Lmmmy und Insrrucrion. Vol. 7, No. 3, pp. X5-275, 1997 0 1997 Elxvier Science Ltd

All nghts reserved. Printed in Great Britain 0959.4752/97 tl7.00 + 0.00

Pergamon

P1I:S09594752(96)00028-X

A RECONCEPTUALISATION OF THE RESEARCH INTO UNIVERSITY ACADEMICS’

CONCEPTIONS OF TEACHING

DAVID KEMBER

Coordinator of the Action Learning Project, Hong Kong Polytechnic University, Hung Horn, Kowloon, Hong Kong

Abstract

Articles about the conceptions of teaching of university academics are reviewed. There is found to be a high level of correspondence between largely independent studies reported in I3 articles. An attempt to synthesise the body of research places conceptions under two broad orientations characterised as teacher-centred/content-oriented and student-centred/learning-oriented. Each orientation has two associated conceptions. A transitionary category, labelled student-teacher interaction, links the two orientations. The five conceptions beneath the orientations are visualised as well-defined points within a continuum, and there is some evidence of lecturers shifting beliefs across the spectrum over time. Teaching conceptions have been shown to be related to measures of the quality of student learning, so are modelled as influencing teaching approaches which in turn effect student learning approaches and learning outcomes. It is argued that measures to enhance the quality of teaching should take account of teaching conceptions if they are to be effective, as teaching approaches are strongly influenced by the underlying beliefs of the teacher. 0 1997 Elsevier Science Ltd.

Research Into Beliefs About Teaching

Around the early 1990s several groups of researchers were working quite independently to examine the beliefs about teaching of university academics. Several of the groups had noted that research into student learning had established a relationship between student conceptions of learning, learning approaches and learning outcomes. The search for a parallel relationship between lecturers’ conceptions of teaching, teaching approaches and possibly student learning approaches and outcomes appeared to be a logical development.

There had been earlier research into the beliefs about teaching of school teachers (see Clark & Peterson, 1986; Pajares, 1992, for reviews), but many university lecturers consider themselves a breed apart from school teachers. Indeed many university academics hardly consider themselves “teachers” at all, instead visualising themselves more as a member of their discipline (Becher, 1989). Universities also operate under quite different value systems and traditions from schools. While some elements of the school-based research were likely to be applicable to university beliefs, it was probable that there would be dimensions which were not apparent in school teachers.

255

256 D. KEMBER

Papers from the various groups appeared mainly in the period 1992-l 994. As the groups had worked independently, and given the timescale for work to be published, there was little opportunity for comparison of results in the published papers. It is now possible to review the outcomes to look for commonality.

Literature for this review was initially accessed through contact with some of the other research groups, and a more systematic search was subsequently made through the ERIC index. Searches were made on terms such as conceptions of teaching and beliefs about teaching. It is possible that there are further articles on this topic which have not been identified. The terminology used to identify studies on this topic is not well defined and accepted. The ERIC index is not comprehensive, so conference papers or book chapters, for example, might have been missed.

Efforts have been made to track down what has been written, but if the odd article has been missed this is not seen as problematic. The aim of this article is not. to produce a comprehensive review of everything which has ever been written on this topic. It is rather to analyse and compare the findings of a substantial number of largely independent studies to see what commonality there is in the findings.

Anticipating the latter parts of the paper, I claim that there is a high degree of com- monality. Categories used to describe conceptions of teaching show a high degree of cor- respondence across the research studies. Given this finding, the paper became an attempt to synthesise a set of descriptive categories which would satisfy the findings of all studies.

The Studies

The search located 13 studies, which are summarised in Table 1. A paper by Biggs (1989), which categorises conceptions of teaching, was also discovered. This, though, was not included, as the descriptions appeared to have been generated from theory rather than empirically derived.

All of the samples were drawn from university academics except that of Pratt (1992) which focuses upon adult educators. This study was included, though, as at least some proportion of the sample were university academics and adult students are commonly found in higher education. The studies were located in diverse countries, but Australia was particularly strongly represented.

Terminology

Before proceeding further it is necessary to mention terminology. Pajares (1992) argues that lack of clear definition and inconsistent use of terminology has been a major impedi- ment to progress in research into the beliefs of school teachers. Several terms have been used by the various authors including orientations, conceptions, beliefs, approaches and intentions, but few of the studies give a definition of the terms used.

The most commonly used term in the papers is conceptions of teaching. Pratt (1992) is one of the few to define this term, so his definition is given below.

Conceptions are specific meanings attached to phenomena which then mediate our response to situations involving those phenomena. We form conceptions of virtually every aspect of our perceived world, and in

ACADEMICS CONCEPTION OF TEACHING

Table 1 Research into Conceptions of Teaching

251

Author(s) Date Sample Location Comments

Dali’ Alba 1991

Dunkin 1990 1991

Dunkin & 1992 Precians

Fox 1983

Gow & Kember 1990 1994

Martin & BaRa 1991

Martin & Ramsden

Pratt

1992

1992

Presser et al. 1994

Samuelowicz 6% 1992 BaitI

Trigwell et al. 1994

20 teachers in 4 subject areas

Australia (?)

55 new lecturers Sydney

12 award winners Sydney

?

39 lecturers

13 staff enrolled in course for academics

13 staff enrolled in course for academics

253 adults and teachers of adults

24 teachers of 1st year physics and chemistry

13 academics in science and social sciences

24 teachers of 1st year physics and chemistry

U.K. Develops metaphors

Hong Kong Developed questionnaite for larger sample

Melbourne

Canada, China, Hong Kong, Singapore, U.S.A.

Australia

Australia and U.K. University was U.K. distance teaching one

Australia

5 case studies followed, 3 presented

Concentrates on adult education

Focus on conceptions of science teaching

Concentrates on intention and approach

so doing, use those abstract representations to delimit something from, and relate it to, other aspects of our world. In effect, we view the world through the lenses of our conceptions, interpreting and acting in accord- ance with our understanding of the world (p. 204).

