Relationship between academics' educational beliefs and conceptions and
their design and use of computer software in higher education
John Bain, Griffith University
Carmel McNaught and Gillian Lueckenhausen, La Trobe University
and Colleen Mills, Griffith University
Paper presented at 'Researching Education in New Times'.
AARE annual conference.
30 November - 4 December, 1997. Brisbane
The research reported in this paper was supported by grant A79601676
from the Australian Research Council for the period 1996-1998.
Understanding the influence of information technology on student
learning cannot be accomplished without reference to the
epistemological and educational assumptions of the academic teachers
who design and use computer-facilitated learning (CFL) programs.
The focus of the paper is the first stage of an ARC project in which
thirty-six technology-based CAUT projects from a number of disciplines
were examined. This study was based on archive material only (the
initial application and final report for each project). Projects were
sorted into categories in each of which the educational presumptions
and practices were similar. Categories were then compared and refined
so as to reveal their major sources of similarity and difference. The
resulting framework is one in which the use educational technology in
higher education can be interpreted in terms of several key qualitative
dimensions which reflect academics' beliefs about the origin of
knowledge, the learning framework, control of the direction of
learning, the nature of the knowledge and of the learning process.
The research reported here is the first of two studies concerned with
the relationships between academics' educational beliefs and
conceptions and the ways in which they design and incorporate
computer-facilitated learning (CFL) into their course units.
The impetus for the work came from recent interest in the beliefs and
understandings that academics bring to their teaching in general (Fox,
1983; Samuelowicz & Bain, 1992; Trigwell, Prosser & Taylor, 1994) as
well as to their use of educational technology (Bain & McNaught, 1996;
Laurillard, 1993). The importance of educational beliefs and
conceptions in framing educational practice is now becoming apparent in
the school sector, despite the countervailing influence of educational
training (Clark & Peterson, 1986; Kagan, 1992; Pajares, 1992), and it
is to be expected that similar belief framing will be observed in
higher education, particularly as most academics have not undertaken
educational training and do not refer to educational literature when
developing their teaching practices (Laurillard, 1993; Ramsden, 1993).
There is some research suggesting that academics' educational beliefs
do frame their teaching practices (Ballantyne, Bain & Packer, 1997;
Trigwell & Prosser, 1996), but more work is needed to uncover the
details of the relationships involved, particularly in relation to the
use of educational technology (Reeves, 1992).
An additional motivation for the present research derives from the
urgent need to understand the influence of CFL on student learning
(Laurillard, 1993; Wills & McNaught, 1996), acknowledging that the
evaluation of any educational innovation is complex (Reeves, 1997) and
is not fully addressed by methods designed primarily to establish that
CFL is superior to conventional techniques (in which context it has
been argued that CFL has no influence beyond the pedagogy realised in
the technology--Clark, 1992). Although the research reported here is
not directly concerned with the evaluation of learning outcomes, it is
concerned with the educational context in which technology is used,
because, as many have observed, the impact of technology cannot be
understood without reference to its place in the whole
teaching/learning environment (Crook, 1994; Elton, 1988; Laurillard,
1993; Reeves, 1997; Wills & McNaught, 1996).
Two literatures are of immediate relevance to the present research:
studies concerned with academics' educational beliefs and conceptions
(Dall'Alba, 1990; Fox, 1983; Kember & Gow, 1994; Martin & Balla, 1990;
Prosser, Trigwell & Taylor, 1994; Samuelowicz & Bain, 1992) and recent
work exploring pedagogical frameworks for interpreting and evaluating
the role of CFL in education (e.g., Perkins, 1991; Reeves, 1992).
We'll briefly consider these in turn.
Academics' educational beliefs and conceptions
Research into academics' educational beliefs and conceptions is of
relatively recent origin and has not yet converged on a common
framework, although there is some agreement that a major difference
between academics can be captured by the distinction between
teacher-centred and student-centred views (Gow & Kember, 1993; Prosser,
Trigwell & Taylor, 1994; Samuelowicz & Bain, 1992). In the framework
reported by Samuelowicz and Bain, for example, five different sets of
beliefs were found to be sufficient to describe the educational
orientations1 of academic staff sampled from science and social science
disciplines. As shown in Table 1, educational beliefs are arranged as
contrasts on qualitative belief dimensions in this framework, while
orientations are the distinguishable patterns of beliefs expressed by
groupings of academic staff. The five orientations described by
Samuelowicz and Bain indicate the range and character of views which
range from imparting information (teacher-centred) to supporting
student learning (student-centred). Similar patterns of
conceptions/beliefs have been reported by Prosser, Trigwell and Taylor
(1994) for science academics2.
Table 1
Teaching/learning orientations reported by Samuelowicz & Bain (1992)
This table can be viewed at the following web site:
http://www.meu.unimelb.edu.au/staffwebsites/dkennedy/AARE/BAIN134.html
The teacher-centred/student-centred ordering of academics' orientations
to teaching and learning would appear to correspond with the contrast
between instructional and constructivist approaches to CFL (Boyle,
1997; Reeves, 1992; Reiber, 1992). Instructional CFL, for example,
emulates teacher-driven learning episodes in which there is limited
scope for the student's understandings and interests to be
accommodated. Constructivist CFL, on the other hand, emphasises the
role of the student in structuring and managing the learning experience
and in building an adequate understanding for themselves. However, as
we will note in more detail below, the picture may turn out to be more
complex than this juxtaposition would suggest.
Some frameworks for interpreting CFL
We will not pretend to do justice in this section to the wealth of
sophisticated analysis of educational technologies which is now
available (e.g., Boyle, 1997; Crook, 1994; Dills & Romiszowski, 1997;
Jonassen, 1996; Khan, 1997; Laurillard, 1993; Müldner & Reeves, 1997).
Rather, our purpose is to characterise some useful frameworks for the
research we are to report.
