indicators summarized in Fig. 4.3. I would opt for the following dimensions:

• pro-active and retro-active regulation in teaching (as a content oriented dimension)

• structure and independence in teaching (as a central process dimension)

• classroom management (as a comprehensive summary dimension of effective teaching interventions).

The work on pro-active and retro-active regulation in teaching, in relation to curriculum and instructional alignment prompted me to distinguish a fourth dimension, indicated as “the elementary parts of teaching”.

The overview presented below provides an advance organizer of the structure and content of the ensuing sub-sections, in which the building blocks will be described in more detail.

On Content

On content two related dimensions are considered, “the elementary parts of teaching” and pro-active and retro-active regulation in teaching.

The first considers the matching of content elements, psychological operations, and didactic considerations as the elementary parts of teaching. Distinguishing these “elementary components” of teaching, is in fact a general analytic definition of what teaching is. The theoretical potential of this dimension is that it provides a decisive distinction for the debate on subject matter-based objectives and “subject matter free” skills. In addition, it provides a basic “grammar” for defining curriculum alignment, not solely a matching of content but also of psychological operations.

The second dimension associated with content is labeled pro-active and retro-active regulation in teaching and deals with instructional alignment and learning from feedback. Here the eye-opener is the potential of retro-active regulation as an efficient lever of student achievement, particularly in settings where pro-active approaches are constrained because of vague objectives, lack of standardized methods and pedagogical principles (“open” education).

On Process

The third dimension, “structure and independence in teaching” sees this distinction as a continuum which might serve in overcoming entrenched positions between behavioristic and “constructivist” teaching strategies. The process of “fading”, in the sense of gradually diminishing structure in a course or teaching sequence is seen as a bridging principle. Optimization in the degree of structure could be considered from the perspective of contingency theory, where effectiveness depends on student and situational characteristics.

On Classroom Management

The fourth building block refers to classroom management, with managing learning opportunities, classroom ecology and climate, and cognitive and emotional support as sub-categories. Unlike the three other dimensions, described in the above, “classroom management” is just an overarching label. The common element is that all sub-categories are malleable and, in principle, controllable by teachers. An alternative might be to consider the three sub-categories as dimensions in their own right. It is also debatable to put cognitive and emotional support together in one sub-category. Classroom management in the sense of a general collective concept could be analyzed from the perspective of control theory, and organization theory. Optimization is best perceived at the level of the sub-categories separately.

4.3.1 The Elementary Parts of Teaching: Matching of Content Elements, Psychological Operations, and Didactic Considerations

In prescriptive formulations, which concern for example the structure of educational objectives, two dimensional classifications are usually proposed (e.g. De Corte et al., 1976; Bloom et al., 1971). Subject matter content and psychological operations are the two basic dimensions. This perspective entails breaking down the contents, of e.g. a geography lesson in smaller units, and, for each content unit specifying the cognitive, or affective behavior/dispositions that should be acquired. Taxonomies of educational objectives have specified continua of operations that ascend in complexity. For example, in the cognitive domain: perception of information, recognition of information, reproduction of information, interpretative production of information, convergent production of information, evaluative production of information and divergent production of information.

Elements defined by the two basic dimensions (content and psychological operations) are used to indicate and specify learning tasks and achievement test items.

In the act of teaching, specific presentation forms and media should be applied when introducing the prime didactic elements as described in the above. The quality of teaching would thus depend on: adequate selection of content; indicating target psychological operations (e.g. cognitive behavior); knowledge about creating tasks; instructional knowledge; a repertoire of presenting and guiding the execution of learning tasks, knowledge about students, and typical behaviors for the learning task in question, including frequently made mistakes. The first three characteristics could, in principle, be taken care of outside the direct teaching situation, by curriculum experts and designers of teaching methods, in cooperation with panels of teachers. Yet, knowledge about content, about tasks as independent tools and about student thinking would constitute a basic teaching competency that was described as pedagogical content knowledge by Shulman (1986). Baumert et al. (2010) found evidence of considerable impact of this variable on student achievement.

Combination of content units and psychological elements bearing in mind pedagogical and didactic considerations are at the heart of instructional sub-disciplines as: curriculum development, teacher training and teacher professional development, lesson preparation, actual teaching and the monitoring of students’ reactions to teaching (e.g. time on task), assessment of learning outcomes, and providing feedback on the basis of assessment.

When it comes to the coverage of the indicators in Fig. 4.2 by the “elementary parts of teaching” the sub-areas “professional knowledge”, “pro-active strategies” and retro-active strategies can be subsumed under this content dimension. Taxonomies of educational objectives provide ranked continua.

A final observation is that didactic analysis offers an analytic structure that is helpful to the “skills debate” (e.g. Weinert, 2001). This debate addresses the question of the place of subject matter content in the teaching of skills. As other analysts, Weinert rejects the position that skills are “content free”.

4.3.2 Pro-Active and Retro-Active Regulation in Teaching

Among the set of educational sub-disciplines that were mentioned when we dealt with the “elementary parts of teaching”: curriculum development, teacher training and teacher professional development and lesson preparation could be said to have a pro-active orientation. Of actual teaching one could say that it has an interactive orientation while assessment and providing feedback have a retro-active orientation.

In this section pro-active and retro-active regulation will be compared; it should be noted that pro-active and retro-active regulation have a clear interpretation in core processes at school management and system level governance as well.

The ideal of “synoptic” planning is to conceptualise a broad spectrum of long-term goals and possible means to attain these goals. As such it contains the basic logic of social engineering and planned change, in our case design of teaching and learning situations. In models of planned change the various aspects of synoptic planning are usually structured as phase models; which basically distinguishing goal specification, means specification, implementation, evaluation and feedback, Ackoff (1981, 74, 75).

