INT. J. SCI. EDUC., 2002, VOL. 24, NO.

4, 357–368

RESEARCH REPORT

Students’ understanding of the role of scientific

models in learning science

David F. Treagust, Gail Chittleborough and Thapelo L. Mamiala, Science

and Mathematics Education Centre, Curtin University of Technology, GPO
Box U1987, Perth, Western Australia; e-mail: [email protected]

Scientific models are used routinely in science not only as learning tools, but also as representations of
abstract concepts and as consensus models of scientific theories. Students’ experiences with scientific
models help them to develop their own mental models of scientific concepts. This paper discusses
the development and evaluation of an instrument to measure secondary students’ understanding of
scientific models. The results of a study with 228 secondary science students identify five themes
about students’ understanding of scientific models: scientific models as multiple representations;
models as exact replicas; models as explanatory tools; how scientific models are used; and the changing
nature of scientific models. The results highlight the need for greater emphasis on the teaching of the
role and purpose of the concept of scientific models in science.

Introduction
As an integral part of the scientific process, models are used in a variety of ways
within the science classroom. Teachers use models as aids to help explain scientific
phenomena and students often make their own models of scientific phenomena to
display their understanding. Indeed, scientific models are often the only way to
explain an abstract scientific theory and scientists’ consensus models are taught as
fact as a result of being the accepted model of a scientific theory, for example, the
model of the atom. Personal mental models are constructed from all the informa-
tion assimilated and understanding is conveyed via each person’s expressed model
(Gilbert et al. 1998a). This diversity in the way in which scientific models are used
and perceived is significant.
Scientific models are an important part of the scientific process and although
the role of the model and the scientific process are not always taught directly, the
concepts are shown through examples in many different topics across the science
curriculum. Scientific models have long been used and appreciated as useful tools
that enhance learning; however, most elementary and junior high school students
regard scientific models as concrete replicas of the real thing, with few students
regarding scientific models as representations of ideas or abstract entities
(Grosslight et al. 1991, Ingham and Gilbert 1991). Grosslight et al. (1991)
found that students have ‘conceptions of scientific models that are basically con-
sistent with a na¹̈ve realist epistemology. They are more likely to think of scientific
models as physical copies of reality that embody different spatio-temporal
International Journal of Science Education ISSN 0950–0693 print/ISSN 1464–5289 online # 2002 Taylor & Francis Ltd
http://www.tandf.co.uk/journals
DOI: 10.1080/09500690110066485
358 D. F. TREAGUST ET AL.

perspective than as constructed representations that may embody different theor-

etical perspectives’ (p. 799).
Learning is an active process and requires students to construct their own
personal schema to assimilate new concepts (Yager 1991). To assist in this con-
structivist process, scientific models are valuable tools because they can be used to
‘make sense of abstract, difficult and non-observable science concepts to accom-
modate the explainer, the audience, the content and the context’ (Treagust and
Harrison 1999: 4). However, scientific models used superficially without under-
standing have been shown to hinder understanding (Cosgrove and Schaverien
1997). Two primary functions of scientific models in teaching are the predictive
power of the model and the ability to provide insight into the fundamental nature
of the phenomenon (Bhushan and Rosenfeld 1995). Scientific models are tools for
prediction and correlation, although their potential is not always fully utilized in
the classroom. Nevertheless, recent technological advancements have brought
computer modelling and simulations into the realm of everyday life so many
students experience simulation games, test their own ideas in strategy games,
and drive cars or fly planes by means of computer simulation. These new experi-
ences, including extensive exposure to visual stimuli, will impact on a student’s
conceptual understanding of what models are and the way models work in learning
science.
The objective of this study was to gain some insight into students’ under-
standing of the role of scientific models in learning science. The instrument
Students’ Understanding of Models in Science (SUMS) has been designed to
achieve this. Students have their own personal and unique understanding of the
role of scientific models in science built up through their life experiences. These
understandings may not always be scientifically correct and may lead to alternative
conceptions; teachers’ assumptions about the degree of students’ understandings
of scientific models also may not always be correct. Consequently, a more accurate
picture may be obtained through the administration of a pencil-and-paper instru-
ment so that science teachers can be more aware of the range and variety of
students’ understandings of the role of models in learning science.