One approach to the analysis of CFL is to group applications into
functionally similar categories and highlight the learning processes
and outcomes which are likely to be 'afforded' by each category. The
instructionist/constructivist contrast noted above is one such
grouping, and of course it can be elaborated to take account of
important variants (e.g., Boyle, 1997, p. 108). Perkins (1991), who
was concerned with this contrast, analysed the resources typically
found in classroom learning environments, resulting in a classification
that includes educational technology:
* information banks (provide content information, such as textbooks,
databases);
* symbol pads (general purpose tools that support cognitive tasks, such
as calculators and wordprocessors);
* construction kits (domain relevant tools such as lab equipment, and
authoring and modelling software);
* phenomenaria (packages that make phenomena accessible to scrutiny and
manipulation, such as videos, microworlds, simulations); and
* task managers (what runs the learning process, such as teachers,
students, study guides, computer-based instruction).
Although this scheme is valuable for the sorts of purposes that Perkins
was pursuing, it has the disadvantage of all such schemes of focussing
exclusively on the prototypical properties of CFL. This may be
appropriate in circumstances where the learning environment is largely
defined by the technology, but these circumstances are rare. In higher
education CFL is usually only one of the teaching/learning resources
deployed, and the method of its use may depend both on the other
resources available and the particular purposes to which the CFL is
directed. For example, an academic located towards the student-centred
end of the Samuelowicz and Bain (1992) scheme might use a drill-based
CFL (a task manager in Perkins' scheme) as a complement to more
'constructive' learning activities which themselves may or may not be
implemented as CFL (cf. Crook, 1994, pp. 14-15). On the other hand, a
teacher-centred academic might use drill-based CFL as one of several
teacher-centred methods. The learning processes and outcomes in the
two contexts are unlikely to be the same in character. And, more to
the point of the present research, without attending to the broad
character of the teaching/learning environment, it might be difficult
to fully understand the connections between academics' educational
beliefs and their CFL practices.
Consider another hypothetical example of an academic who develops a
simulation to demonstrate the properties of a system that is difficult
to study in reality. This CFL might only be used in lectures, or it
might be made available to students to use in association with
exercises and assessments. Again, operating only on the prototypical
features of the CFL, we might infer that its potential influence on
learning is quite substantial, and that the thinking behind its
development was prototypically student-centred, but neither may be
correct.
In short, although the various CFL classification schemes now available
draw attention to the potential properties of CFL packages, we are
interested in how CFL is actually used, partly because from an
educational perspective, that is what we need to know to understand its
likely impact on student learning, but also because we suspect that we
will not adequately understand the linkages between academics'
educational beliefs and their use of CFL without close attention to how
the CFL is deployed. However, as will be noted in detail in the
Methods section, we have not abandoned the use of classification; on
the contrary, categories of CFL practice are an important feature of
our method. Nevertheless, the classifications we have adopted are
grounded in the practices of academics rather than being determined a
priori by the prototypical properties of CFL packages.
An alternative approach to the analysis of CFL is to evaluate each case
against suitable criteria. Laurillard (1993), for example, argues for
twelve processes derived from a conversational analysis of effective
teaching and learning in higher education. The presumptions of this
analysis are that learning occurs as students are assisted (through
conversation and related supports) to translate fluently between
descriptions of the world and the abstract conceptualisations favoured
by academic analysis. This is both a teacher-driven and student-driven
process in which both participants share their understandings
(conceptions) and dialogue is adapted by the teacher to ensure that the
student's understanding approaches that of the teacher. This certainly
is not a process in which students are left to their own devices, nor
is it one which relies on one-way instruction, and hence it has the
potential to reflect the range of belief orientations described in the
literature. Moreover, it would be possible to apply Laurillard's
criteria to whole learning environments rather than to classes of
educational media (as she did), but we elected to adopt another
approach, about which more will be said in a moment.
An alternative to Laurillard's approach is the 14 pedagogic dimensions
proposed by Reeves (1992). Whereas Laurillard's criteria derive from
conversation analysis, Reeves' bipolar dimensions derive from the
constructivist literature and tend to oppose characteristics associated
with constructivist and instructivist approaches (examples are:
epistemology (objectivism vs constructivism); pedagogical philosophy
(instructivist vs constructivist); goal orientation (sharply focussed
vs unfocussed); value of errors (errorless learning vs learning from
experience); structure (high vs low); learner control (non-existent vs
unrestricted); user activity (mathemagenic vs generative)). These
dimensions are not intended to be independent or exhaustive, but they
are continua rather than qualitative categories. Reeves provides
analyses of two CFL packages to demonstrate how each has a distinctive
'profile' across the dimensions, one 'located' towards the
instructivist end of the continua, the other towards the constructivist end.
As with Laurillard's criteria, we could have adapted Reeves' dimensions
to our purposes by rating the overall CFL environment of each case in
our sample, but our preference was to allow the dimensions (in addition
to the categories) to emerge from the data rather than be overly framed
by any one theoretical perspective. We also did not wish to be
constrained by assumptions about the nature of the dimensions. Those
used by Samuelowicz and Bain (1992), for example, involve qualitative
distinctions (different beliefs), those advocated by Reeves (1992) are
continua, but we wanted to keep the options open. It would be
disingenuous of us to imply that we approached the analysis from a
fully neutral position (see Bain & McNaught, 1996, for example), but
our intention nevertheless was to develop categories of CFL practice,
and the dimensions needed to describe that practice, as much as
possible from the cases in front of us. Moreover, given our overall
intention to seek the connections between academics' educational
beliefs and CFL practices, we anticipated that the categories and the
dimensions would incorporate aspects of educational beliefs.
Method
Sample
This study was based entirely on archival material and was designed in
part to provide the sampling plan for a second study drawing upon
extensive interviewing as well as archival material. The cases
included in this study (and in the second one, which is now in
progress) were funded by the Committee for the Advancement of
University Teaching (CAUT), a federal government organisation which
operated in Australia from 1993 to 1996. The sample was restricted to
CAUT-funded projects because the documentation required by CAUT
required academics to attend to the educational framing of their
projects as well as to the learning processes and outcomes involved.
Thus there was some prospect of gaining access to information about
educational beliefs as well as about CFL design and implementation.
