Response—Discussion Paper on Higher Education Research and Research Training

On 27 July 1999, the Australian Academy of Science made the following response to the discussion paper on higher education research and research training.


Recommendations

The Academy supports the Discussion Paper’s broad aims of promoting diversity and excellence in our higher education research. Australian Research Council (ARC)

We support

However, we recommend the following changes or additions:

    • the appointment of a part-time Chair and full-time Chief Executive Officer and Program Managers
    • the use of Discipline Panels, and peer review by the best local and overseas researchers, to assess applications for research grants
    • more frequent rounds of applications.
    • there should be a clear separation of ARC’s policy-advising and program-administration roles, with both remaining within the ARC
    • ARC should develop and implement priorities for allocating funding between broad discipline areas
    • talented young researchers should receive Large Grants from which they are now, in practice, excluded
    • the use of the proposed readers should be integrated with the process of peer review.

The Discovery and Linkage programs

The allocation of funds between the Discovery and Linkage categories should be within the remit of the ARC itself.

Linkage should be interpreted in a broad sense to ensure that collaborations, network-creation and international scholarships continue.

Major national and international facilities require funding contributions from several portfolios and research agencies, not just a matching of ARC’s contributions.

We support a single program to cover the existing Key and Special Research Centres, and recommended that it is administered flexibly to create critical masses of outstanding researchers and facilities. Research infrastructure

We support strongly the inclusion of a realistic research overhead component in competitive research grants but new money will be essential to support our researchers at international standards. This will require an infrastructure component of 50% of research grants. Research operating grants

A research assessment exercise should be established, in consultation with the Academies, to assess universities’ research performance and prospects on a departmental level.

The 60% load factor funding should be moderated by a quality factor derived from a research assessment exercise.

Externally sourced funds should only be used in calculating research operating grants if they are applied to research, according the internationally-accepted definition of the term and not if they are for expert services that do not entail the production of new knowledge.

The value of research student places allocated to universities should include a weighting in favour of disciplines in which research is inherently more costly. A weighting factor of 2:1 for disciplines like science and engineering compared with arts is needed. Research training

Holders of postgraduate awards

  • should be free to change from the institution at which their grant was made to another after at least six months with the first institution,
  • should be provided with a 10% loading on their stipend if they choose to move to another institution for their higher degree, and
  • be allowed a full three years after moving to complete their work.

The Academy strongly recommends that the whole research student allocation should be portable, including the weighting factor related to the differential costs of research.

The Academy does not support the one-year delay period for a student to change institutions and prefers the present six-month delay period.

Differential stipends should be used to encourage growth of research in areas of defined national need.

Introduction

The Academy supports the broad principles of the Discussion Paper. These include the need to aim for international excellence, create diversity in the higher education system, increase the mobility for students, encourage strong international links, strengthen the role of the Australian Research Council and, as part of wider government policy, develop a business tax regime which promotes research investment. We support the process of getting the structures right.

However, it is necessary to pursue these goals with greater concern about potentially undesirable consequences than is evident in the Discussion Paper. Otherwise Australia’s research performance may be reduced rather than enhanced.

In addition, there appears to be insufficient attention to the consequences of the fact that our universities are Australia’s major source of basic research and scholarship. Australian Research Council

The Academy supports strongly the enhanced role proposed for the Australian Research Council (ARC). It must be given responsibility for its own administration. This should involve, as the Discussion Paper proposes:

  • the appointment of a part-time Chair who would also be a distinguished scholar with high standing in the community generally,
  • the appointment of a chief executive officer who is an established, mid-career scholar, but looking to move into research administration,
  • Program Managers who are highly competent researchers, appointed full-time but for a limited term,
  • assessment of proposals by Discipline Panels using peer review by the best local and overseas researchers,
  • receiving applications more than once per year to reduce delays in research programs.

In addition, we make the following points:

  • We commend the clear separation of the policy-advising and program-administration aspects of ARC’s work proposed in the new arrangement. Both must remain under the umbrella of the ARC whose membership should be broad and expert enough for the tasks, but not larger than about twelve persons.
  • It is essential that the ARC develop a process for establishing and implementing priorities for allocating funding between the disciplines for which it is responsible.
  • The ARC needs to develop the potential of talented young researchers by awarding them Large Grants early in their careers. At present, beginning researchers are, in effect, excluded from Large Grants by the low success rate.
  • An advisory committee might be a suitable mechanism to recommend the Chair of the ARC, and the Academy would be prepared to comment further on this.
  • The proposal for the ARC to employ paid, discipline-specific readers should be integrated with the process of peer review of grant applications, which should still involve subject committees and referees of the highest national and international standing. It will be important for the revised arrangements to avoid some of the pitfalls encountered by the Engineering and Physical Sciences Research Council in the United Kingdom.