Applying feedback turns the sequence in steps into a circle that can go on and on. Many authors, including Ackoff, do not take the sequence of phases too seriously and say that they take place in any order. Others, however, see the way one “steps into” the planning, implementation and feedback circle as non-trivial. Borich and Jemelka (1982) see the planned change process as society’s attempts to “maintain equilibrium when the system threatens to become disadvantageously influenced by forces whose effects were previously neglected or would have been difficult to predict” (ibid, 216). Next to the traditional pro-active approach, they discern a retro-active orientation.

In a retro-active regulation of teaching the assessment instrumentation, for example a large item-bank, could be legitimized as the intended curriculum. In the ideal situation of an exhaustive item-bank, “teaching to the test” could be seen as a legitimate and recommendable activity. A similar kind of reasoning could be applied to formative assessment, where assessment is expected to feed into ongoing teaching activity. The theme of proactive and retroactive control will be elaborated by referring to curriculum alignment and summative assessment, and to learning from feedback in formative assessment.

Instructional Alignment

The core element of educational alignment, either in the sense of curriculum alignment, or instructional alignment, is best known as “opportunity to learn” (OTL), defined as the matching of taught content with tested content. As such OTL is part of the larger concept of curriculum alignment in educational systems. When national educational systems are taken as multi-level structures, alignment is an issue at each specific level, but also an issue of connectivity between different layers. General education goals or national standards are defined at the central level (the intended curriculum). At intermediary levels (between the central government and schools) curriculum development, textbook production and test development take place. At the school level, school curricula or work plans may be used, and at classroom level, lesson plans and actual teaching are facets of the implemented, or enacted, curriculum. Test taking at individual student level completes the picture (the realized curriculum). This process of gradual specification of curricula is the domain of curriculum research, with the important distinction between the intended, implemented and realized curriculum, as a core perspective. This perspective is mostly associated with a proactive logic of curriculum planning as an approach that should guarantee a valid operationalization of educational standards into planning documents and implementation in actual teaching (Kurz, 2011; Luyten & Scheerens, 2021).

De Groot (1986) defined an overarching model of “didactic and evaluative specification of educational goals”. Educational goals are based on perceptions of pupils’ needs and societal needs. He then distinguishes two kinds of operationalizations, one leading up to the construction of curriculum products, such as learning programs, textbooks, and teaching methods, and the other to the development of examination and assessment programs. De Groot’s framework underlines the analogy between curriculum and test design and offers criteria to determine the quality and alignment of these two construction processes. According to De Groot evaluative operationalization should happen first because curriculum design needs verifiable learning effects to adequately resolve issues of instrumentality, in other words, constructing means that are adequate to reach goals and intended effects. If the evaluative specification would follow the didactic specification there would be too big a chance of pressure to adapt tests to preferred methods, which he considers a form of goal displacement. Dilemma’s for trying to accomplish curriculum and instructional alignment to enhance students’ opportunity to learn in an organizational structure that is loosely coupled are discussed in Scheerens (2017). As an extreme measure to avoid all kind of intermediary control-fuzz and coordination, educational systems could forego curriculum development and concentrate on high quality high stakes tests and examinations. The invisible hand of “teaching to the test”, would steer all intermediary players in a context of maximal autonomy (apart from the output control inherent in the setting of the high stakes test).

Learning from Feedback

In a context of formative assessment, the key mechanism linking assessment and teaching is feedback. The term feedback stems from control theory, with the functioning of the thermostat as the classical example to illustrate it. When the measuring device indicates that the room temperature is below a certain level, the regulating mechanism switches on the heating (De Leeuw, 1990, p. 126). Feedback loops can be positive or negative. An example of positive feedback would be the case when good results increase positive expectations about students’ learning, which, in their turn lead to setting higher standards, a more optimistic, achievement-oriented climate, more self-confidence and achievement that is further increased). An example of negative feedback would be a teacher needing to increase his or her energy in keeping order, when the students’ behaviour worsens (Clauset & Gaynor, 1982).

In a review of the impact of formative assessment Black and Wiliam (1998) conclude that, across the board, formative assessment and feedback are positively associated with student achievement. However, it is often difficult to separate the impact of assessment-feedback from other regulatory mechanism that are also active. This is illustrated in their analyses of feedback within the framework of Mastery Learning; a form of structured teaching comparably to direct teaching, as referred to in a previous section. Likewise, in reviews and meta-analyses effects of quantitative and qualitative aspects of feedback are sometimes not sufficiently separable.

In search for further insights into the specific characteristics of effective feedback from student assessment, Kluger and DeNisi (1996) underline the importance of instrumental feedback. The idea of instrumental feedback assumes that targets are identified as learning gaps and that there are ideas about mechanisms, means or techniques to bridge learning gaps. Experiencing of learning gaps is closely related to the role of standards and achievement expectations in teaching. Research on standard setting points out that learning gaps should neither be unattainably high nor low (cf. De Vos, 1989). The assumption of instrumentality and mechanisms to close learning gaps, is closely related to matching task characteristics to psychological operations of learners, and knowledge about frequently made mistakes.

In providing instrumental feedback teachers have the choice between providing complete solutions, heavily cued hints towards the correct solution, or an adaptive “scaffolding” response, in simpler terms students receiving as much help as they would need to solve the problem on their own.

4.3.3 Structure and Independence in Teaching

Having defined the nature of core instructional elements and the various sub-disciplines in which they are given shape, it is time to turn back to the core idea of seeing teaching as a set of conditions that should facilitate and “boost” student learning. “Normal” structured teaching could be seen as compensating for lack of student control in learning, while a more reflective and process oriented teaching style could venture to actually teach student control strategies.

