Enabling critical reflection on research supervisory practice

This article was downloaded by: [Dalhousie University]On: 25 October 2012, At: 18:37Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office:Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

International Journal for AcademicDevelopmentPublication details, including instructions for authors and subscriptioninformation:http://www.tandfonline.com/loi/rija20

Enabling critical reflection on researchsupervisory practiceMargot Pearson a & Carole Kayrooz ba The Australian National University, Australiab University of Canberra, Australiac Centre for Educational Development and Academic Methods, TheAustralian National University, Linnaeus Cottage, Building 96, Canberra ACT0200, Australia E-mail:

Version of record first published: 17 Feb 2007.

To cite this article: Margot Pearson & Carole Kayrooz (2004): Enabling critical reflection on researchsupervisory practice, International Journal for Academic Development, 9:1, 99-116

To link to this article: http://dx.doi.org/10.1080/1360144042000296107

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions

This article may be used for research, teaching, and private study purposes. Any substantialor systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, ordistribution in any form to anyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representation that thecontents will be complete or accurate or up to date. The accuracy of any instructions, formulae,and drug doses should be independently verified with primary sources. The publisher shall notbe liable for any loss, actions, claims, proceedings, demand, or costs or damages whatsoever orhowsoever caused arising directly or indirectly in connection with or arising out of the use of thismaterial.

http://www.tandfonline.com/loi/rija20

http://dx.doi.org/10.1080/1360144042000296107

http://www.tandfonline.com/page/terms-and-conditions

International Journal for Academic DevelopmentVol. 9, No. 1, May 2004, pp. 99–116

ISSN 1360–144X (print)/ISSN 1470–1324 (online)/04/010099–18© 2004 Taylor & Francis LtdDOI 10.1080/1360144042000296107

Enabling Critical Reflection on Research Supervisory Practice

Margot Pearson

a

*

and Carole Kayrooz

b

a

The Australian National University, Australia;

b

University of Canberra, Australia

Taylor and Francis LtdRIJA100098.sgm10.1080/1360144042000296107International Journal for Academic Development0000-0000 (print)/0000-0000 (online)Original Article2004Taylor & Francis Ltd91000000May 2004MargotPearsonCentre for Educational Development and Academic MethodsThe Australian National UniversityLinnaeus Cottage, Building 96Canberra ACT [email protected]

This paper describes the development of an instrument—The Reflective Supervisor Questionnaire (RSQ).The RSQ maps the domain of research supervisory practice as a facilitative process involving educationaltasks and activities. It is designed to assist research supervisors explore, by means of self-reflection and reflec-tion on feedback from others, how they practise supervision. In developing the RSQ 58 items were generateddescribing 5 hypothesised constructs derived from prior research. The resulting instrument was tested onpostgraduate research students in 2 institutions. The questionnaire correlated highly with an establishedquestionnaire supervision scale and with an overall satisfaction measure. Four factors identified in an explor-atory analysis closely approximated the hypothesised constructs and extended the theoretical frameworkbeing developed. These 4 factors identified 4 subsets of facilitative supervisory practice: Progressing theCandidature, Mentoring, Coaching the Research Project, and Sponsoring Student Participation in Academic/Professional Practice. Issues in the interpretation of the findings and the possible usage in academic develop-ment programs of an instrument based on them are discussed.

Introduction

Effective supervision is considered a critical issue in students’ satisfaction with their postgrad-uate experience and the successful completion of their degree (for example, Harris, 1996;Latona & Browne, 2001). In Australia this issue has been taken up at the policy level inconjunction with a shift in government emphasis on to timely completions (Fells, 2004;Gallagher, 2000; Kemp, 1999). At the institutional level, there is now more attention tomonitoring student progress and providing student induction and other support activities(Neumann, 2003). At the same time the growing numbers and diversity of the postgraduatestudent population, and changes in the research environment, are creating pressures forchange and flexibility in supervisory practice in Australia as has occurred in other countries

*

Corresponding author. Centre for Educational Development and Academic Methods, Linnaeus Cottage,Building 96, The Australian National University, Canberra ACT 0200, Australia. Email: [email protected]

Dow

nloa

ded

by [

Dal

hous

ie U

nive

rsity

] at

18:

37 2

5 O

ctob

er 2

012

100

M. Pearson and C. Kayrooz

such as the US and the UK (Green, 2002; LaPidus, 1997; Pearson & Ford, 1997; Pearson,2000).

In response, there has been increased interest in supervisor development and the introduc-tion of various supervisor development programs and resources (Pearson & Cryer, 2001;Pearson & Brew, 2002; Kayrooz & Kiley, 2002). Such programs provide the means for super-visors to learn more about supervision, supervisory strategies and issues. The emphasis is onreflective practice by supervisors, more awareness of responsibilities in the current highereducation climate, and the needs and interests of research students. Many of the existingprograms and resources (for example, the “for Improving Research Supervision and Train-ing” (fIRST) website) provide supervisors with “tools” for such reflection and for structuringinformal conversation with their students, some supported by published accounts of theirdevelopment and use (Gurr, 2001; Wisjesundera, Hicks, & Mann, 1996).

Constraining such academic development activity is the lack of a robust conceptual under-standing of what supervision involves (Evans & Pearson, 1999; Pearson, 1996; Pearson,2001). Terms such as “apprenticeship”, “mentor” and “coach” are used frequently in discuss-ing supervision, but often in ways that perpetuate the mystification of the process, which thusstays private. Although research into supervision is a burgeoning field of study, much of thescholarship has been focused on particular aspects of supervision and research education, forexample, specific student groups, disciplinary variations, supervisory relationship stories andnew doctoral degrees and policy. Attempts to theorise supervision pedagogy have been limited(Green & Lee, 1995). As Smith (2001) explains, the different interpretative frames that areavailable give different perspectives on research supervision, leading to different aspects beingscrutinised and questioned. He gives as examples how an administrative framing leads to apredominant focus on the procedural responsibilities of supervisors; and the concept of“culture” contributes to understanding how students come to fit the disciplinary, research andorganisational cultures of university faculties (pp. 25–26). No one framing, he concludes,captures the entire supervision picture “…all framings vary in what they can reveal and what,as a consequence, they tend to conceal” (p. 27).

