CRICOS Provider No. 00103D Page 0 of 40
The aims of this discussion paper are to (a) describe the current research
challenges Federation University Australia faces in the light of recent
Federal government policy changes, (b) to benchmark research at FedUni
with other universities, and (c) suggest possible strategies to meet the
challenges and improve research performance at FedUni.
Research Directions for Federation University Australia
Adiscussionpaper
Professor Frank Stagnitti Deputy Vice‐Chancellor Research and Innovation September 23, 2014
CRICOS Provider No. 00103D Page 1 of 40
Research Directions for Federation University
The aims of this discussion paper are to (a) describe the current research challenges
Federation University faces in the light of recent Federal government policy changes, (b) to
benchmark research at FedUni with other universities, and (c) suggest possible strategies to
meet the challenges and improve research performance at FedUni.
1. Implications of Federal Government Policy Changes
The Federal Government provides a definition of research for the purpose of the Higher
Education Research Data Collection (HERDC). Research is defined as the creation of new
knowledge and/or the use of existing knowledge in a new and creative way so as to
generate new concepts, methodologies and understandings. Research encompasses a
broad range of activities including pure and strategic basic research, applied research and
experimental development. Applied research is original investigation undertaken to acquire
new knowledge but directed towards a specific, practical aim or objective (including a client-
Research income at the former University of Ballarat has dropped by 40% since 2007 or on average by 8% per annum
Our Commonwealth Research Block grant funding is threatened by under performance and recently announced changes to Higher Education policy
We invest heavily in research academic staff salaries and have sector-benchmarked moderate teaching loads
Last year, approximately 21% of the salaries budget or $2.40M out of a total $11.4M was allocated by the Faculties and Schools to support the research activity of approximately 50% of staff that produced less than 1% of the research wealth of the University in the past 3 years
CRICOS Provider No. 00103D Page 2 of 40
driven purpose). Within the definition of research, the Federal Government recognises the
activities that support the development of higher degree by research students.
Research, as defined above, and undertaken by Australian Universities is currently
evaluated in the Excellence for Research Australia (ERA) exercise.
1.1 Research Income and the Research Block Grant (RBG)
Research in Australian universities is supported with funds from two major sources; (a)
external research income derived from research granting bodies such as the Australian
Research Council (ARC) or National Health and Medical Research Council (NHMRC) or
communities, industries, and philanthropic organisations and/or government or semi-
government instrumentalities such as CSIRO, research development corporations (RDCs)
and cooperative research centres (CRCs), and (b) direct funds from the Commonwealth
Government to the University called the Research Block Grant (RGB).
The major schemes that the Commonwealth Government currently use to allocate research
funds to Universities to support the indirect costs of research include the Sustainable
Research Excellence (SRE), Joint Research Engagement (JRE), Research Infrastructure
Block Grant (RIBG) and Research Training Scheme (RTS). Collectively these schemes form
the Research Block Grant (RGB). Common to all the Commonwealth Government funding
schemes are three important ingredients: Research Income, HDR training and Research
Publications.
The Research Training Scheme (RTS) is based on a formula using two-year averaged data
from the annual Higher Education Research Data Collection (HERDC). The RTS is based on
the following formula: 50% research higher degree completions, 40% research income and
10% research publication. FedUni currently receives about $2M p.a. in RTS and the RTS is
the biggest component of the Research Block Grant (RGB) received by the University.
These funds are used to support HDR students and scholarships.
There is currently much discussion around the sector regarding tying RTS to the ERA. The
Go8 strenuously argue that RTS places should only be supported in areas of research that
CRICOS Provider No. 00103D Page 3 of 40
can demonstrate world class standard or better as measured by the ERA. Should this
proposition prevail, then it would virtually wipe out all our RTS funding.
The Joint Research Engagement (JRE) is based on the formula: 60% research income
(Categories 2 to 4), 30% HDR student load (most recent year available) and 10% research
publications. JRE recognises and funds only four HERDC publications categories: journal
articles, books, book chapters and full refereed published conference papers. JRE income at
FedUni has been falling in recent years.