In a way student control strategies are the pendant of the main features of “structured teaching” and direct instruction, where it is the teacher who actively manages and controls the teaching and learning situation. When putting these two orientations next to one another, structured teaching on the one hand, and students effectively employing control strategies on the other, the following types of associations can be discerned:

• structured teaching happens as a substitute for student control strategies

• structured teaching happens as an additional support for student control strategies

• structured teaching happens as a model and example to enhance student control strategies (meta-cognition)

• structured teaching happens as a suppressor of student control because students are not given sufficient leeway to develop and manifest this behaviour themselves.

Weaker students in primary and secondary education are more likely to benefit from the first two alternatives, whereas the last two alternative combinations are more probable when dealing with better students in secondary education.

The above interpretations suggest a reconciliation in the controversy between structured and more open, discovery-oriented teaching approaches, by making it conditional on student aptitudes (cf the literature on aptitude treatment interaction research, Cronbach & Snow, 1981). Further on, in Sect. 5, didactic methods are discussed in which more structured and “open” learning arrangements are placed in sequence, with a gradual “fading” of structure as a bridging principle.

It is beyond the scope of this chapter to go into any kind of detail in contrasting the traditions of structured teaching, mastery learning and direct instruction on the one hand, with “constructivist ideas” about teaching and learning. See for instance, Gage’s comparison of Progressive-Discovery-Constructivist models and Conventional-Direct-Recitation models (Gage, 2009, p. 62, 79; Kirschner et al. 2006; Messner & Blum, 2019). Earlier treatments of the issue suggest two outcomes:

• 1. Structured approaches, like direct teaching, have repeatedly been shown to be more effective than “open”, “constructivist” teaching (Gage, 2009; Van der Werf, 2005; Hattie, 2009; Stockard et al. 2018). Brophy and Good (1986, p. 367) confirmed that highly structured teaching worked equally well for acquiring complicated cognitive processes in secondary education.

• 2. Elements of constructivism, like teaching learning strategies, and “cognitive activation” have blended in overarching conceptions of instructional effectiveness (e.g. Klieme, 2012; Praetorius et al., 2020).

We shall return to the issue by looking for a theoretical explanation of the success of direct teaching, further on (in Sect. 5), and leave it now by citing the main characteristics.

• 1. Teaching goals are clearly formulated.

• 2. The course material to be followed is carefully split into learning tasks and placed in sequence.

• 3. The teacher explains clearly what the pupils must learn.

• 4. The teacher regularly asks questions to gauge what progress pupils are making and whether they have understood.

• 5. Pupils have ample time to practice what has been taught, with much use being made of “prompts” and feedback.

• 6. Skills are taught until mastery of them is automatic.

• 7. The teacher regularly tests the pupils and calls on the pupils to be accountable for their work (Doyle, 1985).

The dimension structured independence covers “preferred teaching strategies” and “structure and independence in teaching”, included as indicators in Fig. 4.2. It was argued that the dimension can be seen as a continuum that bridges guided “direct” teaching and more open instructional invitations to self-regulated learning.

4.3.4 Classroom Management

The three dimensions of teaching that were discussed in the previous paragraphs are abstract dimensions underlying important choices in teaching. “Classroom management” is to be seen as an overarching term to characterize teaching as a set of controlling and responding actions by teachers. Facets of classroom management can be thought of as implementing choices regarding educational objectives, content and skills to be acquired, degrees of structure in teaching and the application of planning and evaluation methods. Apart from an overall characterization and historical development specific attention will be given to two major facets: the management of classroom ecology and providing support in teaching.

Development of the concept of classroom management; from time-management to a comprehensive implementation of broad set of effectiveness enhancing teaching conditions.

A most relevant and influential model for research on teaching is the Carroll model, originally presented in Carroll (1963), and actualized by the author in Carroll (1989). The management of time is a key issue in the original model (Carroll, 1963; 1989).

The mastery learning model formulated by Bloom in 1976 was largely inspired from Carroll’s model, and it is also related to ‘direct instruction’, as described by Rosenshine in 1983. Moreover, the model is often seen as the basis for more comprehensive models of instructional effectiveness, like Walberg’s model of educational productivity (Walberg, 1984). In subsequent developments Carroll’s original conception of “opportunity to learn” became a combination of time investment and “content covered”, where the latter term became the more common interpretation. Some authors started to use the term “quantity of teaching” as the time spent on well-chosen content elements (i.e Creemers & Kyriakides, 2008).

A relatively narrow perspective of classroom management, true to the original Carroll model, is concentrated on the management of time and learning opportunities, with a specific emphasis on the prevention of disturbances and the monitoring of classroom rules.

The quality of instruction was relatively left undefined in the original Carroll model, but became gradually associated with models of structured teaching, like mastery learning and direct instruction. This led to a more comprehensive interpretation of classroom management, beyond time management. Baumert et al. (2001) mention clearly defined rules and procedures, prevention of disturbances, effective responses to critical events, routinization of basic social acts in the classroom, as well as providing aligned content and adequate pacing of instruction, and finally choosing adequate level of difficulty, clarity and structure in the presentation of material, adaptivity and individualization of instruction, and monitoring of student activities. In this way classroom management became associated with teachers’ implementing a set of effectiveness enhancing conditions, confirmed by empirical research. Differentiation and adaptive teaching can also be discerned as challenges of classroom management.