The Discovery and Linkage programs

While this proposed arrangement would enhance the comprehensibility of the ARC’s programs, it should not be allowed to drive the allocation of ARC funding. Particularly, the allocation of funds between the Discovery and Linkage categories should be within the remit of the ARC itself, subject to the ARC’s accountability to its Minister.

The Academy recommends that linkage should be interpreted in a broad sense to ensure that collaborations, the creation of networks and the international postgraduate research scholarships continue. National and international facilities

For Australia’s research base to continue to meet Australia’s needs in a rapidly changing world, a framework is required that provides support for investment in research infrastructure of national or international significance. These facilities are generally of such a scale and cost that they are too expensive to be provided by any single research organisation and, in the case of very expensive facilities, by a single country. However, many different researchers and research organisations need to access them.

These facilities are more appropriately supported through collaboration involving consortia of research organisations, including overseas collaborators in the case of major international facilities.

The proposal that major national and international facilities be supported by refocussing the Research Infrastructure Equipment and Facilities Program on a matching funding basis is unsatisfactory for the establishment and running of major national and international facilities. The resources available under RIEFP are inadequate for the purpose. Major national and international facilities require the cooperation and funding contributions from several government portfolios and research agencies. Single Centres of Excellence Program

The Academy agrees with the proposal for a single program to support Key and Special Research Centres spanning Discovery and Linkage elements. The Paper envisages that the maximum government funding would be $1 million for seven years with matching contributions from partners.

The Academy recommends greater flexibility for the Program. Decisions on maximum funding, the life of a Centre and the requirement for matching funding should be consistent with the aims and needs of a Centre and an assessment of national benefit. The principal aim of the Program should be support of Centres to achieve a critical mass of outstanding researchers, and state of the art facilities. Research infrastructure – the granting agencies’ role

The Academy strongly supports the inclusion of a realistic research overhead component in all competitive research grants. However, at the present inadequate level of funding for grants, the Academy does not believe that it would be feasible to fund infrastructure in this way by transfer of funds already in the hands of the ARC. New funds will be essential.

The report of the Boston Consulting Group documented in 1993 a backlog in university infrastructure needs that has only been partly remedied to date. Since 1993, the infrastructure flagfall cost for competitive research in science and technology has increased significantly.

To implement the funding of infrastructure through research grants, the ARC, and other government bodies contracting research to universities, will need additional funds.

We consider that, because of the importance of the principle involved, the incorporation of infrastructure funding with research grants should proceed, beginning with the transfer to the ARC and the National Health and Medical Research Council (NHMRC) of the $80 million of infrastructure money currently within block grants. This will cover infrastructure costs at 20% of the value of the research grant, far below the 50% level that the Academy recommends. There needs to be additional funding to allow granting agencies to attach the correct 50% level of infrastructure support to their grants. We estimate that the cost of this would be about $120 million per annum.

The Academy recognises the need for universities to maintain infrastructure such as libraries, information networks, equipment, laboratories, safe working environments and technical support on a long-term basis. The funding for this research infrastructure needs to be brought into alignment with international best practice. Mechanisms must be devised at the university level so that the arrangements above support this process, and that they mesh with the extra support needed from the university block grants.

A policy developed coherently in this way also provides the greatest leverage for attracting to universities the necessary infrastructure component in grants from industry and other sources. Research operating grants to universities

The Academy is very concerned that the proposed funding formulae will not deliver the Government’s objectives, especially those aimed at differential support of excellence and promotion of diversity.

We have reservations about all of the three key parameters used for the funding proposal, that is, the 60% load factor, the 40% external funding (including consultancies) factor and the formula associated with research student allocations.

The methods proposed for allocating block research grants to institutions are defective in the following respects.

  • Assessment of quality of outcomes, and of externally funded activities, plays too small a part.
  • They inject a bias against more costly research, and discourage mobility, by valuing all research student positions equally.
  • The fact that the criteria are so abbreviated is likely to promote less rather than more differentiation among universities.

The 60% load factor is of concern not only because of the magnitude of this contribution in the formula but chiefly because there is no external contestability based upon the quality of the research outputs to which this funding is related. The internal training and other matters referred to for university scrutiny are process control matters and, although reviewed from year to year by the Department, are not in themselves a guarantee of the quality of the output. The Academy is of the view that the load factor should be linked closely to a research assessment exercise. The results of that assessment exercise should be publicly available so that students can have additional information about the choice of the institutions at which to do research.

