A research model for the study of science teachers’ PCK and improving teacher education

ARTICLE IN PRESS

Teaching and Teacher Education 23 (2007) 885–897 www.elsevier.com/locate/tate

A research model for the study of science teachers’ PCK and improving teacher education Esther M. van Dijka,, Ulrich Kattmannb a

Didaktisches Zentrum, Carl von Ossietzky University Oldenburg, D-26111 Oldenburg, Germany Institute of Biology and Environmental Sciences, Biology Education, Carl von Ossietzky University Oldenburg, D-26111 Oldenburg, Germany

b

Received 29 January 2006; received in revised form 7 May 2006; accepted 9 May 2006

Abstract In this article, a new research model for the study of pedagogical content knowledge (PCK) is presented which aims to improve teacher education. This model called ‘‘educational reconstruction for teacher education’’ (ERTE) represents the framework for an integrative approach to the study of science teachers’ topic specific PCK, which is a largely unexplored field. By integrating the PCK concept, originating in the American Curriculum tradition, into the German (Fach)didaktik tradition, this model adds a new perspective to (Fach)didaktik. This paper, therefore, also aims to clarify the PCK concept and its relation to Fachdidaktik. r 2006 Elsevier Ltd. All rights reserved. Keywords: Pedagogical content knowledge; Science teachers; Teacher education; Research model; Teacher knowledge; Educational reconstruction

Subject Didaktik and Curriculum research, as represented by [among others] Lee Shulman y are dealing with the same set of questions. What all these efforts have in common is the strong belief that we need an integrative approach y that can do justice to each corner of the Didaktik triangle: the teacher, the content, and not least, the learner (Hopmann & Riquarts, 2000, p. 10).

Corresponding author.

E-mail address: [email protected] (E.M. van Dijk).

1. Introduction The main purpose of this article is to introduce a new research model for the study of science teachers’ pedagogical content knowledge (PCK). This model, which is called ‘‘educational reconstruction for teacher education’’ (ERTE), has been developed as the basis for a further research project on science teachers’ PCK. It represents the framework for an integrative approach to the study of PCK which aims at improving and designing teacher education. This model can be used to explore secondary school teachers’ (1) knowledge and beliefs of students’

0742-051X/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.tate.2006.05.002

ARTICLE IN PRESS 886

E.M. van Dijk, U. Kattmann / Teaching and Teacher Education 23 (2007) 885–897

pre-scientific1 conceptions, (2) knowledge and beliefs of representations of the subject matter, and (3) ‘subject matter knowledge for teaching’, in relation to (a) the design of learning environments or teaching–learning sequences, (b) the study of students’ pre-scientific conceptions, and in relation to (c) a subject matter analysis. The ERTE model is based on an established research model within the German Fachdidaktik2 tradition, the model of educational reconstruction (ER) (Didaktische Rekonstruktion). This latter model has been developed in the biology education group at Oldenburg University in cooperation with the department of physics education at the Institut fu¨r die Pa¨dagogik der Naturwissenschaften (IPN) in Kiel (Duit, GropengieXer, & Kattmann, 2005; Kattmann, Duit, & GropengieXer, 1998; Kattmann, Duit, GropengieXer, & Komorek, 1997). The ERTE model constitutes an example of how two traditions within educational research, the American Curriculum tradition and the German Didaktik tradition, can be mingled to the benefit of both. Therefore, we will begin this introduction with a very short historical, comparative description of the different roles that the teacher and the teaching content, two essential elements of this paper and within educational research, have played in the American Curriculum tradition and the German Didaktik tradition. We are not aiming to compare or even evaluate the two traditions as wholes. Rather we are aiming to identify two different ways of treating the role of the teacher and teaching content in educational studies. This limited and therefore necessarily simplified overview is meant to provide the context and starting-point for this 1 We use the term ‘pre-scientific conception’ introduced by Good (1991; see also Wandersee, Mintzes, & Novak, 1994), instead of the commonly used term ‘alternative conception’. In our view this former term better describes the place of these conceptions on the continuum of scientific understanding. 2 Because of language differences and fundamental cultural differences the German terms Didaktik and Fachdidaktik are difficult to translate into English. The term didactics is not a good translation because ‘‘didactics has a negative valuation in the Anglo-American mind’’ (see Hamilton, 1999, also for an excellent description of the historical development of the terms: didactics and curriculum). German Didaktik is characterized by a differentiation between the general Didaktik and the specific, that is focusing on specific teaching subjects (Fachdidaktik). Didaktik is a theory of learning and teaching that deals with the following questions: what is to be taught and learned? How and why is the content to be taught and learned? Traditionally Didaktik has been a more philosophical than empirical field (Ku¨nzli, 2000. pp. 43–44).

paper. To understand the issues within the field of educational research it is important to be aware of the fundamental differences between the AngloSaxon Curriculum tradition and the Central and North-European Didaktik tradition. Until recently, there has never been a comparison of these two traditions. The first comparison was made in the project called ‘‘Didaktik meets Curriculum—Didactical and Curricular Theories and Patterns: An International Comparison (see Gundem & Hopmann, 1998, for a description of this project). This ‘dialogue’ has shown to stimulate international understanding and opportunities for cooperation. That such a comparative discussion can be useful can be concluded, for example, from the article by Gudmundsdottir and Grankvist (1992). According to these researchers the German Didaktik tradition is a well-kept secret within American educational research. Only recently leading researchers like Lee Shulman have rediscovered Didaktik and recognized the relationship with their own ideas. Since the 17th century, the days of Comenius and Ratke, Didaktik has been the way to plan, to enact, and to think about teaching in most of northern and central Europe (Hamilton, 1999; Hopmann & Riquarts, 2000). In American educational research, which has its roots in the 19th century, the various themes within the Didaktik field are addressed in the two separate fields of Curriculum and Instruction (Hopmann & Riquarts, 2000). According to Westbury (2000, p. 21) the ‘‘most dramatic difference in viewpoint’’ between both traditions, the Didaktik and Curriculum tradition is ‘‘their respective views of the teacher, and the role the teacher is given within their theoretical and institutional systems.’’ Lagemann (2000) observes that the relation between scholars and practitioners of education that has developed in the history of American educational research was hierarchical. It has been assumed that knowledge for teaching should be generated at the university and then used in the schools (CochranSmith & Lytle, 1993). Westbury (2000) even states that the teacher was seen as the passive ‘conduit’ in the school system who implements the system’s, teacher-proof, curricula. He observes that in the German Didaktik tradition, in contrast, the teachers ‘‘are guaranteed professional autonomy, ‘freedom to teach’, without control by a curriculum in the American sense’’ (Westbury, 2000, p. 17). The state curriculum (Lehrplan) in Germany prescribes the content for teaching but is not meant to explicitly direct a teachers’ work. The teacher works within