Additional Facets of Classroom Management

Due to lack of space only a cursory review of other facets of classroom management, additional to the management of time and learning opportunities, will be presented (a more elaborate description is given in Scheerens, 2016).

The Management of Classroom Ecology

The management of classroom ecology deals with partly malleable composition effects of classrooms, like class size, SES composition, ability grouping, and the matching of teachers and classes (Opdenakker & Van Damme, 2001; Baumert et al., 2005; Willms, 2004; Luyten et al., 2005; Slavin, 1996; Monk, 1989, 1992).

Classroom Climate

Classroom climate can be defined as the general atmosphere in the classroom. When further analysed the major facets of a favourable, effectiveness enhancing climate are a supportive style in teacher student interactions, achievement orientation, clear disciplinary rules, and good student-student interrelationships. Some of these facets relate to more overtly “managed”, “institutionalized”, and “planned” aspects of teaching, others are more interactionist and “emergent” and part of the school culture (Maslowski, 2001; Scheerens, 2016, pp. 90–94).

Cognitive Support in Teaching

Cognitive support in teaching is associated with active teaching and cognitive activation. Active teaching is about providing a varied repertoire of presentation forms, alternative ways to group students, and different presentation media (Boekaerts & Simons, 1993; Slavin, 1995; Seidel et al. 2005, 129). Cognitive activation calls for stimulating higher order thinking, deep understanding of content, learning from mistakes, meaningful contexts, authentic instruction, relevance of content, and appropriate and high level of language (Klieme & Rakoczy, 2003).

Motivation and Emotional Support in Teaching

Motivation and emotional support in teaching are particularly sensitive in association with teacher expectations (e.g. the Pygmalion effect) and performance feedback (cf. De Vos, 1989). Paying attention to social-emotional attributes in teaching in a more general sense is addressed in Corcoran et al. (2018) and Scheerens et al. (2020), This theme will be taken up again in Sect. 5, as one of the examples, when searching for explanatory mechanisms in more partial theories on teaching.

Classroom management is the comprehensive orchestration of a broad set of facets of teaching, ranging from time and content management, aspects of classroom climate, achievement orientation, discipline, support, and ethos. It covers all teaching processes listed in Fig. 4.1 and all issues related to climate and student support. It is a collegiation of discrete elements, not a ranked continuum. It should be noted that classroom management as discussed in the above is considerably broader than ‘classroom management aimed at optimizing active learning time and opportunity to learn’, which is included in Fig. 4.1.

4.4 The Theoretical Meaningfulness of the Dimensions

The central question in this part of the chapter is whether summary categories of operational variables in teaching effectiveness can be interpreted as meaningful dimensions for a general theory on teaching. In the text-table below 3 main criteria to assess theoretical meaningfulness are listed, and afterwards explained.

Comprehensiveness is a criterion that should be applied to the whole set of dimensions. The expectation is that the three dimensions that are considered provide an exhaustive coverage of teaching effectiveness.

• (a) The comprehensiveness and parsimony of the complete set of dimensions

• (b) The theoretical meaningfulness of each dimension (over and above a)

  • – Potential for generating fundamental research

  • – Linkage to more established theory

  • – Potential for practice-oriented valorization

• (c) Interconnectivity of the dimensions

Parsimony would be challenged if the set of operational variables covered by each dimension would lack coherence, so that additional dimensions would be needed.

Potential for generating fundamental research is a demand that clearly goes beyond description and categorization. Ideally each dimension should be able to generate sufficient challenge and even controversy to suggest critical research questions.

Linkage to more established theory is an advantage if it helps in identifying explanatory mechanisms and covering laws.

Potential for practice- oriented valorization. Given the complexity and diversity of research outcomes in the field of educational effectiveness, evidence supported synthetic concepts are considered important in the communication with policy makers and practitioners.

Below, by way of a finger exercise, a first exploration will be made with respect to the feasibility of this approach to theory formation. This will be done by checking the criteria and applying them to the whole set of building blocks, or to separate dimensions.

The Comprehensiveness and the Parsimony of the Complete Set of Dimensions

The expectation is that the four building blocks provide an exhaustive coverage of teaching effectiveness. The comprehensiveness of the chosen set is challenged by Gage’s inclusion of a dimension “student motivation and learning”, and by Praetorius et al’s incorporation of a dimension indicated as “students’ direct learning-oriented reactions to teaching interventions”. This latter dimension has indicators like “time on task” and “depth of processing”. It could be argued though that teaching is about teaching interventions, while overt student reactions are learning activities. In a previous presentation I had subsumed such reactions under the heading of learning processes in a model of student learning (Scheerens, 2016, p. 29).

Parsimony would be challenged if the set of building blocks could be replaced by an even leaner set of dimensions. A possible step in this direction might be to set look at the integration of “the elementary parts of teaching” and “proactive and retroactive structuring. We shall return to this issue when addressing the interconnectivity of the set of building blocks.

