Chapter 10 - Teaching and learning: graduate education

I: Introduction

10.1 Since the early 1960s, there has been a huge increase, both in Oxford and nationally, in the number of students on graduate courses.[118] This expansion of graduate education has occurred in both taught and research degree courses and has arisen in response to a variety of impulses from the government, employers, graduates and the academic community:

• the government policy of the 1960s, the early 1980s and the early 1990s to expand overall student numbers;
• the increasing need for those wishing to pursue an academic career to have a doctorate;
• the expansion of scientific research in industry, with its need for highly qualified laboratory workers;
• an increase in the number of specialist careers in industry and government requiring graduate study;
• the perceived advantage of a graduate qualification in a highly competitive graduate job market;
• the increasingly frequent requirement from funding bodies that graduate students should undertake a taught course before embarking on a research degree;
• in universities, the development of the essential role of graduate researchers in scientific research, the beneficial cross-fertilisation between some graduate teaching and academics' own research, and the financial benefit of the income from graduate student fees (especially from overseas students).

10.2 As was recognised by the Franks Commission,[119] Oxford has been well-placed to respond to the increased demand for graduate education, as it has much to offer graduate students : a strong research base and the opportunity to be supervised by leading researchers; an extensive library system, with many unique special collections; opportunities to be involved in scientific projects in well-equipped laboratories and at the leading edge of strategic research; and participation in a highly talented, international student body. The provision of advanced level taught courses and research degrees is a coherent and logical development of Oxford's traditional teaching role, aimed as it is at extending knowledge, stretching the capabilities of the most academically able and developing a range of analytical and research skills applicable to many careers. In return, graduate students offer other members of Oxford's academic community a valuable stimulus and fresh approach to their own study and research. Research students also support the work of leading academics, especially in the natural sciences.

10.3 Oxford has therefore planned and experienced a marked increase in its number of graduate students, both in absolute terms and as a proportion of overall student numbers in the University. This development is illustrated in Figure 10.1. These changes have had a significant impact on the pattern of the University's activities. Our survey of academic staff in Hilary Term 1996 revealed that 75 per cent of academic staff were supervising graduate students and that, on average, each had 4.8 supervisees. In comparison, the Franks Commission's survey of staff in 1964 indicated that 67 per cent of academic staff were supervising graduate students, but that each had an average of only 2.5 supervisees. Furthermore, Figure 10.1 demonstrates not only that there has been overall growth, but also that the growth has been most pronounced in the sciences and medicine, and in taught courses. This shift of emphasis has been significant both because of the amount of academic staff time demanded by taught courses and because of the equipment and space required to support research students in the sciences and medicine.

Figure 10.1: Numbers of graduate students: Oxford University: 1965-6 and 1995-6

10.4 In Oxford as elsewhere, this rapid change in the composition of the student body has necessitated a number of adaptations to a system which had evolved primarily to accommodate undergraduate teaching and research. The changing requirements led to a number of national reviews of the position of graduate students, chief amongst which was the Review of Postgraduate Education (the 'Harris Report') published in 1996. The main concerns of that report were the nature of the funding regime for graduate taught and research courses, the impact of graduate expansion on undergraduate funding, the need for clarification and standardisation of graduate qualifications, and the standards of teaching and facilities for graduate students. National bodies such as the Higher Education Quality Council, the research councils and the British Academy have attempted to identify the academic and other needs specific to graduate students, and the funding of institutions which provide graduate courses is increasingly contingent on their ability to answer those needs. Both nationally and in Oxford these changes have led to increased attention to issues such as training for graduate students, the availability of library, computing and laboratory facilities, admissions procedures and the organisational structures most appropriate for graduate education.

10.5 These issues have particular significance in Oxford for a number of reasons. First, Oxford has a relatively high proportion of graduate students: 30 per cent of its total student body as opposed to 14 per cent nationally. Secondly, nearly all Oxford's graduate students study on a full-time, residential basis, and therefore have high expectations of full academic and social integration into the University, of the quality and intensity provided for undergraduates. Thirdly, and perhaps most significantly, the collegiate nature of the University has given an added dimension to all these issues, and in particular has raised the question of the extent to which colleges can and should be involved in academic provision for graduate students. Oxford itself has therefore undertaken a number of reviews of its own provision for graduate students, the principal ones of which may be summarised as follows.

10.6 The Franks Report.[123] As we noted in paragraph 1.24, ^C publication, in 1963, of the Franks Commission was set up shortly after the Robbins Report on higher education in the UK. It recognised that Oxford was in a particularly good position to contribute to the considerable expansion in graduate student numbers recommended in the Robbins Report, and predicted and endorsed an increase in Oxford's graduate student numbers to between 3,500 and 4,000 by 1985, which would represent 30 per cent of the total student body. This prediction proved fairly accurate, and more or less reflects the current position.

10.7 To ensure that this increasing number of graduate students was well-provided for in Oxford and to ensure that graduate students could be fully incorporated into the collegiate University, Franks made a number of recommendations, which included:

• parity of esteem for graduate and undergraduate education, which were to be provided by the same academic staff;
• the establishment of faculty centres for all the major humanities and social sciences faculties;
• more systematic and consistent training for graduate students by way of seminars, background lectures etc;
• the provision of private funding to encourage more graduates from other UK and overseas universities to come to Oxford;
• the rationalisation of degree nomenclature;
• measures to ensure standards and the quality of examining;
• the continuation of compulsory college membership for graduate students, with a corresponding obligation on colleges to take on some academic responsibility (chiefly through the provision of a college adviser, on which see paragraphs 10.38 to 10.40) and a recognition that graduate students had as much claim on college facilities as undergraduates had;
• changes in the University's governance and administration aimed at bringing graduate studies into a more prominent position in the University's thinking.

10.8 The Franks' recommendations were endorsed by the University, but their implementation was partial and slow. This lack of consistent progress prompted the University to undertake three further reviews, which we outline below.