ARTICLE IN PRESS E.M. van Dijk, U. Kattmann / Teaching and Teacher Education 23 (2007) 885–897

the framework of the curriculum and translates this curriculum into teachable contents. Didaktik in all its forms is, in contrast to the Curriculum tradition, more teacher than system centered. Westbury (2000, p. 35) observes further that in recent years there has emerged ‘‘a new focus on the role of the teacher as an active maker of the classroom curriculum’’. From a present-day perspective a second difference between the two traditions seems to be the focus on the teaching content in educational research. The observed difference between the two traditions concerning the focus on the content being taught appears to be foremost a difference in timing. The Sputnik shock, in 1957, led to a reform of natural science curricula in Great-Britain and America. In Germany, under the influence of the American Curriculum tradition and the aforementioned reforms of the natural science curricula, a renewed focus on teaching content in Didaktik emerged. This led to the establishment of the different Fachdidaktik (subject specific Didaktik) fields (Hedewig, 1990; Hopmann & Riquarts, 1995, 2000; Sandfuchs, 1990). As Bayrhuber (1995, p. 335) for example observes: ‘‘In the mid-1960s, y the Institut fu¨r die Pa¨dagogik der Naturwissenschaften (IPN in Kiel, Germany) was commencing its work in the wake of the ‘‘Sputnik shock’’ and a new phase of focused and lasting support for biology, chemistry and physics didaktik in Germany had thus begun y’’. The Education Reconstruction program which aims to design learning environments in relation to the empirical study of students’ preconceptions and a thorough analysis of the subject matter is an example of the (design) research that is being done within the domain of Fachdidaktik. Shulman (1986) observes in his analysis of the major research programs in American educational research that the teaching content has rarely been given any serious attention. According to Doyle and Westbury (1992; see also Doyle, 1992; Westbury, 2000) this lack of attention for teaching content has resulted in the drifting apart of Curriculum and Instruction. Without the teaching content to bind them together Curriculum and Instruction have become separate fields within educational research. They observe that the field of Instruction research has been focused on quality control through measuring the effectiveness of teachers and teaching methods without paying attention to the subject content being taught. Curriculum research has been dealing foremost with curriculum implementation and with the construction of a curriculum on the

887

institutional level—defining what should be taught and why. But also in curriculum research there was no serious attention for the classroom curriculum, the subject content that was actually being taught in class. It is on this level, the classroom level, that Curriculum and Instruction come together in the development of learning environments by the teacher. Shulman’s work on professionalizing teaching and PCK is but one example of a research program which focuses on the domain in which Instruction and Curriculum intersect. (Doyle, 1992; Doyle & Westbury, 1992). As Gess-Newsome (1999) stated: the introduction of PCK in 1985 has led to a renewed interest in subject specific teaching fields such as science education. The developments of the last decades in American and German educational research could offer opportunities for cooperation between the two very different traditions. The new focus on the teaching content in both traditions and the changed role of the teacher in the American Curriculum tradition can create new possibilities for cooperation and comparison within education research (compare Hopmann & Riquarts, 1995, 2000). Especially the concept of PCK and the German idea of Fachdidaktik seem to have a strong relationship, since Fachdidaktik and PCK can both be seen as knowledge domains that consist of a combination of subject matter knowledge and general pedagogical knowledge (Gudmundsdottir & Grankvist, 1992). A good example of what the mingling of the two traditions can look like is a study of Reinertsen, Nordtømme, Eidsvik, Weidemann, and Gudmundsdottir (1996). They used Klafki’s (1996, 2000) Didaktik Analysis model for lesson planning, a very influential model for Didaktik in Germany (Hopmann, 2000; Ku¨nzli, 2000), as a conceptual framework for the empirical research on certain aspects of PCK (see Gudmundsdottir, Reinertsen, & Nordtømme, 2000, for a description of this research project). Klafki’s model consists of five steps for lesson planning: (1) the contemporary meaning of the content for the children, (2) the future meaning of the content for the children, (3) the content structure, (4) the exemplary value of the content, and (5) the pedagogical representations of the ideas. Reinertsen et al. (1996; see Gudmundsdottir et al., 2000) show in their case study that a teacher’s description of his lesson planning concerning the teaching of the Norwegian constitution with the Didaktik Analysis model can provide a picture of his PCK.

ARTICLE IN PRESS 888

E.M. van Dijk, U. Kattmann / Teaching and Teacher Education 23 (2007) 885–897

We think that the integration of PCK in the ER model can provide us with a new perspective on the study of PCK and the design of learning environments. Furthermore, we think that the concept and study of PCK can add a new perspective to the Fachdidaktik research domain. As Van Driel, Veal, and Janssen (2001, p. 985) pointed out, ‘‘the ‘Fachdidaktik’ tradition has not provided detailed insight into y how they [teachers] relate their transformations [of their subject matter knowledge] to student understanding during classroom communication in order to overcome student misconceptions, thus contributing to more meaningful learning.’’ A study of teachers’ PCK can provide insight in this instruction process. Furthermore, several recent articles mention that there is a lack of studies on science teachers’ topic specific PCK (Bucat, 2004; Van Driel, Verloop, & De Vos, 1998). Van Driel et al. (1998) observe in their study of science teachers’ PCK with respect to the specific topic of chemical equilibrium that the study benefited from incorporating research on student learning of chemical equilibrium. They state that in view of the vast amount of research on student conceptions with respect to specific science topics, studies on the same topics from the teachers’ perspective would be welcome. As mentioned at the beginning of this introduction, the main aim of this paper is to present a new model for an integrated approach to the study of science teachers’ PCK. But before we can introduce the new ERTE research model (in Section 3.4), we shall in Section 2 address the question: ‘‘What is PCK?’’. The history (Section 2.1), the nature (Section 2.2) and the sources (Section 2.3) of PCK will be discussed in order to develop a clear idea of PCK. After this discussion the research model of ERTE for the study of PCK will be introduced and described in Section 3. In Section 4, the conclusion and an outlook are presented. 2. Pedagogical content knowledge 2.1. PCK as category within a knowledge base for teaching In 1983 (see Shulman, 1999), Lee Shulman declared at a national conference at the University of Texas that an element is missing within educational research, especially the research on teacher cognition: ‘‘Where the teacher cognition program has clearly fallen short is in the elucidation of

