Chapter 3

MODELS AND MODELLING IN CHEMICAL

EDUCATION

Rosaria Justi 1, John Gilbert2

1 Universidade Federal de Minas Gerais, Brazil; 2 University ofReading, UK

INTRODUCTION

As chemistry is concerned with the properties and transformations of

materials, chemists are essentially modellers of the substances that constitute
such materials and of their transformations. In doing so, they seek to make
predictions about the necessary conditions both for the occurrence of a
desirable transformations and for the avoidance of undesirable ones.
Chemists model both the phenomena they observe and the ideas with which
they try to explain such phenomena - that is, at both the macroscopic and the
sub-microscopic levels (Johnstone, 1993) -by the use of analogy with what
they already know. The outcomes of such a process are expressed in a
concrete, visual, mathematical and/or verbal mode of representation,
sometimes by using special symbols that constitute chemical language (e.g.,
formulae of compounds). Moreover, chemists are able to transform models
in one mode of representation into equivalent representation in other modes
(Kozma & Russel, 1997). Such models may be focused on different
attributes of representation (their relationship to quantification, their
behaviour through time, and the reproducibility of their behaviour (Boulter
& Buckley, 2000)). Thus chemical knowledge about a range of phenomena
is produced and communicated with the use of several models, which evolve
and are changed as the field of enquiry advances. In the second half of the
last century, such a central role for the production and use of models in the
growth of chemical knowledge was generally recognised by chemists,
philosophers of chemistry, and by chemical educators (Bailer-Jones, 1999;
47
J.K. Gilbert et al. (eds.), Chemical Education: Towards Research-based Practice, 47-{;8.
© 2002 Kluwer Academic Publishers.
48 Justi, Gilbert

Erduran, 2001; Francoeur, 2000; Giere, 1999; Hardwicke, 1995; Luisi &
Thomas, 1990; Nersessian, 1999; Ramberg, 2000; Suckling et al., 1980;
Tomasi, 1988; Trindle, 1999). Modelling is so common in chemistry that it
has become 'the dominant way of thinking' (Luisi & Thomas, 1990, p. 67) as
the subject has matured, something that chemists do 'without having to
analyse or even be aware of the mechanism of the process' (Suckling et al.,
1980, p. 26). This seems to be so because the explanations of the natures of
substances and of their transformations are essentially abstract. Thinking
with models enables chemists to visualise the entities or processes that are
being enquired into, to more readily plan experimental activities on them
(Tomasi, 1988), and to support the processes of reasoning and constructing
knowledge (Nersessian, 1999).
It seems likely that chemical ideas have been visually, mathematically,
or verbally modelled since they began to be produced. However, the
production of the first concrete models for atoms by John Dalton at the
beginning of the nineteenth century was a landmark in the way that models
have contributed to the development of chemical knowledge. Following him,
leading chemists, such as Kekule, Van 't Hoff, Pauling, Watson and Crick
have made increasing use of concrete models to present, develop and discuss
their ideas about molecular structures. This enabled them to predict the
behaviour of the substances they were modelling and to speculate about the
spatial arrangements of atoms and functional groups in their structures
(Francoeur, 2000). Molecular models thus became obligatory tools in the
study of the stereochemistry, properties, and reactivity of substances which,
in turn, corroborated the atomic theory (Francoeur, 1997).
In recent years, computational models and modelling have been
comprehensively established in chemical research. Two factors seem to have
contributed to this. First, the study of the dynamics of chemical reactions -
the mechanisms through which they occur - required the production of more
complex models once static and rigid molecular models, as well as their two-
dimension representations (formulas and equations - even when curly
arrows are used), were shown to have a limited utility for this purpose.
Second, the introduction of quantum mechanics provided chemistry with a
new research programme (in Lakatosian terms) which allowed chemists to
go beyond qualitative descriptions and even to predict the properties of
materials which have not yet been synthesised (Erduran, 2001; Mainzer,
1999). The possibility of both combining a large amount of data, from
several sources and about different aspects of materials, and of moving
readily between different levels of representation (Ealy, 1999), has made
computational models and modelling an essential tool for investigating
known and new substances and their transformations. The approach is also
 Models and Modelling in Chemical Education 49

vital in probing the properties and uses of new materials - undoubted one of
the currently most important areas of chemical research.

MODELS AND MODELLING IN CHEMICAL EDUCATION: THE

LEARNER'S PERSPECTIVE

The essential roles of models and modelling in chemistry ensure that they
must not be neglected in chemical education. On the contrary, learning
chemistry involves: (i) coming to know the major models already produced
by chemists, as well as the scope and limitations of such models; (ii)
appreciating the role of models in the accreditation and dissemination of the
products of chemical enquiry; and (iii) creating and testing chemical models
produced by an individual and/or a group (Justi & Gilbert, in press a).
Therefore, a comprehensive understanding of models and modelling is
essential for students' learning of chemistry.
From an interview-based enquiry involving mixed-ability seventh and
eleventh grade students in the USA. who had not been taught about models
(but not exclusively as part of their chemical education), Grosslight, Unger,
Jay & Smith (1991) proposed that students' notions of the 'nature of model'
formed a distinct hierarchy of stages. Students in Level 1 thought of models
either as toys or as copies of reality. These sometimes have aspects or parts
of the real thing omitted and are produced just to provide copies of objects or
actions. Students in Level 2 thought of models as being created for a
purpose. The emphasis on some components is therefore altered, but the
template of reality still predominates. The model is tested solely in terms of
its fitness for the predetermined purpose. None of these students reached
Level 3, which had been identified in corresponding interviews with 'experts'
(educated adults with an interest in the area of models). A Level 3
understanding was found to have three components: a realisation that a
model is created to test ideas, rather than as a copy of reality; an acceptance
that the modeller has an active role in its construction for a specified
purpose; and the view that models can be tested and changed in order to
inform the development of ideas.
These levels were used by Harrison & Treagust ( 1996) whilst
investigating students' mental models of atoms and molecules. Most of the
students in their sample (58%) were classified at Level 1 because they
thought that there was a strong correlation between the structure of the
models and reality. For instance, almost all of these students said that 'an
atom is like a ball or sphere'. The other 42% of the students were said to be
at Level 2 because they were able to express some difference between
50 Justi, Gilbert