10.9 The 'Roberts Report' (The Report of the Committee of Inquiry into Provision for Graduate Students - 1987). The Roberts Report was entirely concerned with graduate education, and was able to go into many issues in more depth than Franks. Its recommendations covered many aspects of graduate education, prompted by a recognition that Oxford had still not adapted sufficiently to take account of the increased role of graduate studies, especially in the humanities and social sciences. Amongst the issues covered were:

• The admissions system - Roberts made recommendations intended to reduce the delays inherent in the admissions system for graduate students, to make it easier to compare applicants with one another in a 'gathered field', and to improve the information given to applicants about the choice of colleges.
• Induction and training - Roberts recommended that Oxford should give more attention to the induction of graduate students, training in research methods, and teaching to enable them to maintain a broad knowledge of their subject.
• The role of the colleges in graduate provision - Roberts, like Franks, believed that it was beneficial to graduate students, and crucial for Oxford, that colleges play an academic role in the education of their graduate students, as well as providing a variety of social facilities.
• Supervision - The report drew attention to significant concern in some subjects about the system for allocating applicants to supervisors, the performance of supervisors and the monitoring of supervision. One reason for the concern which the report identified was 'a lack of agreement, because of a lack of knowledge, of what the supervisors' duties should be'.[124] The report therefore set out the responsibilities of a supervisor in the form of draft guidelines, which are now embodied in the Memorandum for Supervisors and Research Students issued by the General Board.[125] The Roberts Report also considered that the distribution of the responsibilities of supervision amongst members of academic staff was too uneven, so that some staff were overloaded and unable to give adequate attention to each supervisee. The report recommended that faculties should set norms and maxima for the number of supervisees for each member of staff.

10.10 The 'Southwood Report' (the report of the Working Party on Graduate Provision - 1990) and The Graduate Studies Committee's follow-up to the Working Party on Graduate Provision (1994). Both these reports sought to clarify what type of academic provision colleges were best able to make, and to assess the extent to which colleges had enhanced their provision in line with earlier recommendations. Amongst their conclusions were a clarification of the roles of the tutor for graduates and the college adviser, encouragement to colleges to hold 'graduate collections' (i.e. progress meetings with graduate students) and the identification of a minimum set of facilities which colleges should provide for their graduate students.

10.11 As a result of these reviews, Oxford has now added greatly to its provision for graduate students, and is widely acknowledged as providing a very high quality of graduate education, in terms of both teaching and facilities. However, the process of bringing graduate work fully into the mainstream of academic activity in all subjects in Oxford has not yet been completed, and we have formed the impression during our consideration of this issue that the success of Oxford's graduate courses sometimes still depends too heavily on the goodwill of particular academic and other staff, rather than on sound organisational structure. We are aware, from both our own surveys and from other sources, that implementation of the recommendations of earlier reports is not complete either on the college or on the university side. It is also clear to us that these reports did not address certain broader management issues, such as quality control systems, planning and resource allocation mechanisms and the position of graduate teaching in the joint appointments system. In the following discussion and in our recommendations we have worked on the basis that the recommendations made in the Roberts Report and in subsequent reports are still desirable and should be implemented wherever this has not yet occurred. We have seen our primary task in this area as that of seeing how the overall framework for graduate education could be improved. We begin by considering whether it would be beneficial for Oxford to abandon its existing structures for graduate provision altogether, and replace them with a 'graduate school'.

II: A graduate school for Oxford?

10.12 The graduate school, whilst well-established in a number of North American universities, is a relatively new concept in the United Kingdom, but has been developed by various institutions in response to the expansion in graduate student numbers. There are already several different models in operation or under consideration, which range from institution-wide research and graduate schools, to individual, programme-based graduate schools, with faculty-wide graduate schools between these two extremes. In its 1995 publication, 'Graduate Schools', the UK Council for Graduate Education identified a number of key roles which graduate schools could play to the benefit of the institution and of graduate education: championing the cause of graduate education; pooling expertise and allowing economies of scale in administration, research training etc; delivering quality assurance and control; managing research; providing an interface with the outside world (i.e. potential students, funders and industry); and facilitating new forms of academic collaboration.

10.13 Whilst we see some potential for setting up small graduate schools to foster teaching and research in interdisciplinary subjects, we do not believe that the establishment in Oxford of a single university-wide graduate school would be either helpful or desirable. There are a number of reasons for this conclusion. First, Oxford, as one of the pre-eminent research institutions in the world, can already offer many of the perceived advantages of a graduate school system. Strong links with industry and other funders already exist because of the University's research activities. In many subjects, departments provide a strong research-orientated, academic environment and social focus for graduate students grouped by discipline, whilst college membership offers graduate students the opportunity to mix with staff and students from a broad range of subjects. To the extent that university-wide coordination and administration of graduate studies is desirable, the Graduate Studies Office within the central administration exists to provide them. Secondly, the establishment of a graduate school might further harden the line between undergraduate and graduate courses and isolate undergraduates from the research side of the University's work, at a time when some faculties, especially in the sciences, see value in bringing a research element into undergraduate courses (for example the four year courses in materials, biochemistry and chemistry). Thirdly, the development of a graduate school would introduce another layer of administration into an institution which is already administratively complex because of its collegiate structure. Fourthly, it might lead to a separation of academic staff between those who do research and graduate teaching and those who do undergraduate teaching, as happens in several institutions in North America. Such a separation would violate a fundamental principle of Oxford that undergraduates as well as graduate students should benefit from receiving the bulk of their teaching from members of staff who are active in research.

III: Academic appointments and governance in the context of graduate provision

10.14 Having argued above that graduate provision should be kept within the mainstream of Oxford academic life, we now consider whether, and if so what, reforms of existing structures are necessary. Later in this chapter, we suggest some reforms which relate specifically to the organisation of graduate teaching and support. However, it is also important to note that some of our broader proposals on the joint appointments system (see Chapter 7) and on the University's governance structure (Chapter 5) would have a significant and beneficial impact on graduate provision and we therefore wish to highlight these here.

The joint appointments system

10.15 Whilst graduate teaching now occupies a significant proportion of most academic staff members' time, the organisation of the current joint appointments system for CUF lecturers and, to a lesser extent, university lecturers, is still shaped by the needs of undergraduate teaching. This situation, together with the fact that the demands of regular undergraduate teaching and administration tend to be more urgent and visible than those of graduate teaching, results in a potential for the latter to be marginalised, although the goodwill of academic staff in general ensures that this is not the case. We have therefore proposed a number of reforms to the joint appointments system which are intended to ensure that the organisational arrangements reflect the reality of the position of graduate teaching in Oxford.