The Sustainable Research Excellence (SRE) initiative was introduced in 2010 with the
objective to address an identified shortfall in the funding available to meet the indirect costs
associated with Australian Competitive Grants Register (ACG) or Category One grants such
as ARC and NHMRC. The SRE comes in parts – base funding and threshold funding.
Threshold funding is allocated on the basis of universities' relative share of ACG income,
moderated by a proxy performance factor based on the ratio of full time equivalent (FTE)
research staff (teaching & research and research only) to weighted publications. In future
years the SRE will also be moderated by the ERA. Threshold SRE funding cuts in at $2.5M
of Category One income annually. FedUni does not currently earn a sufficient quantum of
Category One income and therefore does not qualify for Threshold SRE funding. FedUni is
only one of three Universities in Australia that does not receive SRE Threshold funding – the
others being University of Notre Dame and the Australian Catholic University. The University
of the Sunshine Coast met Threshold for the first time this year. Their Category One income
has more than quadrupled since 2007. Since 2007 Category One income at University of
Notre Dame and ACU has doubled in each institution but at the former UB it fell by almost
50% from $1.4M (2007) to $711k (2013).
FedUni is a small university operating in thin markets and in an increasingly competitive
tertiary education system. Consequently, the ability for FedUni to support research with
internal funding is very limited.
The introduction of the deregulated system, the proposed reduced level of Federal
Government funding to universities, the likely reduced levels of research funding, the
increased competition for research funds and the proposed changes to the Research
Training Scheme will significantly further challenge Feuding’s ability to support research.
FedUni must therefore develop very highly focused strategies to improve HDR and
publication outcomes and research income generation.
CRICOS Provider No. 00103D Page 4 of 40
2. Benchmarking Research at FedUni
2.1 FedUni Research Income
The following table presents the sector average, the RUN average (excluding UB) and UB
external research income for the past 5 years.
The growth or decline in research income in the previous table is plotted below.
CRICOS Provider No. 00103D Page 5 of 40
External research income across the sector has grown at a rate of about 8% per year since
2007 or a 40% increase since 2007. The research income across the RUN group of
Universities (excluding UB) increased on average by about 13.5% per year or 67% above
the 2007 figure. In contrast research income at UB has dropped by 40% since 2007 or on
average by about 8% per annum. As a result of the drop in external research income, our
UB/FedUni’s Commonwealth Research Block Grant (RBG) revenue has dropped from a little
more than $6.7M in 2005 to a little more than $3.5M today.
The following figure presents research income since 2007 normalised per EFT staff.
2.2 FedUni Research Publications
The former UB’s publications volume remained approximately uniform with a slight increase
of 10% in 2012 compared to 2007. Publications for the RUN group, excluding UB, shows
steady growth with a 30% increase in 2012 compared with the volume in 2007. Overall the
sector is steadily increasing the volume of publications by about 5% per annum, a 25%
CRICOS Provider No. 00103D Page 6 of 40
increase since 2007. Whilst publications volume at the former UB has remained constant,
the number of articles published in journals has increased slightly and the number of articles
published in conferences has decreased. The average citation rate for articles published in
journals, however, remains low with more than 50% of published journal articles in the past
five years receiving one or less citation.
The following graph presents changes in publication rates since 2007.
2.3 FedUni Research Training
UB maintains a steady completion rate of approximately 25 to 30 HDRs per annum. The
2012 figure shows a slight improvement over the 2007 figure (5%) but the years 2008 to
2011 were below the 2007 figure. The overall sector shows steady growth in HDR load and
improvement in HDR completions by about 3% per year. The RUN group (except UB) is
somewhat volatile – 2010 to 2012 numbers are well below those of 2007, a decline of about
10%.
CRICOS Provider No. 00103D Page 7 of 40
The graph below presents HDR completion rates since 2007.
Whilst UB’s completion rate is marginally better than the RUN average, the time-to-complete
a HDR at UB is higher than that for the sector. A longer time-to-complete rate is of concern
for a number of reasons. First, it requires staff to be engaged with HDR students for longer
periods of time which has implications for workload. Second, an overtime student is taking
an RTS place and/or scholarship away from another student. Overtime students adversely
impacts on RTS revenue.