Instrument analysis

Subjects
This study with 228 students from two government, non-selective, co-educational
high schools in Perth, Western Australia, involved 69 (30.3%) Year 8 students (age
13 years), 44 (19.3%) Year 9 students (age 14 years), and 115 (50.4%) Year 10
students (age 15 years). The students had received no special teaching about
scientific models in science, so the responses reflect their understanding based
on the general science curriculum they have experienced.

Instrumentation
The items in the instrument Students’ Understanding of Models in Science
(SUMS) have been written based on data from a study into the use of chemical
models in teaching organic chemistry (Treagust et al. 2001) and from Grosslight
et al.’s (1991) study into students’ understanding of models and their use in
STUDENTS’ UNDERSTANDING OF THE ROLE OF SCIENTIFIC MODELS 359

science. The instrument was designed to gain some insight into students’ under-
standing of what a model is, the role of models in science, including how and why
models are used and what causes models to be changed. The SUMS instrument is
a 27-item pencil-and-paper questionnaire which requires students to respond on a
five-point Likert-type scale, with a choice of responses: strongly disagree (1),
disagree (2), not sure (3), agree (4) and strongly agree (5). The Statistical
Package for Social Scientists (SPSS) (Coakes and Steed 1996) was used to analyse
the quantitative data.

Factor Structure
Factor analysis using a varimax rotation identified five distinct factors in the items
of the SUMS instrument which are described as five scales in the instrument: the
‘Models as multiple representations’ (MR) scale (factor 1); the ‘Models as exact
replicas’ (ER) scale (factor 2); the ‘Models as explanatory tools’ (ET) scale (factor
3); the ‘Uses of scientific models’ (USM) scale (factor 4); and ‘The changing
nature of models’ (CNM) scale (factor 5) (see table 1).
The MR scale explores students’ acceptance of using a variety of representa-
tions simultaneously, and their understanding of the need for this variety.
Examples of items from this scale are: ‘Many models may be used to express
features of a science phenomenon by showing different perspectives to view an
object’ (item 1); and ‘Many models represent different versions of the phenom-
enon’ (item 2).
The ER scale refers to students’ perceptions of how close a model is to the real
thing. Examples of items in this scale are: ‘A model needs to be close to the real
thing by being very exact in every way except for size’ (item 13); and ‘A model
should be an exact replica’ (item 9).
The ET scale refers to what a model does to help the students understand
an idea. This scale includes providing visual enhancement, generating a mental
model or providing a concrete representation. Examples of items in this scale
include: ‘Models help create a picture in your mind of the scientific happening’
(item 18); and ‘Models are used to physically or visually represent something’
(item 17).
The USM scale explores students’ understanding of how models can be used
in science, beyond their descriptive and explanatory purposes. Examples of items
in this scale are: ‘Models are used to help formulate ideas and theories about
scientific events’ (item 22); and ‘Models are used to make and test predictions
about a scientific event’ (item 24).
Finally, the CNM scale addresses the permanency of models. Examples of
items include: ‘A model can change if new theories or evidence prove otherwise’
(item 25); and ‘A model can change if there are new findings’ (item 26).
The range of items in the SUMS instrument attempts to identify the breadth
of students’ understanding of particular aspects of models. Each item attempts to
identify the details of students’ understanding by asking about particular aspects of
models that are categorized as scales of the SUMS instrument. A number of items
for each scale help ensure consistency of results. Three items load into two factors:
item 5, ‘Many models may be used to show different sides or shapes of an object’ is
loaded into the MR and the ET categories. This is not surprising as the models
take on these complementary roles simultaneously. Item 14, ‘A model needs to be
360 D. F. TREAGUST ET AL.

Table 1. Factor analysis of the 27-item instrument Students’ Under-

standing of Models in Science (SUMS) (n = 228).

Factor Loadings

Factor 1: Factor 2: Factor 3: Factor 4: Factor 5:

Models as Models as Models as Uses of Changing
multiple exact explanatory scientific nature of
Item representations replicas tools models models
number …MR† …ER† …ET† …USM† …CNM†

1 0.75
2 0.62
3 0.61
4 0.60
5 0.59 0.48
6 0.57
7 0.52
8 0.50
9 0.80
10 0.67
11 0.65
12 0.64
13 0.60
14 0.55 0.51
15 0.50 0.45
16 0.47
17 0.66
18 0.66
19 0.61
20 0.45
21 0.41
22 0.83
23 0.70
24 0.69
25 0.70
26 0.67
27 0.47
% variance 33.2 8.6 5.2 4.8 4.1
Eigenvalue 8.97 2.32 1.40 1.28 1.11

Factor loadings less than 0.4 omitted.

close to the real thing by giving the correct information and showing what the
object/thing looks like’, and item 15, ‘A model shows what the real thing does and
what it looks like’, are loaded into the ER and ET scales. These items reflect both
of these aspects of models. These three items are discussed for each of the two
scales that they represent.