10.16 First, we regard it as an out-dated anomaly that time spent on graduate supervision is not included in the university 'stint' of holders of joint appointments but is treated as an additional duty for which a fee of £30 per student per term is paid by the University to the supervisor. This is a prime example of reliance being placed on the willingness of academic staff to accept additional burdens and to give them equal priority to their other responsibilities. Whilst for the most part the system works well and research students receive high quality supervision, some of the graduate students who responded to our survey felt that undergraduate teaching, administration and research took priority over graduate teaching in Oxford; some research students complained that they either were unable to see their supervisor as often as they needed to, or that they felt inhibited from doing so, because of the pressures of the supervisor's other work. What is needed is formal recognition of the fact that graduate supervision is now one of the core duties of the majority of academic staff. We have therefore recommended (Recommendation 38) that time spent supervising graduate students should be explicitly counted in the university teaching 'stint'. Where, for example, a lecturer is contracted to undertake six units' teaching per week for the faculty or department and 12 for the college, an hour's graduate supervision would be counted against those six units rather than being an additional duty. Furthermore, separate supervision fees would cease to be payable.

10.17 Secondly, we have responded to concerns amongst members of academic staff that the contractual obligations laid down for joint appointees are unnecessarily inflexible in terms of the scope given to staff to alter the balance of their teaching duties in either the short- or long-term, for example in order to undertake more graduate teaching. Although any flexibility must of course be subject to the constraints of the overall requirements of the University and the colleges, we have proposed a number of financial and contractual reforms to the joint appointments system, with the aim of giving those staff who seek a somewhat greater degree of flexibility the opportunity to alter the balance of their teaching duties (see paragraphs 7.28 to 7.34 and Appendix C).

10.18 Thirdly, we have noted concerns that the selection procedures for CUF lecturers tend to give priority to the academic needs of undergraduate teaching (i.e. the ability to teach across a broad subject base), and may fail to provide for more specialist graduate subjects. We accept the force of these concerns and we have recommended a revision of the selection system for joint appointees (see paragraphs 7.35 to 7.36), so that, even for posts which carry predominantly college (i.e. undergraduate) teaching responsibilities, the needs of the University and of the college would have to be balanced in making the choice of appointee.

Planning and resource allocation

10.19 In Chapter 5, we have proposed the introduction of planning and resource allocation procedures which we believe would have particular advantages for graduate studies. First, they would make it easier to ensure that new taught courses were set up only when there were sufficient staff and other resources available, such as extra funding for libraries, and training in languages or other specialised skills. Secondly, the new procedures would provide a clearer framework within which faculties could plan and control numbers of graduate admissions, ensure that supervision duties were properly distributed and allow staff sufficient time not only to meet their students at appropriate intervals to discuss their work but also to fulfil related administrative tasks, such as report writing.

The establishment of the three academic boards

10.20 Our proposal for the division of academic planning and administration under three academic boards (see paragraphs 5.104 to 5.144) has, amongst other advantages, the benefit of reflecting the significant differences between graduate studies in the humanities and social sciences, the sciences and medicine. In medicine, for example, graduate education includes clinical training for doctors. In the sciences, there are relatively few graduate students on taught courses, and most research degrees are conducted as part of a team project under an academic project leader. In the humanities and social sciences, a considerable number of taught courses as well as research degrees are offered, and research students usually work independently rather than as part of a team. Moreover, the facilities, such as libraries, laboratories, archives and museums needed by students in each subject group also differ. The procedures for admissions also vary, dictated as they are by the policies and schedules of the appropriate grant-making bodies (primarily the research councils and the British Academy). For all these reasons, it would be logical for many strategic decisions relating to graduate provision to be taken at the level of the academic board.

10.21 In addition, the academic boards should play an important role in promoting interdisciplinary studies, a feature of graduate work and research which a number of staff and students have complained is inhibited by rigid faculty and departmental structures. Members of the boards would be in a position to take an overview of all subjects under their board, to assess where there was scope for interdisciplinary or collaborative work, and to promote structures which would support this, over-arching existing departments or faculties but under the aegis of the board. Cross-board interdisciplinarity could, as now, be dealt with at the inter-faculty level, but academic boards, and in particular their graduate studies committees, should be responsible for ensuring that the division between boards did not prevent graduate students from benefitting from this sort of academic cross-fertilisation.

Recommendation 57
We recommend that the academic boards should be given explicit responsibility for developing effective frameworks for interdisciplinary graduate studies, both within and across the boards.

IV: The organisational framework for graduate education

Faculties and departments

10.22 On the university side, a faculty, through its faculty board, has formal responsibility for all academic matters relating to its graduate students. These include admissions, the appointment of supervisors, the consideration of supervisors' progress reports, the approval of applications for transfer of status and the appointment of examiners for research degrees. Most of the routine aspects of these functions are in practice delegated to the graduate studies committee of the faculty board, and this is chaired by the director of graduate studies, who deals with day to day matters. In those faculties which have sub-faculties, there is a parallel structure to deal with graduate studies at the sub-faculty level, and routine matters are delegated to the sub-faculty's graduate studies committee. In those faculties which do not have a departmental structure, the faculty board or sub-faculty board is also responsible for ensuring that students have access to adequate facilities and that relevant training and background instruction is provided.

10.23 In departmentally organised subjects (most of which are in the sciences and medicine), the department is the main focus of a graduate student's work and many academic matters, such as the provisional admission of graduate students, the appointment of supervisors and the provision of research training, are in practice delegated by the faculty board to the head of department. The department is also expected to take overall responsibility for each student's education by ensuring that there is an effective network of academic support to supplement the supervisor's role and that students are given the opportunity to develop a broad knowledge of their subject. In addition, the department is responsible both formally and in practice for the physical plant, facilities and support staff needed to carry out research. Only the most formal of procedures, such as applications for transfers of status, extensions of time and appointment of examiners, together with formal responsibility for monitoring students' progress and receiving supervisors' reports, remain in the hands of the faculty board (and its graduate studies committee).

10.24 Although within this dual structure of faculty and department the University generally achieves a high standard of provision for graduate students, we believe that there is a danger in a few subjects that the focus of responsibility for the overall education of each student might be unclear, and that there might not be a well-defined structure for dealing with those problems which do occur from time to time, such as a breakdown in a relationship with a supervisor, a serious sense of isolation or very delayed completion of a thesis. In order to achieve clarification of where such responsibilities lie, it is important to recognise the different ways in which graduate studies are handled in departmental and non-departmental subjects.