This paper reports on the generation of an instrument designed to assist in conductingacademic development programs on supervisory practice—The Reflective Supervisor Ques-tionnaire (RSQ). The RSQ provides an instrument with which research supervisors canreflect critically on their practice and interpret student and colleague feedback, but onethat is framed explicitly to gain feedback on specified elements of supervision. The RSQdraws on research and scholarship carried out by the authors and others to provide aresearch-based conceptual framework for describing the operational domain of postgradu-ate research supervisory practice. The framework features research supervisory practice as afacilitative process involving educational tasks and activities that comprise the work ofsupervision. This paper outlines an investigation of the efficacy of that conceptual frame-work and the constructs on which the questionnaire was based. The results show that thestructure of the framework is supported in the main. There is also a very strong relation-ship between the domain as described by the items in the questionnaire and students’overall satisfaction with their supervisors’ practice. The study confirms that the RSQ mapsthe domain of research supervisory practice as operationalised and as carried out by indi-vidual supervisors, supervisory panels or committees. As constructed, the RSQ does notassume that the work of supervision is the responsibility of one supervisor alone. The study

Dow

nloa

ded

by [

Dal

hous

ie U

nive

rsity

] at

18:

37 2

5 O

ctob

er 2

012

Enabling Critical Reflection

101

also gives substance to the claim that traditional notions of intellectual development areseen as important by students for research supervision, and are connected to the emotionaldimension. The analysis provides grounds for distinguishing this intellectual and emotionalsupport from professional networking, career advising and other such activities that areoften included together under the term mentoring. Issues in how to interpret the resultsand how this instrument might be developed further for use in academic developmentprograms with both inexperienced and experienced research supervisors are discussed.

Issues in Designing the RSQ

In contrast to the experience of undergraduate quality assurance and improvement, there hasbeen little attention to the enhancement of supervisor practice through the use of studentfeedback. The literature that does exist describes informal, and sometimes structuredexchanges, between individual supervisors and their students on one hand; or on the other,institutional level surveys of the quality of the student experience such as the PostgraduateStudent Course Experience Questionnaire (Guthrie & Trembath, 1998). This is unsurprisingas introducing any type of evaluative activity or feedback with research students and supervi-sors is highly problematic. The intimacy of the circumstances, and students’ fear of negativecareer consequences, are very important in relationships marked by differences in power,status and dependence. These differences compound difficulties arising from the variability ofexpectations of supervision and research, of learning styles, career goals, and stages in super-vision, in addition to variations in age, gender and life experience (for example, Brew, 2001;Malfroy & Webb, 2000; Pearson & Ford, 1997; Wisker, 2000).

A further issue concerns the extent of an individual supervisor’s responsibility. Variationin practice is to be expected where there are panels, committees, and work-based supervi-sors among whom responsibilities are distributed (Pearson, 2000). At the same time thegrowth of research student numbers and the move to ever further sub-specialisation, cross-disciplinary and interdisciplinary research means more supervisors are likely to find them-selves responsible for students who are seeking specific expertise they do not have. Studentsmeanwhile are learning from a variety of people, various academics, technicians and fromone another (Pearson, 1996). Research on both apprenticeship learning and laboratoryprocesses confirm research and research training as a social practice (Delamont et al., 1997;Lave & Wenger, 1991). In such complex situations flexibility, greater coordination of effortand more explicit allocation of responsibilities are demanded. Exhortations to “good prac-tice” that set up individual supervisors to assume the burden of unrealistic expectations maycause resentment, and will certainly not advance their openness to further professionaldevelopment.

Yet, without feedback, supervisors are denied access to data about their practice asperceived by others, an important “reality check”. They also need to know more than thespecifics of poor practice that is so disastrous as to surface as visibly messy disputes and fail-ures. They need to know more about the more usual story of supervisors who are seen to bedoing their best by the students, but could be doing better, or perhaps differently for differentstudents (Cullen et al., 1994; Malfroy & Webb, 2000). As argued by Pearson (1999a) feed-back of this kind will not come from institutional level surveys. To give guidance in what theymight improve, supervisors want and need feedback that will enable them to understand the

Dow

nloa

ded

by [

Dal

hous

ie U

nive

rsity

] at

18:

37 2

5 O

ctob

er 2

012

102


more nuanced features of their practice. For this, supervisors need conceptual frameworks astools for interpreting their feedback, and for enabling critical reflection on their practice. Suchan approach is in contrast to more prescriptive and bureaucratic guidelines, such as forexample, on how often supervisors and students should meet. As professionals, supervisorsshould also be able to expect that any such conceptual frameworks and instruments areresearch-based and open to critique.

The Conceptual Framework for the RSQ

Critical to the design and construction of the RSQ is a conceptual framework for “unpack-ing” research supervision being developed by one of the authors drawing on a number ofresearch studies (Cullen, Pearson, Saha, & Spear, 1994; Pearson & Ford, 1997; Thompson,Pearson Akerlind, Hooper, & Mazur, 2001). The development of the RSQ was alsoinformed by the experience of both the authors as researchers, professional developers,postgraduate supervisors, and that of one author as a professional and clinical supervisorand research course coordinator (Kayrooz, 1999; Kayrooz & Kiley, 2002; Kayrooz &Trevitt, 2004).

The framing of supervision for the RSQ entails representing supervision as a “job” or work,which includes a set of educational tasks/activities from which flow particular roles andresponsibilities that need to be attended to (Connell, 1985; Pearson, 1999a). This is incontrast to schemas that focus on supervisor characteristics and attitudes (Fraser & Mathews,1999). Similarly, in contrast to approaches and models focused on the management aspects ofsupervision (for example, Vilkinas, 1998), the supervisor(s) is construed as a facilitator ofresearch student learning, given that research education is a form of experiential learning inwhich the research student learns how to do research, how to manage a research project andhow to write a thesis etc. (Cullen et al., 1994; Pearson, 1996; Pearson & Brew, 2002). Fromthis role definition follows expectations of behaviour established as features of effective facili-tation, for example, facilitation involving challenge and support to address the intellectual andemotional dimensions of the student experience (Cullen et al., 1994; Pearson, 1999b; Reason& Marshall, 1987), and the expectation that the process extends the “curriculum” beyondcompleting a research project or thesis, to broader outcomes (Pearson, 2001; Pearson &Brew, 2002; Pearson, in press).