models and reality and to recognise that different models represent different
aspects of the same phenomenon.
However, the association of the word 'reality' with 'models' by a
chemistry student may not mean that s/he is at Level 1. In a study that
investigated the range of uses that UK chemistry university students make of
molecular models, Ingham & Gilbert (1991) found that, for almost all the
students, a model 'corresponds to reality' (p. 197). It may be that this was
because the models they were dealing with - different types of molecular
models - were being deliberately used to represent the structure proposed for
the organisation of atoms in a given substance. It is precisely through this
representational process that a given structure becomes 'visible' (Francoeur,
2000). So, if students believed (if only for the sake of argument) that
'reality' might actually have the proposed structures, it would then be
possible for them to establish a correspondence with the representation.
Moreover, most of these students also pointed out some functions for models
(e.g., their use to facilitate the visualisation of abstract entities, to provide the
basis for scientific thinking) which were not characteristics of Level I
modellers.
Another important aspect that students should understand about
molecular models is that they are not scale models (Hardwicke, 1995). It is
with an understanding that they are analogue models with different scopes
and limitations that students will be able to use them in explaining and
predicting the diverse properties of substances.
The way students deal with multiple models has also been discussed, for
instance, by Ingham & Gilbert (1991 ), Harrison & Treagust (1996, 2000a),
and Coil & Treagust (200 1). They agreed that the use of a diversity of
models for a given reality is generally challenging and confusing for
students. As a consequence of the, often indiscriminate, use of a diversity of
models, some students: do not clearly realise the scopes of different types of
molecular models and the limitations of each of them (Ingham & Gilbert,
1991); prefer concrete and very familiar models for atoms and molecules
when presented with a range of possible models (Harrison & Treagust,
1996); prefer a simple electrostatic model based on the octet rule for
explaining a ionic bonding, even when they had previously studied more
complex models (Coli & Treagust, 2001). The advantages of having
different aspects of a given target represented by different models are not
usually appreciated by students.
This is one of the areas in which computer-based models - a kind of
teaching model that is becoming more frequently used (Ealy, 1999) - may
help students. This is so because the representational and computational
capabilities of computers can be used in designing multiple, co-ordinated,
 Models and Modelling in Chemical Education 51

representations. The use of such computerised models has been advocated as

a way to improve students' understanding of chemical phenomena by
translating information expressed in different representations (Copolo &
Hounshell, 1995; Kosma & Russel, 1997; Russel et al., 1997; Treagust &
Chittleborough, 2001). Moreover, studies of students' (and sometimes
experts') use of chemical computer-based models have shown that they can
also improve students' visualisation in chemistry (Barnea, 2000; Ealy, 1999;
Fleming et al 2000) as well as students' perception of the nature and role of
models (Barnea, 2000).
From the results of their studies, the authors mentioned above strongly
recommended that students should: learn about the nature of models and
their use as thinking tools; learn about the scope and limitations of specific
chemical models; be encouraged to use multiple models for a given
phenomenon. Such recommendations are also found in studies of science
education from other perspectives (Erduran, 2001; Gilbert, 1997; Gilbert &
Boulter, 1997; Grosslight et al., 1991; Halloun, 1996). However, it does not
seem that students readily change their views about the nature of models as a
result of the direct transmission of formal definitions or from looking at a
particular model (Bent, 1984; Clement, 1989; Morrison & Morgan, 1999). In
order for students to form a more comprehensive understanding of the nature
of chemical models, it would seem necessary that they are both involved in
modelling activities and come to understood how and why different chemical
models are constructed (Clement, 1989; Cosgrove & Schaverien, 1997;
Erduran, 200 I ; Gilbert, 1993, 1997; Greca & Moreira, 2000; Halloun, 1996;
Harrison & Treagust, 2000a, b; Mayer, 1989; Morrison & Morgan, 1999).
This would allow students to progress in their learning of chemistry far
beyond the level normally attained: as a set of concepts and rules to be
memorised and used to solve standard problems. Assuming that modelling is
one of the main processes in the development of scientific knowledge
(Gilbert, 1993), the involvement of students (at all educational levels) in
modelling activities would seem to be an essential part of a more
comprehensive approach to learning.
The introduction of modelling activities into chemical education is an
issue that must be the focus of future research. The possibility of improving
students' understanding of chemistry and models, as well as of their ability to
produce their own chemical models, suggests the outcomes of this research
may well be the basis for a revolution in chemical education. However, lying
between researchers' idealised propositions and the actual improvement of
students' learning are the teachers of chemistry. As they must be the agents
of such a revolution, it would first be necessary to convince them of the
necessity of changing their practices and to prepare them to do so.
52 Justi, Gilbert