10.25 Departments, with their clear physical location and well-structured internal organisation, are well placed to deal with all day to day aspects of graduate education, and we believe that the formal role of the faculty board, and sub-faculty, in graduate education in departmentally organised subjects makes the system unnecessarily complex and disperses responsibility. We therefore propose that in all departmentally organised subjects the department should carry primary responsibility for graduate provision. Nearly all the formal roles of the faculty board should explicitly be given to the department, a change which, for the most part, would only reflect and institutionalise current practice. Thus heads of department should be given full responsibility for admitting applicants, appointing supervisors etc. and should be given a more direct and effective role in monitoring the progress of graduate students by a requirement that supervisors' reports were to be sent direct to the department's director of graduate studies (who would be responsible for ensuring that reports were completed on time and were sufficiently informative). The functions of the body immediately above the department, whether it be a faculty board or an academic board, would be restricted to a broad supervisory and regulatory role, for example setting criteria for admissions, drawing up regulations for transfers of status, appointing examiners and dealing with serious complaints.

Recommendation 58
We recommend that, in departmentally organised subjects, formal responsibility for the organisation and delivery of graduate education should be transferred from faculties or sub-faculties to heads of department.

10.26 For a number of reasons, the organisation of graduate provision in faculty-based subjects (i.e. most of the humanities and social sciences) is less straightforward. In faculties without a physical base, supervisors and other members of academic staff may well not come into regular contact either with research students or with each other, and tend to divide their work and time between faculty and college. In some cases, these factors can make it difficult to identify who, or what body, is responsible for particular tasks, such as following up unsatisfactory progress reports or ensuring that graduate students have adequate opportunities to meet other members of academic staff. Whilst we do not believe that a structural change is essential to achieve greater clarity (although it might be desirable), we do think that it is important to identify clearly where responsibility for graduate education lies in each of the non-departmentally organised faculties, mirroring as far as possible the clarity which is inherent in the departmental structure.

Recommendation 59
We recommend that the academic board for humanities and social sciences should ensure that there is a clear focus of responsibility for graduate education in each of the faculties under its aegis.

Research supervision

10.27 A particularly important aspect of a faculty's or department's role is the provision of a supervisor for research students. The relationship between a research student and the supervisor is a crucial element both in the success of the student's academic work and, even more, in the student's feelings about his or her overall experience of graduate work. Whilst most supervision works well, when the relationship fails it can have particularly serious consequences for the student, because of the heavy dependence which many students have to place on a single individual for academic guidance throughout the three or more years of their studies.

10.28 The University has already recognised the potential problems that research students may face if supervision is not adequate and has clarified the responsibilities of supervisors by publication of the Memorandum for the Guidance of Supervisors and Research Students (issued by the General Board following recommendations in the Roberts Report.)[126] We fully endorse the provisions of the Memorandum, but have noted that it appears that not all supervisors are familiar with its contents. One reason may be that it is not given sufficient prominence. The Memorandum is not distributed separately as a matter of course to all new supervisors, but is contained in an insignificant section of the Examinations Decrees and is included as an appendix to faculties' graduate handbooks, where it is prefaced by a statement of students' and supervisors' responsibilities with regard to health and safety: these are very important, but should be issued separately from guidance on academic matters.

Recommendation 60
We recommend that the Memorandum of Guidance for Supervisors and Research Students should be published as a separate document and distributed by the Graduate Studies Office to all newly appointed supervisors.

10.29 There is also clearly a need to ensure that supervisors are not simply overburdened by the amount of supervision required of them. As was recommended in the Roberts Report, faculties have already been asked by the General Board to set norms for supervisors in terms of the numbers of graduate students that they should be asked to supervise, but our inquiries have indicated that some faculties either have not set norms or do not observe them. However, we welcome the fact that some other faculties are already rigorous in refusing to admit an applicant if the appropriate supervisor is overloaded.

Recommendation 61
We recommend that the academic boards should review the practice of the faculties or departments under their aegis with regard to the setting of and compliance with norms for the numbers of supervisees allocated to a supervisor, and ensure that faculties or departments are rigorous in enforcing them.

Other sources of academic support and monitoring in the University

10.30 Even when the relationship between graduate student and supervisor is working well, a heavy reliance on an individual supervisor is not ideal. First, the opportunity to have contact with other leading researchers in one's field and to explore different points of view should be one of the key attractions of Oxford for graduate students. However, our graduate survey indicated that only 18 per cent of research students thought that their opportunities for contact with academics in their field were very good, whilst 48 per cent thought that they were adequate, and 33 per cent that they were poor (the views of science students were only slightly more favourable than those in the humanities and social sciences). In the light of these results, we wish to endorse the points already set out in the Memorandum of Guidance for Supervisors, that

'the supervisor is expected to identify (in conjunction with the director of graduate studies for the faculty, sub-faculty, or department) two colleagues where possible, but if not, one colleague, for limited consultation by the student during his/her first year of research'[127] and that 'the supervisor should arrange for students to have the opportunity to discuss his/her work with other staff and students in the subject area.' [128]

10.31 A related but separate point is the importance of ensuring that there is a sense of collective responsibility for individual graduate students in each faculty or department, to support both supervisor and supervisee, and to ensure that the student's progress is effectively monitored. Some faculties and departments have already established clearly defined intermediate structures, between the level of the supervisor and the graduate studies committee, for supporting and monitoring individual students' academic progress. These structures vary from faculty to faculty, depending on the types of graduate courses offered and on whether or not there is a departmental structure. For example, in some subjects, such as modern history and economics, each student is allocated to an 'interviewer', who is a member of the faculty's or sub-faculty's graduate studies committee but who is not the supervisor; the interviewer provides a direct link with the graduate studies committee and is available for advice, for example if a student's relationship with the supervisor breaks down. In biochemistry, in addition to a supervisor, each student is assigned to a departmental adviser and to a group of four 'senior teaching associates' (senior postdoctoral staff), who are closely involved in the assessment of graduate students and are generally available to provide support when required.

10.32 Whilst we do not wish to be prescriptive about particular mechanisms, we believe that it would be beneficial for all faculties and department to examine what they do in this area to ensure that it conforms with best practice.

