Res Sci Educ

DOI 10.1007/s11165-016-9530-1

Learning About Semi Conductors for Teaching—the Role

Played by Content Knowledge in Pedagogical Content
Knowledge (PCK) Development

Marissa Rollnick 1

# Springer Science+Business Media Dordrecht 2016

Abstract This study focuses on how teachers learn to teach a new topic and the role played by
their developing content knowledge as they teach. The paper is based on seven high school science
teachers’ studies on the teaching of semiconductors, at the time a new topic in the curriculum.
Analysis of artefacts such as teacher concept maps, video recordings of lessons, journals and other
classroom-based evidence shows how the extent and type of teachers’ content knowledge
informed their choice of teaching approaches and how their learning of content took place
alongside the development of teaching strategies. The development of content knowledge was
combined with increased understanding of how to teach the topic in almost all cases. Evidence of
development of teachers’ PCK was found in their increased ability to design teaching strategies,
and their use of representations and suitable assessment tasks for their lessons. Some specific
common teaching strategies were identified across the teachers. These strategies could add to the
canon of teachers’ topic - specific professional knowledge for semiconductors. The study provides
increased understanding of how teachers simultaneously master content and its teaching and how
mediated self-reflection is a fruitful approach for assisting teachers to learn to teach a new topic.

Keywords Pedagogical content knowledge . Teacher cognition . Professional development .

Content knowledge

Introduction

Few would argue with the assertion that a thorough understanding of content is an essential
prerequisite for teaching a topic. However, there is little consensus on the nature and depth of

* Marissa Rollnick
[email protected]

1
Marang Centre for Mathematics and Science Education, Wits School of Education, Wits University,
PO Wits, Johannesburg 2050, South Africa
 Res Sci Educ

the content knowledge required or how it should be acquired (Adler et al. 2009). It is generally
agreed that a teacher needs more extensive knowledge and understanding of content than the
students she is teaching, but it is difficult to establish consensus on the quantity and quality of
knowledge required. In science teaching, content knowledge is one of the central anchors of
the teacher’s professional knowledge base. Kind (2014) cites several sources that show how
rich and flexible content knowledge enhances teachers’ confidence in teaching their subject
and is essential to ensure that students experience positive learning outcomes. With the
boundaries of scientific knowledge continually expanding and changing, high school science
teachers increasingly have to take on the role of learner, needing a rich and flexible under-
standing of science and its principles to learn new material in order to teach it.
This study is located in a context where a demanding high school curriculum was
introduced requiring teachers to master new content while simultaneously changing their
conceptualisation of content and the way it is taught and assessed (Department of
Education 2003; Rogan 2004). Amongst the new topics introduced was that of
Bsemiconductors^, a topic which was totally new to most teachers. Few of them would
have encountered it even in content courses in their initial training. On the face of it, the
topic of Bsemiconductors^ appears to be applied and lacking in conceptual content.
However, many physics and chemistry concepts related to conduction of electricity and
chemical structure are dealt with in this topic. There are also challenges in sequencing the
content and decisions to be made about the depth of engagement with ideas, for example,
about conduction bands.
Although Bsemiconductors^ falls within the broad content area of physics and chemistry, it
is unfamiliar territory for most teachers. This situation is similar to the problem of teaching
outside one’s area of specialism, which is a global challenge. For example, Kind and Kind
(2011) note that for the foreseeable future, the majority of chemistry teachers in state-funded
schools in the UK will be graduates who have majored in a subject other than chemistry. Out-
of-field teaching of this nature has also been found to be commonplace in the USA (Ingersoll
1998; Luft et al. 2013) where Carlsen (1993) has found such teachers to be less interactive and
to ask lower cognitive level questions, a finding endorsed by Sanders et al. (1993). While
characteristics of teachers teaching unfamiliar topics have been identified, the emphasis in
these studies has been on the challenges posed by lack of content knowledge (CK) rather than
pedagogical content knowledge (PCK), and there is little reported on how teachers learn to
teach new topics and how their CK and PCK changes in the process.
This study thus follows closely the process of scaffolding teachers as they simultaneously
master a new topic while teaching it. The findings from this study may assist professional
development practitioners in other contexts in providing teachers with the tools to accelerate
their induction into teaching topics outside their comfort zone, particularly out - of - field
teachers. Such an investigation will also open new routes for teachers to follow in embarking
on teaching new topics.
The paper explores how high school science teachers learn and apply in practice new
subject matter using the construct of PCK first described by Shulman (1987) as one of seven
knowledge bases for teaching requiring the transformation of CK. The content focus is
semiconductors, which was taught to students in grades 10 and 11 (15 to 17 years old). The
following research questions guided this study:

1. How does teachers’ CK develop while engaged in a project developing their teaching of
semiconductors?
Res Sci Educ

2. What evidence of topic-specific professional knowledge and PCK in practice emerges in

this process?
3. What evidence exists for the growth of teachers’ PCK and what is its relationship to CK of
the new topic?

Theoretical Framework and Literature Review

The focus of this study was better understanding of teacher knowledge, in particular, the
development of teachers’ subject matter knowledge and its enactment in teaching, pointing to
PCK (Shulman 1986, 1987) as a suitable theoretical framework. A necessary precursor to PCK
is CK. This review thus first establishes a rationale for the nature of CK in teaching and then
looks at models of pedagogical content knowledge (PCK) and its connection with CK. The
latter part of the review examines tools for capturing PCK and empirical studies on PCK and
CK.

Content Knowledge

Differing terms for content knowledge are used in the literature in both mathematics
and science education. In his first account of PCK, Shulman talks of three kinds of
content knowledge—Bsubject matter content knowledge (sic), pedagogical content
knowledge and curricular knowledge^ (Shulman 1986, p. 9). Later, in a more com-
prehensive account, he lists CK rather than Bsubject matter content knowledge^ as one
of seven categories for a teacher’s knowledge base, again alongside PCK and
curriculum knowledge. In mathematics education, Ball et al. (2008) interpret content
knowledge as knowledge of the subject and its organising structure. Their interpreta-
tion includes Schwab’s distinction of syntactic and substantive knowledge structures
(Schwab 1978 in Shulman 1986) as well as later work by Grossman et al. (1989),
which brings in a further two categories—knowledge of content and beliefs about a
discipline. Ball et al. (2008) consider it difficult to untangle the kind of mathematics
used in teaching from mathematics content per se (Shulman 1986) and evolve a
special brand of CK, termed mathematical knowledge for teaching, which they refer
to as specialised content knowledge. Their concept is closer to topic-specific profes-
sional knowledge, discussed below.
In science education, a useful breakdown emerges from Cochran and Jones (1998, p. 708)
who reviewed research on subject matter knowledge for pre-service teachers. They suggest an
umbrella conception of subject matter knowledge which includes CK (considered as the facts and
concepts of subject matter knowledge), substantive knowledge (explanatory structures or para-
digms of the field), syntactic knowledge (methods and processes of generating new knowledge in
the field) and beliefs about subject matter. This distinction is also favoured by Kind and Kind
(2011) while Abell (2007) chooses to combine substantive and content knowledge.
In this paper, syntactic knowledge and beliefs about science are not under consideration, as
the focus is on teachers’ understanding of Bcentral ideas, relationships, elaborated knowledge
and reasoning ability^ (Abell 2007, p. 1110), so the term Bcontent knowledge^ is used
collectively for CK and substantive knowledge for the topic of semiconductors as CK, using
Abell’s description above. We classify the science equivalent of mathematics for teaching (Ball
et al. 2008) as a type of PCK discussed below.
 Res Sci Educ

Conceptions of PCK

Shulman is regarded as the originator of the term PCK although the concept has been in
existence far longer (Klafki 1958). His original conception (Shulman 1986), as indicated
above, was that PCK was a form of what he called subject matter content knowledge. In a
later paper (Shulman 1987), he was more explicit about the place of PCK as one of seven
teacher knowledge bases, with content knowledge as one of the other six. He explicitly
mentioned representations of content, students’ prior knowledge and powerful teaching
strategies as components of PCK which he regards as products of transformation of content
knowledge to make it teachable (Shulman 1986).
Two reviews with different foci (Abell 2007; Kind 2009) provide useful insights into later
thinking about PCK. Both reviews attest to the multiplicity of representations and terminology
in the area and both agree that thinking about PCK has yet to unify into a single paradigm.
Despite lack of agreement amongst researchers, PCK remains a useful construct in Kind’s
view. Four important points of disagreement emerge. The first is whether content knowledge is
a component of PCK or a separate knowledge base. Kind reviews a number of models, some
of which include CK as a component of PCK and others which consider it as a separate
component. Models including CK as part of PCK are regarded as integrative while those
considering it as a separate component are referred to as transformative (Kind 2009). There is,
however, agreement that CK is a necessary pre-cursor of PCK.
A second point of disagreement is whether PCK is personal or canonical (or collective).
Smith and Banilower (2015) argue that both forms of PCK exist but that they feed into each
other in a cyclic manner. The model described below used in this study concurs with this
viewpoint. The third debate relates to whether PCK is expressed in what teachers know or
what they do. We argue below for differing terminology to describe knowledge and practice.
Fourthly, debates arise about whether PCK is topic specific, domain specific or subject
specific. Veal and MaKinster (1999) consider several levels of PCK, from discipline to topic,
regarding topic specific PCK as the most specific and novel level.
Despite these differences, there is a remarkable level of consensus on certain components,
especially those enumerated by Shulman above.