CRICOS Provider No. 00103D Page 8 of 40
2.4 FedUni Research Workload and Resourcing
Deloitte Access Economics was recently commissioned to conduct a study of resource
allocation for the newly formed Federation University Australia (FedUni). The study
incorporated the latest data (2013) for both the former University of Ballarat and Monash
Gippsland campuses. The study showed that the average workload allocations for academic
staff at FedUni are: 49% teaching, 27% research and 24% other duties. Teaching allocation
for the former UB academics was on average slightly higher (54%) compared to former
Monash Gippsland-based academics (41%). Research allocation for Ballarat academics was
on average lower (23%) compared with Gippsland staff (34%). Measured in terms of actual
salary costs, 40% of the total salary bill for the Faculties/Schools was allocated to teaching
duties, 26% to research and 34% to other duties.
CRICOS Provider No. 00103D Page 9 of 40
The figure below illustrates the Research “Wealth” at FedUni (including all Ballarat and
Gippsland staff). The Research “Wealth” is an amalgam of research income, publications,
HDR load and completions averaged over 3 years from 2011 to 2013. The revenue streams
(RBG) that come to the University from the Federal Government resulting from the HERDC
collection (based on research income, publications, and HDR load and completions) will
differ from year to year but the revenue returned on average is as follows: each $1 of
external research income returns approximately $0.20 to Schools and Faculties; 1 weighted
publication point (based on books, book chapters, journal articles and refereed conference
papers) returns $577 per point; 1 EFT HDR load returns about $638 to Schools/Faculties. 1
HDR completion returns approximately $20,000 to Schools/Faculties and $20,000 to support
FedUni HDR scholarships. (Note: FedUni uses 50% of the RTS to fund internal FedUni HDR
scholarships. The remaining 50% of the RTS is returned to Faculties and Schools).
CRICOS Provider No. 00103D Page 10 of 40
The research wealth and the proportion of staff generating that wealth averaged over 3 years from 2011 to 2013 is presented below. This table includes both Ballarat-based (B) and Gippsland-based (G) staff.
The previous figure and table show that
Approximately 6% of staff (25 staff) generated approximately 50% of the research
“wealth” of the University. These staff generated about a 1/3 of the publications, 2/3
HDR completions, supervised 1/3 of HDR students and produced almost 80% of
research income. These staff consumed less than 15% of the total salary allocated
by the Faculties and Schools to support research.
About ¼ of staff produced 90% of the research wealth, about ¾ of the publications,
almost all HDR completions (97%), taught 93% of HDR students and produced 98%
of the research income. In contrast almost ¾ of academics (73%) generated 10% of
the research wealth.
Approximately 21% of the salaries budget or $2.40M out of a total $11.4M was
allocated by the Faculties and Schools to support the research activity of
approximately 50% of staff that produced less than 1% of the research wealth of the
University.
CRICOS Provider No. 00103D Page 11 of 40
2.2.1 Do FedUni staff have high teaching loads and is this a factor in research
performance?
The following most recently available HERDC data (2012) suggest that the student to
student to staff ratio (measured in terms of EFTSL to FTE staff) for the former University of
Ballarat is one of the highest ratios in the nation: UB was slightly higher than the national
average (8.9) and RUN average (9.7). Swinburne, RMIT and Macquarie have high EFTSL to
EFT staff ratios, yet have high profiles in research.
There is however an anomaly in the HERDC data. As much as a third of the HE education
students are taught by private providers under the so called “moderation” model. These
students graduate with a UB (/FedUni) degree and therefore count in the HERDC data as
our students even though they are largely not taught by our staff. When the data are
adjusted to reflect this the student to staff ratio drops dramatically. The Good Universities
Guide in 2012 ranked UB’s student to staff ratio with the following universities: Adelaide,
Melbourne, Sydney, Tasmania, Canberra and CDU.