All 36 projects included in the present study involved the
implementation of CFL, and were in progress during 1993-1994. Later
projects could not be included (but are being sampled for the second
study) because we required both the project application and the final
report for our analyses. Final reports were not due until 18 months
after commencement, but some 1994 project reports were not available
until late 19953. The discipline into which the CFL was embedded was
not a feature of project selection for the current study. An analysis
of disciplines across the sample indicated that 21 were from science
disciplines, 8 from medical disciplines and 7 from humanities
disciplines. Six projects were omitted because the technology was used
as a measuring instrument, design tool or training tool, rather than as
a tool to enhance teaching and learning.
Analysis
The two documents for each project (grant application and final report)
were analysed jointly with the 'conceptual field' method of Samuelowicz
and Bain (1992). Ideally, the first phase of this procedure involves
the formation of categories of projects based on global impressions
formed through the constant comparative method (Glaser & Strauss,
1967). In practice, because of the wealth of detail involved, it was
necessary to make notes to summarise the salient details of each
project, and this inevitably drew attention to their properties.
However, the main focus at this stage remained on forming stable
impressions of the global similarities of cases within a nascent
category, and the global differences between categories. Once stable
categories were formed, emphasis shifted to the second stage which
involved a systematic comparison of the categories to reveal the major
bases (dimensions) upon which they were similar and different. Note
that this process operated at the category, not the individual project
level, with the consequence that similarities and differences that
might have been apparent between individual projects were disregarded
unless they applied at the category level as well. The third phase
required that each project be checked against the dimensions to ensure
that the descriptions applying to its category also applied to it
individually.
The above characterisation of the analysis method, while broadly
correct, is something of an idealisation of what actually happened. We
began the first phase with two coders working independently to gain
some sense of how 'compelling' the categories might be. There were
many similarities in the two categorisations, but there also were
differences, from which we inferred that there were conspicuous and
subtle differences between the projects, the latter lending themselves
to different emphases. An intensive two-day meeting resulted in a
substantially agreed set of categories and first approximations to the
dimensions, but the process was a complex one in which discussion
shifted between categories, dimensions and cases in a type of mutual
constraint satisfaction process. Subsequently, during detailed
checking of cases against the descriptive framework (third phase), we
found that a few cases did not adequately fit the category into which
they had initially been placed and this resulted in the a rechecking of
the descriptive framework before cases were reassigned. Interspersed
amongst all these deliberations were repeated attempts at refining our
terminology to ensure that it adequately characterised all cases in the
sample. So, in reality, the phases were more like the emphases of an
iterative process aimed at satisfying four constraints: the categories,
their dimensions, the category membership of each project, and the
terminology used to describe the salient features of categories. We do
not claim that the framework we report here is the only way to
interpret the data, but we are confident that it is defensible,
reproducible and a credible description of the data available to us.
Findings
Preliminary observations
Nothwithstanding the reporting requirements of CAUT, the documents
available to us varied considerably in the extent to which they made
the academic's teaching and learning assumptions clear, and in the
detail they provided about the teaching/learning environment in which
the CFL was deployed. These limitations probably resulted in fewer
overt belief dimensions than might be obtained from an interview study
(such as we are now conducting), and we tended to emphasise properties
of the CFL more than anticipated. However, it should be noted in
relation to the last point that it was not the type of the CFL that
distinguished between the categories (Table 3, last column), but rather
the way in which the CFL was used by students, and the structure and
support associated with its use, whether 'built in' or supplied in
other ways.
As far as we were able to discern, the CFL in most of the projects was
congruent with the teaching, learning and assessment practices of the
course unit. That is, most academics in the sample designed CFL that
supported their expressed educational framework and objectives,
although in some cases the implementation fell short of desired goals
for technical or resource reasons, or it was altered in response to
student evaluation. There were few instances in which the CFL was used
'out of character' as a complement to the prevailing teaching/learning
environment.
Three major categories emerged from the conceptual field analysis,
within each of which there are a number of subcategories. A broad
outline of the major categories follows, and the detailed categories
are given in Table 3.
Category 1 is characterised predominantly by the fact that students are
encouraged to construct their own defensible interpretations of the
knowledge base through examining evidence and considering different
perspectives. Students follow their own lines of reasoning within the
framework provided by the academic which is designed to foster the
development of critical reasoning skills.
Category 2 is characterised by teaching and learning situations in
which students are given some freedom to explore concepts and ideas
within the domain structure. The notion that students should actively
engage with and explore concepts in order to enhance or change their
existing conceptual understanding is embraced by the academic. Tasks
generally are open-ended in nature with the academics providing
varying amounts of guidance and structure to the students' own
construction process.
Category 3 is characterised by teaching and learning situations where
the academic takes responsibility for stucturing the domain concepts in
the way in which they should be received by students. Activities are
usually closed in nature, focussing on students being able to review
and reproduce the understanding modelled for them by the academic.
The dimensions
Five qualitative dimensions emerged that both united projects within a
specific category and kept them distinct from other categories. The
dimensions identified are sketched below, while details are provided in
Table 2:
1. The learning framework refers to the degree of structuring provided
by the teaching/learning environment, ranging from facilitated (in
which the support is akin to 'semiotic mediation'--Brooks, 1994, pp.
85-88), through guided (a stronger sense of what is required), to
structured.
2. The origin of the knowledge (student/collaborative,
academic/discipline) refers to the source of the knowledge base itself
and its openness to alternative defensible interpretations.
3. Learning directions (student-managed/ teacher-managed) refers to
the extent to which students have control in selecting their own
pathways through the knowledge base.
4. Knowledge focus refers to the focus of the teaching and learning
environment (conceptual/procedural reasoning, conceptual/procedural
knowledge, case-based reasoning).
5. The learning process refers to the design and implementation of the
teaching activities and learning opportunities provided.
The classification framework
Table 3 summarises the classification framework developed during this
study. The table shows the combination of dimension descriptions which
define the categories and sub-categories.