This definition seems to be largely consistent with the meaning attached to’the term by the other authors. Belief is a less commonly used term, but where it is used the meaning seems to be largely synonymous with the above definition of conceptions.

Orientation is taken as a broader level of categorisation encompassing two or more conceptions. This relationship utilises the common conceptual model of an inverted tree with main categories having a number of subordinate categories. An analogy to quantita- tive research would be the relationship between second- and first-order factors.

The term teaching approaches has obvious parallels with the now widely used term student learning approaches. These have been characterised as having motive and strategy

258 D. KEMBER

components (Biggs, 1987) which presumably influenced Trigwell, Prosser, and Taylor (1994) to analyse teaching approaches in terms of strategies with associated intentions. Pratt (1992, p. 206) sees a similar set of elements in “a dynamic and interdependent trilogy of Actions, Intentions and Beliefs.”

Method

All of the studies used a qualitative approach for gathering data. One study (Gow & Kember, 1993; Kember & Gow, 1994) developed a questionnaire from the original qualita- tive data.

Interviews

The research all appears to have been conducted under an open naturalistic framework so there were no preconceived hypotheses of conceptions of teaching. Rather, the research- ers expected the descriptions or classifications of teaching conceptions to emerge from the data. The aim was to report the perceptions of the teacher rather than impose the researcher’s framework. The reported outcomes would, therefore, claim to be from a second-order perspective and not a first-order researcher perspective.

This position governed the nature of the interviews. To allow the interviewees position to emerge, the questions needed to be open-ended within a semi-structured framework. The interviewer would use an initial question with a follow-up probe if necessary. Some of the papers describe the interview as resembling a conversation.

Much of the questioning seems to have been indirect, although Samuelowicz and Bain (1992) used a direct question, “what is teaching,” and the remainder of the interview consisted of indirect questioning. Partly as a result of the indirect nature of the questions, most of the interviews were not strictly confined to conceptions of teaching but addressed teaching and learning more widely. Some studies, such as Prosser, Trigwell, and Taylor (1994) did this explicitly to examine relationships between aspects of beliefs about teaching and learning. Pratt (1992) assumed that actions, intentions and beliefs were related so asked about these three dimensions of teaching. As a result of the indirect questioning and the breadth of the interviews they were quite long. The reported lengths were in the range of 30-90 minutes.

Analysis

All studies, except possibly that by Fox (1983), tape recorded the interviews and produced full transcripts for analysis. Analysis followed the standard qualitative research technique of dividing the identified beliefs into a limited number of categories. The number of identified conceptions ranges from four to nine.

Consistent with a naturalistic framework, the analyses aimed for categories to emerge from the data using a grounded approach (Glaser & Strauss, 1967). As a methodological

ACADEMICS’ CONCEPTION OF TEACHING 259

safeguard the majority of the studies report independent analysis of part of the data by two or more researchers and an iterative process of identifying and defining categories.

As discussed below, the element of commonality, between studies, within the identified categories lends credibility to the analyses. There are, though, issues concerning the deline- ation of categories, the relationship between them and the use of other dimensions to define categories over which there are divergencies within the analyses. These issues are discussed in the following sections.

Comparison of Categories

Comparison of the labels and descriptions given for the categories indicates a high degree of commonality in the categories. Precisely the same wording or terminology may not be used, but it seems reasonable to interpret categories as common if the descriptors appear to have the same or similar meanings. The finding that a number of largely independent studies has a high degree of consistency in identified categories considerably strengthens the credibility of the research.

Figure 1 presents an attempt to compare the conceptions of teaching identified by the individual studies. The organising framework for the figure is based on the relationships between the teacher, students and the content. The conceptions identified in the various studies are arranged under a framework ranging from a teacher-centred/content-oriented pole to a student-centred/learning-oriented pole.

Vertical relationships in the figure between conceptions indicates perceived similarities between the categories for describing beliefs derived in the studies reviewed. Direct equivalence is not necessarily implied as reducing the figure’s elements to the author’s category label inevitably results in the loss of the subtlety of the accompanying description. There is obviously an element of subjectivity in judging the equivalence of described categories and compiling such a table. Some of the authors themselves, though (e.g., Prosser et al., 1994; Samuelowicz & Bain, 1992) have remarked upon the relationship of their categories to those of some subset of the other authors’ work in the figure.

A discussion of the characteristics of the categories of description is given after the sec- tion which attempts to synthesise a framework of category descriptions derived from the work reviewed in Fig. 1. By first synthesising a common model, it is then possible to use illustrative quotations given by authors to illuminate its elements.

There is one identified conception in the study by Pratt (1992) which does not fit on the continuum because the focus is an ideal which over-shadowed the elements of the teacher, the student and the content, chosen for the framework in Fig. 1. Pratt discovered a variety of beliefs such as religious doctrines or political ideologies which underpinned this concep- tion. In each case, though, the strength of conviction led to the ideal dominating any other element in the teaching process. Pratt’s study was with adult educators and it seems unlikely that this conception would be common in universities. Basing teaching upon a single fervently held ideal would normally be seen as inconsistent with the goals of universities, founded on Western models, which stress critical thinking and encourage the plurality of viewpoints.

260 D. KEMBER

TEACHER -CENTERED I CONTENT-ORIENTED

devebping capacily

to be expefi

STUDENT-CENTERED / LEARNING-ORIENTED

expbring WYS

Of unders&mdilg

Figure 1. Comparison of the categories of conceptions of teaching.

ACADEMICS’ CONCEPTION OF TEACHING 261

Relationship Between Categories

Pratt (1992) seems to see no ordering to the five conceptions he identifies. Fox (1983) orders his four categories in a 2x2 matrix delineated by simple/developed theories and content/student-centred orientations. The remaining studies present their categories in a linear sequence consistent with that presented in Fig. 1. There is also an implication, which is often made explicit, that the conceptions towards the student-centred end of the continuum are superior.