teachers’ cognitive understanding of the subject matter content and the relationship between such understanding and the instruction teachers provide for students.’’ (Shulman, 1986, p. 25). This observation led Shulman, in his Presidential Address to the American Educational Research Association in 1985, to distinguish three categories within the domain of content knowledge: curriculum knowledge, subject matter knowledge and a new category named PCK (Carlsen, 1999). This was the first time PCK was mentioned and it was described as ‘‘the particular form of content knowledge that embodies the aspects of content most germane to teachability’’ (Carlsen, 1999, p. 137). Through addressing this goal for better research Shulman also aimed at addressing the political goal of professionalizing teaching (Carlsen, 1999). The claim that teaching is a profession is based on the belief that there is a knowledge base for teaching (Shulman, 1987). In Shulman’s key publication titled ‘‘Knowledge and teaching: foundations of the new reform’’ (1987) PCK is included in a knowledge base for teaching that consists of seven categories: (1) content knowledge, (2) general pedagogical knowledge, (3) curriculum knowledge, (4) PCK, (5) knowledge of learners and their characteristics, (6) knowledge of educational contexts, (7) knowledge of educational ends, purposes and values, and their philosophical and historical grounds. A quotation from this publication can help us to understand what PCK, as part of the knowledge base for teaching, is: [PCK] represents the blending of content and pedagogy into an understanding of how particular topics, problems, or issues are organized, represented, and adapted to the diverse interests and abilities of learners, and presented for instruction (Shulman, 1987, p. 8). Since then, Shulman and others have proposed variants of this model of the domains of teacher knowledge (for example Carlsen, 1999; Magnusson, Krajcik, & Borko, 1999; Tamir, 1988). The definitions of the components within the different models vary. Grossman (1990, p. 5) has tried to remedy this situation by distinguishing ‘‘four general areas of teacher knowledge [that] can be seen as the cornerstones of the emerging work on professional knowledge for teaching: general pedagogical knowledge, subject matter knowledge, PCK, and knowledge of context.’’ (See also Fig. 1). Within Grossman’s knowledge base for teaching, general

ARTICLE IN PRESS E.M. van Dijk, U. Kattmann / Teaching and Teacher Education 23 (2007) 885–897

subject matter knowledge

pedagogical knowledge

PCK

knowledge of context

Fig. 1. The model of teacher knowledge. In this model PCK is presented as a unique knowledge domain.

pedagogical knowledge is defined as knowledge concerning learning and learners, knowledge of general principles of instructions, knowledge related to classroom management, and knowledge about the aims and purposes of education. Knowledge of context includes knowledge of school setting, for example culture, and knowledge of individual students. One aspect of PCK as defined by Shulman remains unclear, since there seems to be no distinction between PCK as an educational concept, an abstract idea used in teacher education and textbooks, and PCK as a subjective representation—an element of teachers’ professional knowledge (Bromme, 1995). Although Hashweh (2005) states that the PCK concepts needs to be further conceptualized, he does not solve the aforementioned conceptual problem. Hashweh (2005, pp. 277–278) defines PCK as personal and private knowledge and says that ‘‘Efforts by some researchers to capture and represent PCK, etc. can transform it into more public knowledge.’’ But he does not mention that it is necessary, for empirical research on PCK, to distinguish between the educational ideas that concern the integrated area of content and pedagogy that can be used in teacher education and its representation and transformation within a teacher’s mind. This is necessary because, when we study a teacher’s PCK, we should realize that the knowledge that the teacher has acquired during his or her teaching career can differ from the available theoretical concepts within the educational field. Only if the researchers take this difference into consideration the real practical ‘wisdom’ of the teacher is discovered (Bromme, 1995). Bromme (1995) describes an interesting example of a study in which the researchers were not fully aware of the aforementioned difference. Carpenter, Fennema,

889

Peterson, and Carey (1988) aimed to study teachers’ PCK concerning their knowledge of children’s addition strategies. The teachers in the study were asked to assess the degree of difficulty of various task types. Although the majority of assessments were correct, the teachers did not refer to students’ individual solving strategies but instead the teachers referred to the difficulties students experience in finding out which type of task they have to work on. Carpenter et al. (1988) concluded that the teachers lacked PCK. Carpenter et al. (1988) did not fully recognize the teachers’ knowledge of the problems that students have with identifying the required operation as adequately developed PCK (Bromme, 1995). Another reason for making the distinction between PCK as an educational concept and PCK as a knowledge domain in a teacher’s mind with respect to teacher training is that prospective teachers can not learn directly from expert teachers’ PCK. The teachers’ PCK has to be structured or ‘reconstructed’ in order to be transferred from expert teachers to other, prospective teachers in teacher education. The resulting educational ideas, are different from teachers’ PCK since they are extracted from their context: the knowledge, beliefs and experiences of the individual teacher. These educational ideas can then be used to ‘reconstruct’ teacher education. In our view the notion of PCK should be understood as a knowledge domain and not as an educational concept or idea. The model for ERTE, that we shall introduce in the next section, explicitly uses PCK as such. This more limited conceptualization of PCK influences the relation between the concept of PCK and the German idea of Fachdidaktik. In the introduction it was mentioned that Fachdidaktik and PCK can both be seen as knowledge domains that consist of a combination of content knowledge and general pedagogical knowledge (Terhart, 1995). Gudmundsdottir and Grankvist (1992; see also Kansanen & Meri, 1999) for this reason see PCK and Fachdidaktik as almost identical ideas. However, as has been shown, PCK is in the first place personal and private knowledge and if Fachdidaktik is defined in the same way, as a personal knowledge domain that is, then we get a too narrow description of Fachdidaktik. Fachdidaktik can also be seen as— and is usually seen as—a research domain, a scientific discipline that tries to answer questions that refer to the domain in which educational science and, for example, biology come together