Category 1 is distinguished from the other categories because students
are required to develop their own interpretations or arguments assisted
by the CFL. Thus the origin of knowledge is the student (or the
student in collaboration with the teacher) and the learning process is
one which involves challenge to existing interpretations as new ones
are constructed. Within this process there are different levels of
teaching support sufficient to distinguish two subcategories, namely 1a
in which the learning framework is facilitated (in the sense of
'semiotic mediation' emphasised by Crook, 1994, pp. 85-88), and 1b in
which guidance is provided through questioning, feedback or good
practice exemplars.
Category 2 differs from Category 1 in focussing on discipline
knowledge, often with less scope for full knowledge construction,
although students manage the directions of their learning, as in
Category 1. Three subcategories derive from the learning framework,
the learning process and the focus of knowledge. In 2a for example,
the CFL and associated pedagogy is designed to facilitate learning
(there is 'semiotic support'), whereas in the other subcategories a
clearer sense of the correct or preferred interpretations is conveyed
by such means as feedback or model answers. The other subcategories
differ in the learning process they encourage, partly as a consequence
of the focus of the knowledge (case-based reasoning in 2b, conceptual
and procedural knowledge in 2c).
Category 3 differs from the other categories because its learning
framework is more structured, the learning directions are substantially
determined by the CFL, and the learning process is more passive. The
subcategories differ in their learning processes, 3a being concerned
with the emulation of discipline ideas and skills, 3b with the uptake
of factual knowledge.
Tables A.1-A.4 in Appendix A provide brief descriptions of projects
which illustrate four of the subcategories. These descriptions are
organised in terms of their constituent dimensions, but in
project-specific detail and terminology.
Table 2
Details of the qualitative dimensions
Learning Framework
Facilitated
Learning opportunities are created in which students are encouraged to
actively explore the subject/content matter in their own way and
build/challenge their own knowledge representations.
Guided
Learning opportunities are provided in which students are able to
explore the subject/content matter in their own way but the process is
actively guided through feedback, model answers or good practice
exemplars.
Structured
The learning opportunities provided are highly structured. Information
is provided and students are given set tasks to perform using the given
information.
Origin of Knowledge
Student/Collaboration
Knowledge resulting from the reasoned interpretation of information.
Different equally valid interpretations of the same information are
possible.
Academic/Discipline
Knowledge that is drawn from a well-defined discipline base with a
received interpretation.
Learning Directions
Student-Managed
The student is given freedom or opportunity to explore her/his own
lines of reasoning or questioning within the knowledge domain.
Teacher-Managed
The teacher controls the flow of information, questioning and
directions pursued within the program. A student may be free to review
an aspect of choice but then, within that area the paths are laid down
by the teacher.
Table 2 continued
Knowledge Focus
Conceptual/Procedural Reasoning
The development of higher order thinking, reasoning and metacognitive
skills used in conjunction with discipline concepts, principles and
procedures.
Conceptual/Procedural Knowledge
The content, concepts and principles of a discipline and the associated
procedural skills.
Case-Based Reasoning
The development of professional reasoning or decision-making skills in
the application of knowledge to case-based problems.
Learning Process
Knowledge Construction/Challenge
Students are challenged to consider presented information from
different perspectives or reconsider their own understandings so as to
construct new interpretations .
Knowledge Elaboration/Challenge
Students are provided with learning opportunities that extend and/or
challenge their existing conceptual understanding or interpretive
skills often by allowing them to explore the consequences of their
interpretations.
Knowledge Synthesis/Elaboration
Students are required to synthesise knowledge from a variety of sources
often to solve case-based problems. Through this process their
conceptual understanding may be elaborated.
Knowledge Elaboration
Students are invited to explore nuances of concepts, find new examples
and extend their existing understanding of the concepts.
Knowledge Emulation
Ideas/concepts are connected and understandings developed in line with
the received wisdom of the discipline base. The aim is for students to
be able to emulate expert understanding and thinking.
Table 2 continued
Knowledge Assimilation
Factual knowledge is presented in a fairly fragmented way with little
structuring, elaboration or transformation required of the students who
instead are to assimilate the knowledge.
Table 3: The categories described in terms of their distinguishing and
unifying dimensions
This table can be viewed at the following web site:
http://www.meu.unimelb.edu.au/staffwebsites/dkennedy/AARE/BAIN134.html
Discussion
The categories
One feature of the categories reported above is that they represent a
range of views between teacher-centred (Category 3) and student-centred
perspectives (categories 2 and 1), consistent with the research on
academics' educational beliefs and conceptions (Kember & Gow, 1993;
Prosser, Trigwell & Taylor, 1994; Samuelowicz & Bain, 1992). The
numbering system we have used implies an ordering along such a
continuum, but at best that ordering is a simplification of the complex
patterns of beliefs and practices constituting the differences between
categories. For example, part of the teacher- vs student-centred
variation is reflected in the learning framework in which the CFL is
embedded (whether it is structured, guided or facilitated by the CFL or
associated support), but that dimension alone does not distinguish
between the three major categories. The origin of knowledge is
relevant (1 vs 2 and 3) as are the ways in which the directions of
learning are managed (3 vs 1 and 2) and the way in which the learning
process is conceived (1 vs 2 vs 3). One conclusion we draw from this
is that no single continuum, like teacher-centred vs student-centred,
or Reeves' (1992) pedagogical philosophy (instructivist versus
constructivist), is likely to characterise the differences between
academics' beliefs and practices adequately. These involve a mapping
of multiple dimensions onto one, a convenient shorthand which captures
some but not all of the differences of importance4.