The Theoretical Meaningfulness of Each Dimension

Potential for generating fundamental research is a demand that clearly goes beyond description and categorization. Ideally each dimension should be able to generate sufficient challenge and even controversy to suggest critical research questions. The formal characteristics of the dimensions in terms of unidimensional continua, bipolar scales or categorical sub-components may have implications for generating research questions. Teaching can be more or less content focused and address psychological operations of increasing complexity, as in taxonomies of educational objectives. Likewise, structure vs independence in teaching is a continuum with intermediary positions, like a gradual fading of structuring. Optimization of taking certain positions on the continuum can be seen as depending on situational characteristics, like student characteristics and school organizational conditions. This represents a perspective in line with “contingency theory” (see footnote 3), which is testable in what is known as a differential effectiveness approach. Pro-active or retro-active emphasis in educational alignment could-be addressed in a way that goes beyond a simple choice for the one or the other. Proactive and retro-active components will tend to be intermittently present in sequences of teaching events, and hypotheses may be posed about the preferable order and the predominance of the one or the other (Compare De Groot’s position that evaluative specification should precede didactic planning, De Groot, 1986). Classroom management as presented here is a multi-dimensional “container”, for which three sub-dimensions were proposed (management of learning opportunities, management of classroom ecology and climate and support (cognitive and emotional). Research hypotheses will most likely be addressed at the level of the sub-dimensions. An example is given in Sect. 5 of this chapter on “partial” theories of teaching, where we will look at dynamic interaction between emotional and cognitive support. Still classroom management could also be studied at the dimension level, in hypotheses about the required dosage of components, and the effectiveness of control theoretical variations in management approach.

Linkage to a More Established Theory

The linkage to a more established theory is an advantage if it helps in identifying explanatory mechanisms and covering laws.

The degree to which teaching is more or less focused on content and/or psychological operations originates from didactic analysis, and is applied in taxonomies of educational objectives, like Bloom’s taxonomy and more recent forms as the RTTI approach (Drost & Verra, 2019). The debate on the feasibility of “content free” skills, versus the view that teaching is always associated with content is a fundamental debate in education (cf Weinert, 2001). When it comes to conceiving content-free skills, association with theories of intelligence and personality is relevant (Scheerens et al., 2020).

With respect to the dimension structure and independence in teaching a relevant attempt at linking “structuring” to covering laws is made by Gage (2009). In Sect. 5 of this chapter on partial theories we address linkage to learning theory and the information processing approach to cognition.

Pro-active or retro-active emphasis in educational alignment can be linked to variations of the rational planning model, like “bounded rationality” and the cybernetic principle (Scheerens et al., 2003, 2011).

As far as Classroom management is concerned, linkage to more established theory seems to make more sense for the sub-dimensions, and to partial theories associated with them. Development of the construct of cognitive activation would be an interesting case. In Sect. 5.2 we provide an example with respect to emotional support.

Potential for Practice Oriented Valorization

Potential for practice- oriented valorization has to do with the dissemination and communication with policymakers and educational practitioners and is here based on the believe that evidence supported synthetic concepts are important in these processes of dissemination and communication.

From my perspective on the international state of the art on effectiveness enhancing educational policies there are some major issues at stake for which the dimensions are relevant. I should say that my perspective is subjective and colored by the way I perceive developments in the Netherlands. Cryptically indicated, these issues are the neglect of the curriculum dimension in improvement models, resistances to external evaluation and assessment procedures, and an underscoring of the potential of the evaluation and feedback mechanism, the “traditional versus progressive education” debate and a phenomenon, which might be indicated with the label “unhyping soft skills” (Scheerens et al., 2020). It is beyond the scope of this chapter to discuss these observations in more detail.

The Interconnectivity of the Building Blocks

The two content dimensions, namely “the elementary parts of teaching”, and “proactive and retroactive structuring” might be considered for further integration. The distinction of the elementary parts of teaching defined as the matching of content and psychological operations, could be seen as providing a grammar for defining taxonomies of educational objectives and educational alignment. In what is noted as “horizontal alignment” the matching between the formulation of educational standards and objectives on the one hand and high stakes assessment on the other is considered. The structure of content and psychological operations of objectives matches test matrices in which test items have a referent to content and psychological operations at a certain level. Tentatively a comprehensive content dimension labelled as “pro-active and retro-active structuring of the elementary parts of teaching” could be proposed. Alignment and learning from feedback could then be seen as partial theories within this dimension. This is visualized in Fig. 4.4. Other connections between the building blocks also suggest that the elementary parts of teaching function as a common referent among the other dimensions. When didactic operations are brought into the picture, the “elementary parts” are constituent elements of the concept of “pedagogical content knowledge”, which could be given a “process interpretation” when teaching episodes are seen as “enacted pedagogical content knowledge”. Didactic episodes, in their turn, are essentially determined by their degree of structure in teaching, in terms of opportunity to learn, and in terms of cognitive and emotional support. In this way, the three dimensions “pro-active and retro-active structuring”, “structure and independence” and “classroom management” are united in being rooted in the dimension on the elementary parts of teaching.

Fig. 4.4

A hierarchical organization of dimensions and partial theories

In the discussion section of this chapter, I will try and make an overall assessment of the usefulness of this attempt at formulating building blocks for a general theory on teaching effectiveness.

In the next section we shall turn to partial theories of teaching.

5 Examples of Partial Substantive Theories of Teaching Effectiveness in the Domains of “Direct Teaching” and “Social Emotional Support”

In Sect. 4, I described a conceptual map of main components of teaching, which was considered as a contribution to a general theory on teaching effectiveness.

In this section I will turn to partial, substantive theories, defined within components of the conceptual map. Given the scope of the field, it will only be possible to give a few examples. Corresponding to the empiricist nature of the teaching effectiveness paradigm, each of these examples of partial theories is directly related to empirical research outcomes. The selected examples are theoretical conjectures in relationship to substantive results concerning “direct teaching” and “emotional support in teaching”.

5.1 Direct Instruction

Overview

Direct instruction is a form of structured teaching that has repeatedly been shown as producing medium to large effects on student performance. (See the references in Sect. 1, and Zhang et al. (2021) for a more recent review. In this section I will show that the success of the approach can be explained by the information processing theory of cognition, more precisely the matching of teaching interventions to learning challenges derived from this theory.