ARTICLE IN PRESS 890

E.M. van Dijk, U. Kattmann / Teaching and Teacher Education 23 (2007) 885–897

and that focuses on the transfer of the subject matter to the learner (Bayrhuber, 1995; GropengieXer & Kattmann, 2006). For example, biology Didaktik empirically studies, among others, prior knowledge and pre-scientific conceptions from pupils; learners’ motivation with respect to biology and biology education; the effects of the implementation of media and methods; and ER of biological contents (GropengieXer & Kattmann, 2006; Hedewig, 1990). If Fachdidaktik is considered to be, in the first place, a field of science, then the study of PCK should be seen as a research field, among many other fields within the Fachdidaktik research domain (for example Bayrhuber, 1995; GropengieXer & Kattmann, 2006; Hedewig, 1990; Sandfuchs, 1990). In contrast to Gudmundsdottir and Grankvist’s view, PCK and Fachdidaktik thus are not identical. PCK is a teacher’s personal knowledge domain and the study of teachers’ PCK is a field of research within the research domain of Fachdidaktik. 2.2. The nature and key elements of PCK There appears to be widespread agreement among education researchers on the nature of PCK as a unique knowledge domain. PCK refers to specific topics, and is therefore to be discerned from the general knowledge of pedagogy, and PCK concerns the teaching of specific topics, and will therefore differ considerably from subject matter knowledge (Van Driel et al., 1998, p. 677). PCK is a unique domain that is informed by, but does not completely subsume, the other knowledge areas. There seems to exist a reciprocal relationship between PCK and the foundational knowledge domains, subject matter, pedagogy, and context (Fig. 1). The foundational knowledge domains inform PCK and PCK influences the teacher’s knowledge, of the subject matter, pedagogy, and the context (Gess-Newsome, 1999). There is no universally accepted conceptualization of PCK (Hashweh, 2005; Van Driel et al., 1998). Differences occur with respect to the elements that scholars include or integrate in PCK and with respect to the descriptions of these elements. However, all scholars seem to agree with Shulman that the understanding of students’ specific learning difficulties and the knowledge of representations of the subject matter to overcome these difficulties are two essential elements of PCK (Van Driel et al., 1998). During their teaching career

teachers acquire knowledge of strategies and representations for teaching particular topics and knowledge of students’ conceptions of these topics. This PCK enables teachers to anticipate students’ problems with respect to a specific topic. However, according to Ball and Bass (2000), who were considering mathematics education ‘‘a body of such bundled knowledge [bundles of knowledge consisting of the two key elements of PCK] may not always equip the teacher with the flexibility needed to manage the complexity of practice. Teachers also need to puzzle about the mathematics in a student’s idea, analyze a textbook presentation, consider the relative value of two different representations in the face of a particular mathematical issue. To do this, we argue, requires a kind of mathematical understanding that is pedagogically useful and ready’’ (Ball & Bass, 2000, p. 88). Based on their research Ball and Bass (2000) suggest that there is a difference between knowing the subject matter and using this subject matter knowledge for teaching. Therefore we suggest that PCK includes a third element.3 This element of PCK which we will simple call ‘subject matter knowledge for teaching’ enables the teacher to react flexible in different and unanticipated situations. 2.3. The sources of PCK Van Driel, De Jong, and Verloop (2002), Van Driel et al. (1998) state in their review of the literature on PCK that studies of teachers’ PCK indicate that subject matter knowledge is a prerequisite for the development of PCK and that PCK develops in the actual teaching practice of teachers. That teaching experience is essential for the development of PCK implies that beginning teachers usually have little or no PCK at their disposal.4 They further observe that specific courses or workshops have the potential to influence the development of PCK. 3 ‘Subject matter knowledge for teaching’ is comparable with what Ball and Bass (2000, p. 89) describe as ‘pedagogical useful mathematical understanding’. The main difference with our conceptualization of teacher knowledge is that according to the view of Ball and Bass (2000) this knowledge is not included in PCK. However, we suggest that ‘Subject matter knowledge for teaching’ consists of a blending of content and pedagogy and should therefore be included in PCK. 4 See, for example, the comparative studies of the PCK of experienced and novice teachers by Gudmundsdottir and Shulman (1987) and Clermont, Borko, and Krajcik (1994).

ARTICLE IN PRESS E.M. van Dijk, U. Kattmann / Teaching and Teacher Education 23 (2007) 885–897

These three sources of PCK are also identified by Grossman (1990) in her book ‘‘The making of a teacher’’. Complementary to these three sources she identified a fourth possible source from which PCK can be developed, namely apprenticeship of observation—the memories that prospective teachers have of their experiences as a student shape their expectations of learners. She (Grossman, 1990, pp. 105–111) observed in her detailed study, using contrasting case studies with a focus on professional coursework, six novice teachers of English who are all well prepared in their subject matter. The teachers, three having done professional coursework and three having no professional coursework experience, report to have learned most of what they know about student understanding from their teaching experience. However, the teachers without professional training find it very difficult to anticipate students’ prior knowledge and are surprised about what the students do and do not know. They use memories of their own experiences at school to shape their expectations. Although they learn, from their experience, what topics students find difficult, they do not learn why the students find these topics difficult and how they could integrate this newly acquired knowledge of student understanding in their teaching. Grossman (1990) states that through professional training beginning teachers can acquire a framework that shapes their learning from experience. Van Driel et al. (1998) showed in their study of experienced chemistry teachers’ PCK with respect to the specific topic of chemical equilibrium that the teachers were aware of existing misconceptions concerning the dynamic conception of chemical equilibrium. However, most of the teachers themselves struggled with the abstract nature of the dynamic conception of chemical equilibrium: ‘‘These teachers lacked theoretical arguments to promote student understanding. They admitted that their usual arguments are weak and not very convincing for most of their students. The best they felt they could do was to demonstrate the dynamic equilibrium conception with help of metaphors or analogies’’ (Van Driel et al., 1998, p. 686). The chemistry teachers discussed the strengths and weaknesses of each other’s favorite analogies: ‘‘Remarkable in the discussion on the strengths and weaknesses of each others favorite analogies was the focus on the chemical validity, the students’ perspective was not taken in to account.’’ (Van Driel et al., 1998, p. 686). These results are confirmed by Magnusson et al. (1999) in their