MODELS AND MODELLING IN CHEMICAL EDUCATION: THE

TEACHER'S PERSPECTIVE

In order to teach chemistry in the way that we have been advocating,

teachers need to have a clear and comprehensive view of: (i) the nature of a
model in general; (ii) how their students construct their own mental models
and how the resulting expressed models can be constructively used in class;
(iii) how to introduce scientific consensus models in their classes; (iv) how
to develop good teaching models - those that are created with the specific
purpose of facilitating students' understanding of scientific consensus
models; and, finally and most significantly, (v) how to conduct modelling
activities in their classes (Gilbert, 1997).
Van Driel and Verloop (1999) conducted two allied studies into
experienced secondary science (Biology, Chemistry and Physics) teachers'
understanding of the nature of models. In an open-item questionnaire study
(n = 15), inspired by Grosslight et al.'s (1991) work, they found that, in
general, teachers subscribed to the view that 'a model is a simplified or
schematic representation of reality'. Two subgroups of teachers were
identified with a spread of epistemological orientation from logical
positivism to constructivism. Models were seen to have a range of
characteristics, to be encountered in a range of modes of representation, and
to serve a range of function from description to explanation but rarely
prediction. In a Likert-type questionnaire study (n = 71) they identified three
scales that confirmed the above results. The first concerned the relation
between a model and the target it represents: the extent to which models are
seen as a simplified representation of reality. The second concerned the
physical appearance of models: whether they could be met in a range of
modes of representation. The third concerned the social context of model
construction: whether models are the product of human creativity and
communication.
Harrison (2000) interviewed ten experienced high school science teachers
(two of them were chemistry teachers) about their understanding of both the
nature of models and some of the teaching models they use in providing
explanations for their classes. As part of the interview, the teachers were
asked to comment on Gilbert's assertion that
'models are one of the main products of science, modelling is
an element in scientific methodology, [and] models are a
major learning and teaching tool in science education'
(Gilbert, 1993, p. 5).
 Models and Modelling in Chemical Education 53

All the teachers agreed that models are major tools of science, whilst six
of them (including both the chemistry teachers) also said that models are
main products of science. From the analysis of the overall interview data,
Harrison's classification of teachers' understanding of models according to
Grosslight et al.'s (1991) Levels showed that one of the chemistry teachers
was in the transition from Level 1 to Level 2 whilst the other was at Level 2.
The chemistry teachers both volunteered the fewest models when asked to
comment on the teaching of concepts they thought difficult to explain and
were unwilling to manipulate models.
More recently, Harrison (2001) interviewed 17 Australian teachers (5 of
them of chemistry) to investigate the differences between experienced and
pre-service teachers' pedagogical content knowledge with respect to the
models they use to explain science. One of the findings of this study was the
contrast between the few teaching models volunteered by chemistry teachers
and the overabundance of such models found in the chemistry textbooks to
which their students have access.
As part of a broad inquiry into teachers' understanding and use of
models and modelling, 39 Brazilian science teachers working in
'fundamental' (6-14 years) and 'medium' ( 15-17 years) schools, student
teachers, and university teachers were interviewed by one of us (RJ). Within
this sample, 21 teachers were from a chemistry background (1 from the
fundamental level, 6 from the medium level, 5 student teachers and 9
university teachers). Seven 'aspects' to their notions of model were
identified: the nature of a model, the use to which it can be put, the entities
of which it consists, its relative uniqueness, the time span over which it is
used, its status in the making of prediction, the basis for the accreditation of
its existence and use (Justi and Gilbert, in press a). Any individual' s profile
across the seven 'aspects' was usually both complex and fairly unique. On
account of this, this study provided no support for the notion of a 'hierarchy
of levels' in the teachers' understanding of the notion of 'model' taken as
some kind of entity. When we tried to build up a 'profile of understanding'
for each teacher, it was not possible to define patterns that corresponded to
Grosslight et al.'s (1991) Levels. On the other hand, the data showed that
only teachers with a degree in chemistry or physics were able to discuss the
notion of model in a way that was consistently close to the accepted
scientific viewpoint. For instance, 86% of the chemistry teachers included
the idea of a model being a mental image in their view about the nature of
models and recognised that models may be the basis of predictions.
Moreover, among the whole sample, chemistry teachers were 57% of those
who included 'ideas' as entities capable of being represented in a model, 60%
54 Justi, Gilbert

of those who considered that 'a given model as one of several in a historical
sequence' and 50% of those who considered that 'a given model as only one
of several available'. None of the teachers from a chemistry background
assumed that 'a model cannot be changed'. These results must be seen as
encouraging for the widespread and systematic teaching of models and
modelling in chemical education.
In another part of the investigation with these same teachers, they were
asked questions around the theme of: 'What are the knowledge and skills
that a person should have in order to successfully produce a scientific
model?' (Justi & Gilbert, 2002).
In general, teachers were convinced that successful modelling involved a
clear perception of why a model was being produced and of the audience
that would make use of it. At the same time, the scope of use for the model
must be established and as high a degree of similarity as possible to the
phenomenon being represented achieved. They identified experience of a
phenomenon as a prerequisite for the successful modelling of it. They also
saw the value of experience with other, allied, phenomena and with models
of them. Moreover, they saw the value of analogy in the construction of
models and also the importance of clarifying the nature and degree of the
similarity being proposed. Van Driel et al. (1998) have pointed to the
centrality of this skill of the selection of source to the evolution of teachers'
pedagogic content knowledge (Shulman, 1987) in a given subject area. This
would also suggest that teachers should be overtly taught how to draw
analogies (Treagust et al., 1996).
It is interesting .that 57% of the chemistry teachers saw scientists to have
different abilities to other people when it comes to constructing mental
models. They saw models to be produced either by a process of induction, or
of deduction, or of a mixture of the two. Successful modellers, they opined,
must have a personal interest in doing so, be creative in their endeavours,
and persistent in their application to the task. Much to our surprise, the
teachers made no reference to the non-empirical testing of models, perhaps
because this process is not commonly given a separate ontological status in
discussions of science education. This neglect of thought experimentation
generally in science education had been previously noted by one of us (JG)
(Gilbert & Reiner, 2000; Reiner & Gilbert, 2000).
The teachers noted as important the need for a flexibility of approach in
the presentation of a given model to diverse audiences. However, they did
not emphasise either the need for a consideration of the scope and limitations
of models during the process of modelling or the importance of the
discussion of such matters during the presentation of any model to students.
 Models and Modelling in Chemical Education 55