Stockard et al. (2018), present a meta-analysis based on research outcomes collected over a 50-year period, with mean effect sizes in reading, mathematics, language and spelling that range from .50 to .66. Stockard et al. (ibid) discuss the “theoretical base” of direct instruction, by explaining the underlying philosophy and definitional characteristics. They say that direct instruction depends on the assumption that all students can learn with well-designed instruction. A first specification of what they mean by “well designed instruction” is that new material can be learned by students when (a) they have mastered prerequisite knowledge and skills and (b) the instruction is unambiguous. The underlying belief is that “students are inherently logical beings” (ibid, p. 480). Barbash (2012) speaks about direct instruction being based on an optimistic perspective. The main definitional characteristics that are outlined by Stockard et al. (2018, pp 480/481) are the following:

Mastery learning is a key element of direct instruction. “DI theory posits that when students become fluent in a new task, fully grasping a new concept or skill, it becomes part of an existing repertoire”.

Implications are the assumption that new learning is easier when it can be grounded in relevant subject matter that is fully mastered earlier and the observation that it is easier to learn a new concept, than to “unlearn” a faulty conceptualization.

A well-sequenced curriculum is an important basis for mastery learning.

A step-by step approach creates the possibility for continuous positive reinforcement throughout the instruction process.

Curricular materials are expected to provide highly structured guidance to teachers in the wording, sequencing, and review of material presented to students.

Curricula follow a tracked design in which “discrete skills and concepts are taught in isolation but are then brought together in increasingly more sophisticated and complex applications”.

Placement tests are included, and student progress is closely monitored.

Anchorage in Cognitive Information Processing Theory

It is quite interesting that Stockard et al. (2018) refer to Barbash for an explanation of the theoretical basis of direct instruction, because Barbash depicts it as opposed to “vague theories” of teaching and states that “Engelmann (as the founder of direct instruction, JS) did not formulate these principles from books or from abstract speculation about the way children learn”. He formulated them through a painstaking process of trial and error in the classroom, then applied them to create a series of unique programs that outperformed others in their power to teach many different subjects, to all kinds of children. So direct instruction was developed as an empirically supported practical program. Theorizing that connected the principles of structured education approaches as mastery learning and direct instruction to more encompassing concepts and established theory happened at a later stage.

Kirschner et al. (2006) offer a theoretical explanation based on the cognitive architecture of the human memory function. The relations between the short time (or ‘working’) memory and long-term memory, in conjunction with the cognitive processes that support learning are considered of critical importance. Guided instruction to novice learners, like explicit direct teaching, should facilitate storage in the working-memory. In the case of unguided ‘open’ instruction there is a great risk of cognitive overload. Cognitive overload prevents mobilizing the functioning of the long-term memory, in other words, that learning takes place. Learning is defined as a change in long-term memory. After guided instruction has helped in overcoming the information processing limitations of the short-term memory, learned information, stored in long-term memory, can be brought back from long term memory to working memory, more easily and over long periods of time. Open “constructivist” teaching which stimulates free exploration, on the other hand, may generate a heavy working memory load that is detrimental to learning. The more so in the case of novice learners, who lack proper schemas to integrate the new information with their prior knowledge.

Cognitive theory provides a basis for defining hypotheses about effective teaching interventions to stimulate student learning. These appear as the two sides of one coin¹: when limited memory capacity is referred to as a relevant condition of learning, teaching is framed in a way to do justice to this principle, by not creating “cognitive overload”, by means of a careful and prudent introduction of new subject matter, and ample opportunity to exercise and become familiar with the new content (cf Sweller, 1988). Similarly, the property that, once information is stored in long-term memory, it can be transferred back to the working memory relatively easily, supports tackling more complex learning tasks, A recent review by Kirschner et al. (2019) further illustrates how explicit direct teaching accommodates these more complex learning tasks, while matching the principles of cognitive information processing theories. They discuss increasingly complex cognitive processes, elaboration and transfer, problem solving and meta-cognition (Geary, 2008; Newell & Simon, 1972; Chi et al., 1981). The “answer” of teaching to facilitate these levels of learning is basically the same as in the case of avoiding cognitive overload in memory and reproduction of content: providing pre-structuring and support to learners. Rothkopf (1970), mentions orientation (attracting attention to what should be learned), selection (providing focus with respect to content and processes) and processing (calling on previous knowledge, raising questions). Dunlosky et al. (2013) discuss practices to enhance memory functions and retention, learning to learn effective approaches, repeating the subject matter of yesterday’s lessons, the use of “exercise tests” and distributed practice, as effective teaching approaches. Ausubel ‘s well-known concept of “advance organizers” emphasizes the importance of activating previously acquired knowledge. as well as the importance of longitudinally connected content sequence (Ausubel, 1960). Rothkopf (1966) discusses the use of “test-like events” and asking questions to stimulate learning, Hattie and Timperley (2007) refer to the “power of feedback”, and Black and Wiliam relate feedback to formative assessment. Clark (1989) connects assessment of entrance behavior to teaching that is adaptive to initial differences.

All of these contributions (collected in Kirschner et al., 2019) underline the importance of the kind of structured teaching that characterizes direct instruction.