As presented in Pearson and Brew (2002) the educational tasks and activities of supervi-sion can usefully be distinguished. In earlier work (Cullen et al., 1994) one key supervisorytask was identified as facilitating the student’s management of their candidature over itsduration from its initial stages to completion. Later Pearson (2001), drawing on notions ofcognitive apprenticeship, argued for articulating the activity of coaching given its importancein developing the student’s research capability. This in turn led to further clarification of thetask of mentoring students by supporting their development in the context of their evolvingpersonal and career goals (Pearson, 2001; Pearson & Brew, 2002). An additional activityseen as allied to mentoring is that of sponsoring access to resources (which includes exper-tise and introductions to networks). This activity is critical to the learning required of afuture professional researcher to operate effectively in a dynamic research environment asargued by Pearson and Brew (2002), and of greater significance as we learn more aboutapprenticeship learning in communities of practice (Pearson, in press). In integrating these

Dow

nloa

ded

by [

Dal

hous

ie U

nive

rsity

] at

18:

37 2

5 O

ctob

er 2

012


103

tasks and activities in a comprehensive description of supervision practice Pearson and Brew(2002), drawing on the work of writers such as Barnett (1990) and Senge (1990), and onBrew’s research on conceptions of research, added the need for critical reflection by thesupervisor on their own research and supervisory practice. Finally in distributed supervisionit is important that one supervisor—usually a designated principal supervisor—is identifiedas the institutional link to university policies, guidelines, and expectations (Evans & Pear-son, 1999; Pearson, in press).

The Statistical Framework

The RSQ initially consisted of a set of 58 items structured in 5 constructs. Exploratoryanalyses were undertaken to validate the instrument and to test the hypothesized constructstructure for the purpose of further developing the theoretical framework. The analyses aimedto:

●

assess coverage of the domain of supervisory practice as determined by student satisfactionwith the supervision they were receiving from their designated supervisors;

●

provide an indicator of convergent validity by comparing the established supervisionsubscale of the Postgraduate Research Experience Questionnaire (PREQ) with the RSQscale; and

●

examine the structure of the five hypothesized constructs with an exploratory factoranalytic procedure.

Scale Construction

Key statements describing supervisory behaviours were generated to provide coverage of thedomain of supervisory practice that was construed as a facilitative process that includedCoaching, Mentoring, Sponsoring, Facilitating the candidature, and Reflective practicebehaviours by supervisors. For generating the initial items we drew on data (much of it frominterview and focus group material) gathered over time in our research and developmentalwork and experience, and including all major fields of study (Cullen, Pearson, Saha, & Spear,1994; Ford, 1996; Pearson & Ford, 1997; Thompson, Pearson, Akerlind, Hooper, & Mazur,2001), and used a review of pre-existing supervisory questionnaires as a further check oncoverage (Edwards, Aspland, Litster, O’Leary, Ryan, Southey, & Timms, 1996; Mullins &Hejka, 1995; Wijesundera et al., 1996). The resulting five sets of items operationalised facili-tative practice as shown below. Ratings on each item were designated on a 5 point Lickertscale “strongly agree—strongly disagree” with an additional point on the scale designated“not applicable”. The five hypothesised constructs of facilitative supervisory practice were asfollows and items related to each construct given in Table 1.

Expert coaching.

The supervisor as expert and coach provides expertise on the research topic,research methodology, and in writing the thesis. Effective coaching includes assisting studentsto become aware of their own processes in problem solving and to develop their professionalrepertoire and judgement as a research practitioner. For example, “…assists me to formulatemy research topic”.

Dow

nloa

ded

by [

Dal

hous

ie U

nive

rsity

] at

18:

37 2

5 O

ctob

er 2

012

104


Table 1. The RSQ: Hypothesised constructs and items

EXPERT COACHING

challenges intellectuallyencourages take ownership of thesis•gives constructive feedback on ideas in thesis drafts•assists formulate research topichelps plan and refine the projectcomments on thesis drafts in a constructive fashion•comments on thesis drafts in a timely fashion•encourages develop/evaluate own ideasprovides specialist/technical expertiseintroduces to relevant current literatureencourages writing earlyadvises on problem framing +solvingintroduces me to ethical practices formal and informal•advises me on critical aspects of researchprovides advice on the logistics for producing a thesis document

FACILITATING

negotiates programme of study and researchdisplays sensitivity to gender differences in interactions•directive when needednegotiates explicit ground rules for supervisionensures official requirements are met, eg ethics clearance, annual reportsconsults on the selection of examiners•willing to help me without being intrusive•devotes sufficient timepromotes good interaction and learning among students and stafflistens with attentionrespects the knowledge and expertiseapproaches supervision flexiblyputs effort into a good startprovides information on availabilityinitiates contactnegotiates their availabilitydisplays sensitivity to cultural differences•

MENTORING

demonstrates genuine interest in well-beingapproachable, responsive, and affirmingencourages publishingdirects me to leading researchersencourages networking within the universityintroduces to professional networksconveys enthusiasm for project/research•helps arrange for presentations at seminars and conferencesrefers me to relevant professional assistance•collaborates leading to joint publications•conveys enthusiasm for the research•assists progress career goalsadvises on opportunities for relevant experience

Dow

nloa

ded

by [

Dal

hous

ie U

nive

rsity

] at

18:

37 2

5 O

ctob

er 2

012


105

Facilitating the candidature.

The supervisor as facilitator provides guidance that enables thestudent to manage their candidature through the three stages identified in Cullen et al. (1994,p. 96), negotiating a program of study and research that is appropriate for the student. Itincludes ensuring that the student meets official requirements such as ethics clearance andannual reports, and that progress is being made in a timely fashion. For example, “…negoti-ates explicit ground rules for supervision from the beginning.”

Mentoring.

The supervisor as mentor supports the student’s development in the context oftheir evolving personal and career goals, and links them to appropriate research, academicand professional networks. A mentor can offer both personal and professional support. Forexample, “… introduces me to relevant professional networks…”

Sponsoring.

The supervisor as sponsor assists the student to gain the necessary access toresources and opportunities for their learning and research activity. Such resources includetechnical support, funding and introductions to those with specialist expertise. For example,“…assists me obtain resources to attend relevant seminars and conferences.”

Reflective practice.

The supervisor as a reflective practitioner models openness to new ideasand encourages critical discussion on research practice including their own, and the supervi-sory process. For example, “… is open to different research approaches.”