Recommendation 62
We recommend that the academic boards should consider the effectiveness of the mechanisms which each faculty board or department under their aegis has for supporting and monitoring the progress of graduate students, and should ensure that these conform with best practice.

10.33 We recognise that in some humanities and social sciences subjects a lack of opportunity to mix in an academic or social context with fellow research students or academics in their own discipline is exacerbated by the absence of a departmental focus. Although it has to be recognised that, for many such students, research is bound to involve fairly isolated work in archives and specialist libraries, nevertheless to a considerable extent the root of the problem lies in the nature of the faculty buildings. Many of these were designed to house only a library and administrative offices and not to promote human interaction, which might be achieved by providing more space for academic staff and graduate students to work there, and by improving social facilities such as common rooms. Moves are gradually being made in the direction of developing faculty centres and providing more facilities in faculty buildings, but progress is slow, not least because of financial and space constraints. As we discuss in paragraphs 11.24 to 11.28, in the context of staff engaged in research, we believe that further development of faculty centres is to be encouraged. This should include specific provision for graduate students.

Recommendation 63
We recommend that in its long-term plans the University should aim to ensure that each of the humanities and social sciences faculties can make at least the following minimum provision:

• a working space for each graduate student, with a desk and a place to leave papers and personal belongings, usually in a library or a shared office in the relevant faculty centre or department;
• computing facilities for graduate students appropriate to the nature of the work within that faculty;
• attractive common room facilities, open to both academic staff and graduate students.

10.34 There remain two further issues which we wish to address, both of which relate to the subject of the monitoring by the University of the progress of graduate students.

10.35 First, we consider that a system of self-reporting by graduate students can be useful. Although this has been advocated in the past, it has not become systematic. For example, in Trinity Term 1997, the General Board's Graduate Studies Committee announced the establishment of a centrally administered self-reporting scheme for all graduate students, but it is not compulsory for faculty boards to adopt this scheme, and the report forms are not issued to students until their sixth term. We welcome the principle behind this move, but believe that the provisions could and should be more far-reaching.

Recommendation 64
We recommend that all faculty boards and departments should be required to adopt a self-reporting scheme for graduate students and that all graduate students should receive self-reporting forms every Trinity Term, so that students in each year, including the first year, of their graduate work would have the opportunity to comment on their progress and on their experiences of Oxford.

10.36 Secondly, even with thorough and effective monitoring procedures in place, research students are sometimes allowed to continue working on their theses long after the due time for completion. Whilst this delay is often the result of a sympathetic approach by supervisors and the relevant graduate studies committee, it must be recognised that it is often not in a student's best interests in the long term and represents a considerable waste of staff time and resources. It also leads to difficulties in meeting submission rate targets laid down by the research councils and the British Academy.

Recommendation 65
We recommend that faculty boards and departments should adopt clear guidelines for determining the circumstances under which research students can be required to give up their studies on academic grounds (subject to an appropriate appeals procedure).

The colleges' role in graduate education

10.37 Having explored various aspects of the University's responsibilities for graduate teaching, we now turn to the colleges' role in graduate provision. College membership is mandatory for all graduate students, a position which we strongly endorse. On the social and pastoral side, the colleges have and discharge a significant role in providing opportunities for social contact with staff and students outside academic boundaries, and they provide many graduate students with accommodation for one or more years. With regard to academic provision, there is considerable variation in what colleges offer, and less agreement on the role which the college should play. This is because the specialised nature of graduate students' work can make it difficult for colleges to make a significant contribution to the studies of every graduate student therein, through the provision of teaching, supervision or academic facilities. We have concluded that further attention needs to be paid to these matters.

College advisers and tutors for graduates

10.38 An important academic function of the college is to allocate each graduate student to a 'college adviser', who would normally not be the student's supervisor, and whose role was set out in the Southwood Report of 1990:

'First, the adviser should be in a position to discuss the student's academic work, which presupposes a certain knowledge of the subject area in which the student is working but normally should not necessarily require the student and adviser to be working in the same department or faculty. He or she would be expected to monitor a student's progress; to hold at least one meeting a term with the student to discuss the supervisor's report; and to be available at other times for consultation on academic or other matters which a student felt could not be taken to a supervisor. It should be emphasised that the adviser would not in any way be expected to replace a supervisor or to act in his or her stead, and would not therefore be expected to give the same detailed guidance and direction.'

10.39 In addition, all 'mixed' (i.e. undergraduate/graduate) colleges are expected to appoint a tutor for graduates, who allocates college advisers to individual students, ensures proper liaison between the college and the faculty, sub-faculty, or department, and receives the college copies of the supervisors' reports. The tutor for graduates organises discussions with students on their progress, and is available to give students pastoral advice.

10.40 We place a high value on the role of the tutor for graduates and of the college adviser in monitoring graduate students' progress and in being available to give advice and support on academic matters. It is a particular strength of the collegiate system that graduate students have access to such support outside the formal structures of their own faculty or department. We are therefore concerned that 14 per cent of respondents to our graduate survey said either that they did not have a college adviser or did not know if they had one, and that of those graduate students who had a college adviser, only 57 per cent expected to see their adviser at all in the term in question. We urge colleges to ensure that all their graduate students are allocated to a college adviser, and that both the students and the college advisers understand what the role of the adviser is, in terms both of monitoring and of being available to give advice. We believe that it is important to emphasise to students the accessibility of their college adviser; the corollary to that is that the colleges must ensure that college advisers are not so burdened with other college duties that they cannot devote adequate time to this role. We also endorse the recommendations, made in the earlier reports on graduate studies, that regular meetings ('collections') should be arranged at which the student can discuss his or her report with the college adviser and/or tutor for graduates, and that tutors for graduates should have a high status within their colleges.[129]

Recommendation 66
We recommend that colleges should, as a matter of urgency, give renewed attention to ensuring that there is in place a system of college advisers and 'collections' for graduate students and that the effectiveness of this system is monitored.

Provision of academic facilities

10.41 We support the conclusions of the earlier reports on graduate studies that it is right in a collegiate university to expect colleges, as well as faculties and departments, to play a significant role in providing graduate students with access to certain academic facilities, such as libraries, computer equipment and photocopying. This is particularly important where there is no departmental structure, and in any case can help to integrate students into college life, with the social and intellectual benefits that this can bring for both the student and the college. Division of responsibility between University and college can, of course, lead to gaps in provision, if it is unclear which body is responsible for providing a particular facility, and the comments received in our graduate survey indicated that this is sometimes the case. It is therefore essential that, within our proposed planning and resource allocation procedures, each faculty or department should be expected both to specify the level of facilities to which graduate students in a particular subject should have access and to agree with the colleges collectively as to the particular facilities which each side should provide.