A Constructive Solution to the Confrontation with Constructivism

Direct teaching is often contrasted with “open”, teaching, discovery learning and self-regulated learning (e.g. Vander Werf, 2005; Kirschner et al., 2006). According to Stockard et al. (2018, 482) “. the theory underlying DI lies in opposition to developmental approaches, constructivism, and theories of learning styles, which assume that students’ ability to learn depends on their developmental stage, their ability to construct or derive understandings, or their own unique approach to learning”. Still there is some common ground as there is room for independent practice in the direct instruction model, and according to Stockard et al. “like the constructivist approach, DI assumes that students make inferences from examples that are presented to them” (ibid, 482). When teaching moves to providing episodes of more independent learning this needs to be prepared by structured teaching in the form of worked out examples, partially worked out examples, and with a gradual “fading” of support and scaffolding. This gradual fading of support can be seen as a bridge between structured teaching and independent and discovery learning. In the same vein Messner and Blum (2019) discuss what they call “the myth of open education” and conclude that “openness” and structuring should not be considered as complete opposites, but as inherently connected, since “open” “independent learning” in academic subjects without initially guiding instruction by the teacher, is completely unfeasible. The realization that structured teaching can also be seen as supportive of higher order learning (Brophy & Good, 1986, p. 367) further enforces the relevance of the approach. Still, it seems a better perspective not to depict “structured” and “open” learning as diametrical opposites but rather as a continuum, where moments of direct teaching, an independent learning can have different emphasis, depending on all sorts of deliberate considerations and contextual conditions. This is why I proposed “structure and independence in teaching” as one of the basic defining dimensions of teaching.

5.2 Emotional Support in Teaching

In their book “Soft skills in education. Putting the evidence in perspective” Scheerens et al. (2020) discuss the research evidence with respect to social emotional attributes in education. On the one hand these are seen as outcomes in their own right, and featuring as the dependent variables in intervention studies, and on the other hand, they are treated as instrumental to academic outcomes. An intriguing finding was that social emotional learning programs were shown to have effect sizes on cognitive outcomes as large as, or even larger than, dedicated cognitive interventions. Although bias in the empirical studies could not be ruled out as an explanation for this surprising finding, the authors felt stimulated to look for theoretical explanations related to the “workings” of social emotional interventions. In doing so they focused on the interplay of cognitive teaching and stimulation of socio-emotional learning. After considering distinct explanations, one based on econometric modelling by Cunha et al. (2010) and from OECD (2015, p 39), and another on Bandura’ theory on self-efficacy (Usher & Pajares, 2008; Bandura, 1986, the authors thought an approach from social psychology particularly interesting. Yeager and Walton (2011) reviewed empirical studies that had shown that seemingly “small” social-psychological interventions in education—“that is, brief exercises that target students’ thoughts, feelings, and beliefs in and about school—can lead to large gains in student achievement and sharply reduce achievement gaps even months and years later” (ibid p. 267) Inspired by modelling studies by Cohen et al. (2009), Yeager and Walton conclude as follows. “A key to understanding the long-lasting effects of social-psychological interventions is to understand how they interact with recursive processes already present in schools, such as the quality of students’ developing relationships with peers and teachers, their beliefs about their ability, and their acquisition of academic knowledge. It is by affecting self-reinforcing recursive processes that psychological interventions can cause lasting improvements in motivation and achievement even when the original treatment message has faded in salience” (ibid, 268). Scheerens et al., 2020 conclude that the study by Jaeger and Walton reflects two features that are most interesting for embedding social emotional learning in the every-day school context. Firstly, social psychological learning is targeted to the social emotional facets of school life and school learning. Secondly the way they are seen as interacting with regular content related teaching offers a tentative explanation for the finding that some evaluations of SEL programs and intervention showed significant improvement of academic outcomes.

In the previous sections I have approached the theme of “theorizing teaching” from three angles: meta- theory (the educational effectiveness research paradigm), general theory of teaching (major dimensions of teaching as building blocks for such a theory), and partial theories. The approach is in line with the tradition of research on teaching and research on educational effectiveness, in which theory development is explicitly rooted in empirical research. The main function of theory is to generate conjectures that provide explanations for the empirical findings (why what works), and which are expected to drive further empirical research.

It is important to note that “theory” is sometimes conceived in an entirely different way. One example is when issues of educational quality and equity are approached from the perspective of theories of social justice (Francis et al., 2017; Kelly, 2020). Another example are different brands of “critical theory”, which, like the social justice perspective, focus on value-laden, normative, and even political facets of effective schooling and effective teaching (e.g Holborow, 2018). A third example is “non-affirmative theory”, a pedagogical theory developed by Uljens and Ylimaki (2017).

6 Discussion

In this chapter the question “What is a theory of teaching” was answered by distinguishing three levels of theory: meta-theory, general theory, and partial theories (as explanations of empirical evidence). The follow-up question on what theory at each level should contain has different answers for each of the levels. At the level of meta-theory “teaching” was framed in accordance with the educational effectiveness research paradigm. This choice yielded a conceptual ground structure, based on a model from systems theory and reference to the scientific method as the epistemological and methodological background. The level of general theory was conceived as containing a potentially exhaustive limited set of sub-theories of effectiveness enhancing teaching processes. The third level, indicated as “partial theories”, refers to more specific explanatory mechanisms intricately linked to empirical research outcomes. Theory at each of these levels was considered generalizable across major subject matter areas, like reading, language, mathematics, and spelling. The empirical evidence is seen as supporting the assumption that the foundational concepts and explanatory mechanisms work about equally well across these subject matter areas (Scheerens, 2016). This choice was pragmatic and should not be read as a denial of the relevance of studying eventual subject specific conditions (and differences between primary and secondary schools), it was seen as beyond the scope of this chapter. Before addressing the yield and potential products of theory at each of these levels in more detail, a general observation should be made.