Teaching Load measured as EFTSL/ FTE staff for 2012
EFTSL/FTE EFTSL/FTE
Top 10 with highest teaching load Top 10 with lowest teaching load
Swinburne University of Technology 11.96 University of Tasmania 6.21
RMIT University 11.90 Monash University 6.15
University of Ballarat 11.15 The University of New South Wales 5.56
University of Canberra 10.48 The University of Adelaide 5.45
Australian Catholic University 10.38 The University of Sydney 5.44
Macquarie University 10.31 The University of Western Australia 5.07
University of Western Sydney 10.31 The University of Melbourne 4.82
Central Queensland University 10.24 The University of Queensland 4.81
Murdoch University 10.06 Melbourne College of Divinity 4.02
Victoria University 9.82 ANU 3.26
Sector Average 8.03
RUN Average 9.43
CRICOS Provider No. 00103D Page 12 of 40
3. Possible Strategies for Improvement
3.1 Research Income
The former UB experienced the most dramatic decrease in research income compared to all
other universities over the past six years. This has also resulted in a dramatic decline in
RBG funding.
The ERA and HERDC provides the best source of information to benchmark research
performance across the national sector. Appendix A was constructed using sector-
averaged research income data for each 2-digit cluster reported in the last ERA
assessment (2012) and assumes that an average academic is allocated a 40% research
loading. The data was then extrapolated to provide indicative research income generation
for academic staff operating with a 10% to 70% research workload. This table provides
useful information as to the indicative levels of achievement that an academic should sustain
if they wish to remain competitive in the sector at their current appointment level and
classification. Table 1 below has been constructed from the ERA data in Appendix A. Table
1 contains indicative research income levels for academics averaged across relevant
Faculties at FedUni. The table does not distinguish research income generated from
different HERDC categories (see Appendix B).
The information presented in Appendix A highlights an issue with health and medical
research. The NHMRC predominantly funds health and medical research. The NHMRC
budget exceeds that of the ARC. It is not unusual for the NHMRC to fund multi-million dollar
research grants in medicine. However it is almost impossible to achieve the same funding
levels in the ARC or indeed in the NHMRC for other health disciplines such as preventive
health, allied heath, para-medicine, nursing, etc. The NHMRC tends to distort the arithmetic
average. For this reason Medical and Health Sciences has been placed into two bands: The
high band; Band 1 is the national sector average for Medical and Health Sciences. Band 1
should apply for all disciplines other than those placed in Band 2. Band 2 is 50% of Band 1.
Band 2 should apply to the following disciplines: Complementary and Alternative Medicine,
Human Movement and Sport Science, Nursing, Nutrition and Dietetics, Medical Physiology,
Public Health, Health Services, Psychology and Cognitive Sciences.
CRICOS Provider No. 00103D Page 13 of 40
Science and Technology has been placed into two bands for similar reasons. Band 1 should
apply to all research disciplines other than those listed in Band 2. Band 2 should apply for
the following disciplines: all mathematical sciences and all information and computer
sciences.
CRICOS Provider No. 00103D Page 14 of 40
CRICOS Provider No. 00103D Page 15 of 40
3.2 Scholarly Research Outputs/Publications
Appendix A presents sector-averaged publications reported in the last ERA
assessment (2012). This table provides useful information as to the indicative levels of
publication expected from an academic at a certain stage in her/his career. Table 2 has
been constructed from the ERA data in Appendix A. Table 2 presents indicative publication
levels for FedUni academics averaged across relevant Faculty disciplines and according to
career level. Unfortunately in this data there is no measure of the quality of publications. The
ERA uses a complex set of processes to measure quality of research outputs – these
processes are different for peer-reviewed disciplines compared with metric-driven
disciplines. It is too difficult at this stage to incorporate a quality indicator but advice to staff
is that at all times, they should attempt to publish work in formats that gain the most
exposure to their research and hence enhance the impact of their research.
The most common form of research communication is publication in journals. Academics
should target the best quality journals in their field and aim for highest possible citation rates.
The links below may be of assistance in choosing journals.
Assessment of quality for non-traditional research outputs such as creative arts, plays,
etc. is difficult but may be based upon demonstration of quality outlets and reception.