The 40% external funding factor also suffers from the defect of no assessment of the quality of the outputs. Here the quality assessment might be broader in scope to take account of demonstrated value to the external funding source. But the Academy believes that the strict definition of research as used by the OECD and by the Australian Bureau of Statistics should be the yardstick. The Academy has considerable reservations about the use of consultancies as part of the measure. Just as the Department of Education and Youth Affairs had extreme difficulty using publications as a measure of quality, so there will be an even greater difficulty in judging the quality of unpublished outcomes from internal reports and consultancy documents. In the latter case there may even be a restriction on access.

The $13,756 research student value attributable to universities is a matter of concern, although the Green Paper indicates that this will be reviewed to take account of the different costs of research in, for example, the sciences and the arts. The Academy thinks that appropriate adjustments are essential. There is also the need to ensure that the whole of the research student allocation is portable between the universities.

To remedy these defects, the Academy recommends that

  • the 60% load factor funding should be moderated by a quality factor derived from a research assessment exercise which will establish the quality of the research output;
  • externally sourced funds should only be used in calculating research operating grants if they are applied to research, according the internationally-accepted definition of the term research and development work, that is as '…creative work undertaken on a systematic basis in order to increase the stock of knowledge…' (OECD)(1). This excludes expert services that do not entail the production of new knowledge;
  • the value of research student places allocated to universities should include a weighting in favour of disciplines in which research is inherently more costly. A weighting factor of 2:1 for disciplines like science and engineering compared with arts is needed.

Whatever method is selected for the allocation of block research grants to institutions, the Academy considers that the process should include an assessment of an institution’s performance on a field-of-research basis. The Academy’s recommendations for research assessment were discussed in its response to the West Report and are outlined in Attachment A.

It is, therefore, recommended that

  • a process should be established for assessing universities’ research performance and prospects on a departmental level, and
  • the Academies be asked to develop, in collaboration with DETYA and the ARC, a model for the way in which the evaluation is conducted.

Research training

The length of study for which scholarship support is given should be benchmarked against arrangements in countries where broad training and temporary placement in foreign laboratories is the norm. Three and a half years may be impractically short.

The Academy supports the Discussion Paper’s goal that research students should be encouraged and enabled to move to the university that can provide the training most suitable for them.

The Academy does not support the one-year delay period for a student to change institutions and prefers the present six-month delay period.

There needs to be a positive incentive for this to happen. As a basis for discussion, the Academy recommends that holders of postgraduate awards

  • should be free to change from the institution at which their grant was made to another after at least six months with the first institution,
  • should be provided with a 10% loading on their stipend if they choose to move to another institution for their higher degree, and
  • be allowed a full three years after moving to complete their work.

The results of the research assessment exercise should be available to potential students to facilitate their choice of research environment. Without this, or an equivalent, there will be a failure to couple students to good supervisors. We reject the institution’s training plan as the means for achieving this.

The Academy strongly recommends that the whole research student allocation should be portable, including the weighting factor related to the differential costs of research.

The system has no mechanism to facilitate growth of research in areas of defined national need. Differential stipends would be one way to address this.

Footnotes

  • (1) OECD. The measurement of scientific and technical activities, (Frascati Manual). OECD, Paris, 1981, Cited in Australian science and resources brief, 1994, p. 69.

Attachment A

Research assessment

The Academy recommended to the West Review the concept of a funding scheme based on individual Departments’ performance with the university doing a pre-examination to propose its best cases in the competition. This process reduces centralist administrative loads and satisfies the 'focus' ideas of the Discussion Paper. The Academy advocates that a research assessment be undertaken every five years, based on retrospective and prospective performance. In the initial round all of the fields would be assessed concurrently.

As concerns process, universities should be required to meet particular standards of supervision, infrastructure, a minimum number of research student completions and particularly a minimum number of research-active staff. The quality of research outcomes would be the key impact. Performance thresholds would be specific to a particular field of research but set broadly. The assessment teams would comprise external expertise, including international participants.

Based on the outcome of the assessment process, university research would be funded according to five grades and there would be a minimum performance level. Universities would receive some base level funding for the output of all research-active staff under this model.

The Academy believes that the research assessment process in the United Kingdom would be too costly for the size of Australia’s research effort. However a regular process modelled on the Research Quality exercise in 1993 could be used. This could be a role for the reformed ARC who would give advice to the Department on the research quality component for the university funding exercise. The 1995 Quality round provides a useful starting point and the Academy would be willing to work alongside the Department of Education, Training and Youth Affairs (DETYA) to develop a model suitable for Australia.

We consider this a much stronger process, and more likely to ensure long-term excellence and diversity across the whole unified national system, than the internal generic processes proposed in the Discussion Paper.

© 2017 Australian Academy of Science

Top