Recommendation 67
We recommend that:

(a) each faculty or, where applicable, department should draw up a schedule of the academic facilities which are to be made available to graduate students in each particular subject, and decide in consultation with the Committee of Tutors for Graduates whether these are to be provided by the faculty/department or college;

(b) the academic boards should monitor whether the agreed levels of provision are being maintained for graduate students in the subjects for which they are responsible;

(c) faculties should inform graduate students at the beginning of their course what facilities they can expect, and whether these will be provided by the faculty or the college.

Specialised colleges and college research centres

10.42 The graduate colleges which specialise in particular subject areas make a very valuable contribution to the academic and social life of their students. Although we do not consider it to be desirable for all colleges to specialise in their graduate intake (and this is especially so in the case of colleges which also admit undergraduates, because of the consequences for the composition of the fellowship and the effect on the undergraduate body), we do consider that specialisation in some graduate colleges is fully to be supported. In addition, as we discuss in Chapter 11, we would encourage the development of discipline-specific research centres in colleges, whether graduate only or 'mixed', as such centres provide a positive role for the colleges in academic graduate provision, provide a focus for research students in particular disciplines and often foster interdisciplinary research outside the confines of the faculty structure.

V: Graduate taught courses

10.43 During the post-war period, the number of students on taught masters' courses, particularly in the humanities and social sciences, has risen markedly (see Figure 10.1 on p. 174). In 1996-7, 30 M.Phil/B.Phil courses and a wide range of one year masters' courses were on offer. The latter include 38 M.Sts., 25 taught course M.Scs., the Master of Business Administration, the M.Juris. and the M.Theol courses. There are also the longer-established BCL course and a number of other graduate taught courses. Many of these courses can be used as stepping stones to D. Phil work, as can the two year M. Phil courses (with the important consequence that students can count the fees paid for the taught course towards their three years' total of fees for a D. Phil.).

10.44 There are various reasons why the number of graduate students on taught courses has risen and is likely to continue to do so. In the humanities and social sciences, the trend has largely been driven by the policies of the Humanities Research Board of the British Academy and the Economic and Social Research Council (ESRC), which are the principal sources of public funding for British graduate students in these subjects. Both these bodies regard the methodological training provided in graduate taught courses as very important, and have made it clear that they will be more likely to fund research students if they have taken taught courses first. In the academic year 1996-7 the Humanities Research Board of the British Academy made 484 awards to candidates wanting to read for research degrees, and of these only nine were offered to candidates who had not already had 'graduate experience', which in practice means a master's degree. The ESRC, for its part, allocates a proportion of its awards for specific masters' courses.

10.45 From the students' point of view also, taught courses hold a number of attractions. Many self-funded graduate students see the advantage of a 'building block' approach to graduate work which enables them to attain at least a master's degree if their finances run out or if they feel that their careers would be better served by leaving before completing a research degree. Increasingly, too, masters' courses are an end in themselves, as many graduates seek a second qualification in order to help them gain access to specialist careers in industry, international organisations, international banks, business consultancies, and the civil service. For example, in the sciences there is an increasing demand for graduate taught courses (such as Oxford's recently introduced M.Sc. in Neuroscience) which teach subjects not usually offered at an undergraduate level or which cross traditional subject boundaries. Such qualifications can give access to careers to which the subject matter of the course is directly relevant, without the need for a research degree.

10.46 In the light of these developments, it is important that the University's procedures for quality control and planning take account of the significant but distinct demands of graduate taught courses. There is a need to ensure that in all the faculties and departments which offer taught courses it is clear where responsibility for a particular course lies, and also that attention is given to identifying the most appropriate teaching methods and the proper mechanisms for monitoring students' progress on taught courses.

Recommendation 68
We recommend that the academic boards, in conjunction with the Educational Policy and Standards Committee, should give specific attention to the teaching and administrative arrangements for taught graduate courses, and ensure that best practice is disseminated and observed.

VI: Graduate admissions and studentships

10.47 Concerns have been expressed to us about the related issues of the graduate admissions process and the shortage of funding even for the best candidates. These concerns are not felt equally across all subjects, and predominate in the humanities and social sciences, although concerns about a lack of funding are acute in some science departments as well. The current admissions procedure does not always make it easy for faculties to select the best out of the range of applicants and that the University's inability to offer a wide range of studentships of its own may well handicap the University in attracting the best possible candidates in the face of international competition. Above all, it is a source of concern that a shortage of funding is preventing some very strong candidates from both the UK and overseas from taking up graduate places at Oxford. In the following paragraphs, therefore, we propose revisions to the admissions system for graduate students, and recommend the establishment of a university scheme for graduate studentships.

The admissions system

10.48 Amongst some members of the humanities and social sciences faculties, there is a sense of frustration that the University has not yet succeeded in establishing a system for considering all applications for a particular course together in a 'gathered field' after a set deadline for receipt of applications, rather than assessing each individual application as it arrives against an abstract set of standards, as often happens at present. This issue does not arise so acutely in the sciences, as decisions on graduate admissions have to be made in quite a short space of time in the late spring as soon as the departmental quota grants for research council studentships for the coming year are known. The following points therefore apply only to the humanities and social sciences.

10.49 The desirability of operating a 'gathered field' has already been considered on a number of occasions, for example in the Roberts Report in 1987 and most recently by the Committee of Tutors for Graduates and the Graduate Studies Committee in 1995 and 1996. As the Roberts Report succinctly put it:

'The argument for such a procedure is not only one of administrative neatness but is also partly academic: it would allow comparison amongst applicants so that no one is admitted in February who might not be admitted in July.'[130]

10.50 Despite this cogent reasoning, a uniform deadline was seen as impracticable. One difficulty is that some of the universities in the UK and overseas with which Oxford is competing for graduate students do not operate a gathered field system; it is therefore argued that Oxford might be disadvantaged in trying to recruit the best applicants if it could not offer them assured places until after a deadline of, say, 31 January, when other universities might be offering good applicants places in the preceding November or December. It has also been argued that Oxford would put itself at a disadvantage by setting too early a deadline, as under current practice undergraduates often do not decide to go on to graduate study until fairly late in their final year, and sometimes only after their final examination grades are known.