Dimensions

Qualitative and phenomenographic research has a methodological issue in determin- ing how many categories exist and establishing demarcations between categories. When multiple categories are proposed, are they discrete entities or shades of adjoining categories or sub-categories of hierarchically superior categories? There appears to be no firmly established procedures to provide answers to these questions, akin to those used in quantitative research, such as factor analysis and testing for significant differ- ences.

Several authors rely on substantiating their categories with quotations from interview transcripts. This approach can lead to substantial numbers of categories, but the distinc- tions between them can be quite subtle phrasing within the category description and exemplar quotation. Whether the interviewees would themselves have recognised these minor distinctions in wording is often questionable. In several cases the terminology for adjacent categories appears to be plausibly interpreted as within the normal variations of alternative phrasing in everyday speech. In interpreting categories it should be borne in mind that even educational researchers often do not define their terminology as rigor- ously as is the case in many other disciplines, and other academics are not a party to the conventions which are established.

Pratt (1992) demarcated categories conceptually by showing that each of his five categories showed a different relationship between educator, students, content and in some cases the ideal to be taught and the context in which it was taught. Each category is illuminated with a diagrammatic representation of the relationship between these ele- ments. Fox (1983) distinguishes his categories and sub-categories with delightful metaphors. The study which moved on to develop a quantitative instrument (Gow & Kember, 1993; Kember & Gow, 1994) was able to use factor analysis to confirm the qualitatively derived categories.

The most common approach to category definition has taken the form of revealing dimensions which distinguish the categories (Fox, 1983; Martin & Ramsden, 1992; Samuelowicz & Bain, 1992). Again there are major elements of commonality between the studies reviewed. Table 2 presents dimensions which can be used to identify the five conceptions included in the synthetic model developed later in this paper. These dimen- sions are largely derived from the descriptions of the conceptions.

262 D. KEMBER

Table 2 Dimensions Used to Delimit Conceptions of Teaching

Dimension

Imparting information

Transmitting structured knowledge

Teacher- student interaction

Facilitating understanding

Conceptual change

Teacher presenter

Teaching transfer of information

Student passive recipient

Content defined by curriculum

Knowledge possessed by lecturer

presenter

transfer of well structured information

recipient

lecturer needs to order and structure material

possessed by lecturer

presenter and tutor

interactive process

participant

defined by teacher

discovered by students but within lecturer’s

facilitator

process of helping students to learn

lecturer responsible for students’ learning

constructed by students within teacher’s framework

constructed by students

change agent/ developer

development of person and conceptions

lecturer responsible for student development

constructed by students but conceptions can be changed

socially constructed

Ordered or Hierarchical?

There are differing views as to the relationship between the categories of conceptions. Biggs (1989), Dall’Alba (1991) and Martin and Balla (1991) use the term hierarchical in discussing the relationships. Samuelowicz and Bain (1992) suggest that a hierarchical order- ing implies that characteristics present in lower order categories are present in all other categories. It is not always clear that the other authors intended to use the term in this way, but if this definition of hierarchical is accepted, then there are findings in the literature which do not seem to be consistent with hierarchical ordering.

Samuelowicz and Bain (1992) compare quotations from two teachers at opposite ends of the spectrum about five dimensions used to delineate conceptions. None of the five quota- tions from the higher conception contains any element present in the corresponding one from the lower conception. The two sets of quotations are better interpreted as qualitatively different categories.

Martin and Ramsden (1992) developed case studies of lecturers in a course on teaching and learning at tertiary level. They believed that the development of the lecturers was characterised by changes in their conceptualisation of the processes of teaching, the nature of their subject matter and the ways students learn. The lecturers were developing through stages corresponding to qualitatively different conceptualisations.

ACADEMICS’ CONCEPTION OF TEACHING 263

McKay (1995) produced a case study of a major curriculum change within an academic department which resulted in significantly better outcomes with respect to the quality of student learning. The changes were contingent upon the development of a cohesive identity within the department and changes in beliefs about teaching by several members of the department. The difficulty of and the time taken for this process are indicative of conceptual change akin to that described in the literature on science concepts (e.g., Champagne, Gunstone, & Klopfer, 1985; Strike & Posner, 1985).

The cited research suggests that the alternative conceptions of teaching are better regarded as an ordered set of qualitatively differing conceptions. When change does occur, lecturers seem to move from one belief to another, and do not retain all elements of previ- ous beliefs. Further, that development or progression through ordered categories does hap- pen but is unlikely to be a rapid or easy process.

Modelling Category Relationships

The studies by Martin and Ramsden (1992) and McKay (1995) which show lecturers changing or progressing through a series of conceptions of teaching, question to some extent the characterisation of conceptions as a set of discrete categories. A significant proportion of the authors specify their methodology as phenomenography. Marton (1981, 1988) describes this method as searching for a limited number of categories of description of the subjects’ perception of a phenomenon. Others, such as Samuelowicz and Bain (1992), describe their method in terms of a search for emerging categories.

Establishing categories is a useful analytic technique for reducing a mass of data or individual descriptions to more readily understood generalisations. Quantitative research has its parallels in techniques such as factor analysis. There are, however, other methods for reducing or analysing qualitative data and displaying the outcomes in forms which are readily understood. Miles and Huberman (1984) discuss a wide variety of forms for presenting results, while Eisner (1991) eloquently advocates the case for other quite differ- ent descriptive outcomes.

If a researcher sets out to discover categories of description, it must come as no surprise that the results appear as categories of description. There must remain the possibility, though, that the phenomenon could be better explained by some alternative approach to presenting the outcomes. It seems unlikely that all scenarios are best understood by the reader if portrayed as a list of categories in hierarchical order.

Categories need to be characterised as distinct from each other which creates the, perhaps unintended, impression that there are rigid and well-defined boundaries between them. There are several threads of evidence to suggest that this is not the case with concep- tions of teaching. Most of the studies reported that individuals could display evidence of more than one conception. Martin and Ramsden (1992) have case studies of lecturers developing through conceptual positions at different rates. Research into changing students’ conceptions of scientific phenomena characterise cognitive change in terms of advances, retreats and periods of indecision (Strike & Posner, 1985).