The Procedure

After gaining ethics clearance from the relevant university Ethics Committees, the question-naire was developed for web-based usage. The draft questionnaire was circulated amongst

Table 1. Continued

REFLECTIVE PRACTICE

models effective research practiceopen to different research approachesopen to critical discussion on research practiceperiodically reviews our supervisioncarries out supervisory duties professionallyrespects as professional and a learner•encourages open/critical discussion on research practices

SPONSORING

assists obtain resources for seminars and conferencesassists to meet administrative requirements in an efficient and timely mannerkeeps informed about procedures and issues re intellectual property rightsassists me to access essential technical supportadvises about resources and other funding sourcesensures that I can gain access to required facilities (eg fax, phone, computer)•

•indicates items deleted in preliminary screening

Dow

nloa

ded

by [

Dal

hous

ie U

nive

rsity

] at

18:

37 2

5 O

ctob

er 2

012

106


other higher educational theorists and practitioners for comment regarding content coverageand specific wording. The authors also interviewed a small student sub-sample (n = 8,including science and non-science disciplines) who commented on question wording andthe meaning of the items. Several items were changed to achieve linguistic and semanticclarity. Care was taken to construct simple items because questionnaires of this sort requireconcentrated thinking and are susceptible to fatigue. This was especially an issue becausethe final instrument for the pilot included both a section asking for biographical detail, andthe supervision scale from PREQ, adding up to 78 items. The format required allrespondents to indicate the status of the supervisor in question (principal supervisor,co-supervisor, other) but they were not asked to identify the supervisor by name or anyother characteristic (submission of responses was only possible if this item was completed).In addition, students were asked to rate their overall satisfaction with their supervisor’spractice as follows:

“Overall, the supervision I receive from my supervisor meets my needs”.

Allresponses were collected, recorded and analysed using the Statistical Package for the SocialSciences (2001).

Sampling

An invitation to respond to the questionnaire, complete with the site address, was circulatedby email, with the endorsement of the student association, to 1080 research students(doctoral and masters-by-research students) in a research-intensive Australian university inmid-2002. A total of 314 research students (29%) responded to the questionnaire. As set outin Table 2 most in our sample were studying at the doctoral level, and most were studying fulltime at the time of the questionnaire. Over a third were commencing students. Women were asmall majority. The majority gave Australia as the place of residence, though the percentageof those in the sample not giving Australia as their place of residence indicates a relatively highproportion of overseas students. Interestingly only three-quarters of the sample listed Englishas their first language (76%). The majority (70%) were studying in the broad fields of study ofSociety and Culture (that is social sciences and humanities) and the Natural and PhysicalSciences (that is biological sciences, physical sciences).

A comparison of the sample profile with the national profile for 2002 showed that while theproportion of doctoral level students was greater in our sample than for research studentsnationally (nationally 76% were doctoral/masters students by research), similarities betweenthe sample and national profile existed for gender, and to a somewhat lesser extent the desig-nation of Australia as the first place of residence (nationally 85%). Gender representation was

Table 2. Sample research student profile (N=314)

Characteristic %

Doctoral award level 95Commencing study 39Primarily full-time enrolment 90Gender: Female 54Australia as place of residence 78

Dow

nloa

ded

by [

Dal

hous

ie U

nive

rsity

] at

18:

37 2

5 O

ctob

er 2

012


107

similar in both the sample and national profiles (49% nationally for women). However unlikethe national profile, our sample research students in professional fields of study were not wellrepresented. A later invitation in 2003 to respond to the questionnaire was carried out atanother institution with more research students in professional fields of study and the dataincluded in a retest of the factor analysis. There were 59 respondents in this second survey ofwhom 86% were studying at the doctoral level, 53% were part-time, and 53% male. A total of37% were studying in the fields of education, management and commerce. In this instance aswith the initial sample group most respondents nominated their principal supervisor as thesubject of their feedback.

Analyses and Findings

Data Screening and Reliability Testing

Initial analysis included examination of the distributions of the items to assess approximationto normality and the degree of spread. Also, cases with missing values were examined to seeif there were any systematic patterns in missing values. Fifteen items were identified forexclusion from subsequent factor analysis on the basis of being either non-normally distrib-uted, failing to discriminate and not being essential on theoretical grounds to contribute totheir constructs (Table 1). There seemed to be a systematic pattern to non-responses, in thatmany were from students in the first twelve months of their candidature. This group alsotended to rate their supervisors better on a large number of items. However, examination ofthe correlations between the ratings of the items within each construct indicated theirpattern of correlations were the same as those students further into their candidature. There-fore, it was decided to include them in the subsequent analyses. The correlations within thehypothesized scales were consistently high (Cronbach alpha .8 to .9).

Convergent validity.

Analysis confirmed the convergent validity of the RSQ with the PREQsubscale. A correlation matrix between the two questionnaires showed that every item in theRSQ correlated with every item in the PREQ subscale (.3 or above).

Construct validity.

A regression analysis of the hypothesized scales with student satisfactionestablished that the RSQ scale explained 81% of overall satisfaction demonstrating thatthe items of the RSQ covered the domain of satisfaction as experienced by the students.Additionally all 58 initial items (including those excluded for the factor analysis)correlated at

α≤

.001 level with student satisfaction and supported the theoretical approachto defining supervision for the purpose of this instrument. This regression analysis alsoestablished that the items included in the hypothesized scales of Facilitation and Coachingwere sufficient to explain most of the correlation with satisfaction. However as studentsatisfaction is only one measure in the overall quality depicted in our conceptual frame-work, we used a factor analysis to explore the data structure further. Prior to this factoranalysis, principal components analysis was used to detect the presence of outlyingvariables, absence of mulitcollinearity, and factorability of the correlation matrices(Tabachnick & Fidell, 2001).

Dow

nloa

ded

by [

Dal

hous

ie U

nive

rsity

] at

18:

37 2

5 O

ctob

er 2

012

108


The structure of the five hypothesized constructs.

The five hypothesised constructs weresubjected to an exploratory factor analytic procedure because there has been relatively littleempirical research that has attempted to describe the domain under question. A maximumlikelihood procedure (which focuses on shared variation of items rather than unique itemvariation) was used to develop, rather than test, the theory. As the goal was to produce aneasily understood model that could be used by supervisors in reflection on their practice,without the need for complex statistical understanding, the varimax rotation method wasused as it simplifies the interpretation of the factor structure and also facilitates theconstruction of factor scores (Tabachnick & Fidell, 2001, p. 622). Note, though, that asalready described, there is also evidence of a single underlying factor, as all variables corre-lated with student satisfaction. The resultant solution using maximum likelihood extractionwith varimax rotation showed the items loading on five factors with eigenvalues greaterthan one.