Another feature of the categories, one which is most apparent in the
dimensions, is that they were not formed to capture the academics'
educational beliefs (specifically), nor to capture their CFL and
related practices (specifically). Rather, the categories were formed
to represent the overall similarities and differences between the
educational contexts described in the archival documents. Thus the
categories are composites of beliefs and practices, as indeed is the
set of pedagogical dimensions proposed by Reeves (1992). From the
standpoint of positivist research this could be seen as weakening our
ability to make any claims about the relationships between beliefs and
practices. From that perspective we would require operationally
distinct sources of evidence about beliefs and practices to which some
form of relational analysis is applied (cross tabulations, for
instance). While not eschewing such an approach (we will explore it
with the data from Study 2), we also endorse the view articulated by
Marton and Svensson (1979) that it is appropriate to examine the kinds
of relationships of interest here with an 'internal relatedness'
method, one which seeks the best qualitative fit between descriptions
and relations and adjusts both until such a fit is achieved. Although
we did overtly undertake such a process in the present study, we
nevertheless have produced a descriptive framework in which beliefs and
practices are seemlessly knitted together in each category. We do not
pretend to have exhausted either the belief or practice descriptions
that may be necessary to accommodate the full range of academics' uses
of CFL, if only because of the limitations of our data. However, we do
claim that the categories give some sense of the complexity and
diversity that is involved, and how the 'parts' fit together to form
coherent patterns. In short, there are relationships between beliefs
and practices evident in our data even if the dimensions cannot be
labelled as one or the other.
If we move into a form of 'external relatedness' mode for a moment, it
may be worth noting that there are some rough correspondences between
the belief orientations reported by Samuelowicz and Bain (1992) and the
categories reported in Table 3. For example, categories 3a (which
emphasises expertise emulation) and 3b (knowledge assimilation) appear
to share features with the transmitting knowledge and imparting
information orientations respectively of Samuelowicz and Bain. Also,
category 1 of the present scheme is differentiated from category 2
partly in terms of the origin of knowledge (whether it is created by
the student, or the received wisdom of the discipline). This
distinction is the basis of the difference between supporting student
learning and changing students' conceptions in Samuelowicz & Bain's
orientations. In their sample student creation of the knowledge was
linked with postgraduate education, but in our sample it was associated
with undergraduate projects in which alternative interpretations were
acceptable if appropriately defended (for example, the project
summarised in Appendix A, Table A.1.). We regard these possible
connections between our work and Samuelowicz and Bain's study as just
that-- possibilites to be explored once we have appropriate data to
hand.
The last point we wish to make in relation to the categories is that
they do not map in any systematic way onto the types of CFL involved
(Table 3, last column). Although we have characterised the CFL in
terms of Perkins' (1991) scheme, much the same conclusion follows from
other typologies. So, what the categories represent are
teaching/learning environments in which CFL is foregrounded, not types
of CFL. This is a non-trivial point. Most of the available analysis
of CFL is focussed on the prototypical properties of the technology,
with minimal attention to its function and impact in real
teaching/learning environments. Our framework, although more 'messy',
has the potential to describe and interpret the ways in which CFL is
deployed in higher education.
The dimensions
Relative to the 12 processes proposed by Laurillard (1993) or the 14
bimodal dimensions of Reeves (1992), our 5 qualitative dimensions
appear rather frugal. This may or may not be an issue, depending on
one's purposes. The analytic procedure used for this study limits the
number of descriptive dimensions to those needed to characterise the
similarities and differences between categories, not cases. More
dimensions, and distinctions on dimensions, might be appropriate if
cases were the focus of the analysis, a prospect we encountered
repeatedly as we cycled between levels of analysis. Several dimensions
were near to inclusion, but ultimately were omitted either because they
did not operate at the category level or because some case documents
did not include sufficient information to enable a judgment to be made.
Table 4: Comparison of the dimensions of the present study with those
of Reeves (1992) and Samuelowicz & Bain (1992)
This table can be viewed at the following web site:
http://www.meu.unimelb.edu.au/staffwebsites/dkennedy/AARE/BAIN134.html
Despite differences in approach, there are some interesting
similarities between the dimensions that emerged from our analytic
procedure and the dimensions proposed by Reeves (1992). These are
summarised in Table 4 and discussed more fully below, taking each of
our dimensions in turn.
Origin of knowledge: Reeves' epistemology dimension reflects the extent
to which, on the one hand, knowledge exists separately from sentient
beings for whom learning is the acquisition of objective truth, versus
on the other hand the assumption that knowledge is a human construction
which is learnt as each individual makes sense of the world. This is
not a direct parallel of our origin of knowledge dimension, although it
is possible that the underlying presumptions may be those highlighted
by Reeves. In our data, the distinction is much more about whose
knowledge is being studied (the discipline's or the student's) than
about the nature of knowledge itself (the same distinction appears in
the content dimension of Samuelowicz & Bain, 1992). However, the
willingness of academics to adopt a postmodern approach to their
disciplines, and thus allow students to propose and defend alternatives
to received wisdom, probably depends on their basic epistemological
assumptions, to which we did not have access in most cases.
Learning framework: We have tentatively aligned our learning framework
dimension (structured, guided, facilitated) with Reeves' structure
dimension. In Reeves' scheme structure refers to the extent to which
pathways are prescribed by the program. That is part of our learning
framework dimension, especially in relation to structured frameworks,
but in many of our cases learning supports were provided in addition to
the CFL, and these varied in the ways in which they influenced
students' learning. This, therefore, is one of the circumstances
anticipated in the introduction where the learning context, CFL plus
supporting materials and requirements, has been taken into account, not
just the CFL alone.
Learning directions: Our learning directions dimension appears to be a
variant of learner control in Reeves' framework inasmuch as each is
concerned with whether the learner or the CFL controls what material is
studied and in what sequence.
Knowledge focus: There does not appear to be a workable alignment
between our knowledge focus dimension and any of Reeves' (1992)
proposals. It might be thought that his experiential value (concrete
versus abstract) dimension corresponds with the distinction between
case-based tasks (which refer to complex real world instances) and the
others, but this is an inaccurate characterisation of the distinction
we have made, which has more to do with the difference between using
theory to interpret instances (case-based reasoning) in contrast to
using examples to assist with the understanding of theory and the
development of discipline-specific reasoning. Our distinction has more
in common with Samuelowicz and Bain's (1992) nature of knowledge
dimension, although that dimension reflects whether the academic wants
students to view the world in a different way (interpretation of
reality) or to acquire a body of academic knowledge (understanding of
curriculum).