891

review article of the literature on science teachers’ PCK. They observe that, although the amount of research on science teachers’ PCK of students’ understanding is limited, the results of the existing studies are consistent. ‘‘Although teachers have some knowledge about students’ difficulties, they commonly lack important knowledge necessary to help students overcome those difficulties’’ (Magnusson et al., 1999, p. 106). Van Driel et al. (1998) showed that a workshop can enhance this component of PCK. Some participants in their research reported to have extended their repertoire of successful strategies and representations with respect to the topic of the chemical equilibrium. From this short discussion of the literature on the sources of PCK can be concluded that teacher training is necessary for novice and experienced teachers to enable them to learn from their experiences. According to the model of teacher knowledge (Fig. 1) prospective teachers not only need to acquire deep and flexibly organized understanding in the areas of subject matter, pedagogy and context but they also need to develop a framework that enables them to develop their PCK through learning from experiences. This latter aim implies that knowledge should also be taught to teachers in training in a purposefully integrated manner. Then teachers in training should be able to develop more quickly the skills and knowledge they need to be effective teachers. For example Ball and Bass (2000) suggest that analyzing and designing a task can help prospective teachers to develop their ‘subject matter knowledge for teaching’. In the next section, we will introduce the ERTE model that aims to ‘reconstruct’ science teacher education in relation to the design of learning environments and an empirical study of science teachers’ PCK. However, it is not enough to teach solely integrated pedagogical and subject matter knowledge. This would, according to Gess-Newsome (1999), reduce teacher preparation to the presentation of a ‘box of tricks’. It will not enable teacher to learn from their experience and manage their teaching practice flexible. 3. The model of educational reconstruction for teacher education 3.1. Introduction The main purpose of this article is to introduce a research model for the study of science teachers’

ARTICLE IN PRESS E.M. van Dijk, U. Kattmann / Teaching and Teacher Education 23 (2007) 885–897

892

PCK. This new model of ERTE forms the framework for an integrative approach to the study of PCK aimed at improving teacher education. This model is explicitly based on an established research model, the model of ER, that has been developed in the biology education group at Oldenburg University in cooperation with the department of physics education at the IPN in Kiel for the study of students’ pre-scientific conceptions (descriptions of the model can be found in Kattmann et al., 1997, 1998). The research model of ER integrates three wellknown research domains within educational research: the empirical study of students’ conceptions, the analysis of the subject matter and the design of learning environments or teaching–learning sequences (Fig. 2). The research model represents a framework in which the three elements can be studied in relation to each other. The three components of the model are not to be dealt with one after another but influence each other mutually. The first component of the model, design of learning environments, is influenced by the two other components within the model but also influences, from the start, these components—the study of the students’ pre-scientific conceptions and the analysis of the subject matter. This is because these studies are performed with the explicit intention to develop contents for education (see also Duit et al., 2005; GropengieXer, 2001). A second characteristic of the ER model is that the orientation toward the specific science domain and the orientation on the student are brought into balance. The science content and the students’ conceptions are seen as being equally important for ER (see also Duit et al., 2005). A

design of learning environments

subject matter analysis

empirical study of students' pre-scientific conceptions

Fig. 2. The research model of educational reconstruction (Kattmann et al., 1998).

third feature of the ER model is its dual character— it involves both research and development. It provides a framework for designing and evaluating learning environments or teaching–learning sequences (Duit et al., 2005; Me´heut & Psillos, 2004). In Europe and the USA there is a growing interest in design research see for example the theme issues of Educational Researcher (Kelly, 2003) and the International Journal of Science Education (Me´heut & Psillos, 2004). In Section 3.2, a thorough description of the model of ER will be presented followed by a description of the study of PCK in Section 3.3. Finally, in Section 3.4, the new model for the study of PCK, the ERTE model, will be presented. 3.2. The model of educational reconstruction (ER) The model of ER (Fig. 2) is based on two key assumptions. The first assumption is that learning always takes place in particular situational contexts and that these contexts are not prescribed by scientific knowledge. The concepts of scientific theories have to be put in certain contexts: The term reconstruction, therefore denotes that issues that were lost in the process of formation of scientific knowledge have to be reconstructed in order to make the science point of view understandable and meaningful to learners. Because the analysis of content structure is unavoidably influenced by educational issues we use the term educational reconstruction (Kattmann, Duit, & GropengieXer, 1998, p. 255). The second assumption is that the researcher’s or the teacher’s awareness of the students’ point of view may substantially influence the reconstruction of the particular science content. In order to design learning environments, the first element of the model, the science content needs to be reconstructed based on the study of the conceptions of the students in combination with the analysis of the subject matter. Firstly, the similarities between students’ conceptions and the scientific conceptions must be considered mutually. Secondly, the educational objectives and the students’ ideas have to be put in a context that is understandable for students. The abstract scientific conceptions have to be enriched and embedded in the science content for teaching (see also Duit et al., 2005). The relevant questions to the reconstruction of science teaching are: which are the most relevant

ARTICLE IN PRESS E.M. van Dijk, U. Kattmann / Teaching and Teacher Education 23 (2007) 885–897

elements of the students’ conceptions? Which opportunities are opened by certain elements of students’ conceptions or perspectives? In which way do clarified scientific conceptions promote or constrain learning? Which conceptions of students correspond with scientific conceptions in such a way that they can be used for more adequate and fruitful learning? The second component of the ER-model, the empirical research of students’ cognitive and affective conceptions, is based on the important assumption that students’ conceptions are both the starting point for learning and an aid for learning. Learning is considered to be a constructive process. This means that students enter the learning process with their own conceptions and they build their own knowledge based on existing knowledge and experiences. Therefore educational researchers, designers and teachers must be aware of the learners’ prior knowledge or pre-scientific conceptions when they develop learning environments. The study of students’ conception within the ER program aims to answer at least some of the following questions using qualitative methods of empirical research: how are the scientific concepts represented in students’ perspectives? Which conceptions are used by the students? Which perspectives do students have about science? How do alternative conceptions of students correspond with scientific conceptions? The purpose of the analysis of the subject matter, the third component of the model, is to identify the essential concepts and their relationships in the scientific literature—science textbooks and books on the philosophy and history of science. The analysis aims to answer, among others, the following questions: which are the scientific theories and concepts on a specific subject? What is the genesis, function and meaning of the scientific concepts, and in which context are they placed? Which scientific terms are being used, and which of them constrain or promote learning because of their literal meaning? Which ethical and social implications are associated with the scientific conceptions? 3.3. Pedagogical content knowledge studies (PCK-S) The model of ER is designed to better develop learning environments based on the integrated study of students’ pre-scientific conceptions and the scientific conceptions. Complementary to the research on students’ pre-scientific conceptions, it is