The categories of conceptions of teaching are perhaps, then, better portrayed as well- established positions within a continuum. Indeed, Prosser et al. (1994) also talk of a

244 D. KEMBER

Figure 2. A multiple-level categorisation model of conceptions of teaching.

“spectrum” with two ends, while Samuelowicz and Bain (1992) state that there is “broad agreement that these conceptions can be arranged on a continuum” (p. 93).

It is unlikely, though, that all transitions are equally easy or boundaries similarly blurred since there is also evidence of higher level categories. The strongest evidence comes from the broad orientations which emerged from the factor analysis in the Gow and Kember (1993) study. Prosser et al. (1994, p. 228) also talk of “two strongly contrasting subsets. . . The first might be described as a transmission and acquisition category and the second a conceptual development and change category.”

Rather than listing conceptions, implying a single-level structure, it would be more revealing to portray them in two levels of categories. This description is akin to a first- and second-order factor structure.

Figure 2 shows a multiple-level categorisation model which also takes into account the boundaries between the categories. The model posits two broad higher level orientations labelled teacher-centred/content-oriented and student-centred/leaming-oriented. Subordinate to each orientation are two conceptions. The boundary between each pair of conceptions is shown as diffuse, implying a relatively easy development across each pair. Transitions between the two orientations are envisaged as requiring a more significant change. A fifth intermediate conception, in which teacher-student interaction is first recognised as neces- sary, is included as a transitionary bridge between the two orientations and their subordinate conceptions.

Categories for Conceptions of Teaching

The conceptions are placed under two broad orientations. The first orientation is teacher-centred and focuses upon the communication of defined bodies of content or knowledge. The second orientation is student-centred and hence focuses towards the students’ learning. The latter orientation takes a developmental approach towards students and their conceptions of knowledge. It focuses upon their knowledge rather than the lectur- ers.

Each orientation is divided into two subordinate conceptions. A transitional or intermediate conception provides a link between the two orientations and their associated conceptions.

ACADEMICS’ CONCEPTION OF TEACHING 265

Imparting Information

The most teacher-centred conception views teaching as purely presenting information. The student is viewed as a passive recipient of a body of content, if indeed the student appears in the vision at all.

It is common to see teaching defined in this way and several of the papers have interview quotations which virtually paraphrase the conception title.

Pass information onto students, that is the major goal (Kember & Gow, 1994, p. 64). [teaching] means ihat you’re imparting information that they [students] are expected to get because they are enrolled in that course (Samuelowicz & Bain, 1992, p. 101).

Those in this category describe their teaching as lecturing and focus upon the notes they prepare.

I’ll write my notes in such a way so that the students don’t have to decide when to take notes. I’ll tell them to. I’ll dictate to them. I have handouts prepared. I have gaps in them that they fill in and I take that decision away from the students about when and how to take notes (Trigwell et al., 1994, p. 80).

It is not just researchers, but students too, who have observed this conception.

They try to stuff us with information. Employers have complained that we are like robots, lacking in independent thinking and disinterested in the things happening around us (Gow & Kember, 1990, p. 320).

As the focus is upon the lecturer and the knowledge of the lecturer, a good teacher is considered to be one with sound academic knowledge.

knowledgeable, they must be sound. They must convey their field very well. And they must be able to teach, that is, to convey what they have learned to the students. I mean they can’t just go to the lecture and talk nonsense (Kember & Gow, 1994, p. 64).

Of Fox’s (1983) metaphors the broadcast variant of the transfer theory best illustrates this conception.

views teaching as scattering seeds to the wind rather than transferring them to specific containers, All that is required of a teacher is that he deliver himself of his nuggets of wisdom. Whether or not these are relevant or applicable in particular contexts or whether they make any sense to anybody but himself is not his concern. His responsibility is solely concerned with ensuring the purity of the seed (p. 153).

In its extreme form the student hardly enters into the vision of this conception at all. At the most the student is a vessel into which knowledge is poured. Martin and Ramsden (1992, p. 6) observe that one of the lecturers in their course started from this position.

Getting information into their heads Filling in on what’s not there Pouring over it till they know it.

Transmitting Structured Knowledge

The other conception still focuses upon transmitting knowledge but recognises the importance of structuring and arranging the presented knowledge so that the student has more chance of receiving the information. One of the lecturers used as a case study by Martin and Ramsden (1992) envisages the process as developing a picture. Another saw the importance of structuring the material which was being taught.

If they could see the picture even just once, they would be helped. . They have to have an image of where they are going. Once they can see the pathways.

266 D. KEMBER

You learn the theory that underpins the systems and that helps you to see what you’re doing when you’re working with the systems. It’s most helpful if it’s properly structured. There should be some concepts which are fundamental and those have to be dealt with first. If you can set up the structure of that content then it shouldn’t be the problem I found it, for instance. It’s a structured body of knowledge that models the operation of computer programs (p. 8).

The student enters the picture more but is still a passive receiver. The lecturer does at least recognise that there is a receiver out there and does need to catch what is thrown. The ball, then, needs to be thrown carefully and accurately so that it can be caught.

Teaching is transmission of concepts and skills in such a way that students can acquire them that sounds a very rudimentary sort of approach, but I think there is a body of knowledge and skills that students need to start off with (Samuelowicz and Bain, 1992, p. 101).

Sound academic knowledge is still the most important attribute of a good teacher, but there is now more emphasis on the quality of the presentation which can be viewed as a stage performance.

I have a set of lecture notes that’s my safety belt and sometimes I stick very closely to those and other times if the circumstances are right when things click I leave my lecture notes well behind and wander off and march up and down and make jokes fly free so to speak (Samuelowicz & Bain, 1992, p. 101).

Fox (1983) describes the devotees as “conscientious transferrers” who:

spend a great deal of time preparing the material and making sure that it is accurate and up-to-date. Some of them also go to great lengths to develop and refine their method of transfer and they often devise elaborate teaching aids to inject the essence of their subjects accurately into the heart of the container (p, 152).