Data Analysis
The reliability of each scale ranged from 0.71 to 0.84 (see table 2) showing that the
instrument has high internal consistency for each scale; item-to-total correlations
were above 0.45 except for item 16. A bi-variate correlation of the five scales (see
STUDENTS’ UNDERSTANDING OF THE ROLE OF SCIENTIFIC MODELS 361

Table 2. Descriptive statistics and reliability of the five scales of the

SUMS instrument (n = 228).

Cronbach
Number Standard alpha
Scale of items Mean deviation reliability*

Models as multiple representations (MR) 8 3.52 0.63 0.81

Models as exact replicas (MR) 8 3.58 0.71 0.84
Models as explanatory tools (ET) 5 3.58 0.71 0.71
The uses of scientific models (USM) 3 3.41 0.73 0.72
The changing nature of models (CNM) 3 3.73 0.74 0.73

* a measure of the internal consistency of each scale.

Table 3. Bi-variate correlation of the five scales in the SUMS instrument

(n = 228).

Scale 2 …ER† 3 …ET† 4 …USM† 5 …CNM†

(1) Models as multiple representations (MR) 0.61** 0.63** 0.47** 0.58**

(2) Models as exact replicas (ER) 0.49** 0.30** 0.52**
(3) Models as explanatory tools (ET) 0.46** 0.52**
(4) The uses of scientific models (USM) 0.30**
(5) The changing nature of models (CNM)

** Correlation is significant at the 0.01 level (2-tailed).

table 3) shows a high level of correlation indicating that students’ responses to each
scale are related and consistent.
The distribution of scores for each scale of the SUMS instrument is concen-
trated closest to the ‘agree’ elective (table 4). The CNM scale has the most highly
agreed upon response while the USM scale has an even distribution between the
‘not sure’ and ‘agree’ responses. The use of the word ‘phenomenon’ in three items
of the SUMS instrument corresponded to a high ‘not sure’ response indicating
that students were not familiar with the word; consequently, results involving
items using this word are considered guardedly. A one-way ANOVA (Coakes
and Steed 1996) showed no statistically significant differences for any of the scales
between year levels. An independent t-test identified a significant difference in
gender for the ET scale only, with the results indicating that females responded
more positively than males to the items in this scale.

Discussion of the five themes

Through the analysis of the data from the instrument, a number of themes con-
cerning scientific models that related to the five scales identified in the factor
analysis of the SUMS instrument consistently emerged:
(1) Scientific models as multiple representations
(2) Models as exact replicas
(3) Models as explanatory tools
(4) How scientific models are used
(5) The changing nature of scientific models
 Table 4. Results of students’ understanding of models (SUMS) (n = 228).
362