893

also important to study the knowledge and beliefs that teachers have, since the teacher plays an important role in the instruction process. The teacher tries to represent the subject matter to his students in ways he thinks the students can relate to. Therefore, a teacher needs to have some developed PCK. The teacher should have developed ‘subject matter knowledge for teaching’ that enables him or her to follow a students’ reasoning and to analyze representations of the subject matter. He or she must have knowledge of students’ pre-scientific conceptions in order to understand the conceptual difficulties that might arise when teaching a specific topic and the teacher should have knowledge of representations of the subject matter to overcome these difficulties. As discussed in the previous section the term ‘PCK’ refers to a unique knowledge domain within a teacher’s mind. The educational ideas that are the results of what we call pedagogical content knowledge studies (PCK-S), the field within educational research that focuses its research on the PCK that teachers possess, are different from teachers’ PCK since they are extracted from their context: the knowledge, beliefs and experiences of the individual teacher. These results are described, interpreted and abstracted by the researcher and reconstructed for teacher education with the intention that these ideas are internalized by the prospective teachers and form an element of the framework that enables them to learn from their experiences in their own Individual Teaching Practice (Fig. 3).5 Briefly said: teachers acquire PCK from their Individual Teaching Practice; this PCK forms the object for researchers that study teachers’ PCK; these researchers try to construct new educational ideas from the PCK of experienced teachers; these educational ideas, in turn, can be taught to other teachers in workshops and courses and in this way flow into these teachers’ Individual Teaching Practice. 3.4. The model of educational reconstruction for teacher education (ERTE) The model of ER can be easily adapted into a model that includes PCK-S: the original ER model is integrated in the new model that is a copy of this 5 This model was developed in cooperation with Thomas A.C. Reydon, Center for Philosophy and Ethics of Science (ZEWW), University of Hanover.

ARTICLE IN PRESS 894

E.M. van Dijk, U. Kattmann / Teaching and Teacher Education 23 (2007) 885–897

Individual Teaching Practice (ITP) internalized educational ideas

Pedagogical Content Knowledge (PCK)

teacher education

Pedagogical Content Knowledge Studies (PCK-S)

educational ideas Fig. 3. A model which represents the subject, object and purpose of empirical research on PCK. The model aims to clarify the distinction between PCK, the personal knowledge domain of the teacher, and educational ideas, the results from PCK-S.

EDUCATIONAL CONSTRUCTION OF TEACHER EDUCATION

DESIGN OF LEARNING ENVIRONMENTS

subject matter analysis

PEDAGOGICAL CONTENT KNOWLEDGE STUDIES

study of students' conceptions

Fig. 4. The stack model of educational reconstruction for teacher education: In this model, the empirical study of teachers’ PCK is related to the design of learning environments and the construction of teacher education.

triangular model (Fig. 4). The result is a two-layered model or a stack model in which the ER model fills one of the corners of the ERTE model. We assume

that the content for teacher education needs to be ‘reconstructed’ for teaching. In the ERTE model the following research domains are integrated: (1) the

ARTICLE IN PRESS E.M. van Dijk, U. Kattmann / Teaching and Teacher Education 23 (2007) 885–897

design of learning environments, (2) the empirical study of students’ pre-conceptions, (3) the analysis of the subject matter, (4) PCK-S, and (5) the design of teacher education (Fig. 4). These elements are strongly interrelated, they influence each other mutually. Within the framework of the ERTE model the PCK-S will be carried out in relation to the research results of ER with the intention to construct teacher education. In this ERTE design research model the teacher has become an essential element. Lohmann (2006) has recently described a similar enhanced ER model for the empirical study of prospective teachers’ pre-conceptions instead of the PCK of experienced teachers. The purpose of the study of teachers’ PCK is to answer three questions that relate to the three keyelements of PCK6: (1) What ‘subject matter knowledge for teaching’ do teachers have? (2) What conceptions do teachers have of students’ pre-scientific conceptions on the subject matter? (3) What conceptions do teachers have of subject matter representations? These three questions are fully in accord with the elements of the ER model: (1) analysis of the subject matter, (2) study of students’ pre-conceptions, and (3) design of learning environments or representations of the subject matter. The research design for the study of PCK can be sketched as follows: A team of researchers works on the analysis of the subject matter and a study of the literature on students’ pre-conceptions. Some of the researchers focus on the empirical research of students’ pre-conceptions while the others focus on the empirical research of teachers PCK. The interaction between the two studies is essential in this research design. The analysis of the subject matter, the study of students’ pre-conceptions and the process of designing a learning environment or teaching–learning sequences form the reference material for the PCK-S. This reference material is not meant to set the standard on which the teacher’s knowledge is being measured. Rather, the function of the reference material is to provide us with a broad picture of the different conceptions and 6

These three questions are congruent with those of Anne JanXen-Bartels (2003; poster presentation ERIDOB-conference, 2004), who derived her research questions from the model of ER but without considering the concept of PCK.

895

representations of the subject matter that can be used as a context in which the teacher’s conceptions can be fully explored. On the other hand can the results of the PCK-S influence the design process if the teacher has acquired, during his or her teaching career, ways to present the subject knowledge in a for students understandable way. The results of this research process can then be used to improve teacher education. We do not suggest that prospective teachers try out the learning environments that are designed for secondary school students but instead we suggest that the above described research process is ‘reconstructed’ for teacher education. If future PCK studies within the framework of the ERTE model confirm the findings of earlier studies, namely that teachers are aware of the conceptual difficulties of their students but do not have the ability to respond adequately, this integrated exploration of teacher knowledge may be able to provide us with some solutions for this problem. 4. Conclusion and outlook The main purpose of this paper is to introduce the ERTE model. The model forms the basis for an integrated approach of the study of science teachers’ PCK which aims to improve teacher education. Within the framework of the ERTE model the teacher has become an essential element in the design process of learning environments. The usefulness of the model and the proposed research design will have to be substantiated in the future. Our next step will be to use the ERTE model for a qualitative study of PCK of secondary school biology teachers with respect to the topic of evolution. Three research questions have been formulated: (1) What ‘subject matter knowledge for teaching’ do biology teachers have concerning the topic of evolution? (2) What conceptions do teachers have of students’ pre-scientific conceptions on evolution? (3) What conceptions do biology teachers have of subject matter representations of evolution? Within the confines of our planned research project it will be necessary to adapt the above described ERTE research design: In view of the vast amount of research on student conceptions with respect to evolution it is possible to reduce the study of students’ conceptions to a literature study of students’ conceptions (Baalmann, Frerichs, & Kattmann, 1999, studied students’ pre-conceptions with respect to evolution within the framework of the ER model). The analysis of the subject matter

ARTICLE IN PRESS 896

E.M. van Dijk, U. Kattmann / Teaching and Teacher Education 23 (2007) 885–897

will have to be more limited and the process of designing a learning environment will be reduced to the analysis of existing materials in relation to the aforementioned literature study and analysis of the subject matter. The primary purpose of this empirical study of PCK is to formulate recommendations for teacher education with respect to the teaching of evolution.