The importance of arranging the material in a logical manner is recognised and there may be a need to simplify material so that it can be understood by the student. For this aspect of the presentation Fox uses a baby food manufacturing analogy: “the teacher sees his job as one of processing very tough material into more easily digestible nutrient for rather simple minds” (p. 153).

Student-Teacher Interaction

The transitional conception is characterised by a realisation that interaction between teacher and student is important. The following quotation indicates a developmental proc- ess. The current conception of the lecturer appears to have reached the next stage, but clearly went through this transitionary category to reach it.

Initially, I basically talked and they listened - that type of attitude. Now I am trying to get much more, you know, they talk and I listen. And I am there as a guide -to guide them, not to force something down their throats (Kember & Cow, 1994, p. 63).

A further quotation illustrates this position in the context of a science practical class.

But I don’t want the students to take everything I say at face value. I like them to think for themselves so I try to get them to interpret before I tell them, if possible. So if I do a demonstration, I won’t tell them what the result will be I often ask them to predict the result, having given them the principles (Trigwell et al., 1994, p. 80)

This quotation also provides insight into a further facet of this transitionary conception. The lecturer making this statement wants students to examine evidence, but within a body

ACADEMICS’ CONCEPTION OF TEACHING 261

of knowledge and content defined by the lecturer. Even within the short quotation, it is possible to sense a tension between not taking everything at face value and telling them the (right) outcome.

Those in this position face a shifting of ground from focusing almost exclusively on the teacher to recognising the importance of the student, which is manifest in a degree of interaction. This is accompanied by a degree of erosion in the paramount position of the lecturer’s knowledge base, towards a recognition that student understanding and discovery are also significant.

It is not difficult to appreciate the tension which could accompany these shifts. Mezirow (1981) believes that major perspective transformations can only take place if one has become aware of the underlying unconscious assumptions which guide our practice. A precursor to making the transition would normally be a “disorienting dilemma.”

Three studies identify conceptions which see the student as needing to be shaped towards a particular role or form of expertise - the teacher thus has the role of model or expert. It is possible to view these categories as a further aspect of the one intermediate position. Pratt (1992) envisages this conception as a fusion of the teacher and the content - the teacher signifying the knowledge and content to be learnt. There is then a greater stress on learning activities such as experiments or problem solving classes as the student needs to learn how to perform like the master. Dall’Alba (1991) calls this conception developing the capacity to be an expert and illustrates this aspect with the quotation below (p. 295).

One of the barriers you would want to break down is the barrier which says all the time there are authors and there are critics. But never I am an author, I am a critic and as a reader, as a writer, I’m part of this ongoing process And for this reason I proposed a course in which the writing of texts of all sorts is an important element they are involved week by week in being authors and critics themselves.

Facilitating Understanding

Under the student-centred/learning-oriented umbrella, the focus moves away from the teacher towards the student. The role of the teacher shifts towards that of helping the student to learn. The emphasis is on student learning outcomes rather than upon defining content.

There are two or perhaps three conceptions within this orientation. The first conception has a variety of labels but appears consistently in virtually all of the reviewed studies. The common elements are facilitating the development of understanding or conceptions of knowledge, so the label learning facilitation is used.

The role of the teacher is recognised by those holding this conception. They accept that they do have a responsibility towards students learning and that they can influence outcomes. Teaching becomes a process of helping students towards desirable outcomes.

You’ve got to be able to make an environment where students really want to learn. If you do that, they are much more likely to understand why they learn. And then I think after that, the teacher should be a resource person, generally to guide the students, I don’t see it as spoon-feeding (Kember & Gow, 1994, p. 63).

There is a greater understanding of the student displayed by the teacher. They are recognised as individuals rather than as an audience to be lectured to. The teachers recognise that students may not interpret what they were saying in the intended way.

268 D. KEMBER

I’m aware of how much I used to assume. I now try to take nothing for granted and to question my assumptions about what students know and how they see things. Also I don’t expect my sessions to have the effect I intend. I’m looking for evidence of learning before I assume it. I try to be in a position where I expect the unexpected response (Martin & Balla, 1991, p. 301).

The outcome of the teaching process is understanding. The student demonstrates this by applying the knowledge, rather than through regurgitation.

they [students] have learned because they could apply it in some real situation and make sense of it (Samuelowicz & Bain, 1992, p. 100).

Conceptual Changellntellectual Development

Most studies also have a more advanced conception within the student-centred/learning- oriented orientation. These are shown in two columns in Fig. 1 as two distinct labels are used. It is interesting that only the work of Samuelowicz and Bain (1992) has conceptions in both columns. This perhaps suggests that these two identified conceptions are alternative aspects or descriptions of one position. Changing student conceptions is not an easy proc- ess and perhaps needs the establishment of a sympathetic and supportive environment. In this review the two conceptions are, therefore, treated as different facets of the one concep- tion.

The first facet focuses upon changing student conceptions. [Conceptual understanding is developed] by arguing about things, and trying to apply ideas, and being again confronted by differences between what you think and what actually happens . lo get people to make predictions about what’s going to happen, maybe they might backtrack and revise their ideas about things . what’s going on in their heads What we’re trying to achieve in learning physics, is for people to shift their view from the layperson’s view, to what we would call a scientific/physicist’s view view of the world. I think that’s what I’m on about (Prosser et al., 1994, p. 225). What I want to achieve with [these techniques] is confronting students with their preconceived ideas about the subject, which quite often conflict with what we are talking about - the official dogma as it were. So you’ve got to bring out that conflict and make the people aware that what they already know may not be what is the official line (Trigwell et al., 1994, p. 8 I).