Factor{/ %
Item
Number Item Mean …sd† Disagree* Not sure Agree**

MR/1 Many models may be used to express features of a science phenomenon by showing different 3.56 (0.96) 12 27 61
perspectives to view an object.
MR/2 Many models represent different versions of the phenomenon. 3.33 (0.97) 15 41 44
MR/3 Models can show the relationship of ideas clearly. 3.67 (0.97) 11 27 62
MR/4 Many models are used to show how it depends on individual’s different ideas on what 3.56 (0.91) 10 34 56
things look like or how they work.
MR/5 Many models may be used to show different sides or shapes of an object. 3.60 (0.86) 11 25 64
MR/6 Many models show different parts of an object or show the objects differently. 3.44 (0.93) 14 34 52
MR/7 Many models show how different information is used. 3.51 (0.93) 11 34 55
MR/8 A model has what is needed to show or explain a scientific phenomenon. 3.52 (0.93) 13 30 57
ER/9 A model should be an exact replica. 3.14 (1.17) 36 21 43
ER/10 A model needs to be close to the real thing. 3.74 (1.10) 13 18 69
ER/11 A model needs to be close to the real thing by being very exact, so nobody can disprove it. 3.35 (1.09) 23 28 49
ER/12 Everything about a model should be able to tell what it represents. 3.61 (0.94) 14 23 63
ER/13 A model needs to be close to the real thing by being very exact in every way except for size. 3.57 (1.11) 18 20 62
ER/14 A model needs to be close to the real thing by giving the correct information and 3.83 (0.99) 9 16 75
showing what the object/thing looks like.
ER/15 A model shows what the real thing does and what it looks like. 3.69 (0.91) 9 26 65
ER/16 Models show a smaller scale size of something. 3.77 (1.06) 15 15 71
ET/17 Models are used to physically or visually represent something. 3.85 (0.95) 9 17 74
ET/18 Models help create a picture in your mind of the scientific happening. 3.55 (1.06) 16 19 65
ET/19 Models are used to explain scientific phenomena. 3.36 (0.88) 12 43 45
ET/20 Models are used to show an idea. 3.80 (1.02) 12 9 79
ET/21 A model can be a diagram or a picture, a map, graph or a photo. 3.46 (1.11) 20 22 58
USM/22 Models are used to help formulate ideas and theories about scientific events. 3.41 (0.87) 14 37 49
USM/23 Models are used to show how they are used in scientific investigations. 3.46 (0.95) 15 32 53
USM/24 Models are used to make and test predictions about a scientific event. 3.35 (0.90) 14 42 44
CNM/25 A model can change if new theories or evidence prove otherwise. 3.82 (0.90) 6 23 71
CNM/26 A model can change if there are new findings. 3.79 (0.90) 7 22 71
CNM/27 A model can change if there are changes in data or belief. 3.62 (0.90) 10 26 64

{ MR (Models as multiple representations); ER (Models as exact replicas); ET(Models as explanatory tools); USM (The uses of scientific models); and CNM (The changing nature of models).
D. F. TREAGUST ET AL.

* Disagree ˆ Strongly Disagree and Disagree. ** Agree ˆ Strongly Agree and Agree.
STUDENTS’ UNDERSTANDING OF THE ROLE OF SCIENTIFIC MODELS 363

Selected data that support each theme are analysed, presented and discussed. The
consistent and contradictory examples of the data are examined and implications
from the results discussed.

Theme 1: Scientific models as multiple representations

Alternative scientific models can provide a variety of perspectives and appearances
and most students show an appreciation of these (table 4, factor 1, items 1–8).
Consistently, approximately 60% of students agreed that the use of multiple mod-
els is useful to show different perspectives, different views and different versions of
an object. More than half of the students recognized that a variety of scientific
models are useful in catering for individual differences (table 4, item 4). These
consistent results show that more than half of the respondents recognized the need
for multiple scientific models to cater for particular aspects of the item as well as
catering for individual needs of the learner. By recognizing the value of multiple
scientific models we can infer that students have an understanding that a model is
just one representation of an entity and that each representation displays a par-
ticular perspective or emphasis. These results contrast with those of Grosslight
et al. (1991: 816) where ‘very few of the mixed ability 7th and the honors 11th
graders even hinted at the sense of multiple modeling’. Considering the extensive
use of multiple representations in science, the need to recognize multiple repre-
sentations and be able to transfer from one representation is important.

Theme 2: Scientific models as exact replicas

The SUMS instrument showed that 43% of students agreed that a model was an
exact replica (table 4, factor 2, item 9). The responses consistently confirm that
models need to be ‘close to the real thing’ (items 14, 10, 11 and 13). These results
indicate that there is a significant group of students with a narrow and na¹̈ve
understanding of the concept of a model as an exact replica. This corresponds to
scale scientific models that are usually representative of more familiar and better-
understood objects, for example, a model ear or a globe of the earth, for which
accuracy and detail are crucial. Despite this, as many as 20% of the students realize
that there is more to scientific models than just being a copy of the original. When
scientists model abstract and unknown entities, often the actual appearance is not
known or is irrelevant and the representation – which does not have accuracy or
detail – can provide insight into why and how something works the way it does.
In considering the need for scientific models to be accurate and closely repre-
sent the real thing, 75% of students agree that a model needs to be close to the real
thing by giving the correct information and showing what the object looks like
(table 4, factor 2, item 14); 62% agree that a model should be exact in every way
except for size; and 49% agree that the model must be very exact, so nobody can
disprove it (table 4, items 11 and 13). The data consistently confirm that some
students continue to regard scientific models as exact duplicates of reality. Similar
results have been obtained in other studies (Grosslight et al. 1991, Ingham and
Gilbert 1991, Harrison and Treagust 1996). This stumbling block associated with
models is recognized by Hardwicke (1995) who emphasized the role of the teacher
in ‘distinguishing the positive and negative analogies as clearly as possible’ (p. 64)
so that students realize the limitations of the model. Students seem to be faced
364 D. F. TREAGUST ET AL.