References Baalmann, W., Frerichs, V., & Kattmann, U. (1999). How the gorillas became dark—Research in students’ conceptions leads to a rearrangement of teaching genetics and evolution. In O. de Jong, K. Kortland, A. J. Waarlo, & J. Buddingh (Eds.), Bridging the gap between theory and practice: What research says to the science teacher, Proceedings of the 1998 international summer symposium, Utrecht University, The Netherlands (pp. 171–189). Hong Kong: ICASE. Ball, D. L., & Bass, H. (2000). Interweaving content and pedagogy in teaching and learning to teach: Knowing and using mathematics. In J. Boaler (Ed.), Multiple perspectives on mathematics teaching and learning. Westport, CT: Ablex. Bayrhuber, H. (1995). Dimensions of the Didaktik of natural sciences in Germany. In S. Hopmann, & K. Riquarts (Eds.), Didaktik and/or curriculum (pp. 335–344). IPN: Kiel. Bromme, R. (1995). What exactly is ‘pedagogical content knowledge’?—Critical remarks regarding a fruitful research program. In S. Hopmann, & K. Riquarts (Eds.), Didaktik and/or curriculum (pp. 205–216). IPN: Kiel. Bucat, R. (2004). Pedagogical content knowledge as a way forward: Applied research in chemistry education. Chemistry Education: Research and Practice, 5, 215–228. Carlsen, W. S. (1999). Domains of teacher knowledge. In J. GessNewsome, & N. G. Lederman (Eds.), Examining pedagogical content knowledge (pp. 133–144). Dordrecht: Kluwer. Carpenter, T. P., Fennema, E., Peterson, P. L., & Carey, D. A. (1988). Teachers’ pedagogical content knowledge of students’ problem solving in elementary arithmetic. Journal of Research in Mathematics Education, 19, 385–401. Clermont, C. P., Borko, H., & Krajcik, J. S. (1994). Comparative study of the pedagogical content knowledge of experienced and novice chemical demonstrators. Journal of Research on Science Teaching, 31, 419–441. Cochran-Smith, M., & Lytle, S. L. (1993). Inside/outside: Teacher research and knowledge. New York: Teachers College Press. Doyle, W. (1992). Curriculum and pedagogy. In P. W. Jackson (Ed.), Handbook of research on curriculum: A project of the American Educational Research Association. New York: Macmillan. Doyle, W., & Westbury, I. (1992). Die Ru¨ckbesinnung auf de Unterrichtsinhalt in der Curriculum- und Bildungsforschung in den USA. Bildung und Erziehung, 45, 137–157. Duit, R., GropengieXer, H., & Kattmann, U. (2005). Towards science education research that is relevant for improving practice: The model of educational reconstruction. In H. Fischer (Ed.), Developing standards in research on science education. The ESERA Summer School 2004. London: Palmer Press.

Gess-Newsome, J. (1999). Pedagogical content knowledge: An introduction and orientation. In J. Gess-Newsome, & N. G. Lederman (Eds.), Examining pedagogical content knowledge (pp. 3–17). Dordrecht: Kluwer. Good, R. (Ed.). (1991). Editorial. Journal of Research in Science Teaching, 28, 387. GropengieXer, H. (2001). Didaktische Rekonstruktion des Sehens; wissenschaftliche Theorien und die Sicht der Schu¨ler in der Perspektive der Vermittlung. In U. Kattmann (Ed.), Beitra¨ge zur Didaktische Rekonstruktion Band 1. Oldenburg: Didaktisches Zentrum, Carl von Ossietzky Universita¨t. GropengieXer, H., & Kattmann, U. (Eds.). (2006). Fachdidaktik Biologie begru¨ndet von Dieter Eschenhagen, Ulrich Kattmann und Dieter Rodi. Ko¨ln: Aulis Verlag Deubner. Grossman, P. L. (1990). The making of a teacher: Teacher knowledge and teacher education. New York: Teachers College Press. Gudmundsdottir, S., & Grankvist, R. (1992). Deutsche Didaktik aus der Sicht neuerer empirischer Unterrichts- und Curriculumforschung in den USA. Bildung und Erziehung, 45, 175–187. Gudmundsdottir, S., & Shulman, L. (1987). Pedagogical content knowledge in social studies. Scandinavian Journal of Educational Research, 31, 59–70. Gudmundsdottir, S., Reinertsen, A., & Nordtømme, N. (2000). Klafki’s Didaktik Analysis as a conceptual framework for research on teaching. In I. Westbury, S. Hopmann, & K. Riquarts (Eds.), Teaching as a reflective practice: The German Didaktik tradition (pp. 319–334). Mahwah, NJ: Lawrence Erlbaum Associates. Gundem, B. B., & Hopmann, S. (1998). Introduction: Didaktik meets Curriculum. In B. B. Gundem, & S. Hopmann (Eds.), Didaktik and/or curriculum: An international dialogue (pp. 1–8). New York: Peter Lang. Hamilton, D. (1999). The pedagogic paradox (or why no didactics in England?). Pedagogy, Culture and Society, 7, 135–152. Hashweh, M. Z. (2005). Teacher Pedagogical Constructions: A reconfiguration of pedagogical content knowledge. Teachers and Teaching: Theory and Practice, 11, 273–292. Hedewig, R. (1990). Aufgaben einer zeitgema¨Xen Biologiedidaktik. In R. W. Keck, W. Ko¨hnlein, & U. Sandfuchs (Eds.), Fachdidaktik zwischen Allgemeiner Didaktik und Fachwissenschaft (pp. 289–306). Bad Heilbrunn: Klinkhardt. Hopmann, S. (2000). Klafki’s model of Didaktik analysis. In I. Westbury, S. Hopmann, & K. Riquarts (Eds.), Teaching as a reflective practice: The German Didaktik tradition (pp. 197–206). Mahwah, NJ: Lawrence Erlbaum Associates. Hopmann, S., & Riquarts, K. (1995). Starting a dialogue: Issues in a beginning conversation between Didaktik and the curriculum traditions. Journal of Curriculum Studies, 27, 3–12. Hopmann, S., & Riquarts, K. (2000). Starting a dialogue: A beginning conversation between Didaktik and the curriculum traditions. In I. Westbury, S. Hopmann, & K. Riquarts (Eds.), Teaching as a reflective practice: The German Didaktik tradition (pp. 9–11). Mahwah, NJ: Lawrence Erlbaum Associates. JanXen-Bartels, A. (2003). Schu¨lervorstellungen zum Unterricht u¨ber ‘‘Menschenrassen’’. In H. Vog, D. Kru¨ger, & U. Unterbruner (Eds.), Erkenntnisweg Biologiedidaktik (pp. 57–71). Kassel: Universita¨t.