The other envisages a holistic developmental process resulting from the establishment of inter-personal relationships between teacher and student. An analogy to nurturing is made by both Pratt (1992) and Northedge (1976) in the quotation below.

we view the ground as already covered with vegetation (concept systems), some of which is clearly worth retaining and cultivating. In the garden plants will tend to grow quite readily regardless of intervention from the gardener, and it is his aim to encourage certain plants at the expense of others; finding ways of acting as a catalyst in bringing out the best he can from the available ground. The gardener does not work towards a precisely defined end, since the garden is continually changing as different plants come to their prime. He has broad plans as to how he wants the garden to develop (probably rather flexible ones, which change as possibilities within the garden reveal themselves), but he does not attempt to specify the exact dimensions that each plant (or concept structure) is to achieve (Northedge, 1976, p. 68).

Samuelowicz and Bain (1992) found their conception of supporting student learning only in the context of postgraduate teaching. This does not necessarily mean that it is exclusive to this level, though - it may just mean that most academics or courses do not reach these standards of teaching at the undergraduate level.

The Effect on Student Learning

By deriving a questionnaire from their initial qualitative work, Gow and Kember (1993) were able to relate lecturers’ conceptions of learning with measures for changes in students’

ACADEMICS’ CONCEPTION OF TEACHING 269

approaches to learning. Departmental scores were obtained for two main orientations to teaching: knowledge transmission and learning facilitation. These were related to measures of students’ approaches to learning from the Study Process Questionnaire (Biggs, 1987) which gives values for deep, surface and achieving approaches, and can be applied in a repeated measure design to show change in approach during a course. The results showed that underlying conceptions or orientations to teaching had a significant effect on students’ learning approaches. Departments with high mean scores for the knowledge transmission orientation tended to depress the use of a deep approach to learning. Departments more attuned to learning facilitation were less likely to promote a surface approach to learning. By the normal standards of educational research many of the observed correlations were quite appreciable.

Beliefs InyYuence Teaching Approaches

The Cow and Kember study examined the direct correlation between orientations to teaching and student learning outcomes. It is likely that this relationship is mediated by teaching approach as an intervening variable. Trigwell et al. (1994) found a logical relation- ship between diagnosed intentions of lecturers and the strategies they claimed to adopt. Those who intended to transmit information adopted teacher-focused strategies, while oth- ers who believed in conceptual development or change adopted student-focused strategies.

0 institutional influence

Figure 3. The relationship between conceptions of teaching, teaching approaches and learning outcomes.

270 D.KEMBER

Sheppard and Gilbert (1991) found that the development of student epistemology was influenced by the lecturer’s theories of teaching and the students’ perception of the learning environment and learning approaches. The research was limited to case studies of four departments, but did find that meaningful learning outcomes were more likely to be associ- ated with courses which explicitly considered alternative conceptions of knowledge.

Together these studies suggest a relationship from teaching conceptions. through approaches to teaching, to student learning outcomes. At the level of the individual teacher, the methods of teaching adopted, the learning tasks set, the assessment demands made and the workload specified are strongly influenced by the orientation to teaching. These contextual variables in turn impact upon the learning approaches of the students. These relationships are portrayed in a model shown in Fig. 3.

At a departmental level, those departments where the knowledge transmission orientation predominates are more likely to have a curriculum design and employ didactic teaching methods which have undesirable influences on the learning approaches of the students. Whereas departments with a greater propensity toward learning facilitation are more likely to design courses and establish a learning environment which encourages meaningful learning.

Teaching Approaches

While the portrayal of teaching conceptions as a hierarchical set of categories has been questioned in this article, there may be some element of a hierarchical nature in the teach- ing approaches or strategies. There will not always be an automatic relationship between underlying beliefs and observable teaching approaches. Those holding student-centred conceptions of teaching may at times still have to employ approaches which appear inconsistent with that belief.

A lecturer who holds an information transmission conception is likely to rely almost exclusively upon a unidirectional lecture approach. Even classes designated as tutorials are likely to end up largely as monologues. It is hard to see anyone holding such beliefs engag- ing in more interactive teaching methods such as dialogue or role play.

Those holding conceptions towards the opposite end of the framework will need to employ more interactive and student-centred teaching methods if they are to be consistent with their beliefs. They are also likely to use methods such as lecturing and spend a propor- tion of their time communicating content and bodies of knowledge. This does not imply, though, that while they are doing this they have changed or switched off their underlying beliefs. The lecturing would be just one element in a wider effort to facilitate learning.

The extent to which a faculty member’s teaching approach and conception of teaching is a reflection of the individual or the department presumably depends upon the balance of individual autonomy and departmental and institutional pressures. This will obviously vary by individual, department and institution. It is pertinent, though, to note that Marsh’s (1987) review of the student evaluation literature suggests that the effect of the teacher on student ratings is much larger than that of the course being taught. The correlation observed between teaching orientation and the quality of student learning at the depart- ment level may, therefore, be greater at the individual teacher level.

ACADEMICS’ CONCEPTION OF TEACHING 271

Implications for Quality Assurance

The reviewed research has implications for measures to monitor and improve the quality of teaching in higher education. There has recently been an international trend for govern- ment and state bodies either to form regulatory bodies or to require or encourage universi- ties to establish their own quality assurance measures. The result has been the introduction or strengthening of quality control measures such as course approval processes, external examiner systems, teaching evaluation and regular faculty appraisal. These and other procedures imposed by administrative bodies normally act mainly as quality control standards. They aim to ensure that courses and teaching meet some minimum acceptable standard.

From the perspective of the model displayed in Fig. 3, the limitation of these quality control measures is that both the monitoring and any recommendations for remedial action focus upon the teaching approach. It is hard for regulatory bodies to discover, let alone have any influence upon, the underlying beliefs or conceptions of teaching.

In the model in Fig. 3 the path between institutional influence and conceptions is shown as a dotted line to imply that there is uncertainty as to whether and to what extent the institution can influence conceptions of teaching. Each academic’s conception of teaching will have formed through some complex amalgam of influences such as experiences as a student, departmental and institutional ethos, conventions of the discipline and even the nature of the classroom. As teaching is central to the role of academics, conceptions of teaching tend to become subsumed into the unconscious. It, therefore, takes a major perspective transformation to change them (Mezirow, 1991).