Selection of the number of factors to retain was based on the percentage of varianceexplained as there was a sharp drop in the percentage explained between the fourth and fifthfactors (12.6% to 3.4%). Surprisingly, these four factors accounted for a relatively largeamount of the variance (59.5%) that supports the validity of the selected items. A furthertest for reliability was carried out on an extended population comprising the original groupof respondents and the later cohort from the second institution (Set 1, N=166; Set 2,N=193). The original factor structure was confirmed with the total variance explaining60.9%. A comparison of the factors between the two sets indicated no significant differencein the loading of the items. In each analysis 41 items were retained. In selecting items forinterpreting the four factors, loadings under .4 were not included. (In this analysis a furtheritem was dropped in addition to those excluded in earlier screening. Note that in the Set 1analysis the item “assists to meet administrative requirements in an efficient and timelymanner” was not retained, but in the analysis of Set 2 this item was retained and item“encourages writing early” was dropped.) The percentages of variance explained by each ofthe factors are shown in Table 3 (giving the figures for the second analysis from theextended Set 2). The loadings that are unique at the .4 level are given in bold to facilitateinterpretation as shown in Table 3.

The factors that emerged from the factor analytic procedure approximate the hypothesisedconstructs derived from exploratory work and theory posed earlier in this paper. Althoughthese four factors and the hypothesised constructs are not fully aligned, the items mostly clus-ter in the factors in a pattern that meets expectations according to our hypothesised theoreti-cal base (see Table 1). Facilitation items are clustered in Factor 3. Expert coaching items areclustered primarily in Factor 4. Mentoring items are primarily clustered in Factor 2, as is thecase for Sponsoring items, which do not occur in any other factor. Moreover Factor 2confirms our expectation that the constructs Mentoring and Sponsoring as hypothesised arestrongly related.

The unexpected finding is Factor 1 in which Reflective Practice items are clustered prima-rily, but in which there are also significant items from other constructs, with the exception ofSponsoring. Factor 1 includes both a focus on intellectual development and those behavioursaddressing the more personal/supportive aspect of interaction. The inclusion of items relatingmore strongly to personal support and interaction in Factor 1 is contrary to the expectationthat such support might be a feature of the construct Mentoring (Table 1).

Dow

nloa

ded

by [

Dal

hous

ie U

nive

rsity

] at

18:

37 2

5 O

ctob

er 2

012


109

Table 3. The RSQ—Construct/Factor Matrix

Hypothesised Constructs

SET 2 Factor 1 2 3 4

Variance

16.6% 15.3% 15% 12.6%

EXPERT COACHINGchallenges intellectually .41 .47assists formulate research topic

.63

helps plan and refine the project .45 .51encourages develop/evaluate own ideas

.68

provides specialist/technical expertise/or who .43 .51 .43introduces to relevant current literature/or who .55 .47advises on problem framing and solving .41 .59advises me on critical aspects of research

.57

provides advice on the logistics for producing a thesis document

.45

FACILITATINGnegotiates programme of study and research .49 .46directive when needed

.55

negotiates explicit ground rules for supervision

.59

devotes sufficient time .44 .58promotes good interaction and learning among students and staff .46 .51listens with attention .63 .48respects the knowledge and expertise

.76

approaches supervision flexibly

.60

puts effort into a good start

.50

provides information on availability

.66

initiates contact

.65

negotiates their availability

.70

ensures official requirements are met (ethics clearance)

.43

MENTORINGdemonstrates genuine interest in well-being

.53

approachable, responsive, and affirming .62 .41encourages publishing

.47

directs me to leading researchers

.66

encourages networking within the university

.59

introduces to professional networks

.73

helps arrange for presentations at seminars and conferences

.65

assists progress career goals .49 .42advises on opportunities for relevant experience

.54

SPONSORINGassists obtain resources for seminars and conferences

.69

assists to meet administrative requirements in an efficient and timely manner

.49 .41 .41

keeps informed about procedures and issues re intellectual property rights

.47

advises about resources and other funding sources

.49

Dow

nloa

ded

by [

Dal

hous

ie U

nive

rsity

] at

18:

37 2

5 O

ctob

er 2

012

110


Discussion

This study was conducted to investigate the efficacy of a conceptual framework thatattempts to describe the operational domain of postgraduate research supervisory practice asa basis for enabling critical reflection on supervisory practice in academic developmentprograms. The framework posited research supervisory practice as a facilitative processinvolving a range of educational tasks and activities. In the main, the framework is supportedin that the factors that have emerged from the questionnaire approximate the hypothesisedconstructs. There is also a strong relationship between the RSQ and the supervisory sub-scale of the PREQ. Further there is a very strong relationship between the domain asdescribed by the items in the questionnaire and students’ overall satisfaction with theirsupervisors’ practice.

The validity of the structure of the five hypothesised constructs of facilitative supervisorypractice—Expert Coaching, Mentoring, Sponsoring, Facilitating, Reflective Practice—are notconfirmed in full. The four factors that result from the factor analysis are not aligned in allrespects with the original constructs. However, the items cluster in these new factors in apattern that is confirmatory of the validity of the structure of the hypothesised constructsCoaching, Facilitating, Mentoring, and Sponsoring as indicated in Table 3. There are threemain ways in which the items cluster that are of particular interest and suggest a set of inter-pretative labels to assign to each factor. First, Factors 3 and 4 include a strong cluster of theexpected items (3 out of 4) for Coaching with a focus on the research project and Facilitatingthe candidature, making Coaching the Research Project and Progressing the Candidatureself-evident descriptions. Second, Factor 2 confirmed the close relationship hypothesised atthe outset of the two constructs of Mentoring and Sponsoring, of which the Sponsoring itemsare independent of the other factors. Additionally, the wording of the two coaching items,C36 and C37, each with a rider that the supervisor provided expertise or assistance ordirected the student to those who could, suggests the possible reason for their inclusion inFactor 2. Directing students to alternative sources of expertise is consistent with the functionof sponsoring.