ARTICLE IN PRESS E.M. van Dijk, U. Kattmann / Teaching and Teacher Education 23 (2007) 885–897 Kansanen, P., & Meri, M. (1999). The didactic relation in the teaching–studying–learning process. The TNTEE Journal, 2, 107–116. Kattmann, U., Duit, R., & GropengieXer, H. (1998). The model of educational reconstruction. Bringing together issues of scientific clarification and students’ conceptions. In H. Bayrhuber, & F. Brinkman (Eds.), What– Why– How? Research in Didaktik of biology (pp. 253–262). Kiel: IPN. Kattmann, U., Duit, R., GropengieXer, H., & Komorek, M. (1997). Das Modell der Didaktische Rekonstruktion—Ein Rahmen fu¨r naturwissenschaftsdidaktische Forschung und Entwicklung. Zeitschrift fu¨r Didaktik der Naturwissenschaften, 3, 3–18. Kelly, A. E. (2003). Theme issue: The role of design in educational research. Educational Researcher, 32, 3–4. Klafki, W. (1996). Zur Unterrichtsplanung im Sinne kritischkonstruktiver Didaktik. In W. Klafki (Ed.), Neue Studien zur Bildungstheorie und Didaktik: Zeitgema¨X e Allgemeinbildung und kritisch-konstruktive Didaktik (pp. 251–284). Weinheim: Beltz Verlag. Klafki, W. (2000). Didaktik analysis as the core of preparation of instruction. In I. Westbury, S. Hopmann, & K. Riquarts (Eds.), Teaching as a reflective practice: The German Didaktik tradition (pp. 197–206). Mahwah, NJ: Lawrence Erlbaum Associates. Ku¨nzli, R. (2000). German Didaktik: Models of re-presentation of intercourse, and of experience. In I. Westbury, S. Hopmann, & K. Riquarts (Eds.), Teaching as a reflective practice: The German Didaktik tradition (pp. 41–54). Mahwah, NJ: Lawrence Erlbaum Associates. Lagemann, E. C. (2000). An elusive science: The troubling history of educational research. Chicago: The University of Chicago Press. Lohmann, G. (2006). Didaktische Rekonstruktion in der Hochschuldidaktik. Journal fu¨r Lehrerinnen- und Lehrerbildung, 2 (in press). Magnusson, S., Krajcik, J., & Borko, H. (1999). Nature, sources, and development of PCK. In J. Gess-Newsome, & N. G. Lederman (Eds.), Examining pedagogical content knowledge (pp. 95–132). Dordrecht: Kluwer. Me´heut, M., & Psillos, D. (2004). Teaching–learning sequences: Aims and tools for science education research. International Journal of Science Education, 26, 515–535 (Special Issue).

897

Reinertsen, A., Nordtømme, N., Eidsvik, R., Weidemann, E., & Gudmundsdottir, S. (1996). Fagdidaktisk kunnskap i historie belyst grennom undervisning i hendelsene pa˚ Eidsvoll 1814: Tre kasusstudier av lærer i viderega˚ende skole. Trondheim: Tapir. Sandfuchs, U. (1990). Anmerkungen zur historischen Entwicklung und zum gegenwa¨rtigen Stand der Fachdidaktik als Wissenschaft. In R. W. Keck, W. Ko¨hnlein, & U. Sandfuchs (Eds.), Fachdidaktik zwischen Allgemeiner Didaktik und Fachwissenschaft (pp. 289–306). Bad Heilbrunn: Klinkhardt. Shulman, L. S. (1986). Paradigms and research programs for the study of teaching. In M. C. Wittrock (Ed.), Handbook of research on teaching (pp. 3–36). New York: Macmillan. Shulman, L. S. (1987). Knowledge and teaching: Foundations of the new reform. Harvard Educational Review, 57, 1–21. Shulman, L. S. (1999). Foreword. In J. Gess-Newsome, & N. G. Lederman (Eds.), Examining pedagogical content knowledge (pp. ix–xii). Dordrecht: Kluwer. Tamir, P. (1988). Subject matter and related pedagogical knowledge in teacher education. Teaching and Teacher Education, 4, 99–110. Terhart, E. (1995). Didaktik/curriculum in teacher education: Some German complications. In S. Hopmann, & K. Riquarts (Eds.), Didaktik and/or curriculum (pp. 289–300). IPN: Kiel. Van Driel, J. H., De Jong, O., & Verloop, N. (2002). The development of preservice chemistry teachers’ pedagogical content knowledge. Science Education, 86, 572–590. Van Driel, J. H., Veal, W. R., & Janssen, F. J. J. M. (2001). Pedagogical content knowledge: An integrative component within the knowledge base for teaching (an essay review). Teaching and Teacher Education, 17, 979–986. Van Driel, J. H., Verloop, N., & De Vos, W. (1998). Developing science teachers’ pedagogical content knowledge. Journal of Research in Science Teaching, 35, 673–695. Wandersee, J. H., Mintzes, J. J., & Novak, J. D. (1994). Research on alternative conceptions in science. In D. L. Gabel (Ed.), Handbook of research on science teaching and learning (pp. 177–210). New York: Macmillan. Westbury, I. (2000). Teaching as a reflective practice: What might Didaktik teach curriculum? In I. Westbury, S. Hopmann, & K. Riquarts (Eds.), Teaching as a reflective practice: The German Didaktik tradition (pp. 15–39). Mahwah, NJ: Lawrence Erlbaum Associates.