As conceptions of teaching have a distinct influence upon the teaching approach adopted, attention to teaching approaches without a corresponding change in beliefs may negate any educational development initiative. An example of this happening in practice comes from my own institution. The committees which approve and validate new courses initiated a campaign to make teaching less didactic and more interactive. Courses were only approved, therefore, if a component of the contact hours was designated for tutorials or seminars. Several proposals had to be modified to incorporate tutorial sessions and other course teams pro-actively boosted tutorial hours. So on the surface it might appear that there would be greater levels of interaction between faculty and students, which hopefully would go hand-in-hand with enhancements in learning outcomes.

However, observation of tutorial classrooms subsequent to the approval of the new courses revealed a rather different outcome. A high proportion of the “tutorials” which were observed featured very low levels of student participation. They were dominated by exposition from the tutors, who turned the tutorials into mini-lectures.

The reason for this lack of impact was simply that the quality control campaign was directed at teaching approaches without any corresponding initiative to influence beliefs about teaching. Those who held conceptions in the teacher-centred/content-oriented domain might see no reason for interaction with students, even if they possessed the neces- sary communication skills. Instead they would be worried about covering content, so viewed the tutorial as an extra session for transmitting knowledge.

Even with hindsight, it may have been hard for the campaign to make much impact because quality control bodies usually have power to approve or withhold approval, but no obvious mechanism for influencing deeply ingrained beliefs. Underlying beliefs about

272 D. KEMBER

teaching cannot be imposed by regulation. This does not imply that all teaching quality initiatives are futile, rather that quality control initiatives of the type described need to be accompanied by appropriate faculty development programs which are cognisant of the significant influence of conceptions of teaching.

There must be question marks against a wide range of faculty and teaching development measures which focus upon teaching approaches. For example, it is possible that short workshops which focus upon teaching skills or approaches may have limited outcomes if the underlying beliefs of the participants are inconsistent with the conceptual framework of the initiative.

A logical position for improving teaching quality follows from the interpretation of concep- tions of teaching as a continuum, and the degree of evidence which suggests that some lectur- ers do alter their positions on the continuum. If the evidence for the link between conceptions of teaching and the quality of student learning is accepted, then it should be possible to improve the latter by changing beliefs about teaching. Quality enhancement and faculty development measures should incorporate opportunities for participants to shift beliefs along the continuum towards the student-centredneaming-oriented pole.

There appear to have been few documented attempts to promote conceptual change with teachers, especially at the tertiary level. Bowden (1988) describes a workshop which aims to expose disparities between espoused theory and theory in use. Although, as Bowden points out, such workshops are only the start of a process because significant changes to teaching and learning require sustained effort over a lengthy period of time.

Martin and Ramsden (1992) report outcomes from a one-year course for tertiary teach- ing which monitored changes in conceptions of teaching of the paritcipants. The three presented case studies clearly show that the lecturers change their conceptions of teaching during the course. Again, though, it is obvious that change takes time, so Martin and Ramsden recommend that any such courses extend over a period of at least one year.

My own attempts to bring about changes in teaching orientation have followed an action research format (Kember & Gow, 1992; Kember & Kelly, 1993; Kember & McKay, 1996). Action research follows spirals or cycles of planning, acting, observing and reflecting, so there is an opportunity to work with academic staff over an extended period to allow time for conceptual change. The observation or evaluation phase of the action research cycle can be seen as a process for lecturers to confront the outcomes of their own teaching. The initial project cycles often reveal more fundamental problems at the course or department level. This can lead to the core participants trying to alter the teaching orientations of their colleagues.

It should be noted that each of these three approaches operate over an extended period and operate within the framework of either a course or a project. Within these the participants observe or evaluate the impact of their teaching upon their students’ learning. This in itself can cause the teachers to examine the intended or unintended outcomes of their own actions and beliefs. Reflection upon these actions, preferably as part of a small supportive group, can be a mechanism for changing beliefs.

Conclusion

This article has reviewed 13 papers which investigated university academics’ conceptions of teaching. The faculty interviewed came from seven countries and a much larger number of

ACADEMICS CONCEPTION OF TEACHING 273

institutions. The researchers conducted most of the studies with little or no insight as to what the other groups were doing since the bulk of the work was carried out simultaneously.

In spite of the diversity of the samples and the independence of the research studies, the findings show a high degree of commonality. The categories of description derived by the studies are closely comparable. There are differences in terminology and the number of categories varies, but the latter differences can often be minimised if visualised in terms of either groups and sub-groups or overlapping categories.

My attempt to synthesise the body of research places conceptions under two broad orientations characterised as teacher-centred./content-oriented and student-centredneaming- oriented. The five conceptions placed beneath the orientations are then visualised as well- defined categories within a developmental continuum. A transitionary category links the two orientations.

A study of teaching conceptions is seen as important because they have been shown to be related to measures of the quality of student learning. Conceptions are modelled as influencing teaching approaches which in turn effect student learning approaches and subsequently learning outcomes.

An understanding of teaching conceptions then becomes important if measures to enhance the quality of teaching are to have any impact. If teaching approaches are strongly influenced by the underlying beliefs of the teacher, quality assurance measures should take into account conceptions rather than concentrate exclusively upon approaches. Real changes in teaching quality are only likely to be brought about by changes in the beliefs about teaching of faculty.

Implications for Future Research

A substantial body of independent research has found a high degree of consensus on the categories to describe conceptions of teaching in higher education. This research has been conducted in several countries and in a diverse range of universities. There seems to be little value, therefore, in further exploratory studies to classify academics’ conceptions of teach- ing.

There is, though, a need for further investigation of the relationship between categories. Are they adequately described as several discrete categories or does a continuum with well- defined positions provide a better model? Insights into these questions are most likely to come from investigations of the way in which conceptions of teaching change over time. One possible interpretation of the alternative conceptions is that they are seen as a developmental sequence.