Models of PCK

Many models have been employed to trace the transformation of subject matter knowledge to
PCK as manifested in the classroom, the most cited of which is Magnusson et al. (1999) which
itself is derived from Shulman’s (1987) knowledge bases discussed above. Magnusson et al.
consider PCK as a unique domain of teacher knowledge used to transform teachers’ CK for
improved learning outcomes. Their model explicitly includes orientations towards the various
identified components, which include knowledge of representations and instructional strate-
gies, students’ learning difficulties, curricular knowledge and assessment. A different line of
thinking was adopted by Cochran et al. (1993) who used a constructivist approach and referred
to pedagogical content knowing (PCKg), including the components of CK, students’ context,
general pedagogical knowledge, and students’ learning difficulties. These four basic compo-
nents were used by Rollnick et al. (2008) who extended the model to differentiate between
knowledge and practice, referring to the PCK in practice as manifestations of PCK.
Mavhunga and Rollnick (2013) and Rollnick et al. (2013) addressed the topic specific
debate by introducing the construct of Topic Specific PCK (TSPCK) as the specific knowledge
Res Sci Educ

needed for transformation of CK for teaching purposes. TSPCK is regarded as canonical and
composed of five components derived from Geddis (1993). These are students’ prior knowl-
edge including misconceptions, curricular saliency, what makes a topic easy or difficult to
understand, representations including analogies, and conceptual teaching strategies. The model
which has been found to be most useful for this study evolved from a PCK summit held in
Colorado (Gess-Newsome, 2015). The model (see Fig. 1) addresses two of the debates
outlined above—the tension between personal and Bcanonical^ PCK and the interface between
PCK knowledge and practice. Both these debates are addressed by the inclusion of a construct
referred to as topic specific professional knowledge—TSPK. It is topic specific, as its name
suggests. TSPK lies firmly in the domain of knowledge rather than practice and can be thought
of as canonical PCK, signifying that it has accepted indicators of quality. In terms of the
models outlined above, it is similar to Topic Specific PCK. As can be seen, there are overlaps
between the components, for example, representations and instructional strategies. TSPK links
to classroom practice through a series of amplifiers and filters such as classroom context and
teacher beliefs which in turn link to student outcomes, again through a series of contextual
filters. The model is both recursive and dynamic, with all three constructs feeding back into
each other, accounting for teacher learning through practice and interaction with learners.
Thus, the construct of PCK occurs in the classroom context as both a knowledge base and as a
skill (PCK/PCK&S) when involved in the act of teaching (Gess-Newsome 2015). Hence,
TSPK refers to Topic Specific PCK as part of teacher knowledge while the term PCK is
reserved for the personal construct as it appears in practice.
As shown in Fig. 1, CK appears as one of the teacher professional knowledge bases, and as
such is an essential underpinning of TSPK.
Schneider and Plasman (2011) examine trajectories of teacher PCK from novice to adaptive
expert. The concept of adaptive expertise is a useful one to describe teachers who relish

Fig 1 Model of PCK (Gess-Newsome 2015)

 Res Sci Educ

challenges and are continually looking for ways to improve their knowledge and abilities and
thus grow their PCK. Schneider and Plasman regard PCK as domain-specific knowledge of
teaching that is unique to teachers and developed within practice. An important use of adaptive
expertise would be applying it to the teaching of new content, suggesting that PCK is specific
to the topic being taught. Gess-Newsome (1999a) and Park and Oliver (2007) regard PCK as
topic specific, suggesting that new PCK would need to be developed for new topics. The
current study works on this assumption. In addition, PCK is tacit and so is not readily
expressed by teachers, whether novice or experienced (Loughran et al. 2004). Consequently
it is necessary to capture and portray PCK in a way that can be accessed by others.

Empirical Studies on CK and PCK

Several empirical studies reveal useful but contradictory findings about teachers’ CK and its
relationship to PCK. A concept map study by Hoz et al. (1990) showed poor disciplinary
knowledge of biology teachers and geography teachers which did not improve with teaching
experience. These findings need to be viewed in conjunction with Angell et al. (2005) who
reported little difference in the CK of novice and expert teachers. The real differences were
noted in the types of connections and pedagogical skills used by the expert teachers. This
argument is supported by Gess-Newsome (1999b) who cites studies that show that experi-
enced teachers were more accurate in their explanations and selection of demonstrations for
teaching.
As noted in the introduction, difficulties with CK are more generally found in studies
involving teachers working outside their area of expertise (e.g., Kind and Kind 2011; Luft et al.
2013). Another group of teachers experiencing difficulties with CK are those at the elementary
school level (e.g., Appleton 2003; Nilsson and Van Driel 2008). Warren and Ogonowski
(1998) describe how a primary school teacher grappled with CK outside her area of expertise
while teaching and how the teacher’s ideas and experiences combined with scientific ideas and
practices to create learning opportunities for the children. They suggest that learning for
teaching is shaped in practice and that the knowing of the science cannot be separated from
the messy practices of which it is a part. Hence, they found that the learning of the science is
bound up with the task of preparing to teach it. Sperandeo-Mineo et al. (2006), who worked
with Italian pre-service physics teachers, endorsed the view that the value of PCK lies in its
topic specificity and showed that while a sound knowledge of content enables transformation
of CK to PCK, the process also further enhances the development of CK, indicating a two-way
process resulting in deepening of CK and enhanced PCK.
In the context of the current study, CK is found to be a challenge even among secondary
teachers working in their area of specialism, as seen in Rollnick et al. (2008) who found the
high school chemistry teachers lacked adequate CK even in topics they had taught for some
time and that they generally did not adopt a conceptual approach to teaching the content. The
problem investigated in the current study is similar in that experienced teachers lacked the
required CK but unlike in the Rollnick et al. (2008) study, they were conscious of their
knowledge gap and were anxious to address it.

Semiconductors

Although the topic of semiconductors broadly falls in the discipline of engineering, there are
basic physics and chemistry concepts underlying how they work. The concern of the
Res Sci Educ

curriculum in this study is on these concepts and only peripherally on their application. Few
empirical studies exist on the teaching and learning of semiconductor concepts. A search of the
literature revealed only two studies (García-Carmona and Criado 2009; Wettergren 2002),
though some studies were found on the learning of concepts associated with conduction (e.g.,
Guruswamy et al. 1997). Wettergren’s study of Swedish second-year engineering students
aimed at establishing student conceptions on aspects of semiconductors and found that
students’ actual understanding of these concepts bore little relationship to the scores they
obtained in examinations. The study by Garcia-Carmona and Criado was directly relevant to
the study described in this paper as it investigated the teaching of semiconductors in a
secondary school context in Spain. The research uncovered several difficulties for learners
that also emerged in this study, notably understanding the concept of a hole and appreciating
that a doped semiconductor is electrically neutral.
In summary, this review points to a need to examine how teachers learn new
content for teaching purposes. Some findings exist about the difficulties teachers
experience in this area, but there is little information about how the material is learnt.
There is also a lack of consensus on the relationship between the development of CK
and PCK as some studies reveal a two-way relationship between the two while others
suggest that teachers’ CK does not develop through teaching experience. However, it
was not clear in the latter studies whether teachers made a conscious effort to learn
the new content. In this paper, learning new content was a clear agenda for the
participating teachers. Further, there is a dearth of literature on teaching and learning
semiconductors, a topic that has become increasingly important due to its wide
application in modern computer technology.

Capturing and Portraying TSPK and PCK

As mentioned above, one of the challenges of identifying PCK is its tacit nature. In
response to this challenge, Loughran et al. (2006) and Loughran et al. (2004) devised two
complementary tools for capturing and portraying PCK called Content Representations
(CoRes) and Pedagogical and Professional-Experience Repertoires (PaP-eRs), described
as manifestations of TSPK by Gess-Newsome (2015). CoRes were originally developed
by engaging small groups of expert science teachers in activities that were designed to
help them to articulate and share with others how to teach particular science topics,
leading to the identification of Bbig ideas^ for the teaching of these topics and subse-
quently to the development of framing questions. They have since been found useful in a
number of other contexts including with novice teachers (e.g., Hume and Berry 2010).
CoRes are tabulated summaries that focus on teachers’ understandings of those aspects
that represent and shape the content and contribute to TSPK, while PaP-eRs are narrative
accounts of practice designed to bring to life the ideas in the CoRe. An example of a CoRe
can be seen in Table 4 later in this paper. PaP-eRs are instances of teaching that illustrate
the ideas of the CoRe, framed in the form of narratives (Loughran et al. 2004). The uptake
of PaP-eRs in the literature has been far lower than that of CoRes, possibly due to the bulk
of text involved or their less accessible nature. In this paper, PaP-eRs are used as a data
reduction method, a novel application of this artefact.
More recently, CoRes and to a lesser extent, PaP-eRs, have been used in different ways.
Hume and Berry (2010) found them to have important potential for the development of student
teachers’ TSPK. Rollnick et al. (2008) and Davidowitz and Rollnick (2011) used them as data
 Res Sci Educ

analysis tools to capture and portray PCK in practice. Bertram and Loughran (2011) intro-
duced practicing teachers with varying experience to CoRes and PaP-eRs. They found the
process helped the participating teachers to develop an awareness and understanding of TSPK.
In this paper CoRes and PaP-eRs are used as a means of articulating and sharing experienced
teachers’ understanding of CK (in particular conceptual knowledge and subject matter struc-
ture) as well as their TSPK as they learnt to teach the new topic.

The Context of This Study

This study was located in South Africa, a country emerging from a past where many were
denied a quality education. More than 50 % of teachers hold 3-year diplomas from poorly
resourced teacher training institutions (closed down in the 1990s) where they are likely to have
been exposed primarily to transmission teaching with little or no practical work (CDE 2007;
Mda and Erasmus 2008). Teachers entered these programmes with a grade 12 qualification,
generally with poor passes in science, and the level of disciplinary science knowledge reached
in these institutions was roughly equivalent to that of a first year college student in the USA.
Five of the seven teachers in this study came from colleges such as these although all had since
upgraded their qualifications to degree level (though generally majoring in education, rather
than content studies).
The participants in this study were teachers who were part-time students. They made
semiconductors the focus of their research project, an important component of their
programme of study. The research projects were carried out in South Africa during the
implementation period of a new high school physical science curriculum. This new
curriculum was welcomed by many stakeholders as the first major change in the high
school curriculum for over 30 years, but was received with mixed feelings. On the one
hand, the new curriculum represented a fresh start from a rote - based curriculum that
had been in place for many years, but on the other hand it presented a great challenge to
teachers. The new curriculum called for the mastery of new content as well as ensuring
that three learning outcomes — related to process, CK and science and society/nature of
science — are taken in account (Nakedi et al. 2012).
Table 1 presents profiles of the seven teachers. All were teaching in the Gauteng
province in South Africa, an urban province which is a hub of industrial infrastructure in
South Africa and includes the cities of Johannesburg and Pretoria. All, apart from Nkosi,
can be said to be experienced as they have more than 5 years’ teaching experience
(Berliner 2001).
General interest articles on semiconductors and teaching materials available on the
internet (e.g., Jenkins 2005) were found useful for the teachers in their project.
BSemiconductors^ was taught in Grades 10 and 11 using a spiral approach. In grade
10, semiconductors were introduced as materials whose conductivity rises with increased
temperature, whereas in grade 11 the topic was explored in more detail, introducing band
theory, doping and diodes. Teaching semiconductors at school level was new to all
members in the group, including the supervisors. The relevant CK on which the CoRe
was based (see below) was drawn from several curriculum documents (Department of
Education 2003, 2006) and sections in school text books written by commercial pub-
lishers. A summarized statement of the curriculum content as extracted from the official
documents is given in Table 2 below.
Table 1 Profiles of the teachers
Res Sci Educ

Teacher pseudonym Ethnicity and gender Years teaching experience Qualifications Type of school