with a dilemma of trading off the accuracy of the model (exact replica) with the
concept of a model providing insight into specific aspects of the entity, even
though this could mean that the model is not totally accurate. These results dis-
tinguish two types of models: the scale replica, a precise representation, which has
accuracy and detail; and the imprecise representation, which doesn’t have the
accuracy or detail, and may be nothing like the object, but can provide insight
into why and how something works the way it does. Students’ experiences with
everyday models are usually associated with the first type, whereas scientific mod-
els, especially of the more abstract concepts, would more commonly fall into the
latter scale. Students’ awareness of the type of model being used is a most import-
ant issue when considering their understanding of the role of scientific models in
learning.

Theme 3: Models as explanatory tools

Models are often used to represent things that are too small or too big to be seen
with the eye, so, in this way, models are the only visual representation that the
learner sees. The visual aspect of scientific models applies to many forms of scien-
tific models and is described by Gilbert et al. (1998b) as providing a form of visual
explanation, which helps students link the known and the unknown, familiar and
the unfamiliar (Collins and Gentner 1987). The responses to the SUMS instru-
ment revealed that most students valued this descriptive aspect of models with the
majority agreeing to statements such as ‘models are used to physically or visually
represent something’ (table 4, factor 3, item 17, 74%) and that ‘many models show
different sides or shapes of an object’ (factors 1 and 3, item 5, 64%) and ‘a model
shows what the real thing does and what it looks like’ (factors 2 and 3, item 15,
65%).
The results suggest that most students are aware of the value of the visual
representation that many scientific models provide. The ability of the student to
take advantage of the features of a model must be taken into consideration when
assessing the value of a particular model; consequently, with visual representations
and three-dimensional representations, a student’s spatial ability is a significant
factor in the success of using the model. In a study investigating the effects of
visually stimulating computer-generated representations, Barnea and Dori (1999)
showed that there is a strong correlation between spatial ability and achievement in
science, which is not surprising considering the dependence on scientific models in
the form of diagrams, graphs, tables and three-dimensional scientific models. The
close relationship between models and explanations identified by students in their
responses is significant. Students use models to make a connection between the
observed phenomena and the scientific explanation. Through this process,
students generate a mental model as explained by Gilbert et al. (1998a). Many
topics in science require students to generate their own mental models and
students are aware that physical representations can help them to generate their
own mental models and understand new concepts. Teachers routinely make use of
models and representations to assist students to construct their own mental model.
This is particularly relevant and useful for abstract ideas. Students have indicated
a good understanding of this role of models as explanatory tools in their responses
to the SUMS instrument with the majority agreeing that ‘Models are used to show
an idea’ (table 4, factor 3, item 20, 79%), and that ‘Models help create a picture in
STUDENTS’ UNDERSTANDING OF THE ROLE OF SCIENTIFIC MODELS 365

your mind of the scientific happening’ (item 18, 65%). It is interesting that
students are able to recognize this quality of models that enables them to develop
mental models and conceptual models for new concepts (Tiberghein 1994, Duit
and Glynn 1996). Modelling is an instinctive behaviour and the explanatory role
the model adopts is most useful in learning science (Gilbert and Boulter 1998).
The diverse forms that a model may take may include an idea, an object, event,
system or process (Gilbert and Boulter 1998). The variety of forms that can be
represented were appreciated by the majority of students with 58% agreeing that a
model can be a diagram, picture, map, graph or photo (table 4, factor 3, item 21).
However, the remaining 42% of students ‘disagreed’ or responded with ‘not sure’
to this description of a model, indicating that they do not regard all of these par-
ticular items to be models. With respect to scientific models, the implied meaning
of the term model is broad and includes many representations compatible with the
variety of representations used in explaining science; however, the meaning of the
term ‘model’ in general everyday use is narrower and hence may lead to misun-
derstandings. The contextually relevant dictionary meaning of the term ‘model’ is
‘(1) a standard or example for copying or comparison; (2) a representation, usually
on a small scale; (3) an image in clay or wax’ (The Macquarie Essential Dictionary
2000: 508). The discrepancy between the definition, understanding, and use of the
term model could explain the students’ lack of understanding of the concepts of
models and model building in science (Gilbert 1991).

