Students' perceptions of supervisory qualities: What do students want? what do they believe they receive?

Aim/Purpose This paper explores students' perceptions of qualities they believe their ideal supervisor should possess as well as those they see as characterizing their current and past supervisors. Background Over more than three decades, multiple cultural contexts and diverse methodologies, research studies have demonstrated that what person re-lated human qualities in postgraduate research supervision have greater valence for students than does discipline/research expertise. This paper probes why this might be so. Methodology Across 15 Australian universities and all disciplines 698 students partici-pated in an opt-in online survey which invited students to provide de-scriptors of their supervisors' qualities as well as those of their ideal su-pervisor. The survey was student centred in that it required them to nom-inate the qualities of their supervisor/s rather than asking them to re-spond to statements about supervisors/supervision on a Likert scale. Contribution This research which was designed to allow students to characterise their actual supervisors and their ideal supervisor in an unconstrained and anonymous way demonstrated their dominant valuing of, firstly, human traits consistent with emotional intelligence and, secondly, the profes-sional aspects of supervision especially in relation to research process. In providing a snapshot of the janus face of supervision, these uniquely student generated perspectives on supervisory qualities provide data not only supportive of previous studies with very different methodologies but also with implications for supervisor development programs and su-pervisor benchmarking within universities. Findings The resultant student initiated perceptions of positive and negative quali-ties of supervisors support the findings of other studies which show that students value and seek cognitive and affective person related qualities in supervisors over discipline/research expertise qualities. For 25 percent of the sample there were no qualities in common between their principal supervisor and their ideal; this increased to 50 percent with one quality in common. Recommendations for Practitioners In developing and honing individual philosophies of supervision, super-visors should reflect, for example, upon the ways in which they present to and interact with students as individuals, their availability to students, their interest in students' research and career development. Those deliver-ing supervisor development programs should consider the balance in such programs between process- oriented material and human interaction strategies. Recommendation for Researchers Research in the doctoral space has tended to be summative as in post completion evaluations of the experience or cross-sectional sampling of experience or what is valued as in the current study. Longitudinal research which samples perspectives both within and beyond candidature is need-ed. This should thus encompass the experiences of those who complete and those who do not over a period of perhaps six years. Impact on Society Globally since the late 1990s, universities have initiated doctoral training programs and codes of conduct pertaining to the supervisory relation-ship yet evidence suggests that supervision issues remain vexatious. The sector thus needs to address the efficacy of such programs in ameliorat-ing issues raised by students. The silent acknowledgement of late stage doctoral attrition - and the lack of follow up as to the complex interrela-tionship of factors prompting such a personally difficult and societally wasteful decision - remains a besetting problem for the sector. Future Research Two critical issues would usefully guide future research in the doctoral education space. Firstly, the ultimate efficacy of supervisor development programs requires evaluation and follow up. Secondly, the perspectives of those who exit the PhD process virtually without trace need to be inves-tigated and evaluated for policy implications. Further some respondents in this study had supervisory roles themselves and the qualities they at-tributed to self as supervisor were closer to the ideal than those of real supervisors. This suggests that a more extensive investigation of how supervisors see themselves in the supervisory role would be useful as such research would potentially impact on the nature of supervisor de-velopment programs in the future.