Finally, as the last part of this research project, teachers' use of models
and modelling in science education was probed (Justi & Gilbert, 2001). Very
few of those interviewed would base their teaching on unmodified scientific
consensus models. Most of them recognised the need to produce special
curricular or teaching models in order to make such models, indeed science
generally, more understandable to their students. Perhaps because of this
view about models used in teaching, 95% of the chemistry teachers
recognised that the contribution that models and modelling can make to
science education is concerned with learning. This point of view was mainly
justified by emphasising that 'making the abstract more concrete' is essential
to their students. In asserting this, they were probably being influenced by
the nature of their subject itself.
Some other relevant outcomes of this study involving chemistry teachers
were that:
1. 33% ofthem (from all sub-sets concerning their level of employment) did
not know what their students think about the nature of models. This may
be due in some considerable part to the conditions of employment in
many universities: large numbers of students in a class; too much
material that has to be 'covered'; lack of time for discussion.
2. 56% of the university chemistry teachers do not provide opportunities for
the engagement of their students in modelling activities. In some
countries, e.g. Brazil, the subject knowledge of future chemistry teachers
and techniques of how to teach that subject knowledge are taught by the
same person. In other countries, e.g. UK, different people are involved.
Whichever system is used, university teachers educate all school
chemistry teachers; a fact that may be the cause of the gap in that
educational process at school level.
3. 29% of them accepted that, despite the fact that they discuss their
students' models with them, these models are ignored as being just part
of the learning process: scientific consensus models are subsequently
imposed as 'truth'. Again, most of them were university chemistry
teachers or were being educated by such university teachers. As
discussed elsewhere, this view may be a consequence of the fact that,
due to the extensive use of models and modelling in chemistry,
consensus models have become the main tool in thinking about
chemistry (Harrison & Treagust, 2000b; Luisi & Thomas, 1990). In such
a climate, 'reality' and 'consensus model' may converge in the teaching
process.
From such results, we are able to discuss the extent to which teachers'
use of models and modelling in chemical education was influenced by their
56 Justi, Gilbert

own views of the nature of models. As was stressed before, teachers with a
degree in chemistry or physics were able to discuss the notion of model in a
way that was close to the scientifically accepted viewpoint. However, the
way they think about modelling activities in classes leads us to conclude that
a scientifically acceptable view of the nature of models is not sufficient to
make a teacher aware of both the importance of and the ways of using
models and modelling in chemical education.
In trying to develop chemistry pre-service teachers' pedagogical content
knowledge about models and modelling, DeJong & VanDriel (2001) ran a
modular course on 'teaching models and modelling'. Twelve Dutch pre-
service teachers with a 'not very pronounced' (p. 13) knowledge of the
subject were involved in a series of activities. They answered written
questions about their views on models and modelling and their use in
chemical education. There was discussion - based on a piece of the research
literature - about both the characteristics of scientific models and modelling
in science and about the hierarchy of levels of understanding of models. A
chapter of a chemistry secondary textbook dealing with molecular models
and Dalton's atomic model was analysed. The aim here was to investigate
student teachers' knowledge of learning and teaching difficulties about
certain topics. Classes were conducted on a chosen topic from the chemistry
curriculum in which specific models and/or modelling activities were
dominant: reports on their teaching practice were written, followed by
discussion. From the analysis of all the data generated in this course, the
authors concluded that:
1. The course was more efficient in promoting teachers' awareness of gaps
in their knowledge of appropriate models and modelling than in
providing a significant development of their content knowledge of the
area.
2. 'These gaps in their knowledge may have caused the observed
discrepancy between their teaching intentions (models as constructs) and
their teaching practice (models as facts).' (DeJong & VanDriel, 2001, p.
13)
Because of this, the authors recognised the need to revise their module to
emphasise both teachers' initial content knowledge and pedagogical content
knowledge about models and modelling and adequate ways of presenting
models to students.
In summarising the existent literature, we may say that:
1. Chemistry teachers seem 'to focus their practice on the content of
specific models, rather than on the nature of models and modelling' (Van
Driel, 1998) and/or without emphasising the role of models and
 Models and Modelling in Chemical Education 57

modelling in the development of chemical knowledge. Furthermore,

their views on and use of models and modelling in chemical education,
despite some good exemplars, needs to be improved. This must be so if
we want them to be able to make educational provision that will result in
better student understanding both of particular chemical models and of
the role of models and modelling in the production and growth of
chemical knowledge. In particular, the education of university teachers
in this area should be improved due to their influence on the education of
teachers for the other educational levels.
2. There have been very few initiatives to promote the development of
teachers' pedagogical content knowledge in this area. It seems that the
outcomes of the current research on teachers' (and students') views and
uses of models and modelling in chemical education, as well as the
consideration of the role of models and modelling in the development of
specific chemical knowledge, may lead to interesting proposals for
curriculum development.

