Literature Review

To supervise means to direct or oversee the performance or operation, and a supervisor is defined as a person who . . supervises; (in some British universities) a tutor supervising the work, esp. research work, of a student (Collins, 2003). Within these broad definitions, the role of a research supervisor and the nature of the supervision relationship are interpreted very widely and very differently. The role of a supervisor has been seen as: an authority on the particular topic being researched (Zuber-Skerritt & Ryan, 1994); a guardian of standards (Cryer, 1997); a mentor (Zuber-Skerritt & Ryan, 1994); a research role model (Wisker & Brown, 2001); a manager (Vilkinas, 2002); and even a coach (Rochford, 2003).

The nature of the relationship between the supervisor and the student has also been warmly debated, with some supervisors recognising nothing more than a quasicontractual relationship and others who believe that it is also a personal relationship. Grant & Graham (in Zuber-Skerritt & Ryan, 1994,) note that university staff . . . at times vehemently contest the use of that word [relationship] in connection with supervision, insisting that it is an academic matter only. Depending on what a supervisor believes the nature of the relationship to be, he or she would interpret the role and duties of a supervisor very differently. The student expectations of the nature of a supervisors role and duties also differ widely. The role, duties and expectations would, in turn, determine the nature of the training supervisors should receive. The supervisor -student relationship is one of the most important aspects of research supervision. The best supervisor -student relationships approach the platonic ideal of mentor and protg. Good or bad, the quality of the relationship between student and supervisor
1

directly influences the quality of the supervision experience, as research has shown. A positive relationship has positive outcomes for students, including a positive departmental environment (Hartnett, 1976), successful socialization into the department and discipline (Gerholm, 1990; Weiss, 1981), and timely completion of the degree (Girves & Wemmerus, 1988; Long, 1987Lovitts, 2001). An unsatisfactory advising relationship is strongly implicated in many students decisions to leave doctoral study (Golde, 1996; Jacks, Chubin, Porter, & Connolly, 1983; Lovitts, 2001)

Effective supervision requires supervisors to be knowledgeable and skilled in the research field (Donald et al., 1995; McQueeney, 1996; Sheehan 1994).They are also expected to take the lead in establishing a quality of relations which will give their students access to the knowledge and skills they possess (Ballard and Clanchy, 1993) and to have counselling skills (Hockey, 1997; Zuber-Skerrit, 1994). Students not only expect their supervisors to have the knowledge and ability to supervise in a particular area of research but also want them to be reasonable, serious, supportive of their work in good times and bad, and approachable (Moses, 1985). Moses (1992) considers that supervisors standards, similar interests and regular contact According to Moses (1994), supervisors should at least have an equivalent degree to the one the student is studying for and, if this is not the case then, they must have a solid background of research involvement and publications. Brown and Atkins (1988) suggest that, to supervise effectively, one has to be a competent researcher and to be able to reflect o research practices and analyze the knowledge, techniques and methods that make them effective. Frischer and Larsson (2000) and Phillips and Pugh (2000) different view, in that they suggest that students take a slightly are recommended to select a should act as mentors and that a mentoring relationship requires mutual aspect based on high academic

supervisor based on the key factor of whether the latter has an established research record and is continuing to contribute to the development of his or her discipline. This takes account of whether the person has recently published research, holds research grants and is invited to speak at conferences in their own country or abroad. Therefore, an effective supervisor should satisfy such criteria. Spear (2000) supports this statement and adds that
2

often it will be sufficient for the supervisor to be competent in the general area of the students research even if not expert in the detailed area of the thesis topic. Yeatman (1995) gave a similar view, stating that good supervisors must have a track record in successfully bringing through a large number of Ph.D candidates. According to Burton and Brueckner (1995), the primary function of

supervisions of all types is leadership, plus the encouragement and recognition of leadership in other people, either on the professional staff or among community participants. On the other hand, Phillips and Pugh (2000) and Zubir Skerrit (1994) advised supervisors to act as role models. Frischer and Larsson (2000) described three different pattern of leadership, which are called democratic, authoritarian and laissezfaire leader. Cullen et al. (1994) in his study noted indicators of supervisor effectiveness in four major categories. These categories have been supported independently by researchers over the years. For example, the importance of academic standing was highlighted by Moses (1994) and supervisory competence by Zuber Skerritt (1994), ESRC (2001) and Zhao (2003)

What makes for good supervisor? The qualities of a good advisor include supportiveness (Long, 1987), high levels of interaction (accessibility, frequent informal interactions, and connections with many faculty members) (Gerholm, 1990; Girves & Wemmerus, 1988; Hartnett, 1976; Weiss, 1981), purposefully helping the student progress in a timely manner (Heiss, 1970; Lovitts, 2001; Rudd, 1986), providing regular reviews of progress (Hartnett, 1976; Heiss, 1970), and treating the student as a junior colleague (Girves & Wemmerus, 1988).