Factor 1 is the least consistent with expectations. Although the construct items for Reflectivepractice cluster most strongly in Factor 1, it is a mix of construct items concerning reflective

Table 3. Continued

Hypothesised Constructs Factor 1 2 3 4

SET 2

REFLECTIVE PRACTICEmodels effective research practice .46 .41open to different research approaches

.73

open to critical discussion on research practice

.61

periodically reviews our supervision

.75

carries out supervisory duties professionally .45 .42 .46encourages open/critical discussion on research practices

.64

Dow

nloa

ded

by [

Dal

hous

ie U

nive

rsity

] at

18:

37 2

5 O

ctob

er 2

012


111

practice, interpersonal interaction and support. In this mix the items focused on the intellec-tual aspect of the interaction are dominant with the highest loading, suggesting initially thelabel “intellectual leadership”. However “leadership” for many still carries with it culturalbaggage that “strong leaders” direct and impose. Such connotations are contrary to theemphasis on making space and opportunity for open enquiry in the items included in Factor1. They are also contrary to the inclusion of items denoting emotional and personal support inFactor 1 as well as contrary to the expectation that personal support is a feature of theconstruct Mentoring.

A solution to these discrepancies emerged from rethinking the nature of mentoring.Mentoring is a term used frequently in research supervision discussions but usually to encom-pass a range of dimensions such as personal support; career development; and the introduc-tion to academic/professional networking. It was in this sense that it is described in Table 1.In this definition intellectual development is not profiled, although the established relation-ship of the emotional and intellectual dimensions in the literature on experiential learning andfacilitation, as well as in research training, has been established (Boud, Keogh, & Walker,1985; Cullen et al., 1994; Heron, 1989; Reason & Marshall, 1987). On re-examination ofFactor 2 it is of note that all the items included in Factor 2 refer to activities of one sort oranother to introduce the student to resources and networks but do not relate to enculturationinto the intellectual and disciplinary/professional domain. The factor analysis has thusdiscriminated between those aspects of mentoring which are similar to sponsoring, andincluded those items referring to personal support with the items which focused on the intel-lectual development aspect of supervisory interaction to form Factor 1. Following this re-examination we have given the label of Mentoring to Factor 1, and Sponsoring to Factor 2, asolution that contributes to further “unpacking” of research supervision practice, and hasextended our initial conceptual framework.

It is possible to suggest that the resultant factors with their interpretative labels could offer auseful shift in focus in interpretation. The hypothesised constructs were focused more on thetype of engagement a supervisor might have with a research student. The factors give a clearersense as to the context and purpose of the sets of activity that they represent. These sets ofsupervisory activities could be conceived as subsets of the overall task of supervision as facili-tation (or “teaching” following Connell, 1985). A final set of labels reflect this approach:Progressing the Candidature, Mentoring, Coaching for the Research Project, and SponsoringStudent Participation in Academic/Professional Practice. These factor labels give more infor-mation as to context and purpose for the behaviours described in each set of items, and haveobvious value for academic development purposes.

An issue yet to be resolved concerns how to determine the extent of an individual supervi-sor’s responsibility. The development of a model such as that presented in this paper couldlead to the expectation that it describes “best practice” and the goal is to be rated well on allitems, factors and constructs. This would be contrary to our intent to develop an instrumentsufficiently flexible to avoid setting up the supervisor to be responsible for everything—the“Atlas complex”. There is a tension here between avoiding such an outcome on one hand, yetnot reverting to the view that supervisory practice is all a matter of individual styles and atti-tudes. For example, in Brown and Atkins (1988) a list of roles is given that represent sets ofbehaviours (some of which are given as examples), but the roles appear to be in the nature ofoptions to be adopted as a supervisor wishes.

Dow

nloa

ded

by [

Dal

hous

ie U

nive

rsity

] at

18:

37 2

5 O

ctob

er 2

012

112


In contrast is a student view (Pearson & Ford, 1997, p. 87) that “how supervision is done isless important than the fact that it is done”. Or as Mintzberg, in discussing the work ofmanagers (1980, pp. 54–58) states, where roles are conceived as “organised sets of behavioursbelonging to an identifiable office or position”, individuals may vary in how they perform arole “but not that it is performed”. Particularly where a supervisor is the institutional link(usually the designated principal supervisor) there are responsibilities whose performance isessential in enabling students to manage their candidature (Pearson, in press). Even for thosewho may have a less formal role the provision of relevant expertise, assistance in networkingand so on can make the difference between a passable or high quality experience. In situationsof distributed or shared responsibility for supervision a strength of the RSQ could be its use inidentifying how all relevant tasks and activities are being attended to; or conversely a researchstudent might be assisted to find out what is missing and seek that assistance from others.

The initial intention of the authors was to design the RSQ so that it would only be used inan academic development context as a diagnostic tool. Drawing on their experience in senioruniversity leadership and management training (Pearson, 1995; Kayrooz, Pearson, & Quin-lan, 1997) the authors proposed to use the RSQ for a modified 360° feedback approach (forexample, Collins & Saul, 1991). These instruments are used in corporate workplace situa-tions similar to that of research supervision with similar issues as to intimacy and asymmetricpower relations. For the RSQ, the authors suggest key aspects of such a professional develop-ment process could be: to have the supervisor invite others (primarily students though somemay invite peers) to give them feedback; to make self-assessment mandatory; to only giveback feedback from respondents that is aggregated; and to provide guidance in the interpreta-tion of any feedback by explaining the conceptual model on which it is based with reference tothe underpinning research. This leaves the supervisor in charge of the process within limitsthat protect the respondents and the integrity of the process.

The current press for quality assurance however means that any instrument based on thefindings of our study so far, especially in an area so poorly served is likely to be used for vari-ous purposes. The RSQ could be used as an heuristic for inexperienced supervisors, anapproach now under development. It could be used by students to diagnose how satisfactorytheir supervision is with a view to seeking additional supervisory help. It could also be used asthe basis of student and supervisor conversations. Given the convergent validity with PREQ,the instrument might attract interest for institutional performance management purposes,however the very small populations of available respondents would invalidate such use. Onepossible consequence of addressing this issue is for us to consider an alternative five-pointscale other than “strongly agree—strongly disagree” and the generation of numbers whichsuggest an affinity with performance ranking instruments. Some such change could clarify theinstrument’s value in identifying strengths and weaknesses, with a clear expectation that noone supervisor should necessarily cover the domain.