It would also be of value if there were further studies which examine the relationship between the facets of teaching and learning depicted in the model in Fig. 3. The relation- ship between orientations to teaching has been established by one quantitative study (Gow & Kember, 1993; Kember & Gow, 1994) and by case studies of four teachers (Sheppard & Gilbert, 1991). Some of the other elements and relationships within the model have less strong evidence. The model as a whole has not been substantiated.

For the current pressure towards quality assurance in higher education, there are significant implications, from this research, for the way in which initiatives and programs are formulated. If the model in Fig. 3 is accepted, it suggests that measures to improve the

214 D. KEMBER

quality of teaching which focus exclusively upon teaching approaches may have little impact. As many measures do seem to impact upon teaching approaches rather than underlying beliefs, this claim is surely worth further investigation. This in turn would sug- gest a need to examine approaches to changing teaching conceptions.

References

Becher, T (1989). Academic tribes and territories: Intellectual enquiry and the cultures of disciplines. Milton Keynes: SRHE and Open University Press.

Biggs, J. (1987). Student approaches to learning and studying. Melbourne: Australian Council for Educational Research.

Biggs, 1. (1989). Approaches to the enhancement of tertiary teaching. Higher Education Research and Development, 8(l), l-25.

Bowden, J. (1988). Achieving change in teaching practices. In P. Ramsden (Ed.), Improving learning London: Kogan Page.

Champagne, A. B., Gunstone, R. F., & Klopfer, L. E. (1985). Effecting changes in cognitive structures among physics students. In L. H. T. West, & A. L. Pines (Eds.), Cognitive structure and conceptual change New York: Academic Press.

Clark, M., & Peterson, P. (1986). Teachers’ thought processes. In M. Wittrock (Ed.), Handbook of research on teaching (3rd edition) (pp. 255-296). New York: Macmillan.

Dall’Alba, G. (1991). Foreshadowing conceptions of teaching. Research and Development in Higher Education, 13, 293-297.

Dunkin, M. J. (1990). The induction of academic staff to a university: Processes and products. Higher Education, 20, 47-66.

Dunkin, M. J. (1991). Orientations to teaching, induction experiences and background characteristics of university lecturers. Australian Educational Researcher. I8( I), 3 l-52.

Dunkin, M. J., & Precians, R. P. (1992). Award winning university teachers’ concepts of teaching. Higher Education, 24.483-502.

Eisner, E. W. (I 991). The enlightened eye: Qualitative inquiry and the enhancement of educational practice. New York: Macmillan Publishing.

Fox, D. (1983). Personal theories of teaching. Studies in Higher Education, 8(2), 151-163. Glaser, B. G., Kc Strauss, A. L. (1967). The discovery ofgrounded theory Chicago: Aldine. Gow, L., & Kember, D. (1990). Does higher education promote independent learning? Higher Education, 19,

307-322. Gow, L., & Kember, D. (1993). Conceptions of teaching and their relationship to student learning. British Journal

of Educational Psychology, 63, 20-33. Kember, D., & Gow, L. (1992) Action research as a form of staff development in higher education. Higher

Education, 23(3), 297-3 IO. Kember, D., & Gow, L. (1994). Orientations to teaching and their effect on the quality of student learning. Journal

of Higher Education, 65( I ), 58-74. Kember, D., Kc Kelly, M. (1993). Improving reaching rhrough action research. NSW HERDSA Green Guide No.

14. Kember, D., & McKay, J. (1996). Action research into the quality of student learning: A paradigm for faculty

development. Journal of Higher Education, 67(5), 528-554. Marsh, H. W. (1987). Students’ evaluations of university teaching: research findings, methodological issues, and

directions for future research. International Journal of Educational Research, II, 253-388. Martin, E., & Balla, M. (I 991). Conceptions of teaching and implications for learning. Research and Development

in Higher Education, 13. 298-304. Martin, E., & Ramsden, P. (1992). An expanding awareness: how lecturers change their understanding of teach-

ing. Paper presented at the 1992 HERDSA Conference, Gippsland. Marton, F. (198 1) Phenomenography-describing conceptions of the world around us. Instructional Science, IO,

177-200. Marton, E (1988). Phenomenography: Exploring different conceptions of reality. In D. M. Fetterman (Ed.),

Qualitative approaches to evaluation in education: the silent scientific revolution. New York: Praeger. McKay, J. (1995). Promoting rejection within teaching: A case study in educational change within a department

(Unpublished Ph. D. Dissertation). Hong Kong: Hong Kong Polytechnic University. Mezirow, J. (1981). A critical theory of adult learning and education. Adult Education, 32(l), 3-24. Mezirow, J. (1991). Transformative dimensions of adult learning San Francisco: Jossey-Bass.

ACADEMICS’ CONCEPTION OF TEACHING 275

Miles. M. B., & Huberman, A. M. (1984). Qualitative dara analysis: A sourcebook o/new methods. Newbury Park, CA: Sage.

Northedge, A. (1976). Examining our implicit analogies for learning processes. Programmed Learning and Educational Technology, 13(4), 67-78.

Pajares, M. E (1992). Teachers’ beliefs and educational research: Cleaning up a messy construct. Review of Educational Research, 62(3), 301-332.

Pratt, D. D. (1992). Conceptions of teaching. Adult Educarion Quorferlx 42(4), 203-220. Presser, M., Trigwell, K., & Taylor, P. (1994). A phenomenographic study of academics’ conceptions of science

learning and teaching. Learning and Instrucrion, 4, 2 17-23 I Samuelowicz, K., & Bain, J. D. (1992). Conceptions of teaching held by academic teachers. Higher Educution, 24,

93-111. Sheppard, C., & Gilbert, J. (1991). Course design, teaching method and student epistemology. Higher Education,

22, 229-249. Strike, K. A., & Posner, G. J. (I 985). A conceptual change view of learning and understanding. In L. H. T. West, &

A. L. Pines (Eds.), Cognitive structure and-conceptual change. New York: Academic Press.- Trigwell, K., Presser, M., & Taylor, P. (I 994). Qualitative differences in approaches to teaching first year university

science. Higher Education, 27, 75-84.