10.51 We recognise the continuing force of these arguments, but we want to emphasise our belief that, if the University wishes to select the most highly talented graduate students, the gathered field system is preferable to that currently operated in most humanities and social science faculties, provided that it is operated with some flexibility to enable exceptional candidates to be accepted outside the normal time limits. We believe that there should be continuing moves towards such a system and that, if necessary, additional resources should be committed to faculties or the Graduate Studies Office to enable such a system to be adopted.

Recommendation 69
We recommend that the academic board for the humanities and social sciences should encourage the faculties and departments under its aegis to adopt a 'gathered field' system for graduate admissions.

10.52 One factor which we believe would facilitate the operation of a gathered field system is the linking of eligibility for graduate studentships offered by the University (which we discuss below) with a fixed deadline for applications. We also believe that a number of administrative measures could and should be taken to speed up the processing of applications and thus the notification to applicants of the outcome of their applications. It has been suggested, for example, that it might be helpful to require candidates to send in their references in sealed envelopes at the same time as their application form and other documents.

Recommendation 70
We recommend that the University should undertake a review of the administration of graduate applications in both the University and the colleges, with a view to enabling applicants to be informed of the outcome of their applications as swiftly as possible.

The funding of graduate education

10.53 There are now very serious concerns in a number of subjects about the availability of funding for graduate students. Many such students depend chiefly on publicly funded studentships from the research councils (in the sciences and social sciences) or the Humanities Research Board of the British Academy (in the humanities). However, the availability of such funding varies considerably from subject to subject, and the numbers of studentships on offer in some subjects have either fallen or failed to keep pace with the growing number of well qualified applicants.

10.54 In the social sciences, the problem is particularly acute. Changes in policy at the Economic and Social Research Council (ESRC) have caused the number of studentships for advanced (i.e. taught) courses at Oxford to fall from 47 in 1992 to 24 in 1997. There is also some evidence of a decline in the number of research studentships available.

10.55 In the sciences, the situation is much more varied. Most research council studentships are funded through quota grants to individual departments. This has the advantage of allowing departments themselves to determine which students should receive support. However, whilst in some subject areas the quota fairly accurately reflects the level of demand, in others the research councils have imposed reductions in quotas which have meant that some departments' capacity to take on new graduate students has been quite restricted. Furthermore, the situation can vary even between subjects within one research council; for example, whilst the number of engineering research studentships at Oxford provided by the Engineering and Physical Sciences Research Council has fallen from 17 in 1993 to 10 in 1997, the physics department has not suffered any overall decline in the number of studentships from the same research council over that period.

10.56 In the humanities, whilst Oxford has not experienced an overall reduction in the number of studentships available from the Humanities Research Board of the British Academy, either numerically or as a percentage of the total number awarded in the UK, there is nevertheless a real concern that the overall number of studentships available has not been increased although the number of high quality applicants wishing to pursue graduate studies in the humanities has grown markedly. For example, for the 1996 masters' courses competition run by the Humanities Research Board (competition A), the overall success rate in the UK was 19 per cent. 94 per cent of those chosen had first class degrees, but 55 per cent of applicants with first class degrees were turned down. Even this understates the problem, because after the first competition there is a second (competition B) for the remaining three year study grants. In 1996, only 22 per cent of those applying were successful, of whom 98 per cent had graduate experience (ie had done a master's degree) and 81 per cent had originally graduated with first class honours. As this illustrates, the chances of even a first class candidate emerging with enough public funding to complete a research degree are slender.

10.57 There is also some frustration that the Humanities Research Board, unlike the research councils, does not make quota grants for studentships to particular faculties or departments but considers each application individually. As the conditions on which these awards are made are somewhat different from those applied by faculties and departments for admission to Oxford, this means that the acceptance rate for places offered can be quite unpredictable. Consequently, allocating appropriate numbers of supervisees to members of staff and achieving the desired number of students on particular taught courses is difficult.

10.58 From this combination of experiences there emerges a university-wide conclusion that Oxford must take action to ensure that the unavailability or unpredictability of public funding for graduate students does not result in a situation in which either faculties or departments have to place inappropriate restrictions on the number of graduate students they recruit, or less able, but privately-funded, applicants are admitted because some of the most highly qualified candidates have to turn down a place for lack of funding.

10.59 We share these concerns, and believe that it is essential that the University itself should offer a substantial number of full studentships to support outstanding applicants wishing to embark on masters' or research degree courses. Oxford must recognise that the market for recruiting graduate students is highly competitive both within the UK and internationally, and that Oxford is in strong contention with the leading North American universities. As an increasing number of these competitor institutions offer well-funded graduate studentships and teaching assistantships, we regard the establishment of a number of such studentships at Oxford as a matter for the University to tackle with urgency. If needs be, funds should be provided from the University's existing resources, notwithstanding that this will mean some economies elsewhere. However, it must be recognised that the shortage of publicly funded studentships is much more acute in some subjects than others. Therefore, given that the University, even supported by fund-raising efforts, cannot hope to provide funding for all the studentships it would ideally like to offer, its own studentships must be carefully targeted at the areas of greatest need.

10.60 The Graduate Studies Committee of the General Board has recently considered the means of establishing a studentship scheme and has consulted with faculties and colleges. The committee's approach has been endorsed by the General Board and further thought is currently being given to the practical, and in particular the financial, issues involved. We fully endorse the position taken by the Graduate Studies Committee and the General Board on this and we wish to add our support to moves in this direction.

Recommendation 71
We recommend that the University should establish a graduate studentship scheme.

10.61 We would also encourage more colleges to consider offering scholarships to graduate students. It would be important, however, to ensure that these were used to increase the chances of attracting graduate students who would not otherwise be able to fund their studies, and were not just aimed at attracting graduate students from one college to another, as has sometimes happened in the past.

VII: Teaching by graduate students

10.62 A question which often arises in the context of graduate education is the extent to which universities such as Oxford should employ their graduate students to undertake undergraduate teaching. In many North American universities, including Harvard and Stanford, this is common practice, and is often formalised in a teaching assistantship scheme, in which graduate students are contracted to undertake a certain amount of teaching (usually between a quarter and a half of the working week) in return for a payment which goes towards their fees.