Annemarie White female 15 BSc Honours degree (chemistry) Former white only school, with a
Post graduate teaching certificate majority of African students
from a historically white university
Bongani African male 9 Secondary Teacher Diploma African township school
2 year part time adult education diploma
from a distance learning institution
Gugu African female 9 Secondary Teacher Diploma Technical high school in an
Advanced Certificate in Education from informal settlement
a historically white university
Gumani African male 19 Secondary Teacher Diploma High school in an informal
Advanced Certificate in Education from settlement area
a historically white university
Nkosi African male 4 BSc degree (biology) Advantaged former white school
Post graduate teaching certificate
from a historically white university
Phumzile African female 8 Secondary Teacher Diploma Former white school, with
2 year technical degree equivalent from a a majority of African students
historically white university
Tebogo African female 20 Secondary Teacher Diploma African township school
Advanced Certificate in Education from
a historically white university

All 3-year diplomas are in physical science (physics and chemistry) unless otherwise stated
Secondary Teacher Diploma 3-year teaching diploma from a former African teacher training institution, Advanced Certificate in Education 2-year part time education diploma
considered to be a degree equivalent
 Res Sci Educ

Table 2 Prescribed content for teaching semiconductors, grades 10–11

Grade Learners must be able to

10 Describe the difference between conductors and insulators in terms

of how easily electrons can move inside them
Classify substances as metals using their properties
Identify the metals, their position on the periodic table and their number
in comparison to the number of non-metals
Classify substances as non-metals using their properties
Identify the non-metals and their position on the periodic table
Describe semi-metals as having mainly non-metallic properties
Classify semi-metal by their characteristic property of increasing conductivity
with increasing temperature—two elements meet this classification
Identify the semi-metals and their position on the periodic table
11 Explain:
How energy levels of electrons in an atom combine with those of other atoms
in the formation of crystals
How the resulting energy levels are more closely spaced than those in the
individual atoms, forming energy bands; and thus the existence of energy
bands in metal crystals as the result of superposition of energy levels
Contrast the conductivity of conductors, semi-conductors and insulators
using energy band theory
Explain how insulators can break down and conduct under extreme conditions
Explain the process of doping and how it improves the conductivity of semi-conductors
Compare p and n type semi-conductors
Explain how a p–n junction works
Give everyday examples of the application of semi-conductors

Methods

Research Approach

The research was based on the project work of seven teachers registered for a part time
graduate qualification (an honours1 degree) at a historically white research university in South
Africa. Each teacher’s project was styled as a self-study (Bullough and Pinnegar 2001) carried
out on his/her own practice but lacked some of the usual components of a self-study such as a
critical friend though the weekly meetings served as a reference group. The supervisors played
multiple roles beyond that of merely supervising the study. They providing input through
teaching (see week 2 below), they were major contributors to the building of the CoRes (week
3 below), they chaired and offered input during the peer teaching sessions (weeks 5-6 below)
and as would be expected from supervisors, commented on drafts of the teachers’ writing.
Since the study deviated from the normal definition of a self-study it would be more correct to
describe it as a self-reflection after Bengtsson (1995) who asserts that teaching action as a kind
of self-research can be thought of as reflection. At the time of writing he noted that reflection
had become an integral part of teacher education closely linked to teacher competence. He
regarded reflection as empowering, giving teachers control over their professional activities.

1
In South Africa, students do a 1-year equivalent honours’ degree before embarking on masters’ degrees
Res Sci Educ

Loughran (2002) asserted that this level of control can only be achieved through reflection
leading to the ability to see one’s practice through others’ eyes. This reflection is a powerful
tool for teacher development, closely linked to reasoning about teaching and PCK (Shulman
1987). Zeichner and Liston (2013) consider the self as one of the critical elements, together
with students and contexts, that require future reflection. These elements link closely with the
amplifiers and filters (see Fig. 1) connecting knowledge and action especially in the context of
this study where there is such diversity in the schools and the teachers.
Thus, the study for this paper was based on analysis of the seven projects and the data
gathered during the period of the supervision. Hence the research described here was consid-
ered a collective case study of seven teachers’ work (Stake 1995). Stake regards a collective
case study as one where a case is being studied to understand the way a teacher (in this
instance) works with something as well as the case itself. Collective case studies involve
interpretive methods and may allow limited generalisation. With studies of this nature there is a
risk of reduction of objectivity due to the overlap between teaching and research methods as
the two researchers (who were lecturers) also co-supervised the projects. This risk was
minimised through the use of frequent peer validation of data sources as well as triangulation
of data from different sources. Triangulation was observed in some cases, for example,
teachers’ concept maps closely reflected their teaching strategy as was observed in the video
recordings of their lessons. An example of validation of data sources was in the coding of the
concept maps which was carried out by colleagues not connected to the project (See Rollnick
et al. 2013).
In this paper, an overview is provided of the seven teachers while selected teachers
are studied in more detail. The selected teachers were chosen because of the richness
of data available and the variability in insights obtained in order to answer the
research questions.

Data Collection Process

The projects were supervised by two lecturers, a physics and a chemistry educator (referred to
henceforth as supervisors), as the topic of semiconductors straddles the two disciplines.
Although the seven teachers participated in this study in their role as students, they will be
referred to as teachers throughout this paper. The teachers were part of a larger group who
chose this project from a number of other projects on offer because they wanted to equip
themselves by learning to teach a new topic introduced in the secondary school curriculum.
The two supervisors assisted the teachers with the acquisition of new knowledge on semicon-
ductors as part of their supervisory guidance. The teachers and their supervisors held 3-h
weekly group meetings for a period of 12 weeks. During this time, various activities took
place. Key stages are outlined below with data sources in italics:

Week 1: All teachers were provided with a project outline sheet which explained the
objectives of the project and a brief description of the proposed methodology as well as
key readings related to key aspects of the project.
Week 2: The two supervisors gave interactive presentations to the teachers on the topic,
from a physics and chemistry point of view. Teachers were reminded how to construct
concept maps. The teachers drew their first Concept Map after this session.
Week 3: Teachers and supervisors co-constructed a CoRe (Loughran et al. 2004) which
served as an initial portrayal of PCK for the whole group.
 Res Sci Educ

Week 4: Teachers prepared two lessons on semiconductors. The second lesson was
designed to be a discussion of a task that had been reviewed by the teacher.
Weeks 5–6: A key portion of one of the lessons to be taught was peer taught and critiqued.
Peer lessons were audio recorded. The teachers drew Concept Map 2 as a take home exercise.
Week 7: Lessons were revised after peer teaching.
Four teachers (Annemarie, Gumani, Phumzile and Tebogo) and the two supervisors
attended a course on the teaching of semiconductors offered by a nongovernmental
organization.
Weeks 8–9: The teachers drew Concept Map 3 just before teaching.
Each teacher taught and video recorded the two lessons on the topic to their students.
Weeks 10–15: (Write up period). Teachers submitted drafts of their write-up for comment
and feedback. The supervisors also interviewed four of the teachers about how they felt
about taking part in this study.
Week 16: The teachers submitted a written report of their project for assessment. The
appendix material of the report included lesson plans, worksheets, examples of student
answers, and journal notes which were kept throughout the process.
The seven teachers followed the normal ethics process in obtaining approval for their
own projects (e.g., consent from their students to video record their lessons). In addition, the
author obtained permission from the seven teachers to use their material (video recordings
and project report material) for the overall study. The data generated during the above
activities were utilised in relation to the different research questions as outlined below.

Data Sources and Processing

The principal source of data for tracking the development of teachers’ CK (research question
1) was the concept maps, though sources such as journal notes, interviews and project reports
were also valuable. For research questions 2 and 3 regarding the development of TSPK and
PCK in practice, data sources were the video recordings of teachers’ lessons, their lesson plans,
interviews, teachers’ journals, audio recordings of peer teaching and their project reports. The
CoRe designed in the data collection process (Week 3) served as a portrayal of the TSPK for
the entire group and was used as a reference point throughout for both their initial TSPK. For
example, the CoRe was a starting point for the development of the teachers’ lessons. More
information is given on the concept maps and their coding below.

Concept Maps

As the schedule above shows, three concept maps were drawn and submitted by teachers in
weeks 2, 6, and 8 primarily to track development of CK. The teachers were provided with
literature about how to draw concept maps and given instruction in drawing the maps though
most of them were familiar with the technique as they had previously been exposed to the
drawing of both flow diagrams and concept maps. However, they were reminded of the
essential features of concept maps with emphasis on the point that the primary purpose was
a graphical representation of growth of CK. Nevertheless, some of the maps were less than
complete as concept maps in the orthodox sense since they lacked words on the links. Some
took the form of satellites a round the central word Bsemiconductor^ and others resembled net
diagrams without explicit linking words. It would have been ideal to have collected a fourth set
Res Sci Educ

of maps after they had taught the topic but this was not logistically possible due to the tight
deadlines for submission of their project although one teacher (Annemarie) did include a
fourth map in her report. It should be noted that the construction of the concept maps served as
a data source for both the teachers’ self-reflection and this study. This dual purpose meant that
some teachers also used the construction process as a learning tool to master the content.
In the end, 21 concept maps were drawn, three from each of five teachers, four from one
teacher (Annemarie) and two from Nkosi who did not submit his final map. The maps varied
widely in size and substance, from three connected boxes to extensive maps covering two
sheets of A3 paper. The concept maps were studied as complete maps and the more complex
maps were divided into portions that appeared to constitute units of meaning, so that smaller
maps and coherent units of meaning together could be treated as a set of maps for analysis
(Rollnick et al. 2013).
Ultimately, 43 maps and map portions from the 21 maps were individually coded by three
researchers, two of whom were supervisors. The coding was then cross checked for consis-
tency. Where two or all of the three coders disagreed on the coding, consensus was reached
through discussion. During the analysis process various codes emerged which, when clustered,
reflected one of three conceptual orientations that together with a Bno meaning^ category
provided the following four characteristics (A-D):

& A: No meaning—a map with no relevant structuring or explanatory value (e.g., Fig. 2)
& B: Focus on CK—aspects of semiconductors that showed an orientation towards personal
meaning-making, related to CK of semiconductors and linked in meaningful ways (e.g.,
Fig. 3)
& C: TSPK context—aspects that showed an orientation towards mapping the context for
teaching semiconductors, such as necessary background knowledge and particular didactic
aspects of the field (see Fig. 4, where the map contrasts the observable with models/theories
and mirrors the salient aspects of matter, but there is no suggestion of a teaching strategy)
& D: TSPK strategy—explicit aspects of semiconductors transformed into TSPK, including links
that explicitly link them in particular ways and reflect a viable teaching strategy (see Fig. 5,
where a specific teaching order of the concepts is shown with the relevant school grades)

Fig 2 Concept map coded A (Bongani map 1)

 Res Sci Educ

Fig. 3 Concept map coded B (Tebogo map 3)

The codes show that although the concept maps were originally intended to elicit CK, they
also provided rich data about TSPK. Since CK is a necessary precursor for TSPK, the codes C

Fig. 4 Concept map coded C (Annemarie’s third map)

Res Sci Educ

Fig. 5 Concept map coded D (Gumani map 3)

and D imply the presence of CK (Rollnick et al. 2013). Teachers’ progress in development of
CK could thus be tracked both by the codes on the maps and the extensiveness of the map.