Learning process: This dimension includes six variants which vary in
several aspects in tandem: the extent to which knowledge and skill are
to be absorbed rather than constructed by the student; the extent to
which knowledge construction involves challenge more than elaboration
of ideas; and how the progression from simple to complex occurs (in
parts or as increasing refinements of wholes). We elected not to
separate these components so as to retain a global sense of each
learning process, but is is possible that some decomposition would be
profitable in later work. It is perhaps only at the level of the
component distinctions that some correspondences with Reeves (1992)
become apparent. For example, his instructional sequencing dimension
contrasts mastery of component knowledge and skills (reductionist) with
successive refinements of knowledge about the whole (constructivist).
This clearly has to do with how the progression from simple to complex
occurs. Reeves also draws on Hannifin's (1992) work to distinguish
between acquiring representations of knowledge (mathemagenic) versus
creating or elaborating representations for oneself (generative). This
user activity dimension appears to correspond with the difference
between knowledge emulation (mathemagenic) and knowledge extension and
construction (generative).
Missing dimensions: We debated at length the inclusion of two dimensions
that seemed to be 'required' on the basis of previous research but
which did not emerge from our initial analysis of the similarities and
differences between the categories: namely, whether (and how) academics
take account of students' conceptions of phenomena in their
discipline5, and the nature of the understanding and learning being
sought (cf. the students' conceptions and learning outcomes dimensions
respectively in Samuelowicz & Bain, 1992). Although the documents for
some projects would have permitted coding of these dimensions, this was
not generally the case. In other words, neither dimension was
'present' to influence the categorisation process, hence neither could
be extracted in the dimensional analysis. If nothing else, this fact
serves to reinforce our claim that, as far as possible, the framework
was not influenced by our preconceptions.
The advantages of the analysis
What then are the virtues of a grounded description and interpretation
of academics' beliefs and CFL practices if, on the one hand, the
resulting framework can be approximated by distinctions that others
have already made, and if, on the other hand, some established
distinctions are not supported? The answer lies partly in what we have
achieved thus far and partly in what remains to be done. The main
claims that we would make about our framework are that:
a) it represents what academics actually think and do when developing
and using CFL;
b) it gives weight only to those characteristics which distinguish
between broadly different approaches to teaching with CFL;
c) it incorporates not only the immediate properties and consequences
of CFL but also the supporting pedagogy; and
d) it provides an amalgam of educational beliefs and practices in a
form that, with further development, should assist us to understand the
function and impact of CFL in higher education.
On the other hand, largely because of limitations in the project
documentation, we cannot claim to have resolved the categories and
dimensions, nor have we been able to state confidently what the raison
d'etre of each category is--the 'theme' that makes the category
coherent and 'understandable'.
Where to from here?
The second study in this series is now in progress. It draws on
interviews and demonstrations as well as archival material. The sample
comprises projects tentatively categorised according to the framework
outlined in Table 3 (on the basis of CAUT documentation), including
projects from the present study. The interview protocols explore a
wide range of issues, some concerned with the CFL and its associated
teaching/learning environment, some concerned with educational beliefs
and conceptions, some with the connections between CFL development and
scholarship. We expect that our present framework will be refined as a
consequence of having more complete data with which to work. We may
also need to make adjustments once we compare the impressions gained
from an 'internal relatedness' analysis of all the data (as in the
present study), with those provided by a cross-tabulation of separately
coded beliefs and practices. Application of Reeves (1992) scheme may
also assist us to refine the distinctions we need to make. A major
addition to the present analysis will be an attempt to determine the
governing rationale of each category, much as Fox (1983) described for
his four orientations to teaching. If that level of interpretation can
be provided for each category it will not only provide a thematic sense
of why the beliefs and actions are thematically coherent, it will also
corroborate the integrity of the categories.
References
Bain, J., & McNaught, C. (1996). Academics' educational conceptions
and the design and impact of computer software in higher education. In
C. McBeath & R. Atkinson (Eds.), The Learning Superhighway. New world?
New worries?(pp. 56-59). Proceedings of the Third International
Interactive Multimedia Symposium, Perth, 21-26 January.
Ballantyne, R., Bain, J., & Packer, J. (1997) Reflecting on university
teaching: Academics' stories. Canberra: Australian Government
Publishing Service.
Boyle, T. (1997). Design for multimedia learning. London: Prentice
Hall.
Clarke, R. E. (1992). Media use in education. In M. C. Alkin (Ed.),
Encyclopedia of educational research (pp. 805-814). New York:
Macmillan.
Clark, C. M., & Peterson, P. L. (1986). Teachers' thought processes. In
M. C. Witlrick (Ed.), Handbook of research on teaching (pp. 255-296).
New York: Macmillan.
Crook, C. (1994). Computers and the collaborative experience of
learning. London: Routledge.
Dall'Alba, G. (1990). Foreshadowing conceptions of teaching, Paper
presented at the annual conference of the Higher Education Research and
Development Society of Australia, 6-9 July, Griffith University,
Brisbane.
Dills, C., & Romiszowski, A. (Eds.), (1997). Instructional development
paradigms. Englewood Cliffs, NJ: Educational Technology.
Elton, L. (1988) Managing change in the institution. In Miall, D.
(Ed.), Evaluating information technology in the arts and humanities.
Proceedings, (pp. 39-45). Birmingham: Computers in Teaching Initiative
Support Service.
Fox, D. (1983). Personal theories of teaching. Studies in Higher
Education, 8, 151-163.
Gow, L. & Kember, D. (1993). Conceptions of teaching and their
relationship to student learning. British Journal of Educational
Psychology, 63(1), 20-33.
Glaser, B. G., & Strauss, A. L. (1967) The discovery of grounded
theory: Strategies for qualitative research. Chicago: Aldine
Publishing Company.
Hannafin, M. J. (1992). Emerging technologies, ISD and learning
environments: Critical perspectives. Educational Technology Research
and Development, 40(1), 49-63.