10.63 In Oxford, also, graduate students have for many years taught undergraduates. In the Franks Commission's survey of graduate students, 37 per cent of respondents (not including those in their first year of graduate study, who presumably did less) were doing some teaching in the term in question. Our graduate survey[131] (which included first year graduate students) revealed that just over a quarter of graduate students had taught for either the University or a college at some stage in their course, with the highest proportion in the sciences (34 per cent), compared with 24 per cent in the humanities and 25 per cent in the social sciences. Most of those who taught were research students. The teaching which they carried out chiefly consisted of tutorials and (in the sciences) demonstrating. Undergraduates reported that, on average, 10 per cent of their tutorials were given by graduate students (13 per cent in the sciences and 8 per cent in the humanities).[132] However, the arrangements for teaching by graduate students are not systematic or formalised. Training is available, but not compulsory; selection is by informal processes; and arrangements tend to be quite short-term.

10.64 In addition there is some ambivalence about the whole concept of teaching by graduate students. On the one hand, it is recognised as a useful means to relieve the pressures on academic staff. It is also popular amongst graduate students. When those graduate students who had not done any teaching were asked in our survey whether they would like the opportunity to teach, 66 per cent said that they would. (Of these, 84 per cent gave as their reason the wish to gain teaching experience; a large proportion of these intended to pursue an academic career.) On the other hand, there are concerns as to whether the quality of teaching which graduate students can give is sufficiently high and about the effect of doing such teaching on the graduate student's own work. A number of comments on this topic were received in our staff, graduate and undergraduate surveys. We therefore thought it right to give fresh thought to this issue.

10.65 For graduate students, the opportunity to teach:

• encourages them to maintain a broad knowledge of their subject;
• gives the opportunity to acquire work experience, and of a sort which is particularly relevant to those planning on an academic career;
• helps to relieve financial problems.

However, teaching obviously takes a student away from his/her academic work and, therefore, can impair academic work or delay completion of a thesis. It might be noted that the research councils set a maximum of six hours teaching per week for those students funded by their awards.

10.66 For undergraduates, being taught by a graduate student:

• can offer the opportunity to enjoy a fresh approach to teaching and to their subject;
• can give an informal insight into the world of research;
• can give them the opportunity to discuss their academic concerns informally with someone close to them in age and experience.

On the other hand, there is always a concern that Oxford cannot continue to lay claim to excellence in undergraduate teaching if too large a proportion of it is carried out by graduate students whose knowledge of their subject may be insufficiently broad, who may not have any previous teaching experience and who may be unfamiliar with such matters as the requisite course content and styles of teaching.

10.67 Finally, for both the University and the colleges, the use of graduate students as teaching staff can provide a cost-effective means of relieving burdens on academic staff, who thus have more time available for research, for more specialised teaching and for essential administrative and decision-making roles.

10.68 We have weighed these arguments carefully. Our conclusion is that, on balance, it would be beneficial if more graduate students were given the opportunity to teach undergraduates, provided that the hours worked by each graduate student, including preparation time, were not too onerous and that proper support and monitoring were provided. We believe that this experience would prove a positive benefit to graduate students in terms of acquiring new skills. We believe that the concerns about quality can be answered by embedding such teaching in the overall quality control framework which we propose in Chapter 8. For example, it should be compulsory for all graduate students who wish to teach to attend the appropriate training sessions run by the Academic Staff Development Committee (unless they have already received relevant training elsewhere). This would not only help to assure the University of the quality of their teaching, but would also give a valuable training to the graduate students which might prove an asset in their future careers. Graduate students' teaching should be overseen by the relevant college director of studies (see paragraphs 8.12 to 8.13) or the appropriate person in the faculty or department. We also believe that the issuing by faculties of tutors' handbooks would help to keep graduate students who are giving tutorials informed of syllabuses, examination schedules etc, and would help to standardise practice where appropriate. Nevertheless, in order to ensure that undergraduates continue to benefit to a substantial degree from the wider knowledge and greater experience of full-time academic staff, we propose that each faculty should specify a maximum number of hours for which any undergraduate could be taught by graduate students (see paragraph 8.5). This could vary between years of study and between courses, and might include tutorials, demonstrations, classes, seminars and, in some cases, lectures, as appropriate for each subject. It would be the responsibility of an undergraduate's director of studies to ensure that the maximum was not exceeded. The Educational Policy and Standards Committee of the Council would be responsible for monitoring the resultant quality of education which undergraduates receive and might find it appropriate to set an overall university-wide maximum number of hours' teaching to be done by graduate students.

Recommendation 72
We recommend that all faculties, departments and colleges should:

(a) consider increasing the amount of undergraduate teaching delivered by graduate students;

(b) set appropriate norms and maxima for teaching by graduate students in respect of each course and year, bearing in mind the needs of undergraduates, and the abilities of and time constraints on graduate students;

(c) ensure that such teaching is fully supported by compulsory training of the graduate students, effective dissemination of information and good quality control systems.

10.69 One noteworthy point which emerged from the comments returned by respondents to our graduate survey was that there was perceived to be a strong bias in favour of those who had a first degree from Oxford when teaching opportunities were being offered to graduate students. Despite the existence in many faculties and departments of a register of all those graduate students who wish to teach, it appears that there is indeed such a bias. Our survey revealed that 43 per cent of Oxford graduates had done some teaching, whereas only 26 per cent of other UK students and 21 per cent of overseas students had taught. It is not wholly surprising that this is the case, as it is likely that tutors invite students whom they already know to teach their undergraduates. It is, however, an understandable source of grievance for non-Oxford graduates if they are excluded from teaching opportunities in this way. Any concerns that such graduates would not be as good at teaching because they were not familiar with the 'Oxford style', especially in tutorials, could be reduced by the compulsory attendance at training courses. We believe that taking active steps to give more non-Oxford graduates the opportunity to teach is not only a matter of fairness for graduate students, but should also provide a positive benefit for undergraduates, by giving them the opportunity to view their subject from a broader range of perspectives.

Recommendation 73
We recommend that directors of studies and those organising teaching in faculties and departments should, as a matter of course, consult the appropriate register of graduate students wishing to teach before allocating teaching time to graduate students, rather than relying simply on personal contacts.