MODELS AND MODELLING IN CHEMICAL EDUCATION: THE ROLE

OF CHEMISTRY TEXTBOOKS

In discussing chemical education, the analysis of textbooks is of pivotal

importance because they are the most widely and frequently used teaching
aid at all educational levels. Some analyses of science textbooks have shown
that they tend to present science as a collection of true or complete facts and
of mathematical formulations, as if the material had been 'read directly off
from nature' (Stinner, 1995; Sutton, 1996). Such a view means that the
historical aspects of the development of scientific knowledge cannot receive
an adequate emphasis in such books (Chiappetta et al., 1991 ; Kuhn, 1996;
Monk & Osborne, 1997). In the area of models and modelling, chemistry
textbooks do not even discuss the meaning of the word 'model' (Harrison &
Treagust, 2000b; Justi & Gilbert, 2000).
The role of chemistry textbooks in the teaching and learning of models
and modelling may be discussed from two main angles: the way that
chemical models are introduced in them and the teaching models that they
present.
Analyses of how chemistry textbooks introduce the historical and/or
consensus models of 'chemical kinetics' and 'the atom' (Justi, 1997, 2000;
Justi & Gilbert, 1999, 2000) have showed that their authors introduced
'hybrid models' - those formed by merging characteristics of several distinct
historical models in a field of inquiry. In a particularly blatant case, Justi &
58 Justi, Gilbert

Gilbert (2000) found a diagram of an atom in which a single electron

'swam', in a travesty of the Bohr model, in a 'sea' of electron-probability.
Hybrid models do not allow the history and philosophy of science to make a
full contribution to science education. They do this by denying the role of
distinct models in the history of science and of the role of progression
between these models in the philosophy of science. Moreover, when hybrid
models are used, the gaps of validity between the attributes of a given model
cannot readily be addressed, no questions requiring different ways of
thinking about a phenomenon can be raised, and no different approaches to
the interpretation of a phenomenon are possible. All these aspects contribute
to making chemical knowledge lack meaning for students.
On turning to the teaching models found in chemistry textbooks, new
problems emerge. A teaching model is a representation produced with the
specific aim of helping students to understand some aspect of content (Duit
& Glynn, 1996; Gilbert & Boulter, 1997; Harrison & Treagust, 1996; Mayer,
1989). Assuming the abstract nature of chemical knowledge, they are used
very frequently in chemical textbooks, mainly in the form of overt analogies,
as drawings and as diagrams. The extent to which their introduction really
contributes to students' understanding of specific chemical models has been
studied, specifically in relation to 'the atom' (Cordeiro & Justi, 2000; Justi &
Gilbert, 2000), 'chemical bonding' (Batista & Justi, 2000) and 'chemical
equilibrium' (Milagres & Justi, 2001). These studies showed that they
always fail both in respect of their declared aim and in making a contribution
to students' understanding of the meaning of 'model'. Among other specific
reasons, this is because they support the establishment of hybrid models;
introduce incorrect aspects into given models through an inadequate use of
colours and other graphical elements; and are not accompanied by
discussions of their limitations. By combining this fact with the previously
discussed problems over chemistry teachers' views of the nature and use of
models, we see reasons for the general absence of good practice in the use of
teaching models and modelling activities in chemistry classrooms.

SOME FUTURE RESEARCH DIRECTIONS

The need to enhance scientific literacy has been advocated as one of the
ways to improve science education in this millennium (Millar & Osborne,
1999). Assuming scientific literacy must include an understanding of the
nature and processes by which scientific . knowledge is developed, some
researchers (Boulter & Gilbert, 2000; Erduran, 2001) pointed out the
importance of model-based teaching and learning. In the chemical education
 Models and Modelling in Chemical Education 59

context, it seems that one of the key points in introducing such an approach
is teachers' literacy about models and modelling in general. Therefore, one
of the main challenges to chemical education researchers must be concerned
with teachers' education in this subject. Such challenges may be summarised
in some research and development questions:
1. How can teachers' notion on models and modelling, in respect to all the
aspects pointed out in the current literature, be improved?
2. How can teachers' pedagogical content knowledge about models and
modelling be improved? How can they come to understand that
modelling activities are essential for a comprehensive student
understanding of the meaning of 'model', of the role of models and
modelling in the development of chemical knowledge, and in the
learning of specific chemical models?
3 How can teachers' pedagogical content knowledge about models and
modelling be improved by getting them to produce adequate teaching
models - in two-dimensional, three-dimensional and dynamic (mainly
through computational simulations) forms of representation - and to
critically analyse existing ones?
4 How can it be made evident to teachers that the introduction of a model-
based teaching and learning approach can be a way to shift the emphasis
in chemical education from the transmission of existing knowledge to a
more contemporary perspective in which students will really understand
the nature of chemistry and be able to deal critically with chemistry-
related situations?
We believe much of such research would make more sense after a
systematic introduction of modelling activities into chemistry teaching.
Despite both the recognition of the importance of modelling as a way of
constructing chemical knowledge and the conduct of investigations into
learning through modelling in other areas of science (Barab et al., 2000;
Frederiksen et al., 1999; Halloun, 1996), proposals as to how to actually do
this in chemical education have still to be explored. To provide a framework
for such endeavours, we have developed a 'model of modelling' framework
(Justi & Gilbert, in press b) that is substantially based on Clement (1989)
(see Figure 1)
The elements of the model warrant some expansion. All modelling is
undertaken for a purpose, whether it be to describe the behaviour of a
phenomenon, to establish the entities of which it is thought to consist
(together with their spatial and temporal distribution), to ascribe the reasons
for - the causes and effects of- that behaviour, to predict how it will behave
under other circumstances, or several or all of these (Gilbert et al., 1998).
60 Justi, Gilbert

Decide on
purpose
..... ....

.....
~ ..
Select source Have
r--+ for model experience

Reject mental Produce mental

model model

,,
Express in mode(s) of
representation

Consider scope
and limitation of
Modify mental
Conduct model
model

v
thought
experiments
Fail Pass

I
Design and
perform
empirical tests

Fail Pass

...____ _____.I ~ ___..,

,...--.........
Fulfil
purpose

Figure 1. A 'Model ofModelling ' Framework.