“Outlets” refers to the means by which the output appears in the public domain. This
includes but is not limited to art galleries, publishers, theatres and concert halls, networks
of film and television distribution, festivals and events and commissioning bodies. The
quality and prestige of these outlets must be evidenced. “Reception” refers to the extent
and quality of the attention received by the output. Quality can be inferred by, for example,
the review of a novel in highly regarded national/international media outlets, or by the
profiling of an architectural design or project in a leading professional journal. The receipt
of honours, prizes and awards may be used to support a claim of recognition, but is not
evidence of reception per se.
Assessment of the quality of research outputs such as books and book chapters for peer-
review disciplines largely depends on the reputation of the publisher. Authors of books and
book chapters should attempt to publish their work with highly reputable publishers.
CRICOS Provider No. 00103D Page 16 of 40
Conferences are an excellent vehicle for communication of research and learning. Some
conferences publish research as fully refereed conference manuscripts in proceedings and
in some disciplines; for example, Engineering and IT, publications in conference
proceedings are very prestigious. However conference papers, particularly if they have not
been exposed to a rigorous referee process are generally not regarded in the same light as
a fully-refereed research article: Likewise for conference abstracts, notes, corrections,
reviews and editorials. In some Universities conference abstracts, notes, corrections,
reviews and editorials are weighted as a half publication point.
Question 1: Do Tables 1 and 2 provide a useful framework to benchmark research income generation and publication across the sector? If not what other framework with readily available and easily accessible data can be used?
Tables 1 and 2 will be updated with new data following ERA 2015.
3.3 Higher Degrees by Research
The highest source of research revenue at FedUni results from RTS which is largely driven
by HDR load and completion. Currently, 50% of the RTS funds the internal HDR
scholarships. The remaining 50% is transferred to Schools and Faculties to support their
HDRs.
RTS income is under the most threat from recently announced Federal Government policy
changes.
The following table presents the current HDR statistics for FedUni. The statistics were
calculated using a 3 year average from 2011 to 2013 and include UB and ex-Monash
Gippsland load and completions data. It is not entirely accurate but represents the best data
to date.
CRICOS Provider No. 00103D Page 17 of 40
Most of the current load and completions (>80%) is shared by the Faculties of Science and
Technology and Education and Arts. Less than 10% of the load and completions
respectively come from the Faculties of Health and Business. Data on attrition and time to
complete is not yet available for FedUni. However a comprehensive external report
commissioned by the then School of Education and Arts in November 2013 on HDR matters
presented some very interesting findings:
most students in SEA study part time
full-time PhD candidates take much longer (4.8 years) than the 3.5 years that is
funded while part-time students (6.3 years) are under the equivalent time that would
be expected. This confirms anecdotal evidence from elsewhere that shows when
part-time candidates complete, they do so more quickly than their full-time
equivalents
there is a higher rate of withdrawal from part-time candidates compared with full time
candidates
professional doctorates take longer to complete than PhDs
overall average completion rate for students commencing in 2003 to 2005 is about
two thirds while the completion rate for professional doctorates was only 41%
there was an unusual (and unexplained) doubling in withdrawal rates from students
commencing studies in years 2008 to 2010 resulting in a current HDR profile in which
most students are very early in their candidature and completions will decrease over
the next 3 years
if not carefully managed there is significant potential to impact on staff workload and
ability for students to progress satisfactorily
CRICOS Provider No. 00103D Page 18 of 40
A full review of the University’s HDR programs, processes, funding, support, administration,
training, supervision and development will be undertaken soon by an external consultant.
The consultant will also consider the merits of the following propositions.
3.3.1 HDR load
Supervision of HDR at FedUni should be recognised as a contribution to an Academic’s
teaching load. Each full-time HDR student approximately equates to 8 EFTSL revenue per
annum. This contribution to teaching should be apportioned by the number of supervisors
and adjusted pro-rata for part-time students.