Capturing and Portraying the Initial TSPK of the Group

The CoRe was initially drawn up as a communal process in one of the sessions (week 3) with
input from both students and supervisors. The students then added to the initial CoRe as the
study and their own understanding developed, but the communal CoRe remained the dominant
artefact.
The construction of the CoRe was initially used as a developmental tool to start the process
of development of TSPK. The first challenge was identifying the big ideas for teaching. This is
a necessary first step in the construction of the CoRe and requires a sound understanding of the
content and its demands. The teachers and supervisors worked in a single group with the
teachers to extract the Big Ideas. As this was at an early stage of the research, the teachers
struggled to identify the central ideas and tended to contribute topic headings. The supervisors
guided their thinking by drawing their attention to the curriculum documents but this was an
exploratory process for both supervisors and teachers as neither had previously taught
semiconductors. In the end three ideas were agreed on, with considerable input from the
supervisors. Identifying the big ideas was considered an essential step in the process and they
felt they could not proceed with ideas that did not provide conceptual clarity. Each big idea
was then interrogated though eight prompts (Loughran et al. 2006) which are shown in Table 3
alongside their mapping on to professional knowledge bases and TSPK categories of the PCK
model from Fig. 1.
Table 4 shows the initial CoRe developed by the group. The teaching and assessment
procedures which normally constitute prompts 7 and 8 of the CoRe are not shown here as
 Res Sci Educ

Table 3 Relationship between CoRe prompts and TSPK categories in Fig. 1

CoRe prompt Relevant professional knowledge bases

and TSPK categories

1.What do you intend students to learn about this idea? CK

2.Why is it important for students to know this? Curricular Knowledge, Knowledge of students
3.What else you might know about this idea CK, Assessment Knowledge,
(that you don’t intend students to know yet) Knowledge of context operating as a filter,
Knowledge of students
4.Difficulties/limitations connected with Knowledge of students, CK
teaching this idea
5.Knowledge about students’ thinking that Knowledge of students
influences your teaching of this idea
6.Other factors that influence your teaching Pedagogical Knowledge, Knowledge of students,
of this idea Knowledge of context operating as a filter
7.Teaching procedures and Knowledge of Instructional strategies,
8.Specific ways of ascertaining understanding Content representations, Assessment
knowledge, Curricular knowledge

those implemented varied according to each teacher’s context and were articulated later by the
individual teachers as they developed their own teaching strategies.
The articulation of the ideas in Table 4 served as an important starting point for the
preparation of the lessons on semiconductors and the group revisited the CoRe several times
in the weekly meetings as the teachers’ ideas on teaching semiconductors developed. At the
initial stages, not much was known about school students’ thinking about semiconductors
because they had never taught it before and also because of the paucity of literature in the area.
Hence, most of the entries for prompt 4 were added later by the group as the study proceeded.
The CoRe as shown in Table 4 was a group product and was reproduced in all their project
reports as the CoRe.

Collection of Data Related to TSPK and PCK

Teachers made video recordings of the two lessons they taught. Six of the teachers managed to
find an assistant to capture the lesson for them. The seventh teacher did not have access to a
camera and her lessons were captured by one of the supervisors who visited her school by
arrangement. The lessons were captured either on DVD or MPEG and stored on disks. In their
projects, the teachers carried out their own analysis of their lessons. This was captured in their
reports, one of the data sources in this study. The video recordings were further analysed by the
researchers for this study. They were not transcribed in their entirety, but relevant portions were
selected for transcription. The selection of portions was the result of multiple viewings of the
recordings by the researchers first separately and then together. Key episodes were identified
and coded looking for application of the knowledge categories from the upper part of the
model as shown in Fig. 1 (e.g., content representations, instructional strategies) enacted in
practice. For example, if a teacher was making use of a representation drawn on the chalk-
board, this was considered important evidence of enactment in practice and the relevant
portion was transcribed, while periods of individual work where not much could be gleaned
from the video were not transcribed. The knowledge categories as observed in practice were
Table 4 Final CoRe generated by the group

Prompts/big ideas 1. Materials exhibit a spectrum of 2. Valence electrons as vehicles 3. Conductivity is enhanced by
electrical conductivity (Grade 10) for conduction (Grade 11) introduction of impurities (Grade 11)
Res Sci Educ

1. What do you intend students Different materials conduct differently Energy is needed to excite electrons Doping is the introduction of impurities
to learn about this idea? Materials exhibit conductivity ranges to a higher energy level in materials
Differentiate between conductors, Valence electrons are the vehicle of conduction Know how doping improves conductivity
semiconductors and insulators Band theory is a useful way to understand of materials
Test materials for conductivity conductivity in solids Examples of materials that can be used
Give examples & uses of If there are electrons in the conduction band, in semiconductors
conductors etc. the material will conduct electricity Joining an n type to a p type material
Relate conductivity of elements produces a diode
to position on periodic table Explain the properties of p type and
Metalloids are elemental semiconductors n type semiconductors
Temperature dependence of conductivity How a diode works: Mechanism of
conduction in a diode, reverse
and forward bias
Uses of diodes
Insulator Bbreakdown^
2. Why is it important for students To prepare for study of semi To understand why semiconductors conduct Has important implications for new
to know this? conductors in grade 11 electricity at higher temperatures technologies such as cell phones,
To use materials appropriately To connect the world of ideas to the world LEDs etc.
To help classify materials of observables
To revise learners’ dichotomous view To understand diodes and doping later
of the conductivity of matter
3. What else you might know about Band theory, diodes Effect of doping Transistors
this idea Other electronics devices Diodes Zener diodes?
Measurement of conductivity MO theory Fermi level
Rectifier
4. Difficulties/Limitations connected Seeing conduction as a continuum Distinguish between valence band Diode symbol suggests conventional
to teaching this idea Range of magnitude of conductivity and valence electrons current direction
Distinguishing between where electrons Break down voltage
are and their energies Difficulties in conceptualising
Conceptualising the band gap movement of holes
Excitation of electrons
Table 4 (continued)

Prompts/big ideas 1. Materials exhibit a spectrum of 2. Valence electrons as vehicles 3. Conductivity is enhanced by
electrical conductivity (Grade 10) for conduction (Grade 11) introduction of impurities (Grade 11)

Difficulties in conceptualising movement of holes

5. Knowledge about students thinking Abstract concept: Charges cannot be seen Know that electrons occupy quantified Students’ interest in applications
that influences your teaching Refer to grade 8 knowledge of energy levels
conductors and insulators Electron configurations known from Grade 10
6. Other factors influencing teaching The text books are not satisfactory – look for other resources
Interplay between theory and technology
Res Sci Educ
Res Sci Educ

coded by the researchers and analysed for their connection to teacher knowledge bases and
TSPK. Where differences were found, these were resolved by discussion. The same process
was applied to the peer teaching audio recordings.
Interviews were carried out with four students during the write up stage of the project at the
time when students came for consultations. Bongani, Gugu and Nkosi were not interviewed as
they did not engage substantially with the supervisors during the write up process and were thus
difficult to contact. The purpose of the interview was to gain further insight to allow connections
to be made with the rest of the data. The questions were of a general nature and probed the
teachers’ perceptions of how their teaching had changed, what they felt about the idea of self-
reflection, as a way of learning content, TSPK and PCK, and what they perceived as high points in
the process. Each interview was audio recorded and transcribed, and analysed using open coding
(Strauss and Corbin 1990) seeking supporting evidence to triangulate with the other data sources.
The teachers’ project reports were an invaluable source of data as they contained reflec-
tions, their analysis of the lessons and students’ work as well as journal notes, lesson plans,
student worksheets and samples of corrected student work. The text of the reports was
carefully read for perceptions and insights that triangulated with other data. Manifestations
of TSPK and reflections of PCK in practice were also identified and coded on the reports.
To answer research question 2, evidence of TSPK was drawn from lesson plans, student
worksheets and samples of corrected student work supported by the interviews where available
and evidence of PCK in practice was drawn primarily from video recordings of lessons. The
evidence identified in the various data sources was used to identify key points that could be
identified in at least two, but preferably three data sources where transformation of CK was
evident. For this paper, two key episodes were translated into PaP-eRs to show the enactment
of the PCK in the form of narratives.
To seek evidence for the growth of PCK in practice and its relationship to CK (research
question 3), PCK and CK were evaluated at two key stages in the process. The first, termed
early PCK, was at the time of the peer teaching sessions; a second point, termed final PCK,
was at the time of the teaching of the lessons. These were convenient points as data were
available in the form of concept maps, recordings of peer teaching sessions and actual lessons,
as well as lesson plans. The various manifestations of PCK were coded using a rubric designed
by Park et al. ( 2011). This rubric has been validated with a larger sample of teachers and
analyses teacher planning, implementation and reflection and classifies teachers’ PCK as
limited, basic, proficient or exemplary. The criteria are based on requirements for teaching
which are rooted in constructivism and focus on elements such as use of learner prior
knowledge in teaching, probing student understanding and appropriate use of representations,
all of which have a strong agreement with the components used in the model. An extract from
the rubric on implementation is shown in Table 5.
The rubric focuses on implementation of teaching strategies such as questioning to probe
student understanding, recognition of student ideas and difficulties, and how teaching strate-
gies are linked to student need and understanding.
Since the descriptors in the rubric were rather generic, a common understanding
had to be reached through peer validation. The data were coded by the author and
data for two of the teachers were independently coded by a second researcher (not a
supervisor) familiar with the use of the rubric. Agreement was obtained in both cases
on more than 80 % of the classifications. Finally, to determine the link to CK, the
PCK categories obtained for each teacher were juxtaposed with the categories obtain-
ed from the concept maps (see findings in Table 7 below).
Table 5 Extract from Park et al.’s (2011) rubric on implementation

Element Limited Basic Proficient Exemplary

Questioning to probe No questions to probe Few questions to probe Some questions to probe Many questions to probe
student understanding student understanding student understanding student understanding student understanding
Spontaneity to challenge No recognition and/or attempt Attempts to challenge Some attempts to challenge Many attempts to challenge
misconceptions or resolve to challenge student misconceptions student misconceptions student misconceptions or student misconceptions or
learning difficulties or resolve learning difficulties or resolve learning difficulties resolve learning difficulties resolve learning difficulties
encountered discovered during instruction discovered during instruction discovered during instruction discovered during instruction
Rationale for instructional No rationale for instructional Weak rationale for instructional Adequate rationale for instructional Strong rationale for instructional
strategies and strategies and representation strategies and representation strategies and representation strategies and representation
representation in in connection with student in connection with student in connection with student in connection with student
connection with student understanding understanding understanding understanding
understanding
Res Sci Educ
Res Sci Educ

The findings begin below by addressing the first research question, related to the develop-
ment of CK.

