ABSTRACT Postgraduate research student numbers are increasing, the need for us to nurture and develop excellent researchers and the explore the purposes of doing so is in debate (Park 2007). The relevant literature falls into three categories. Firstly that on research supervision is largely practical, focusing on the functions that a supervisor needs to undertake and managing the emotional relationship that can build up between student and supervisor over a number of years (eg Wisker 2005, Taylor & Beasley 2005, Cryer 2000). Secondly there is a large body of literature on academics’ practice as teachers which has been conceptualised as the transmission/constructivism binary. Thirdly there is a literature on academics’ concepts of research which has been undertaken from a phenomenographic perspective. Brew (2003) links these to communities of practice as a way of bringing teaching and research together. Initial interviews with research students suggested that in some cases they resented what they experienced as coercive or negligent supervisory behaviour. Interviews with supervisors suggested that there are a range of conceptions of what research supervision is about. The research data includes the analysis of interviews with twelve supervisors (from across a range of disciplines at a university in the South of England) who between them had experience of supervising over 100 students. It also reflects experiences of these supervisors at a range of eight other UK universities. The methodological approach was based on grounded theory and used a range of qualitative research methods. This paper proposes a model of five different approaches to research supervision. The model aims to provide a neutral language for analysing conceptions and expectations in the supervisory relationship. This is important if it is to be acceptable to supervisors from a wide range of disciplines and to provide an opportunity for co-supervisors to express their beliefs in a non-confrontational way. The model is based upon a review of the literature and the qualitative analysis of interviews with doctoral supervisors, students and supervisors-as-students. A pedagogy of research supervision is beginning to be formed out of a conceptual approach to the subject, this paper contributes to that debate and describes how the proposed model can be used to support the development of supervisors, supervisory teams and supervisor/student relationships. This session will outline the research and then participants will be able to analyse their own experience in the light of the five different approaches to supervision that the research identified.

Qualitative research. Using interviews and focus groups. Analysed into themes and concepts and iteratively reviewed in the light of the literature.

That a conceptual approach to research supervision is important for supervisory development.