3.3.2 HDR completion
Each completion of a PhD student generates approximately $40,000 of Research Block
Grant revenue to the University (and about half of this goes to Schools and Faculties). A
Masters by Research generates approximately $25,000 in RBG revenue. It is difficult to say
with precision how much a completion brings because the amount of RTS and RBG that
FedUni gets each year from the Department is based on the proportion of the total funds
allocated across all Australian Universities. A completion rate and a HDR load that is above
the national trend will result in a greater “relative” share of the total funds distributed by the
Federal Government each year.
A timely HDR completion should be recognised and rewarded. Over-time HDR candidature
should be more carefully managed.
To improve overall HDR outcomes, consider the following propositions:
(a) Each supervisor of a student that successfully completes the requirements for a
PhD on time (i.e. in 3.5 years equivalent full -time enrolment or less) should
attract a share of $40,000 as “notional” revenue that is counted as “credit”
towards reaching the income targets set in Table 1.
CRICOS Provider No. 00103D Page 19 of 40
Masters by Research on time (i.e. in 2 years equivalent full -time enrolment or
less) should attract a share of $25,000 as an “notional” revenue that is counted
as “credit” towards reaching the income targets set in Table 1.
The income “equivalent credit” for completions is apportioned by the number of
FedUni supervisors. Note that the School and Faculty receive the actual income
generated for a completion through the RTS. The supervisors do not receive the
actual revenue from a student that completes but may claim the completion as a
“credit” towards meeting the income benchmark target set in Table 1.
(b) FedUni has a particular problem with HDR attrition and over-time completions. This
adversely impacts on throughput, reputation and revenue. For this reason,
no “income” equivalent credit or “offset” will be applied to supervisors of a student
that does not complete on time;
as an incentive, students who complete within 3.2 years EFT PhD will be offered
a 6 month extension of scholarship to “write up research publications”. No
extension scholarship applies to PhD students that submit after 3.2 years or to
Masters by Research candidates;
students that fail to submit their thesis within 4 years FTE for PhD or 2.5 years
FTE for Research Masters will be required to pay full fees and lose their
scholarships, unless there are mitigating circumstances;
Schools/Faculties will not receive RTS completion funds for students that submit
after 4.5 years EFT for PhD and 3 years EFT for Research Masters
(c) There are many reasons for failing to complete a HDR. Research Services preliminary
investigations indicate that more than half of the candidates experience difficulty in their
HDR programs or who have withdrawn is as a result of poor supervision, poor project
management, lack of guidance, and/or inadequate project support and resources. The
review of HDRs will provide recommendations that will be considered to improve
outcomes.
CRICOS Provider No. 00103D Page 20 of 40
3.3.3 HDR scholarships
The available data indicate that FedUni may be facing a serious crisis with respect to HDR
completions over the next few years. This raises questions as to whether the current merit-
based allocation of scholarships is serving the University well and whether it might be wiser
to prioritise the allocation of student places and scholarships to supervisors that are highly
research active with proven record in research supervision and work in research areas that
have a critical mass of researchers. Greater restrictions should be placed on the allocation
of students and scholarships to supervisors and research areas that have historically
returned relatively poor performance outcomes. Another issue is the quality of the student
intake and ability of a student to undertake a higher degree by research in a timely fashion.
Is there a more suitable research training model that would distinguish FedUni from other
Universities and would mitigate criticisms that students need to be placed in research areas
with proven ERA track record?
One possible approach is to reduce the number of scholarships offered and increase the
stipends as shown in the following model:
CRICOS Provider No. 00103D Page 21 of 40
Introduce a Tier A Scholarship – Industry or Community Named Scholarship
• Minimum $35,000 p.a. tax free for 3.5 years + $5,000 pa (minimum) project
support
• Requires at least $15,000 p.a. cash from industries, community organisations
and/or external grants (greater contributions pay higher awards)
• Scholarships maybe named e.g. the Prostate Cancer Foundation
Scholarship, the Brown Coal Industries Scholarship, the Mary Jones Early
Childhood Education Scholarship, the Gippsland Aviation Industries
Scholarship, etc.