Findings

The cases of four teachers are shown in more detail below, followed by an overview of the
progress of the whole group. The selected teachers are Gugu, Gumani, Annemarie and
Phumzile. These teachers were chosen because of the richness of data available for these
teachers. As mentioned above, two key points in the process were used to evaluate the
development of PCK, the first at the peer teaching stage and the second at the time of teaching
the lessons to the class.
Two of the four cases, Gugu and Gumani, have been constructed as PaP-eRs (Loughran
et al. 2006) below. The PaP-eRs illustrate the teaching of two key concepts, identified in the
CoRe as difficult for school students (see Table 4, prompt 4)—the idea of a spectrum of
conductivity which relates to big idea 1 and conceptualising the movement of holes (García-
Carmona and Criado 2009) which relates to big idea 3. Annemarie and Phumzile’s lessons
both involve teaching grade 11 classes in former white schools, now populated with a majority
of African students bussed in from African townships (see Table 1). Like Gugu, their lessons
focussed on big ideas 2 and 3.

Gugu

Gugu’s initial understanding of semiconductors (CK) was less than rudimentary and she was
able to show very little benefit from the initial presentation in week 2 as shown by her first map
in Fig. 6. Figure 6 also shows that there was not much progress between maps 1 and 2.
However, Gugu’s progress was remarkable between maps 2 and 3. At the beginning of the
process, she was withdrawn and lacking in confidence. In her journal, she recorded
I was very confused about everything they said that day [the first session] because I was
totally blank, because everything said was new to me....... it was so difficult to grasp
everything said, I felt so stupid. (Gugu, journal)

She noted in her project report that 3 weeks later, after submitting her second concept map,
she discovered that the technical subjects taught in her school included electricity. On speaking
to the teacher concerned, she found that he dealt with (and understood) many of the necessary
concepts. Gugu’s final map, drawn shortly before she taught, shows a radical change. Her map
now has a defined structure and integrated elements of pedagogy—which are confirmed by
analysis of her lessons, described below. The progression of Gugu’s three concept maps is
shown in Fig. 6. The increased CK was reflected in her PCK. The teaching strategy she used
reflected her final map of which only a portion is shown in Fig. 6. The PaP-eR below provides
a narrative of the strategy, referred to here as Bthe np Strategy^.

PaP-eR 1: Explaining the Movement of Holes—Bthe np Strategy^ (Gugu)

A common strategy used by the teachers teaching grade 11, also reflected by the flow of ideas
in more than half their concept maps, is exemplified in Gugu’s last map (Fig. 6). In her map,
 Res Sci Educ

Fig. 6 Progression of Gugu’s concept maps

Gugu separates the ideas of n and p doping as parallel concepts, one involving an additional
electron and the other involving a deficient electron in the dopant. This separation in the map
matches with the teaching strategy. In her lesson, Gugu explained the two types of doping (n-
type and p-type) used in semiconductors, closely following her concept map in Fig. 6. The
students were given a group exercise to draw Lewis diagrams showing the electron configu-
ration of atoms, first in an n-type conductor, given feedback and then asked to do the same for
a p-type semiconductor. Based on the feedback given in the first exercise, they were able to
complete the task for the more conceptually difficult p-type case successfully. Gugu ended the
sequence with the powerful explanation below, explaining how conduction occurs in the p-
type case.
So look at this. … There’s a space. You see that there is an empty space here. There is an
empty space, because we have doped this material with an atom which has only 3
electrons. Then it leaves an empty space — which is called a hole. This empty space
here we call it a hole (points to the diagram on the board) and... what happens? Do you
see we have other neighbouring silicon atoms? And what happens here is that these
neighbouring silicon atoms which are bonded here — they can move because of the
empty space here hey? They can jump from here and fill up this hole and when they
have left they leave another hole! And another one comes in and jumps into this hole and
we end up getting — and when you get this picture one might observe as if the holes are
moving. You see? Is it holes that are moving guys? (Chorus NO!!!). Electrons!!! So
these other electrons, they see a chance of moving around. They jump into this hole and
where they have left, they leave another hole. So these holes — it might seem that they
Res Sci Educ

are moving but it is not holes that are moving — they are called positive holes, hey? It is
not holes that are moving but electrons are moving. (Gugu, Lesson Transcript)

Gugu’s PaP-eR shows how she had integrated her newly acquired CK with well-developed
pedagogical practices (pedagogical knowledge in Fig. 1), which for her consisted largely of
whole class teaching, as well as her understanding of the learners and their context. What
became apparent is that she only felt comfortable with her new found knowledge once she
could conceptualise teaching it as demonstrated in her third concept map, which mirrors the
teaching strategy used (Fig. 6).
PaP-eR 1 shows a teaching strategy that was used by several teachers in the study (see both
Annemarie and Phumzile below, who both used the same strategy). It involves teaching the n
type case of doping first, where the semiconductor is doped with an element with atoms that
have an additional electron. The improved conductivity is relatively easy to explain as there is
a surplus electron. Having established understanding of this concept, the teacher then ex-
plained the more difficult p type case where the doped element has one electron less in its
atom. In this case the teacher needs to explain the movement of Bholes^ in the opposite
direction, a concept that students find more difficult to grasp. In most cases teachers found this
strategy to be successful as they were able to use the learning from the n type case to scaffold
the explanation of the p type case. This thinking was also visible in several of their concept
maps.

Gumani

Gumani began the study with reasonably good CK. His first map contained a logically ordered
hierarchical structure of concepts which could be divided into 3 portions and was coded D
according to the scheme developed above. His second map also consisted of 3 portions, using
his understanding from map 1 put into a teaching context. This map was coded B,C, and D.
Gumani’s last map (Fig. 5, coded D—TSPK strategy was similar to map 2, also containing
3 portions but explicitly linked to curriculum and teaching where the same concept is explicitly
developed for teaching purposes, shown in the central portion of the map. This can be seen by
following the links from Bdoping^ to Bpure crystal^). PaPeR 2 below highlights his teaching of
big idea 1 in the CoRe (Table 4) where he addresses students’ conception of conductivity as a
dichotomy.

PaP-eR 2: A Spectrum of Conductivity (Gumani)

Many science concepts are introduced in the elementary and middle school as dichotomies
with no overlap between the categories. This has been recognised by Taber (2001) as a
hindrance to later learning. For example, matter is presented as being divided into conductors
and non-conductors of electricity. The introduction of semiconductors breaks this mould and
Gumani made Bbreaking the dichotomy^ an explicit goal of his instruction. He conducted a
practical exercise with his grade 10 students, giving them five carefully chosen elements,
asking them to comment on their lustre, ductility, malleability and electrical conductivity by
observing and testing them. He selected three metals, zinc, copper and iron; a non- metal,
suphur; and the anomalous element carbon in the form of graphite, which he knew the students
would recognise as a non-metal yet it would conduct electricity. He was limited in the range of
materials available to him as his school, located in an impoverished area, is poorly equipped.
 Res Sci Educ

To facilitate group work, he had to borrow multiple sets of conductivity testing equipment
from a school in a nearby town.
In his discussion with the class after the activity, Gumani established with them the
anomalous property of conduction exhibited by carbon and generated a discussion on whether
carbon should be considered a metal or a non-metal. To conclude, he showed the class an
overhead transparency containing a non-linear graph taken from Jenkins (2005) to convey the
wide range of conductivity exhibited by different materials. Gumani designed a written
exercise containing a series of questions to assess student understanding. One of these
questions is considered here.
The question entailed interpreting the graph shown in Fig. 7. He asked the students to
identify with explanation which line on the graph (A, B or C) represents an insulator, a
semiconductor and a conductor. His intention was to test their understanding that the conduc-
tivity of semiconductors increases with temperature and that of conductors decreases with
increased temperature.
The students’ responses to this question showed Gumani that he had overestimated their
ability to interpret graphs. He deduced that they were relating the lines to only one of the axes.
One student stated that there is no conductivity while another stated that conductivity increases
with increasing temperature. The graphical form of the question clearly presented difficulty for
the students as most were able to answer successfully a second direct question testing similar
understanding.
Although some students seemed to understand the idea of continuity of electrical conduc-
tivity, Gumani was disappointed with his ability to teach this concept and he reflected:
Learners find it difficult to accept that there are semi-metals in between metals and non-
metals as well as semiconductors between conductors and insulators. These dichoto-
mous classifications make it difficult for learners to further understand the continuum
concept. (Gumani, Project report)
PaP-eR 2 exhibits the extent of transformation of Gumani’s knowledge for teaching. He
produced a representation that showed relative conductivity of insulators, semiconductors and
conductors in relation to temperature to enable him to elicit students’ understandings of these
concepts. He realised that the students held a dichotomous view of conductivity (knowledge of
students) and attempted to address their conception through the use of the graphical represen-
tation (content representations in Fig. 1). In his selection of the representation on the overhead

Fig. 7 Gumani’s exercise for

students
Res Sci Educ

transparency, he primarily made use of his CK. He wished to convince the class that
conductivity was continuous rather than dichotomous. To test the effectiveness of his expla-
nation, he made use of the graphical assessment task in Fig. 7. His understanding of students’
prior knowledge turned out to be lacking as he discovered that students had difficulty
interpreting graphs. In his project report he reflected:
I still have to come with a better way of explaining the conductivity spectrum as this
concept forms the crux of electrical conductivity. (Gumani, Project report)
Two important findings emerge through this PaP-eR and Gumani’s subsequent reflection.
Firstly he realises that his strategy has not been successful and secondly the reflection reveals
his strategic thinking about the centrality of Big Idea 1, an important signifier of TSPK being
informed by PCK, showing the recursive nature of the process of knowledge in relation to
practice. He was aware of this close link between CK and PCK as the following interview
excerpts show:

I think I’ve referred to that in my, my project that initially when I started with concept
map — it was basically accumulating or acquiring content knowledge but eh, uncon-
sciously what I picked up is that, along the way, actually I was coming up with the
teaching strategies. (Gumani, Interview)