Jonassen, D. (Ed.), (1996). Handbook of research for educational
communications and technology: A project of the association for
educational communications and technology. New York: Simon & Schuster
Macmillan.
Kagan, D. M. (1992). Implications of research on teacher belief.
Educational Psychologist, 27(1), 65-90.
Kember, D. & Gow, L. (1994). Orientations to teaching and their effect
on the quality of student learning. Journal of Higher Education, 65(1),
58 -74.
Khan, B. (Ed.), (1997). Web-Based Instruction. Englewood Cliffs, NJ:
Educational Technology.
Laurillard, D. (1993). Rethinking university teaching: A framework for
the effective use of educational technology. London: Routledge.
Martin, E., & Balla, M. (1990). Conceptions of teaching and
implications for learning. Paper presented at the annual conference of
the Higher Education Research and Development Society of Australia, 6-9
July, Griffith University, Brisbane.
Marton, F., & Svensson, L. (1979). Conceptions of research in student
learning. Higher Education, 8, 471-486.
Muldner, T. & Reeves, T. C. (Eds.), (1997). Educational
multimedia/hypermedia and telecommunications (Vol 1 & 2). Proceedings,
World conference on educational multimedia/hypermedia and world
conference on educational telecommunications. Charlottesville:
Association for the Advancement of Computing in Education.
Pajares, M. F. (1992). Teachers' beliefs and educational research
Cleaning up a messy construct. Review of Educational Research, 62,
307-332.
Perkins, D. N. (1991). Technology meets constructivism: Do they make a
marriage? Educational Technology, 31(1), 18-23.
Prosser, M., Trigwell, K., & Taylor, P. (1994). A phenomenographic
study of academics' conceptions of science learning and teaching.
Learning and Instruction, 4, 217-231.
Ramsden, P. (1993). Theories of learning and teaching and the practice
of excellence in higher education. Higher Education Research and
Development, 12, 87-97.
Reeves, T.C. (1992). Effective dimensions of interactive learning
systems. Information Technology for Training and Education Conference
(ITTE '92), (pp. 99-113). St Lucia, Brisbane: University of
Queensland.
Reeves, T. C. (1997). Established and emerging evaluation paradigms for
instructional design. In C. Dills & A. Romiszowski (Eds.),
Instructional development paradigms (pp. 163-178). Englewood Cliffs,
NJ: Educational Technology.
Rieber, L. P. (1992). Computer-based microworlds: A bridge between
constructivism - and direct instruction. Educational Technology
Research and Development, 40(1), 93-106.
Samuelowicz, K., & Bain, J. D. (1992). Conceptions of teaching and
learning held by academics. Higher Education, 24, 93-111.
Trigwell, K., & Prosser, M. (1996). Congruence between intention and
strategy in university science teachers' approaches to teaching.
Higher Education, 27, 1-11.
Trigwell, K., Prosser, M., & Taylor, P. (1994). Qualitative
differences in approaches to first year university science. Higher
Education, 27, 75-84.
Wills, S., & Mc Naught, C. (1996). Evaluation of computer-based
learning in higher education. Journal of Computing in Higher Education,
7(2), 106-128.
Appendix A: Some illustrative cases
Table A.1
Illustrative Case of Category 1
(Subject: History, Education and Computer Studies, Griffith University)
Origin of Knowledge - Student/Collaboration
Students are presented with a database consisting of primary source
material, video footage, photographs, newspaper articles, etc. They
are able to manipulate and transform data in the process of generating
new knowledge. They are challenged to apply their own hypotheses and
demonstrate the validity of their own interpretation.
Learning Framework - Facilitated
The CFL supports a historical inquiry approach to teaching, with the
CFL offering an alternative to the expound and discuss method through
direct use within lectures and individual and small group use by the
students. Students are able to actively explore different perspectives
about what it means to be an Australian, they are able to shift the
focus of their investigations to construct their own knowledge
representations. The CFL is claimed to combine information transfer
with constructivism. At the time of the final report the program had
not been incorporated significantly into any specific course.
Learning Directions - Student-Managed
The nature of the interactivity allows students to determine their own
pathways through the material. The learner is given direct control of
the processing of aural, visual and written material.
Knowledge Focus - Conceptual/Procedural reasoning
Students are constantly having to make decisions, evaluate their
progress and apply higher order reasoning skills. The constant
evaluation of information helps students develop the analytic skills
necessary to think critically about information sources.
Learning Process - Construction/Challenge
Dominant cultural perspectives of historical events are challenged by
the presentation of alternative interpretations of events. Students are
encouraged to deconstruct received narratives through active
manipulation of video and audio sequences and actively reconstruct
their own versions of events.
Table A.2
Illustrative Case of Category 2a
(First Year Statistics For Behavioural and Social Sciences, La Trobe
University).
Origin of Knowledge - Academic/Discipline
The course is designed to teach a number of key statistical concepts
which have an accepted discipline interpretation and application.
Examples include normal distributions, the central limit theorem and
confidence intervals.
The Learning Framework - Facilitated
The flexible design of the CFL allows teachers to choose different
emphases and orders of presentation within different curriculum
frameworks. It can be used as a teaching aide in lectures, by groups
of students or by individual students. Students are provided with the
opportunity to explore ideas and use their own initiatives in knowledge
construction. Certain print and on-line materials do provide some
support and guidance for students. The CFL consists of linked
microworlds or playgrounds within which students are able to explore
data representations, inter-relationships between summary statistics
and sampling distributions. Linked multiple representations are
provided.
Learning Directions - Student-Managed
The CFL enables a high degree of interactivity. Students are
encouraged to actively explore the concepts and test intuitions within
a game format. Printed activity sheets guide activities within each
playground. In the sampling playground, all aspects of sampling are by
user choice. Distributions can be manipulated by the user using
on-screen handles and students are able to actively explore different
relationships between certain statistical measures. There is a
distinct sense in the materials provided that the developers would
encourage active manipulation and exploration by the students and the
CFL is designed to allow maximum scope for this.