Conclusion

While the paper outlines in detail the discrepancies between the expected and resultant, itshould be noted that they were minor in this study. On the whole, the study confirms thesignificance of the selected items of the RSQ and their underlying factor structure. The studyconfirms that supervision can be framed as a set of tasks and responsibilities that can be

Dow

nloa

ded

by [

Dal

hous

ie U

nive

rsity

] at

18:

37 2

5 O

ctob

er 2

012


113

clustered and operationalised. In so doing, this study maps the domain of postgraduateresearch supervisory practice in a structured and systematic manner. It explicitly delineateswhat facilitation might mean in the context of research supervisory practice. Structured as it ison key elements of supervision it leaves supervisors to adapt their practice to their varyingcontexts and contingencies. It gives substance to the claim that traditional notions of intellec-tual development and inquiry are important to research supervision and it reasserts the inter-relationship between the emotional and intellectual aspects of facilitative leadership. TheRSQ thus provides the basis for a feedback instrument structured on an explicit model ofsupervisory practice with sufficient explanatory power to provide guidance in interpretation,and to provoke discussion and critique.

Acknowledgements

This paper was presented in an earlier form to the SRHE Annual Conference in Glasgow in2002. We would like to acknowledge the considerable contribution provided by our statisti-cal advisor, Robin Collins of the Centre for Educational Development and AcademicMethods, at The Australian National University. We would also like to acknowledge colle-gial commentary by colleagues at a seminar on this work at the University of Technology,Sydney.

Notes on Contributors

Margot Pearson is active in research on research supervision and research education morebroadly. Her publications draw on studies carried out at the institutional and nationallevel. She is a foundation member of the Steering Group for the Inter University Consor-tium developing a web-based resource for research supervision (fIRST), and teaches aGraduate Certificate unit on research supervision at the Australian National Universitywhere she is a Reader in the Centre for Educational Development and AcademicMethods.

Carole Kayrooz is an Associate Professor in Education in the Division of Communication andEducation at the University of Canberra, and Deputy Head of the Division. At the timeof writing, she was Director of Postgraduate Studies, overseeing the research doctoraland masters programs within the Division. She has taught research methodologystudents at both undergraduate and postgraduate levels and successfully supervised manyresearch students.

References

Boud, D., Keogh, R. & Walker, D. (1985). What is reflection in learning? In D. Boud, R. Keogh & D. Walker(Eds.),

Reflection: Turning experience into learning

(pp. 7–17). London: Kogan Page.Barnett, R. (1990).

The idea of higher education.

Milton Keynes, England: Society for Research into HigherEducation and Open University Press.

Brew, A. (2001). Conceptions of research: A phenomenographic study.

Studies in Higher Education, 26,

271–285.Brown, G. & Atkins, M. (1988).

Effective teaching in higher education.

London: Routledge.Collins, R. & Saul, P. (1991).

Management competencies development workbook

. Roseville, NSW: McGraw-Hill.

Dow

nloa

ded

by [

Dal

hous

ie U

nive

rsity

] at

18:

37 2

5 O

ctob

er 2

012

114


Connell, R. W. (1985). How to supervise a PhD,

Vestes, 2,

38–41.Cullen, D. J., Pearson, M., Saha, L. J., & Spear, R. H. (1994).

Establishing effective PhD supervision.

Evalua-tion and Investigations Program. Canberra, ACT: Department of Employment, Education, Trainingand Youth Affairs.

Delamont, S., Atkinson, P., & Parry, O. (1997). Critical mass and doctoral research: reflections on the HarrisReport.

Studies in Higher Education, 22,

319–331.Edwards, E., Aspland, T., Litster, D., O’Leary, J., Ryan, Y., Southey, G., & Timms, P. (1996).

Trackingpostgraduate supervision

, QUT Postgraduate Supervision Evaluation Project. Brisbane, QLD: QueenslandUniversity of Technology.

Evans, T. & Pearson, M. (1999). Off-campus doctoral research and study in Australia. In A. Holbrook &S. Johnston (Eds.),

Supervision of postgraduate research in education,

Review of Australian Research inEducation, No. 5, (pp. 185–203). Melbourne, VIC, Australian Association for Research in Education.

fIRST. (n.d.).

For improved supervision and research training.

Retrieved 4

th

August, 2004, from fIRST.edu.au.Fells, C. & External Reference Group. (2004). Evaluation of knowledge and innovation reforms. Consultation

Report. Department of Education Science and Training, Commonwealth of Australia. Retrieved 4th

August, 2004, from http://www.detya.gov.au/highered/ki_reforms/Ford (Marck), L. (1996). Clarifying the transmission of research culture at ANU. Paper presented at the

National Conference: Quality in Postgraduate Research: Is It Happening? Adelaide. Retrieved 4th

August, 2004, from http://www.canberra.edu.au/celts/QPR/1996.html.Fraser, R. & Matthews, A. (1999). An evaluation of the desirable characteristics of a supervisor, Australian

Universities Review, 42, 5–7.Gallagher, M. (2000). New directions in Australian research and research training policy – some questions for

researchers. Paper presented at the Annual Conference for the Australian Network for Higher EducationPolicy Research, Australian National University, Canberra, ACT, December.

Green, H. (2002). The quality agenda: the UK experience. In M. Kiley & G. Mullins (Eds.), Quality inpostgraduate research: Integrating perspectives, proceedings of the 2002 International Quality in Postgradu-ate Research Conference, (pp. 4–22). Adelaide, SA.

Green, W. & Lee, A. (1995). Theorising postgraduate pedagogy. The Australian Universities Review, 38(2),40–45.

Gurr, G. M. (2001). Negotiating the “rackety bridge” – a dynamic model for aligning supervisory style withresearch student development. Higher Education Research & Development, 20, 81–92.

Guthrie, B. & Trembath, R. (1998) Researching the researchers: The pilot postgraduate research experiencequestionnaire. In M. Kiley & G. Mullins (Eds.), Quality in postgraduate research: Managing the newagenda, proceedings of the 1998 Quality in Postgraduate Research Conference, Adelaide, April.Retrieved 4th August, 2004, from http://www.canberra.edu.au/celts/QPR/1998.html.

Harris, M. (1996). Review of postgraduate education (2 vols). Bristol: Higher Education Funding Council forEngland.

Heron, J. (1989). The facilitators’ handbook. London: Kogan Page.Kayrooz, C. (1999). Metacognitive methods for training in problem formulation. Doctoral dissertation.

Sydney, NSW: Macquarie University.Kayrooz, C. & Kiley, M. (2002) A developmental framework for postgraduate study. Quality in Postgraduate

Education conference proceedings, Adelaide, South Australia. Retrieved 4th August, 2004, from http://www.canberra.edu.au/celts/QPR/ByAuthor.html.

Kayrooz, C., Pearson, M. & Quinlan, K. M. (1997). Development from within academe: eschewingimperialism, managerialism and missionary zeal? International Journal for Academic Development, 2,64–71.