Annemarie

Annemarie began the project conscious of gaps in her CK and unhappy about her earlier
attempts to teach the topic, as the students had not performed well in assessments. Compared
to the rest of the group, she had a good comprehension of semiconductors as she was able to
identify in her journal notes specific areas of content for which she lacked understanding. As
shown in Table 6, her first map contained a single portion which was coded as showing a
teaching context (C). Essentially the map contained an arrangement of topics to be taught with
the words Bmatter and materials^ at the centre and the words Bconductors, insulators and
semiconductors^ emerging from this. Only the semiconductor concept was broken down into
three sub-concepts. The C coding is confirmed by her own evaluation of the map in her project
report where she refers to the map, making distinctions between the concepts as outlined in the
curriculum documents rather than organization of the content per se (curricular saliency in
Fig. 1). Her third map (Fig. 4) with 2 portions showed conceptual thinking about relating sub-
micro to macro concepts and strong links to teaching (C). Her final understanding of content as
shown by her last concept map (map 4) showed ability to transform her knowledge for
teaching. In her reflection on the maps she identified the growth of hierarchies and richer
cross links.
The rich links reflected in the later maps were not apparent during the peer
teaching exercise in week 5 when she explained n type and p type semiconductors,
beginning with the more difficult p type case, showing a lack of knowledge of student
understanding. She was able to use some representations during the peer teaching
session, but showed the doping by adding the dopant to the original material rather
than substituting it. Her journal notes record feedback on the session primarily related
to content issues though she does note suggested improvements to her representations
and explanations, for example:
 Res Sci Educ

emphasis must be made of the following ideas: (a) the uncertainty when it comes to
classifying an element as a conductor and semiconductor (b) Interpreting reality in terms
of the world of ideas and the world of observables. (Annemarie, Journal notes)
Annemarie’s PCK was classified as basic using Park et al.’s rubric (Table 5), primarily due to her
limited ability to consider students’ prior ideas and her weak rationale for her use of representations.
A turning point for Annemarie was the workshop offered by the outside organization in
week 7. She notes that she suddenly realised the importance of semiconductors in the school
curriculum and their relevance to everyday life, and the importance of practical activities to
link the theory to the observable phenomena. Her video recorded lessons showed well planned
lessons demonstrating curricular saliency leading to well thought out strategies, including the
Bnp^ strategy demonstrated by Gugu above, this time starting with the n type semiconductor.
Lessons included use of sub-microscopic representations of doped crystals and forward and
reversed bias diodes on overhead transparencies and charts as well as a spontaneously
organised role play, designed to address student difficulties in understanding the movement
of holes in a semiconductor. In her interview she noted:

… and then from the comments that the kids make whether I’m asking the questions or
they, I can, immediately see… that there’s a problem and then from there on I adjust and
I think they saw that there was some progress doing it that way. (Annemarie, Interview)
Annemarie’s final PCK was classified as exemplary mainly because of the evidence
of her ability to identify and respond to student learning difficulties and probe student
understanding.

Phumzile

Unlike Annemarie, Phumzile had never taught semiconductors before and her first
concern was the limitation in her CK. Her first concept map had a single portion
linking 5 ideas to the word Bsemiconductors^ with no linking words. The words were
sensibly linked and the map was classified as showing understanding of content but she
did not feel confident. In her journal notes she described the map as a disaster and
described the five words she had put on paper as Bwords that I just heard during the
same session^. Her next concept map had grown in both scope and depth of under-
standing both for personal meaning and for teaching (B and D). She confessed during
her interview that she had spent an entire day devising the second map, which was
produced at the time of the peer teaching session. Her final map was a 4 portion map
covering two A4 pages, two portions showing direct connections to teaching, one
showing understanding and one context.
For her peer teaching session in week 5, Phumzile designed an expository lesson lacking in
conceptual learning. Her PCK at this stage was classified as basic as she had narrow
understanding of student learning difficulties and thus could not accommodate these into
teaching strategies. Two key learning points for Phumzile were the workshop referred to above
and the time she put into reconstructing the original CoRe (she was the only teacher who gave
evidence of this). The video recording showed that in her lesson she was able to make links to
students’ previous knowledge and provide well thought out explanations (e.g., Ba hole acts/
behaves like positive charge^), illustrated with both appropriate pre-prepared and spontane-
ously drawn representations on chalkboard. Like Annemarie, she made use of a role play to
Res Sci Educ

illustrate movement of holes and designed a practical activity with written questions.
In her lesson she built on student ideas and also employed the np strategy. Despite
this, she was critical of the lesson she delivered, identifying shortcomings in her CK
and in her ability to anticipate student learning difficulties. She was classified as
proficient according to the rubric.

Development of the Group Overall

CK

In most cases, the concept maps grew both in scope (as seen by the number of
portions) and depth of understanding, showing growth in depth and breadth of CK.
What the emergent codes show is the presence of elements of pedagogy, showing the
creation of Bschool knowledge^ or Bspecialised CK^ (Ball et al. 2008). A qualitative
summary of the maps drawn at different stages of the research is shown in Table 6,
together with the applicable codes (see data collection section above) and number of
portions per map:
Table 6 shows progress in understanding achieved during the process. A fuller
analysis of the maps appears in Rollnick et al. (2013). Apart from Bongani, who was
the only teacher not to fulfil the requirements to pass the project, all the teachers at
least showed understanding of the content by the end of the process (code B, focus
on CK) and the breadth of their knowledge had also grown as shown by the increase
in concept map portions. Five teachers showed further depth in the CK, including
links in their concept maps to teaching. This was an unexpected finding that emerged
when the maps were being coded.
The most notable finding from this analysis is the difficulty in separating the CK
from the way it is taught, as evidenced by the emergent coding of the map portions.
The higher coding categories reflect the emergence of knowledge for teaching, rather
than simply understanding of the content itself, though portions of the later maps also
show CK alongside content knowledge for teaching, showing deeper and more
flexible CK.

Relationship of CK to PCK

The relationship between the teachers’ CK and PCK early and near the end of the process is
shown in Table 7. The early period was determined as the stage of peer teaching of lessons
(Week 5) as this was the first opportunity to gauge teachers’ early PCK and the final period
was determined as the stage when the lessons were taught (weeks 9-10). The CK level was
determined by the highest coded portion of the concept map at that stage as well as the extent
of the maps as indicated by the number of portions. Teachers whose concept maps incorpo-
rated ideas about teaching the concept were considered to have high quality CK as this would
be a prerequisite to transform it for teaching.
Table 7 shows the close relationship between the two constructs. Teachers whose
understanding of CK is classified as for personal meaning making only (code B) are
generally classified as basic or below with respect to PCK. To achieve levels of
PCK above basic, their thinking about CK needs to show some awareness of the
teaching context.
Table 6 Qualitative summary of maps drawn at different stages of the study

Teacher Map 1 Map 2 Map 3

Annemarie Single portion map showing application Map with 2 portions showing strong Maps 3 and 4 (2 portions each) showed
to teaching context (C) links to teaching (C) increasing emphasis on teaching with
understanding. Map 3 (C) and Map 4 (C and D)
Bongani Maps grew in numbers of nodes and links
with no linking terms but all three
were single portion maps consisting
of unrelated parts, with no explanatory
value (A).
Gugu Both had single portions with unrelated 3 portion map consisting of sections showing
parts as for Bongani (A) understanding and others showing explicit
teaching orientations (D).
Gumani Hierarchical structure of concepts, Understanding from map 1put into a teaching Similar to map 2, also 3 portions but explicitly
3 portions, logically ordered (D) context, 3 portion s (B,C,D) linked to curriculum and teaching (D)
Nkosi Single portion map showing application 2 portion map showing strong links Submitted only 2 maps
to teaching context with linking to teaching (D)
words (C)
Phumzile Single portion map linking the word 5 portion map, 4 showing development 4 portion map covering two A4 pages, two portions
semiconductors to other ideas of ideas for understanding and one showing direct connections to teaching, one showing
showing understanding (B) showing connection to teaching (B, D) understanding and one context (B,C,D)
Tebogo Single portion map showing same ideas Substantially developed 3 portion map showing
as map 1 with further branching (B) development of ideas (B)

Codes (A–D) shown in brackets (see data sources and processing section for explanation of codes)
Res Sci Educ
 Res Sci Educ

Table 7 Relationship between CK and PCK

Teacher Early CK Early PCK Final CK Final PCK

(concept map 2) (peer teaching) (Final Concept map) (Actual lesson taught and lesson plan)

Annemarie C: Focus on teaching context (2 portions) Basic D: Focus on teaching (2 portions) Exemplary
Bongani A: No understanding (1 portion) Absent A: No understanding (1 portion) Limited
Gugu A:No understanding (1 portion) Absent D: Focus on teaching (4 portions) Basic/Proficient
Phumzile B: Understanding (5 portions) Basic D: Focus on teaching (4 portions) Proficient
Tebogo B: Unders tanding (1 portion) Limited B: Understanding (3 portions) Basic
Gumani B: Understanding (3 portions) Basic D: Focus on teaching (3 portions) Exemplary
Nkosi B: Understanding (1 portion) MAP 1 Limited D: Focus on teaching (2 portions) MAP2 Basic
 Res Sci Educ

Discussion and Conclusion

The findings above are illustrative of the growth that the teachers underwent during the self-
reflection process. Six out of the seven teachers in the study can be considered to be
experienced teachers but were teaching a topic that they had never taught and had to learn
themselves for the first time. Thus, all the teachers could be considered to be novices regarding
the teaching of semiconductors.
The first research question interrogated the growth of CK during the process. This question
was answered mainly through a study of the teachers’ concept maps, although other sources
such as their journal reflections played a part. The most significant finding regarding the
development of CK emerged during the coding. Four levels of depth of engagement with
content were identified from the analysis of the concept maps while the breadth of knowledge
was seen through the number of map portions. The development of CK was found to be
closely connected to the teachers’ development of ideas on teaching the content. Many of the
concept maps framed understanding of content in a teaching context, or even more explicitly
outlined a possible teaching strategy, suggesting that the teachers’ learning of content was
closely bound up with their understanding of how to teach it. This relationship is closer than
suggested by either Sperandeo-Mineo et al. (2006) or Warren and Ogonowski (1998).
A further finding was that, as expected, all teachers’ understanding grew through the
process. Even Bongani, whose maps remained at the lowest level, added new vocabulary
though he was unable to link many of the ideas coherently. Common themes also emerged
from the maps, such as the separation of consideration of the n and p type conductors which
were converted into a teaching strategy (see below). This seemingly unconscious inclusion of
teaching ideas into the concepts maps does not seem to have been found elsewhere.
It is also noteworthy that the level of teachers’ early CK (see Table 7) showed a relationship
with their educational background. Annemarie and Nkosi, whose initial tertiary education took
place at advantaged universities rather than disadvantaged teacher training colleges, showed
both a higher initial CK and an advance to the highest CK level. It is misleading to make a
judgment on Annemarie’s initial CK as she had taught the topic before, but Nkosi did advance
to level D in his final concept map. Most of the other teachers had a second qualification from
an advantaged university and were able to gain an understanding of semiconductors reason-
ably quickly. Gugu had great initial difficulty grasping the concepts related to semiconductors
until she engaged in the task of teaching the topic.
The second research question examined manifestations of teachers’ TSPK (topic specific
professional knowledge) and PCK (topic specific PCK in practice) in the new content. This
question was addressed through a study of the teachers’ practice and their lesson plans and
supplemented by data from their reflections and correction of student work. Two PaP-eRs were
constructed to illustrate the teachers’ ability to address key concepts needing attention—
breaking the conductor/non-conductor dichotomy and a strategy to teach conduction in n
and p type conductors (referred to as the np strategy). The PaP-eRs brought to life some of the
issues developed in the CoRe which was developed as a starting point for the whole group.
One common manifestation of PCK in practice which emerged related to big idea 3, the idea
that the conductivity of semiconductor materials is enhanced by the introduction of impurities.
The CoRe (see Table 4) did not suggest any teaching strategies but one which emerged was the
Bnp strategy^—the idea that the conduction mechanism should be first explained using the n
type conductor before attempting to explain the p type conductor. This teaching sequence
emerged because teachers realised the difficulty of explaining the movement of Bholes^ in the
Res Sci Educ