 Models and Modelling in Chemical Education 61

The person forming a mental model of an existing consensus model,

modifying an existing model, or producing his or her own model de novo,
must be clear as to the purpose being addressed. This process of decision
will be enmeshed with some initial, direct or indirect, qualitative or
quantitative, observation of the phenomenon being modelled: having
experience of it. The processes of selecting the source from which the model
is derived and the analogical transfer of the relationships between the entities
within the source to those within the target have been set out in the 'structure
mapping' approach of Gentner & Gentner (1997). Having produced a mental
model in this way, the decision has to be taken about the mode of
representation in which it is to be expressed: material, visual, verbal,
mathematical (Boulter & Buckley, 2000). This process of expression does
seem to be cyclically developmental in respect of the mental model, with the
act of expression leading to a modification in the mental model. Having
produced a model (or having formed an appreciation of a scientifically
accepted, consensus, model), the next step is to explore its implications
through 'thought experimentation ' conducted in the mind. As Reiner &
Gilbert (2000) have commented, it is likely that scientists always mentally
rehearse the design and conduct of empirical experimentation. It is only
when the outcomes of this mental activity seem successful that actual
empirical testing takes place, where this is possible. If the model fails to
produce predictions that are confirmed in the thought experimental testing
phase, then an attempt will have to be made to modify it and to re-enter the
cycle. However, if it passes the thought experimental phase, it can go on the
empirical testing phase. This entails the design and conduct of practical
work, followed by the collection and analysis of data, and finally by the
evaluation of the results produced against the model. This would also mean a
good opportunity to make students understand that experimentation may be
not only a rote recipe-following activity, but an essential activity through
which models are produced and evaluated (Erduran, 2001). If the model fails
at this stage, an attempt also has to be made to modify it and to re-enter the
cycle. However, if it passes the empirical testing phase, the modeller will
feel confident that the purpose for which it was constructed has been
fulfilled. This will be followed by a phase in which attempt will be made to
persuade others of its value. During this process of advocacy, the scope and
limitations of the model will become apparent, leading to a reconsideration
of the earliest elements in the model-production cycle. If the sub-cycle of
model modification and thought and/or empirical testing is repeatedly
unsuccessful, then the model will have to be rejected. This will lead to a
62 Justi, Gilbert

radical reconsideration of the earliest elements in the model-production

cycle.
We believe that if students were introduced to this modelling framework
(as general guidelines) whilst learning specific chemical topics, they would
be able, for instance, to formulate their own models, to evaluate them against
other models (including the scientific consensus models), to understand how
and why chemical models were/are produced. Moreover, it could support
students' understanding of abstract, core, themes in chemistry, like the
interactions between particles, chemical equilibrium, electrochemistry and
many others. Following the introduction of such a framework in chemistry
teaching, its influence on students' and teachers' development of declarative
and procedural knowledge (as emphasised in the above points) might be the
basis for the development of new research.

PRECEPTS FOR 'GOOD PRACTICE'

Until such research and development work is done, the most that can be
provided for chemistry teachers and teacher educators are some precepts
derived from existing research. This suggests that :
Textbook writers and publishers:
• explicitly introduce the representational conventions that they are
using;
• avoid the use of hybrid models, adhering instead to the use of
historical or consensus models and curricular simplifications of these.
Teachers:
• provide opportunities for students to develop and test their own
models;
• provide a phased yet comprehensive introduction to the 'nature of
model';
• state the scope and limitations for all models that they introduce to
students.
Teacher educators:
• provide examples both through their own teaching and through
student activities, of
- the nature of teaching models and how to introduce them to
students;
- strategies for the teaching of modelling skills.
 Models and Modelling in Chemical Education 63

REFERENCES

Barab, S. A., Hay, K. E., Barnett, M., & Keating, T. (2000). Virtual solar system
project: Building understanding through model building. Journal of Research in
Science Teaching, 37(7), 719-756.
Barnes, N. (2000). Teaching and learning about chemistry and modelling with a
computer managed modelling system. In J. Gilbert, & C. Boulter (Eds.),
Developing Models in Science Education, (pp. 307-323). Dordrecht, The
Netherlands: Kluwer.
Batista, A.A., & Justi, R. S. (2000). Modelos de ensino de liga~oes quimicas
[Models in the teaching of chemical bonding], Paper presented at the lOth
Brazilian Conference on Chemical Education, Porto Alegre, 12-15 July.
Bailer-Jones, D.M. ( 1999). Tracing the development of models in the philosophy of
science. In L. Magnani, N. J. Nersessian and P. Thagard (Eds.) Model-based
Reasoning in Scientific Discovery. (pp. 23-40). New York: Kluwer and Plenum
Publishers.
Bent, H.A. (1984). Uses (and abuses) of models in teaching chemistry. Journal of
Chemical Education, 61(9), 774-777.
Boulter, C.J., & Buckley, B.C. (2000). Constructing a typology of models for
science education. In J. K. Gilbert & C. J. Boulter (Eds.), Developing models in
science education (pp. 41-57). Dordrecht, The Netherlands: Kluwer.
Boulter, C.J. & Gilbert, J.K. (2000). Challenges and opportunities. In J K. Gilbert &
C. J. Boulter (Eds.), Developing models in science education (pp. 343-362).
Dordrecht, The Netherlands: Kluwer.
Chiappetta, E.L., Sethna, G.H., & Fillman, D. A. (1991). A quantitative analysis of
high school chemistry textbooks for science literacy: Themes and expository
learning aids. Journal ofResearch in Science Teaching, 28(10), 939-951.
Clement, J. (1989). Learning via model construction and criticism. In J.A. Glover,
R. R. Ronning & C. R. Reynolds (Eds.), Handbook of Creativity (pp. 341-381).
New York: Plenum Press.
Coli, R. K., & Treagust, D. F. (2001). Learners' mental models of ionic bonding: A
cross-age study, Paper presented at the Annual Meeting of the National
Association for Research in Science Teaching, St. Louis, Missouri, 25-28
March.
Copolo, C. F. & Hounshell, P. B. (1995). Using three-dimensional models to teach
molecular structures in high school chemistry. Journal of Science Education
and Technology 4(4), 295-305.
Cordeiro, E. S., & Justi, R. S. (2000). Influencia de Modelos de ensino na
aprendizagem de modelos atomicos de Thomson e Rutherford [Influence of
models in the teaching and learning of the Thomson's and Rutherford's atomic
64 Justi, Gilbert