The presentation underlined what was already concluded in earlier reviews (Scheerens, 2013, 2015), namely that educational effectiveness, including teaching effectiveness is a predominantly empiricist field of inquiry. “Theory” is mostly treated, inductively, as reconstruction “after the fact” instead of deductive in generating hypotheses for further empirical research. Ideally these two orientations should follow-up on one another, but my impression is that theory driven effectiveness research is relatively rare. The interaction of conceptual development and empirical research associated with the Dynamic Model of educational effectiveness by Creemers and Kyriakides (2008), is one of the positive exceptions. Despite occasional calls for more theory driven research (e.g., Reardon, 2011) theory is getting a stepmotherly treatment in the educational effectiveness research community. Hopefully, the present volume will turn the tide! In any case some of the perspectives, documented in this chapter, are seen as useful to furthering theory. Even when theory development remains limited to induction and reconstruction after the fact, it contributes to the “supportive understanding” of empirical results, to strengthening the conceptual identity of the field, to parsimonious summary of major outcomes and as a basis for communication with education practitioners and decision-makers. More specifically the current presentation has tried to provide the following inroads to stimulate a more active use and application of theory in this field.

First, the distinction between meta-theory, general theory and partial theories on teaching is offered as a vehicle to treat theory comprehensively.

Secondly, at the level of meta-theory, the presented research paradigm offers a conceptual ground structure of all possible associations between context-input-process-and outcome components. A worked-out example is provided by Gage, in reference to his “paradigm for the study of teaching”, when he lists the relationships between all possible pairs of categories (Gage, 2009, pp. 55–56). Elsewhere, (Scheerens, 2016) I have discussed the most relevant associations between the components, when the CIPO structure is applied in a multi-level framework.

Thirdly, at the level of the conceptual “building blocks” for a general theory of teaching four dimensions, or building blocks, were distinguished: “the elementary parts of teaching”, “pro-active and retro-active regulation”, “structure and independence” and “classroom management”. An attempt was made to enhance the heuristic function of these core dimensions of teaching, by explicitly addressing their exhaustiveness as a set, the “theoretical meaningfulness” of each dimension and the interconnectivity and internal structure of the set. All this against the background of Snow’s developmental sequence of levels of theory. I found that the dimensions showed potential for generating fundamental research, could incidentally be linked to more established theory, and spoke to important “societal” debates on educational quality and effectiveness. In the process of verifying these criteria, particularly when addressing the inter-connectivity of the dimensions, I came to the conclusion that “the elementary parts of teaching” and “proactive and retroactive structuring” show a particular connectivity, which might be considered as a ground for integrating them in one dimension. (compare Fig. 4.2).

The dimensions that were considered as building blocks for a general theory on teaching could be used as a basis for directing comparative research on dimensions like “structure and independence”, the relative effectiveness of pro-active and retro-active regulations dependent on contextual conditions, curriculum alignment as a content-based encompassing strategy to optimize opportunity to learn, and potential constructive interaction between cognitive and emotional support in classroom management. To the degree that these “building blocks” would be supported as an exhaustive coverage of teaching, they could provide new input to comprehensive school and teaching improvement projects.

Fourth, only two illustrative examples were given of more specific partial theories, intricately connected to empirical research outcomes. Results on structuring and direct teaching could be credibly explained on the basis of principles from the information processing approach to cognition. Insights from social psychology seemed to have potential for explaining interactions between cognitive and emotional support in classroom management.

Despite the potential of developments in the realm of a general theory on teaching, this last level of partial theories is seen as having the most potential in furthering theory formation and application in teaching effectiveness research. This is due to the established empiricist tradition of the field, and the expectation that piecemeal progress is the most realistic expectation in a loosely defined and fragmented, multi-disciplinary research community. And, finally, finding fruitful conjectures to generate new research is also inspired by surprising research outcomes, and this occurs most likely, “close to the data”, in more specific partial domains.

I would like to finish this chapter by addressing the five questions that the editors wanted to be commented on by all contributors to this volume.

What is a theory of teaching? In this chapter this question was addressed by distinguishing meta-theory, general theory, and partial theories.

What should it contain? The answer differs for meta-theory on the one hand and substantive theory (both general and partial) on the other. Meta-theory contains first principles, such as logical ground structures, epistemological preferences, methodologies and ontological considerations (defining characteristics). Substantive theory in relation to the educational effectiveness research paradigm is strongly rooted in empirical evidence, distinguishes descriptive components and relationships between these, as well as explanatory conjectures that explain hypothetical as well as empirically supported relationships.

Can such a theory accommodate differences across subject matter areas and student populations taught? I would say the answer is expected to be yes for a general theory on teaching, while one way of seeing theories as partial is the degree of their being restricted to a specific context.

Do we already have a theory on teaching? There is growing consensus on core sub-theories on teaching in the sense Gage refers to these, while others prefer to speak of core dimensions (examples and references have been given in the text of my chapter). Still some contributions might not be called theories by everyone. In an earlier contribution. I concluded that conceptual maps and dimensional models reflect the state of the art. Snow’s levels of theory development supports calling models, and “summaries” of empirical findings “theories” be it at a low level on his scale. Occasional applications of “eclectic” use of more established theory from basic disciplines is seen as an instance of gradual progress towards a higher level of theory. From the perspective of the educational effectiveness paradigm the key issue is the explanation of the findings by means of a plausible and established causal mechanisms.

In the future, in what ways might it be possible, if at all, to create a (more comprehensive) theory of teaching? I see this as a continuation of a piecemeal, bottom up development, rooted in the analysis and synthesis of empirical research outcomes. Making sense of the enormous quantity of research outcomes by means of meta-analyses and research reviews stimulates reflection on what is generalizable and what is helpful for further research. Last but not least, the answers that policy makers and practitioners want from researchers call for conceptual synthesis and theoretically meaningful interpretation of the evidence. Again: nothing more practical than a good theory.