Theme 4: How scientific models are used

Approximately 50% of students agreed that scientific models are used for making
predictions, formulating theories and showing how information is used (table 4,
factor 4, items 22, 23 and 24); approximately 50% of students either were not sure
or disagreed with this view. In this instance, there is evidence that many students
do not understand how scientific models are used in the development of scientific
ideas and theories. In Grosslight et al.’s (1991) study into students’ understanding
of the role of models in science, similar results were found and it was recom-
mended that students
be provided with some experiences using models to solve intellectual problems. In
this way students would have an opportunity to learn that a model can be used as a
tool of inquiry and that it is not simply a package of facts about the world that needs to
be memorized. (Grosslight et al. 1991: 820)

The application of models beyond a descriptive nature, as described in theme 3,

seems to be beyond the understanding of many of these student respondents. If the
results reflect students’ experiences, then it would be suggested that students have
had experience with scale models and descriptive models but have not used models
in a quantitative or interpretive fashion. Stephens et al. (1999) reported that
model-based reasoning is not practiced in school science and consequently
students have no experience with models being used in a scientific way.
Similarly, Gilbert et al. (1998a) report on the value of mathematical and symbolic
representations which allow predictive, interpretive and causative varieties com-
pared to the more descriptive explanations of visual, verbal and physical models.
Obviously, the descriptive models are most valuable for teaching and are used
366 D. F. TREAGUST ET AL.

extensively; however, there is a need to make more use of interpretive and

predictive models in the teaching process.

Theme 5: The changing nature of scientific models

Over two-thirds of students showed an appreciation that models are constructs to
support scientific theories and that they will change according to changes in scien-
tific thinking. This is supported by strong agreement for statements in the SUMS
instrument, such as ‘A model can change if there are new findings’ (table 4, factor
5, item 26, 71%), or ‘A model can change if new theories or evidence prove other-
wise’ (table 4, item 25, 71%). The consistency of these results confirms that more
than two-thirds of those students have a clear understanding of the changing
nature of scientific models in response to changes in scientific thinking. This
aspect of models introduces students to the important feature of the uncertainty
of scientific knowledge and the nature of science, which has been shown to be
lacking even in Grade 12 chemistry students (Boo 1998). Students’ understanding
of the nature of science has been shown to influence their learning in science
(Songer and Linn 1991), so it is a most important aspect of their learning to foster.

Conclusion
This analysis has focused on the difficulties associated with understanding the
model concept, but this should not detract from the results that have shown that
many students have a good understanding of the role of scientific models in learn-
ing science. Students’ interpretation of the term ‘scientific model’ will depend on
their experiences and personal understanding. Models as multiple representations
(factor 1) were recognized as being necessary and useful by the majority students,
and they appreciated the visual value of scientific models in helping to generate
their own mental models. Students showed a good appreciation for the changing
nature of scientific models (factor 5), which was linked to the changing nature of
scientific knowledge.
However, there are inconsistencies in the percentage of students’ responses, in
that some students clung to the understanding that a model is an exact replica
(factor 2) supporting the descriptive concept of a model. The categorization of a
model as a precise representation or an imprecise representation helps to explain
some of the conflicting ideas that students have about scientific models. When
dealing with more abstract concepts it is assumed that students would adopt a
more abstract nature of scientific models, but this is not necessarily true. While
this study has specifically focused on scientific models, students’ experience with
general models is the starting point in their understanding of scientific models.
General models more commonly fit into the category of scale replica, whereas
scientific models assume many forms and are used more analytically (Hardwicke
1995). By highlighting these subtle differences between different types of models,
they may be used more effectively in teaching and learning science.
The results of this instrument showed that the majority of students under-
stood that scientific knowledge can change (factor 5), with new ideas and theories
resulting in changes to the accepted scientific models. We also can conclude that a
large majority of these students understood the descriptive role of models (factor
3), but there is scope to expand the applicable role of models in scientific ways such
STUDENTS’ UNDERSTANDING OF THE ROLE OF SCIENTIFIC MODELS 367

as making predictions and testing ideas (factor 4). The evaluation and use of
scientific models in this way could improve students’ understanding of the use
of scientific models in the development of scientific ideas as well as developing a
better understanding of the particular content area.
Under a constructivist philosophy, learning in science requires students to
take ownership of an idea or concept, reconstruct it, internalize it and be able to
explain or communicate it to others. Models serve as invaluable tools in this pro-
cess. The links between models and learning are indisputable; however, there is
evidence in these results that many students do not fully appreciate scientific
models. The reason for this could be lack of opportunity to use models effectively
and applicably, or teachers may fail to emphasize the strengths and limitations of
particular models and thereby create misunderstandings in students’ perceptions.
The vast extent to which models are used in the scientific field provides inspiration
to further the use of models in the science classroom to enhance learning in a
scientific manner.