models], Paper presented at the lOth Brazilian Conference on Chemical

Education, Porto Alegre, 12-15 July.
Cosgrove, M., & Schaverien, L. (1997). Models of science education. In J. Gilbert
(Ed.), Exploring models and modelling in science and technology education:
Contributions from the MISTRE group (pp. 20-34). Reading, UK: Faculty of
Education and Community Studies, The University of Reading.
De Jong, 0., & Van Driel, J. (2001). Developing preservice teachers' content
knowledge and PCK of models and modelling, Paper presented at the Annual
Meeting of the National Association for Research in Science Teaching, St.
Louis, Missouri, 25-28 March.
Duit, R., & Glynn, S. (1996). Mental modelling. In G. Welford, J. Osborne, & P.
Scott (Eds.), Research in science education in Europe: Current issues and
themes (pp. 166-176). London: Falmer.
Ealy, J. B. (1999). A student evaluation of molecular modeling in first year college
chemistry. Journal ofScience Education and Technology, 8(4), 309-321.
Erduran, S. (2001). Philosophy of chemistry: An emerging field with implications
for chemistry education. Science & Education, 10(6), 581-593.
Fleming, S. A., Hart, G. R., & Savage, P. B. (2000). Molecular orbital animations
for organic chemistry. Journal of Chemical Education, 77(6), 790-793.
Francoeur, E. (1997). The forgotten tool: The design and use of molecular models.
Social Studies ofScience, 27, 7-40.
Francoeur, E. (2000). Beyond dematerialization and inscription: Does the materiality
of molecular models really matter? HYLE - An International Journal of the
Philosophy of Chemistry, 6(1), 52-69.
Frederiksen, J. R., White, B. Y., & Gutwill, J. (1999). Dynamic Mental Models in
Learning Science: The importance of constructing derivational linkages among
models. Journal ofResearch in Science Teaching, 36(7), 806-836.
Giere, R. N. (1999). Using Models to Represent Reality. In L. Magnani, N. J.
Nersessian and P. Thagard (Eds.) Model-based reasoning in scientific discovery
(pp. 41-57). New York: Kluwer and Plenum Publishers.
Gilbert, J. (Ed.) (1993). Models & modelling in science education. Hatfield, UK:
The Association for Science Education.
Gilbert, J. ( 1997). Models in science and science education. In J. Gilbert (Ed.),
Exploring models and modelling in science and technology education:
contributions from the MISTRE group (pp. 5-19). Reading: Faculty of
Education and Community Studies, The University of Reading.
Gilbert, J., & Boulter, C. (1997). Learning science through models and modelling. In
B. Fraser, & K. Tobin (Eds), International Handbook ofScience Education, Part
1 (pp. 53-66). Dordrecht, The Netherlands: Kluwer.
 Models and Modelling in Chemical Education 65

Gilbert, J. K., & Reiner, M. (2000). Thought experiments in science education:

potential and current realization. International Journal of Science Education,
22(3), 265-283.
Gilbert, J., Boulter, C., & Rutherford, M. ( 1998). Models in explanations, Part 1:
horses for courses? International Journal ofScience Education, 20(1), 83-97.
Greca, I. M., & Moreira, M. A. (2000). Mental models, conceptual models, and
modelling. International Journal ofScience Education, 22( 1), 1-11.
Grosslight, L., Unger, C., Jay, E., & Smith, C. L. (1991). Understanding models and
their use in science: Conceptions of middle and high school students and
experts. Journal ofResearch in Science Teaching, 28(9), 799-822.
Halloun, I. (1996). Schematic modelling for meaningful learning of physics. Journal
ofResearch in Science Teaching, 33(9), 1019-1041.
Hardwicke, A. J. (1995). Using molecular models to teach chemistry Part 1
modelling molecules. School Science Review, 77(278), 59-64.
Harrison, A. G. (2000). How do teachers and textbook writers model scientific ideas
for students? Paper presented at the Annual Meeting of the National
Association for Research in Science Teaching, New Orleans, 29 April - 1 May.
Harrison, A. (2001). Models and PCK: Their relevance for practicing and pre-
service teachers, Paper presented at the Annual Meeting of the National
Association for Research in Science Teaching, St. Louis, Missouri, 25-28
March.
Harrison, A. G., & Treagust, D. F. (1996). Secondary students' mental models of
atoms and molecules: Implications for teaching chemistry. Science Education,
80(5), 509-534.
Harrison, A. G., & Treagust, D. F. (2000a). Learning about atoms, molecules, and
chemical bonds: A case study of multiple-model use in grade 11 chemistry.
Science Education, 84(3), 352-381.
Harrison, A. G., & Treagust, D. F. (2000b). A typology of school science models.
International Journal ofScience Education, 22(9), 1011-1026.
Ingham, A.M., & Gilbert, J. K. (1991). The use of analogue models by students of
chemistry at higher education level. International Journal of Science Education,
13(2), 193-202.
Johnstone, A.H. (1993). The development of chemistry teaching: A changing
response to changing demand. Journal ofChemical Education, 70(9), 701-705.
Justi, R. S. (1997). Models in the teaching of chemical kinetics, Unpublished Ph.D.
Thesis. Reading: The University of Reading.
Justi, R. (2000). Teaching with historical models. In J. Gilbert, & C. Boulter (Eds.),
Developing models in science education, (pp. 209-226). Dordrecht, The
Netherlands: Kluwer.
66 Justi, Gilbert