Kayrooz, C. & Trevitt, C. (2004). Research in organisations and communities: Tales from the real world. CrowsNest NSW: Allen & Unwin.

Kemp, D. A. (1999) Knowledge and innovation: A policy statement on research and research training. Canberra:Department of Education, Training and Youth Affairs.

Lapidus, J. B. (1997). Issues and themes in postgraduate education in the United States. In R. G. Burgess(Ed.), Beyond the first degree. Milton Keynes, England: Society for Research into Higher Education andOpen University Press.

Dow

nloa

ded

by [

Dal

hous

ie U

nive

rsity

] at

18:

37 2

5 O

ctob

er 2

012

Enabling Critical Reflection 115

Latona, K. & Brown, M. (2001). Factors associated with completion of research hgher degrees. Report No. 37,Higher Education Series. Canberra, ACT: Department of Employment, Education, Training and YouthAffairs.

Lave, J. & Wenger, E. (1991). Situated learning: Legitimate peripheral participation. Cambridge: CambridgeUniversity Press.

Malfroy, J. & Webb, C. (2000). Congruent and incongruent views of postgraduate supervision. In M. Kiley &G. Mullins (Eds.), Quality in postgraduate research: Making ends meet. Proceedings of the 2000 Quality inPostgraduate Research Conference (pp. 165–177). Adelaide, SA, April.

Mintzberg, H. (1980). The nature of managerial work. Englewood Cliffs, NJ: Prentice-Hall.Mullins, G. & Hejka, E. (1995). The evaluation of postgraduate supervision. In M. Pettigrove & M. Pearson

(Eds.), Higher education in transition. Proceedings of Higher Education Research and DevelopmentSociety of Australasia Annual Conference, Canberra, ACT.

Neumann, R. (2003). The doctoral education experience. Evaluation and Investigation Programme. Canberra,ACT: Department of Education, Science and Training, Commonwealth of Australia.

Pearson, M. (1995). Contextualising leadership skills for academics. In M. Pettigrove & M. Pearson (Eds.),Higher education in transition. Proceedings of Higher Education Research and Development Society ofAustralasia Annual Conference, Canberra.

Pearson, M. (1996). Professionalising PhD education to enhance the quality of the student experience.Higher Education, 32, 303–320.

Pearson, M. (1999a). The changing environment for doctoral education in Australia: Implications for qualitymanagement, improvement and innovation. Higher Education Research & Development, 18, 269–287.

Pearson, M. (1999b). Rethinking the past and the future of the PhD in the Humanities and Social Sciences, aresponse to: “Forged in Fire: Narratives of Trauma in PhD Supervision Pedagogy”, Alison Lee andCarolyn Williams. Southern Review, 32, 186–190.

Pearson, M. (2000). Flexible postgraduate research supervision in an open system. In M. Kiley & G. Mullins(Eds.), Quality in postgraduate research: Making ends meet. Proceedings of the 2000 Quality in Postgradu-ate Research Conference, (pp. 103–118). Adelaide, SA, April.

Pearson, M. (2001). Research supervision – Mystery and mastery. In J. Higgs & A. Titchen (Eds.), Practiceknowledge and expertise in the health professions (pp. 192–198). Oxford: Butterworth-Heinemann.

Pearson, M. (in press). Changing contexts for research education: Implications for supervision development.In P. Green (Ed.), Postgraduate research supervision: Contexts and processes, theories and practices.Melbourne, VIC: RMIT Press.

Pearson, M. & Brew, A. (2002). Research training and supervision development. Studies in Higher Education,27, 135–150.

Pearson, M. & Cryer, P. (2001). “Training the trainers”: Current initiatives in enhancing research supervi-sion in Australia and the UK. Higher Education Research and Development Society of Australasia News, 23,16–18.

Pearson, M. & Ford, L. (1997). Open and flexible PhD study and research, Evaluation and InvestigationsProgram. Canberra, ACT: Department of Employment, Education, Training and Youth Affairs.

Reason, P. & Marshall, J. (1987). Research as personal process. In D. Boud & V. Griffin (Eds.) Appreciatingadults learning: From the learner’s perspective (pp. 112–126). London: Kogan Page.

Selected higher education statistics, students (2002). Tables. Department of Education, Science andTraining. Retrieved 4th August, 2004, from http://www.dest.gov.au/highered/index.htm.

Senge, P. M. (1990). The fifth discipline. New York: Doubleday.Smith, B. (2001). (Re)Framing research degree supervision as pedagogy. In A. Bartlett & G. Mercer (Eds.),

Postgraduate research supervision: Transforming (r)elations (pp. 25–41). New York: Peter Lang Publishing.Statistical Package for the Social Sciences. (2001). Author version 10.0.8 for the Macintosh. Chicago: SPSS

Inc.Tabachnick, B. S. & Fidell, L. S. (1996). Using multivariate statistics. Northridge: Harper Collins.Thompson, J., Pearson, M., Akerlind, A., Hooper, J., & Mazur (2001). Postdoctoral training and employment

outcomes, Evaluation and Investigations Program. Canberra, ACT: Department of Employment,Education, Science and Training.

Dow

nloa

ded

by [

Dal

hous

ie U

nive

rsity

] at

18:

37 2

5 O

ctob

er 2

012

116 M. Pearson and C. Kayrooz

Vilkinas, T. (1998). Management of the PhD Process: the challenging role of the supervisor. In M. Kiley &G. Mullins (Eds.), Quality in postgraduate research: Managing the new agenda. Proceedings of the 1998Quality in Postgraduate Research Conference, Adelaide, April, 1-13.

Wijesundera, S., Hicks, O., & Mann, S. (1996). Exploring postgraduate supervision: Student and supervisorperceptions. Paper presented at the National Conference: Quality in Postgraduate Research: Is It Happen-ing? Adelaide.

Wisker, G. (2000, April). Postgraduate learning styles and enabling practices: A multicultural action researchstudy. In M. Kiley & G. Mullins (Eds.), Quality in postgraduate research: Making ends meet. Proceedings ofthe 2000 Quality in Postgraduate Research Conference (pp.145–164). Adelaide.

Dow

nloa

ded

by [

Dal

hous

ie U

nive

rsity

] at

18:

37 2

5 O

ctob

er 2

012

Date post:	15-Oct-2016
Category:	Documents
Upload:	carole
View:	212 times
Download:	1 times

Enabling critical reflection on research supervisory practice

Documents