References
Barnea, N. and Dori, Y. J. (1999) High school chemistry students’ performance and gender
differences in a computerised molecular modeling learning environment. Journal of
Science Education and Technology, 8(4), 257–271.
Bhushan, N. and Rosenfeld, S. (1995) Metaphorical models in chemistry. Journal of
Chemical Education, 72, 578–582.
Boo, H. K. (1998) Students’ understandings of chemical bonds and the energetics of chemi-
cal reactions. Journal of Research in Science Teaching, 35(5), 569–581.
Coates, S. J. and Steed, L. G. (1996) SPSS for Windows: Analysis without Anguish
(Milton, Queensland, Jacaranda Wiley).
Collins, A. and Gentner, D. (1987) How people construct mental models. In D. Holland
and N. Quinn (eds), Cultural Models in Language and Thought (New York: University
of Cambridge), 243–265.
Cosgrove, M. and Schaverien, L. (1997) Models of science education. In Exploring Models
and Modelling in Science and Technology Education (Reading, UK: The University of
Reading), 20–34.
Duit, R. and Glynn, S. (1996) Mental modelling. In G. Welford, J. Osborne and P. Scott
(eds), Research in Science Education in Europe: Current Issues and Themes (London:
The Falmer Press), 166–176.
Gilbert, J. K. and Boulter, C. J. (1998) Learning science through models and modelling.
In B. J. Fraser and K. G. Tobin (eds), International Handbook of Science Education
(Dordrecht: Kluwer Academic Publishers), 53–66.
Gilbert, J. K., Boulter, C. and Rutherford, M. (1998a) Models in explanations, Part 1:
Horses for courses? International Journal of Science Education, 20(1), 83–97.
Gilbert, J. K., Boulter, C. and Rutherford, M. (1998b) Models in explanations, Part 2:
Whose voice? Whose ears? International Journal of Science Education, 20(2), 187–203.
Gilbert, S. W. (1991) Model building and a definition of science. Journal of Research in
Science Teaching, 28(1), 73–79.
Grosslight, L., Unger, C., Jay, E. and Smith, C. (1991) Understanding models and their
use in science: conceptions of middle and high school students and experts. Journal of
Research in Science Teaching, 28(9), 799–822.
Hardwicke, A. J. (1995) Using molecular models to teach chemistry: Part 1 using models.
School Science Review, 77(278), 59–64.
Harrison, A. G. and Treagust, D. F. (1996) Secondary students’ mental models of atoms
and molecules: implications for teaching chemistry. Science Education, 80(5), 509–
534.
368 STUDENTS’ UNDERSTANDING OF THE ROLE OF SCIENTIFIC MODELS

Ingham, A. I. and Gilbert, J. K. (1991) The use of analogue models by students of

chemistry at higher education level. International Journal of Science Education,
13(2), 203–215.
Songer, N. B. and Linn, M. C. (1991) How do students’ views of science influence knowl-
edge integration? Journal of Research in Science Education, 28(9), 761–784.
Stephens, S., McRobbie, C. J. and Lucas, K. B. (1999) Model-based reasoning in a year 10
classroom. Research in Science Education, 29(2), 189–208.
The Macquarie Essential Dictionary (2000) (NSW, Australia: The Macquarie Library Pty
Ltd)
Tiberghein, A. (1994) Modeling as a basis for analysing teaching – learning situations.
Learning and Instruction, 4, 71–87.
Treagust, D. F. and Harrison, A. G. (1999) The genesis of effective scientific explanations
for the classroom. In J. Loughran (ed.), Researching Teaching: Methodologies and
Practices for Understanding Pedagogy (London: Falmer Press), 28–43.
Treagust, D. F., Chittleborough, G. and Mamiala, T. (2001) Learning introductory
organic chemistry: secondary students’ understanding of the role of models and the
development of scientific ideas. Paper presented at AERA 2001, Seattle, WA.
Yager, R. E. (1991) The constructivist learning model: towards real reform in science
education. The Science Teacher, 58(6), 52–57.