Justi, R., & Gilbert, J. (1999). A cause of ahistorical science teaching: use of hybrid
models. Science Education, 83(2), 163-177.
Justi, R., & Gilbert, J. (2000). History and philosophy of science through models:
some challenges in the case of 'the atom'. International Journal of Science
Education, 22(9), 993-1009.
Justi, R., & Gilbert, J. (2001). Teachers' views about models and modelling in
science education. Paper presented at the Annual Meeting of the National
Association for Research in Science Teaching, St. Louis, Missouri, 25-28
March.
Justi, R., & Gilbert, J. (in press a). Teachers' views on the nature of models.
International Journal ofScience Education.
Justi, R., & Gilbert, J. (2002). Modelling, teachers' views on the nature of modelling,
implications for the education of modellers. International Journal of Science
Education, 24(4), 369-387.
Koama, R. B. & Russel, J, (1997). Multimedia and understanding: expert and novice
responses to different representations of chemical phenomena. Journal of
Research in Science Teaching, 34(9), 949-968.
Kuhn, T. S. (1996). The structure of scientific revolutions, 3rd. Chicago and
London: The University of Chicago Press.
Luisi, P. L., & Thomas, R. M. (1990). The pictographic molecular paradigm -
Pictorial communication in the chemical and biological sciences.
Naturwissenschaften, 77, 67-74.
Mainzer, K. (1999). Computational models and virtual reality. New perspectives of
research in chemistry. HYLE - An International Journal of the Philosophy of
Chemistry, 5(2), 117-126.
Mayer, R. E. (1989). Models for understanding. Review of Educational Research,
59(1), 43-64.
Milagres, V. S. 0., & Justi, R. S. (2001). Modelos de ensino de equilibrio quimico-
Algumas considera~5es sobre o que tern sido apresentado por livros didaticos
brasileiros destinados ao ensino medio [Teaching models of chemical
equilibrium - some aspects about what Brazilian medium level textbooks have
been presented]. Quimica Nova na Escola, 13, 35-40.
Millar, R., & Osborne, J. (Eds.) (1999). Beyond 2000: Science education for the
future. London: School of Education, Kings' College London.
Monk, M., & Osborne, J. (1997). Placing the history and philosophy of science on
the curriculum: A model for the development of pedagogy. Science Education,
81(4), 405-424.
Morrison, M., & Morgan, M. S. (1999). Models as mediating instruments. In M. S.
Morgan, & M. Morrison (Eds.), Models as mediators (pp. 10-37). Cambridge:
Cambridge University Press.
 Models and Modelling in Chemical Education 67

Nersessian, N. J. (1999). Model-based reasoning in conceptual change. In L.

Magnani, N. J. Nersessian and P. Thagard (Eds.) Model-based reasoning in
scientific discovery (pp. 5-22). New York: Kluwer and Plenum Publishers.
Ramberg, P. (2000). Pragmatism, Belief, and Reduction: Stereoformulas and Atomic
Models in Early Stereochemistry. HYLE - An International Journal of the
Philosophy of Chemistry, 6(1 ), 29-51.
Reiner, M., & Gilbert, J. (2000). Epistemological resources for thought
experimentation in science learning. International Journal of Science
Education, 22(5), 489-506.
Russel, J. W., Kozma, R. B., Jones, T., Wykoff, J., Marx, N. & Davis, J. (1997). Use
of simultaneous-synchronized macroscopic, microscopic, and symbolic
representations to enhance the teaching and learning of chemical concepts.
Journal of Chemical Education, 74(3), 330-334.
Shulman, L. S. (1987). Knowledge and teaching: Foundations of the new reform.
Harvard Educational Review, 57(1), 1-22.
Stinner, A. (1995). Science Textbooks: Their present role and future form. InS. M.
Glynn, & R. Duit (Eds.), Learning science in the schools (pp. 275-296).
Mahwah, New Jersey: Lawrence Erlbaum Associates.
Suckling, C. J., Suckling, K. E., & Suckling, C.W. (1980). Chemistry through
models. Cambridge: Cambridge University Press.
Sutton, C. ( 1996). Beliefs about science and beliefs about language. International
Journal ofScience Education, 18(1 ), l-18.
Tomasi, J. ( 1988). Models and modeling in theoretical chemistry. Journal of
Molecular Structure (Theochem), 179, 273-292.
Treagust, D. F., & Chittleborough, G. (2001). Chemistry: A matter of understanding
representations. In J. Brophy (Ed.), Subject-specific Instructional Methods and
Activities (pp. 239-267). Oxford: Elsevier Science.
Treagust, D. F., Harrison, A. G., Venville, G. J., & Dagher, Z. (1996). Using an
analogical teaching approach to engender conceptual change. International
Journal ofScience Education, 18(2), 213-229.
Trindle, C. (1999). Entering Modeling Space. An apprenticeship in molecular
modeling. HYLE - An International Journal of the Philosophy of Chemistry,
5(2), 127-142.
Van Driel, J. H. (1998). Teachers' knowledge about the nature of models and
modelling in science, Paper presented at the Annual Meeting of the National
Association for Research in Science Education, San Diego, 19-22 April.
Van Driel, J. H., & Verloop, N. (1999). Teachers' knowledge of models and
modelling in science. International Journal of Science Education, 21 ( 11 ), 1141-
1154.
68 Justi, Gilbert

VanDriel, J. H., Verloop, N., & DeVos, W. (1998). Developing science teachers'
pedagogical content knowledge. Journal ofResearch in Science Teaching 35(6),
673-695.