Knowledge Focus - Conceptual/Procedural knowledge
The software and associated learning materials were designed to "assist
students to build correct and strong understanding of the target
concepts."
Learning Process - Elaboration/Challenge
The developers are aware of basic misconceptions in probability and
statistics and target these areas of misconception using a computer
simulation approach. Multiple-linked representations are provided and
student explorations, appropriately framed, are encouraged. The
intended outcome is a better understanding of the concepts involved.
Some simple experiments which students can perform allow certain
misconceptions to be confronted and more appropriate conceptions to be
built.
Table A.3
Illustrative Case of Category 2b
(Physiology, University of Sydney)
Origin of Knowledge - Academic/Discipline
The course integrates knowledge of the basic sciences (physics,
chemistry and biology) and professional disciplines with clinical
situations. Thus, the underlying knowledge base has a received
discipline interpretation.
Learning Framework - Guided
The CFL is based upon a constructivist approach to learning with tasks,
resources and materials provided to lead students to construct their
own knowledge. It uses a problem- based approach to facilitate the
integration of knowledge from the basic sciences with clinical
situations and to enable students to see how the various professional
groups and clinical disciplines interact to solve given problems. The
CFL is used in conjunction with tutorial group discussions, student
collaborative learning groups and practical classes. Knowledge is
"anchored in authentic situations and activities". Examples of
clinical reasoning or thinking processes of practitioners are provided.
Tutorial exercises encouraging the use of 'patient profiles' are also
available. The 'patient' can be examined by a number of different
health professionals which provides the student with multiple
perspectives of one clinical case.
Learning Directions - Student-Managed
Students are given freedom to work through the case study and resources
in any order they wish at their own pace. There is no pre-determined
pathway for a student to follow. The developers claim a
student-centred approach and value student control and ownership of
learning.
Knowledge Focus - Case-Based Reasoning
Knowledge from different disciplines and professional subjects is
integrated through use of the case study format. Examples of the
thinking patterns of professionals are provided with the intention that
students will learn how to generate and test hypotheses in the clinical
setting. The skills in self-directed learning and problem solving are
said to be essential competencies of health professionals.
Learning Process - Synthesis and Elaboration
Solution of the problem requires the application and integration of
knowledge from the various scientific and professional disciplines. It
is hoped that through this process students will be able to perceive
the "integrated nature of the different disciplines" and their
relevance to clinical problem solving. Animations and video clips of
procedures and skills are available as part of the learning module.
Diagrams, text, examples and analogies facilitate mastery of key
concepts.
Table A.4
Illustrative Case of Category 3a
(First year Chemistry, University of Melbourne)
Origin of Knowledge - Academic/Discipline
The scope of this project spans an entire first year chemistry course.
A set of key concepts form the basis for the objectives which define
the course aims and content.
Learning Framework - Structured
The CFL is integrated into a traditional lecture and laboratory
program which consists of 3 lectures, 3 hours of interactive CFL
tutorials, 1 hour face-to-face tutoring in smaller classes (includes a
short answer test) and one 3-hour laboratory class per fortnight. The
program is clearly set out to ensure that the content of the tutorials
is keyed into the previous week's lecture thereby reinforcing the key
ideas and focussing student attention on them through questioning.
Pre-lab CFL tutorials prepare students for the practical and help to
familiarise them with correct laboratory procedures. Each interactive
tutorial comprises a number of cards which follow a pre-determined
sequence, initially stating the objectives of the tutorial, providing
information which could include the use of animations, videos or still
pictures to help illustrate a concept, posing questions and providing
feedback or detailed explanations depending on the students' response
to a question. Hints are available if needed by the student to answer
a specific question.
Learning Directions - Teacher-Managed
The course is highly structured. A set printed notes specifies the
topics to be covered in the lectures and the tutorials each week.
Within the CFL itself the tutorials follow a pre-determined sequence
which the student must follow and the questions asked are closed in
nature.
Knowledge Focus - Conceptual/Procedural knowledge
The objectives upon which the course is built focus on specific key
concepts in chemistry and associated laboratory procedures. A
computer-based diagnostic test is given at the start of the process to
ascertain the students' existing knowledge base and allow corrective
measures to be planned if necessary. The test items in the CFL require
students to concentrate on specific course content and laboratory
procedures.
Learning Process - Emulation
Students are provided with detailed explanations of key concepts
together with appropriate illustrations in lectures. This is supported
by the interactive tutorials and laboratory sessions. The content
presented in the lectures is tested in the interactive tutorials and
the short answer test given every fortnight. The teach-test nature of
the course suggests that the intention is for students to be able to
emulate the received understandings with which they are presented.
Endnotes
1 We are using the term 'orientation' to refer to a coherent set of
beliefs, 'belief' to refer to a disposition to interpret, value and act
in particular ways, and 'conception' to refer to a way of experiencing
or interpreting of a phenomenon (such as learning or teaching).
Samuelowicz and Bain (1992), rather unhelpfully, used the term
'conception' to refer to all of these possibilities.
2 Prosser, Trigwell and Taylor (1994) conducted their research from a
phenomenographic perspective which seeks to reveal conceptions rather
than belief dispositions.
3 We are grateful to the CAUT secretariat for access to the project
applications and final reports. We also wish to thank the academics
involved for permission to include their projects in our research
analyses and publications.
4 This is not to deny that Reeves (1992) proposed pedagogical
philosophy as but one of 14 non-exhaustive dimensions. However, we
would argue that pedagogical philosophy (teaching/learning orientation)
consists of multiple component beliefs about educational purposes and
practices, some of which are accessed by Reeves' other dimensions, some
not. The unidimensionality of his pedagogical philosophy dimension is
thus a convenient simplification.
5 In Bain & McNaught (1996), for example, we distinguished between CFL
that ignores students' understandings, CFL that seeks to preempt
difficulties by providing more effective ways to represent ideas, and
CFL which while achieving the latter also allows students to play with
the ideas and have their understandings probed and stretched through
conversation (Laurillard, 1993; Crook, 1994).
16
19