p type case. If the idea of doping was first understood, the teacher could then focus on the
difficulties of explaining the movement of holes. This strategy was found in Gugu’s lesson, as
illustrated in PaP-eR 2, and was also observed in Phumzile’s and Annemarie’s lessons. It was
also visible in several of the concept maps. For all these teachers a process of identifying the
content, conceptualising the teaching and recognising the student difficulty with the topic
culminated in teaching strategies that took all of this into account. The frequency of use and
success of this strategy suggests that it is a significant addition to teachers’ knowledge bases
for teaching this concept and could form part of the canon of TSPK.
Another manifestation of PCK was Gumani’s emerging ability to use representations to
address student prior knowledge in relation to big idea 1, the idea that there is a spectrum of
conductivity. His use of a non-conventional graph attempted to confront the student conception
of separate conductor and non-conductor categories. In this case, the lesson learnt was through
difficulties encountered in the use of an assessment item used to assess understanding of this
concept (Fig. 7). Students’ difficulties in graphical interpretation impeded Gumani’s ability to
diagnose student learning difficulties. Producing the manifestations used (the conductivity
representation and assessment item) required him to draw on his newly developed CK together
with knowledge of learners for implementation in practice (see Fig. 1). His assessment item
(Fig. 7) was deliberately simplified, demonstrating his understanding of learners and context
(the school curriculum). However he was dissatisfied with the success of the strategy. It is
possible that the students’ limitations in graphical interpretation may also have hindered their
understanding of the initial graph he used in teaching. This would have impeded their
understanding of the big idea.
These two strategies have not been documented elsewhere and would be useful approaches
in teaching semiconductors. As stated above there are few studies on the teaching of semi-
conductors and the np strategy seems to be a good idea for addressing the difficulty of
understanding p-type doping identified by García-Carmona and Criado (2009). Additionally,
the idea of using PaP-eRs as a data reduction method may find applications in other studies.
The level of PCK reached by the different teachers is harder to relate to either their school
or academic background. All teachers apart from Bongani made some advances in their PCK
but the characteristics of the PCK in practice would be moderated by the amplifiers and filters
as shown in Fig. 1, related to the context within which they teach. Gugu and Gumani were
teaching students in an informal settlement area and had to take congnisance of the limited
English proficiency of their students and a lack of resources in the school. They both
developed pedagogies appropriate to their situations and were able to deliver successful
lessons. Annemarie and Phumzile were working in schools where language proficiency was
less of a problem and the schools were much better resourced. Teaching students with adequate
language proficiency was essential for Annemarie who did not have the ability to move into a
local language if necessary.
Research question 3 sought evidence for the growth of PCK and its relationship to CK of
the new topic. To answer this question and provide further answers to question 2, teachers’
PCK and CK was determined in weeks 5 and 9/10. Table 7 shows that all teachers advanced in
their PCK using Park et al.’s (2011) categories and confirms a strong alignment between CK
and PCK, especially when the CK develops with teaching in mind. The two more detailed
exemplars presented show that the teachers were able to bring in richer representations and
show greater consciousness of student prior learning, accompanied by a deeper understanding
of the content in a teaching context. As demonstrated by Annemarie, teachers were able to
make spontaneous moves in the class in response to student needs. Teachers put extraordinary
 Res Sci Educ

effort into the planning of the lessons and were able to move from expository lessons, as
demonstrated in Phumzile’s initial peer teaching, to participatory lessons showing the ability to
use student ideas in their teaching. Hence the observed PCK in practice was produced by
combining knowledge of students, knowledge of content and pedagogical knowledge. Earlier
research (Mavhunga and Rollnick 2013) does show that interventions with pre-service teachers
can improve topic specific PCK (TSPK in this study), and there are findings showing
improvement in CK and PCK using self-reflection on other topics with a lack of evidence
documenting their development (Rollnick 2014). This study is able to show development of
CK in line with advances in TSPK.
The teachers’ initial PCK (albeit communal) was articulated in the first instance through
the co-creation of the group CoRe, together with their supervisors. The first important step in
this process was the elucidation of the big ideas which, though scaffolded by the supervisors,
resulted in coherence with the progression of the content though the curriculum. The extent
to which the big ideas became part of the teachers’ PCK is illustrated through their focus on
the big ideas in their teaching. Those reaching the highest level of proficiency in their PCK
showed strong use of student ideas and thoughtful application of appropriate representations
in their teaching. However, the content of the CoRe has to remain background material for
this study as it was difficult to separate out the contributions of the supervisors and the
teachers, so while the development of the CoRe was an important step in the development of
the teachers’ PCK, it could not be used for evidence of its growth. Nevertheless, the big
ideas themselves are a useful resource for future teaching of the topic and like the np strategy
make an important contribution to the teachers’ TSPK. Another note of caution needs to be
added regarding the dual role of the supervisors in this study. Their lack of distance in the
process needs to be acknowledged. However, the lessons, project reports and concept maps
produced during the study remain the teachers’ artefacts and are evidence of their learning
during the process.
Teachers were aware of this close link between CK and PCK as the following interview
excerpt shows:
So I actually, my concept map actually informed my teaching altogether. (Annemarie,
Interview)
This study has shown that a carefully scaffolded approach to learning to teach a new topic
can result in development of both PCK and CK. There are several elements in the process
contributing to this. Firstly, the process of exposure to the material on semiconductors was
facilitated through a conceptual approach, for example, the use of concept maps for teachers to
determine the growth of their subject matter knowledge and the communal construction of a
CoRe. Such an approach to learning a topic would have been very different to the methods that
the teachers had been exposed to in their earlier education. Secondly, the approach made space
for self-reflection. Finally, the fact that this exercise was part of a small research project played
a big role in gaining the commitment of the teachers in a context where teacher morale is
generally low.
The participants in this study took part in an intervention which lasted over 4 months. It is
important to interrogate to what extent the ability to learn to teach content can be transferred to
other topics. In this regard evidence is beginning to emerge in the context of pre-service
teacher education that such transfer is possible (Mavhunga 2016). However, the relationship
between CK, TPCK and PCK remains complex despite the increased clarity provided by the
model in Fig. 1. Further work needs to be done in this regard.
Res Sci Educ

A further consideration is that this study took place in the context of a developing country
where many teachers have limited CK and resources in the classroom are limited, often
necessitating the borrowing of equipment. Students in several of the classrooms in this study
come from informal settlements and townships where they will not have had much exposure to
applications of science. However, the success of the self-reflection strategy in this environment
shows promise for assisting teachers in other countries who are teaching out of specialism, a
problem identified earlier in this paper.
The circumstances of the study were rather special. However, it does demonstrate that such
change in teacher practice is possible. The question to be asked in further research is whether
these teachers would be able to apply this approach to learning to teach other topics. Despite
the unusual conditions in which the study was conducted, the findings show that if conditions
are created where teachers are encouraged to carry out self-reflection on teaching new topics
when faced with curriculum change, careful scaffolding of the learning process can result in
conceptual learning as well as adoption of new teaching strategies. Outside the formal study
environment this could be achieved through professional learning communities (Hargreaves
2008). However, the support is managed, teacher commitment is a central element for success.
In conclusion, content and pedagogical knowledge appear to develop hand in hand as the
teachers gained confidence with their CK. Thus, professional development programs that
attempt to improve teachers’ CK alone may be simplistic as improvement in content under-
standing needs to take place alongside an understanding of how the content is to be taught.
Professional development that consists of pure presentation of content is insufficient to enable
deep learning of content. Findings emerging from the study also indicate a renewal of interest
in teaching on the part of the teachers and an increased understanding of how teachers’
learning of content is intrinsically bound up with their vision of how it should be taught.

Acknowledgments I would like to note the helpful comments provided by Prof Tony Lelliott in the improve-
ment of this paper.

References

Abell, S. A. (2007). Research on science teacher knowledge. In S. A. Abell & N. G. Lederman (Eds.), Handbook
of research in science education (pp. 1105–1150). Mahwah: Lawrence Erlbaum.
Adler, J., Pournara, C., Taylor, D., Thorne, B., & Moletsane, G. (2009). Mathematics and science teacher
education in South Africa: a review of research, policy and practice in times of change. African Journal of
Research in Mathematics, Science and Technology Education, 13, 28–46.
Angell, C., Ryder, J., & Scott, P. (2005). Becoming an expert teacher: Novice physics teachers’ development of
conceptual and pedagogical knowledge. Paper presented at the European Science Education Research
Association Conference, Barcelona, Spain. http://folk.uio.no/carla/ARS_2005.pdf
Appleton, K. (2003). How do beginning primary school teachers cope with science? Toward an understanding of
science teaching practice. Research in Science Education, 33(1), 1–25.
Ball, D. L., Thames, M. H., & Phelps, G. (2008). Content knowledge for teaching: what makes it special?
Journal of Teacher Education, 59, 389–407.
Bengtsson, J. (1995). What is reflection? On reflection in the teaching profession and teacher education. Teachers
and Teaching, 1(1), 23–32. doi:10.1080/1354060950010103.
Berliner, D. C. (2001). Learning about and learning from expert teachers. Educational Researcher, 35, 463–482.
Bertram, A., & Loughran, J. (2011). Science Teachers’ Views on CoRes and PaP-eRs as a Framework for
Articulating and Developing Pedagogical Content Knowledge. Research in Science Education, 1-21. doi:10.
1007/s11165-011-9227-4
 Res Sci Educ

Bullough, R. V., & Pinnegar, S. (2001). Guidelines for quality in autobiographical forms of self-study research.
Educational Researcher, 30, 13–21.
Carlsen, W. S. (1993). Teacher knowledge and discourse control: quantitative evidence from novice biology
teachers’ classrooms. Journal of Research in Science Teaching, 30(5), 471–481.
CDE. (2007, October 2007). Doubling for growth: Addressing the maths and science challenge in South Africa’s
schools. CDE Research no 15 Retrieved from www.cde.org.za
Cochran, K., DeRuiter, J., & King, R. (1993). Pedagogical content knowing: an integrative model for teacher
preparation. Journal of Teacher Education, 44(4), 263–272.
Cochran, K., & Jones, L. L. (1998). The subject matter knowledge of preservice science teachers. In B. J. Fraser
& K. G. Tobin (Eds.), International Handbook of Science Education (Vol. 2, pp. 707–718). Dordrecht:
Kluwer.
Davidowitz, B., & Rollnick, M. (2011). What lies at the heart of good undergraduate teaching? A case study in
organic chemistry. Chemical Education Research and Practice, 12, 355–366.
Department of Education. (2003). National Curriculum Statement – Grades 10 -12 Physical Sciences. Pretoria:
DOE, South Africa Retrieved from ISBN 1-919975-61-6.
Department of Education. (2006). National Curriculum Statement Grades 10 - 12 (General) Physical Sciences
Content. Retrieved from http://www.thutong.doe.gov.za/Default.aspx?alias=www.thutong.doe.gov.za/
physicalsciences
García-Carmona, A., & Criado, A. M. D2009]. Introduction to semiconductor physics in secondary education:
evaluation of a teaching sequence. International Journal of Science Education, 31D16], 2205–2245.
Retrieved from http://www.informaworld.com/10.1080/09500690802272074.
Geddis, A. N. D1993]. Transforming subject‐matter knowledge: the role of pedagogical content knowledge in
learning to reflect on teaching. International Journal of Science Education, 15D6], 673–683. Retrieved from
http://dx.doi.org/10.1080/0950069930150605.
Gess-Newsome, J. (1999a). Pedagogical content knowledge: an introduction and orientation. In J. Gess-
Newsome & N. Lederman (Eds.), Examining pedagogical content knowledge (pp. 3–20). Dordrecht:
Kluwer.
Gess-Newsome, J. (1999b). Secondary teachers’ knowledge and beliefs about subject matter and their impact on
instruction. In J. Gess-Newsome & N. Lederman (Eds.), Examining pedagogical content knowledge (pp.
51–94). Dordrecht: Kluwer.
Gess-Newsome, J. (2015). Teacher professional knowledge bases including PCK: results of the thinking from the
PCK summit. In A. Berry, P. Friedrichsen, & J. Loughran (Eds.), Re-examining pedagogical content
knowledge (pp. 28–42). Oxford: Routledge.
Grossman, P., Wilson, S. M., & Shulman, L. S. (1989). Teachers of substance: Subject matter knowledge for
teaching. In M. Reynolds (Ed.), The knowledge base for beginning teachers (pp. 23–36). New York:
Pergamon.
Guruswamy, C., Somers, M., & Hussey, R. (1997). Students’ understanding of the transfer of charge between
conductors. Physics Education, 32(2), 91–96.
Hargreaves, A. (2008). Sustainable professional learning communities. In L. Stoll & K. S. Louis (Eds.),
Professional learning communities: Divergence, depth and dilemmas (pp. 181–195). Maidenhead: Open
University Press and McGraw Hill Education.
Hoz, R., Tomer, Y., & Tamir, P. (1990). The relations between disciplinary and pedagogical knowledge and the
length of teaching experience of biology and geography teachers. Journal of Research in Science Teaching,
27(10), 973–985.
Hume, A., & Berry, A. (2010). Constructing CoRes—a strategy for building PCK in pre-service science teacher
education. Research in Science Education, 41(3), 341–355. doi:10.1007/s11165-010-9168-3.
Ingersoll, R. M. (1998). The problem of out-of-field teaching. The Phi Delta Kappan, 79(10), 773–776.
Jenkins, T. (2005). A brief history of…semiconductors. Physics Education, 40(5), 430–439.
Kind, V. (2009). Pedagogical content knowledge in science education: perspectives and potential for progress.
Studies in Science Education, 45(2), 169–204.
Kind, V. (2014). Science teachers’ content knowledge. In H. Venkat, M. Rollnick, M. Askew, & J. Loughran
(Eds.), Exploring mathematics and science teachers’ knowledge: Windows into teacher thinking (pp. 15–28).
Oxford: Routledge.
Kind, V., & Kind, P. M. (2011). Beginning to teach chemistry: how personal and academic characteristics of pre-
service science teachers compare with their understandings of basic chemical ideas. International Journal of
Science Education, 33(15), 2123–2158.
Klafki, W. (1958). Didaktik analysis as the core of preparation of instruction. London: Lawrence Erlbaum
Associates.
Loughran, J. (2002). Effective reflective practice in search of meaning in learning about teaching. Journal of
Teacher Education, 53(1), 33–43.
Res Sci Educ

Loughran, J., Berry, A., & Mulhall, P. (2006). Understanding and developing science teachers pedagogical
content knowledge. Rotterdam: Sense Publishers.
Loughran, J., Mulhall, P., & Berry, A. (2004). In search of pedagogical content knowledge in science: developing
ways of articulating and documenting professional practice. Journal of Research in Science Teaching, 41(4),
370–391.
Luft, J. A., Hill, K., Weeks, C., Raven, S., & Nixon, R. S. (2013). The Knowledge Needed for Teaching Science:
A Study of In and Out-of-Field Science Teachers Paper presented at the National Association for Research in
Science Teaching, Rio Grande, Puerto Rico. http://www.google.co.za/url?sa = t&rct = j&q = &esrc =
s&frm = 1&source = web&cd = 1&ved = 0CCoQFjAA&url = https%3A%2F%2F2.gy-118.workers.dev/%3A443%2Fhttp%2Fwww.researchgate.net%
2Fpublication%2F256493435_The_Knowledge_Needed_for_Teaching_Science_A_Study_of_In_and_Out-
o f - F i e l d _ S c i e n c e _ Te a c h e r s % 2 F f i l e % 2 F 5 0 4 6 3 5 2 3 1 8 c c f e a f 9 a . p d f & e i =
udNgU5WJBoaS7QaSz4CYAQ&usg = AFQjCNHSSJG2-rYCT1cPJ90JWaPTi3JWTQ&sig2 =
T5G67R2hWMIduYzybjkdmg
Magnusson, S., Krajcik, J., & Borko, H. (1999). Nature sources and development of pedagogical content
knowledge for science teaching. In J. G. Newsome & N. G. Lederman (Eds.), Examining pedagogical
content knowledge: the construct and its implications for science education (pp. 95–132). Dordrecht:
Kluwer.
Mavhunga, E. (2016). Transfer of the pedagogical transformation competence across chemistry topics. Chemistry
Education Research and Practice. doi:10.1039/C6RP00095A.
Mavhunga, E., & Rollnick, M. (2013). Improving PCK of chemical equilibrium in preservice teachers. African
Journal of Research in Mathematics, Science and Technology Education, 17(1-2), 113–125.
Mda, T., & Erasmus, J. (2008). DoL Scarce and Critical Skills - Educators. 70. http://www.labour.gov.za/DOL/
downloads/documents/research-documents/Educators_DoL_Report.pdf. Accessed 4 May 2016.
Nakedi, M., Taylor, D., Mundalamo, F., Rollnick, M., & Mokeleche, M. (2012). The story of a physical science
curriculum: transformation or transmutation? African Journal of Research in Mathematics, Science and
Technology Education, 16(3), 273–288.
Nilsson, P., & Van Driel, J. (2008). Do we dare to teach physics? - Primary science student teachers’
development of subject matter knowledge and a positive attitude towards physics. Paper presented at the
ECER 2008, From Teaching to Learning? European Educational Research Association (EERA), Berlin.
Park, S., Jang, J.-Y., Chen, Y.-C., & Jung, J. (2011). Is pedagogical content knowledge (PCK) necessary for
reformed science teaching? Evidence from an empirical study. Research in Science Education, 41, 245–260.
Park, S., & Oliver, J. S. (2007). Revisiting the conceptualization of pedagogical content knowledge (PCK): PCK
as a conceptual tool to understand teachers as professionals. Research in Science Education, 38(3), 261–284.
Rogan, J. (2004). Out of the frying pan…? Case studies of the implementation of Curriculum 2005 in some
science classrooms. African Journal of Research in Mathematics, Science and Technology Education, 8(2),
165–179.
Rollnick, M. (2014). Using self-study to learn new topics in chemistry—a case study of three practicing teachers.
In H. Venkat, M. Rollnick, M. Askew, & J. Loughran (Eds.), Exploring mathematics and science teachers’
knowledge: Windows into teacher thinking (pp. 147–162). Oxford: Routledge.
Rollnick, M., Bennett, J., Rhemtula, M., Dharsey, N., & Ndlovu, T. (2008). The place of subject matter
knowledge in PCK—a case study of South African teachers teaching the amount of substance and
equilibrium. International Journal of Science Education, 30(10), 1365–1387.
Rollnick, M., Mundalamo, F., & Booth, S. (2013). Concept maps as expressions of teachers’ meaning—making
while beginning to teach semiconductors. Research in Science Education, 43, 1435–1454. doi:10.1007/
s11165-012-9314-1.
Sanders, L., Borko, H., & Lockard, J. D. (1993). Secondary science teachers’ knowledge base when teaching
science courses in and out of their area of certification. Journal of Research in Science Teaching, 30(7), 723–
769.
Schneider, R. M., & Plasman, K. (2011). Science teacher learning progressions: a review of science teachers’
pedagogical content knowledge development. Review of Educational Research, 81(4), 530–565.
Schwab, J. J. (1978). Science, curriculum and liberal education. Chicago: University of Chicago Press.
Shulman, L. S. (1986). Those who understand: knowledge growth in teaching. Educational Researcher, 15(2), 4–14.
Shulman, L. S. (1987). Knowledge and teaching: foundations of the new reform. Harvard Educational Review,
57(1), 1–22.
Smith, P. S., & Banilower, E. R. (2015). Assessing PCK: A new application of the uncertainty principle. In A.
Berry, P. Friedrichsen, & J. Loughran (Eds.), Re-examining pedagogical content knowledge (pp. 88–103).
Oxford: Routledge.
Sperandeo-Mineo, R. M., Fazio, C., & Tarantino, G. (2006). Pedagogical content knowledge development and
pre-service physics teacher education: a case study. Research in Science Education, 36, 235–268.
Stake, R. E. (1995). The art of case study research. Thousand Oaks: SAGE Publications.
 Res Sci Educ

Strauss, A., & Corbin, J. (1990). Basics of qualitative research: Grounded theory procedures and techniques.
Newbury Park: Sage.
Taber, K. S. (2001). Building the structural concepts of chemistry: some considerations from educational
research. Chemistry education: research and practice in Europe, 2(2), 123–158.
Veal, W., & MaKinster, J. (1999). Pedagogical Content Knowledge Taxonomies. Electronic Journal of Science
Education, 3(4). http://www.scholarlyexchange.org/ojs/index.php/EJSE/article/viewArticle/7615. Access 14
Sept 2016.
Warren, B., & Ogonowski, M. (1998). From knowledge to knowing: an inquiry into teacher learning in science.
Retrieved from Newton, MA.
Wettergren, J. (2002). Understanding concepts needed for semiconductor physics. European Journal of
Engineering Education, 27(1), 105–111.
Zeichner, K. M., & Liston, D. P. (2013). Reflective teaching: An introduction: New York: Routledge.