• Open to both International and Domestic Students
Tier B Scholarship – Standard FedUni Scholarship
• ~$26,000 p.a. tax free for 3.5 years + $3,000 pa project support
• Open to both International and Domestic Students
International HDR Students
• Eligible for all scholarships (including tier A and B), however students ...
• Must meet entry requirements
• Must pay fees – no waivers
Phase out Tier B Scholarships within 5 Years
• Within 5 years – ALL scholarships will be Tier A (and will be the highest paid
scholarships in the nation)
Question 2: (a) Should supervision of HDR at FedUni be recognised as a contribution to an Academics teaching load? (b) Do you agree with the strategies to improve timely HDR completion? If not how should FedUni deal with over-time candidates? (c) To improve HDR quality, should stipends be increased in value, the number reduced and greater emphasis placed on the allocating students to teams of very active researchers with proven track records?
CRICOS Provider No. 00103D Page 22 of 40
3.4 Evaluating Research Performance at FedUni
The University needs to be able to develop research performance standards that will lead to
improving its relative position in the sector. The information presented in Tables 1 and 2
forms an excellent framework for doing this.
3.4.1 Faculties and Schools
The following illustrates how the information in Appendix or Tables 1 and 2 may be used to
determine performance of Schools and Faculties.
3.4.2 Staff
The following illustrates examples of how this approach may be used to measure the
research activity of academics.
CRICOS Provider No. 00103D Page 23 of 40
Notes: The sector-averaged benchmarks in Tables 1 and 2 were applied using a 3-year rolling average Minimum sector average thresholds for both publications/outcomes and research income generation
must be met Additional publications do not offset deficits in research income and vice-versa HDR completions may be used to offset income generation, as outlined previously described
CRICOS Provider No. 00103D Page 24 of 40
Question 3: The examples use weighted publication points (i.e. total points split amongst all authors). There is a strong argument that this approach inhibits collaboration. Collaborative research is arguably the fastest way to improve research outcomes. Therefore is it better to allocate 1 publication point to every author, regardless of the number of authors? This will alter the HERDC averages but will it promote better outcomes?
3.4.3 Research Active Definition
The University encourages all academics to be engaged in research but recognises that the
contribution academics make to research will vary greatly from academic to academic
depending on stage of career, appointment level and ability/interest, and from discipline to
discipline.
A single, minimum-threshold metric (such as the 40% research, 40% teaching and 20%
other) is unsophisticated, not flexible, often results in a minimal expectation for staff, does
not properly account for the stage in academic career and does not take into account
differences in research culture for different disciplines.
Recognising differences in stage in career and differences in academic disciplines, Tables 1
and 2 present a more useful approach to recognising the level of contributions academics
makes to supporting research in the University.
A research active academic will have at least a publication record (averaged over a three
year period) commensurate with the 10% threshold in Table 2 and generated research
income (averaged over 3 years) commensurate with at least the 10% threshold in Table 1.
An academic that has sustained over a 3 year period a contribution in both publications and
research income commensurate with the 50% levels or better in Tables 1 and 2 could be
described as “highly research active”.
Research-only staff are expected to be “highly research active”.
CRICOS Provider No. 00103D Page 25 of 40
Question 4: Does this definition of research active adequately describe and fairly recognise research contributions across different disciplines and career stages and lift the research aspirations of FedUni? If not, what better method might there be for determining whether a staff member is ‘research active’?
3.4.4 Setting Appropriate University Targets
The table on the next page presents the actual publication and research income data from
the latest 2014 HERDC submission (2013 data). Using a similar approach outlined above,
the expected number of publications and research income generation based on our
academic profile is shown in bold and black. The actual publication count (2013 and 3 year
average) is above the expected sector publication count. Therefore strategies focused at
increasing our publication rate are not required. However as mentioned previously the
quality of our publications as evidenced by citation rates, etc. require attention.
The expected sector-averaged research income ($7.44M) is well below actual income
($4.14M in 2013 or $3.9M for 3 year average), suggesting that FedUni urgently needs to
develop strategies aimed to improve income generation.
Question 5: Do Tables 1 and 2 (on page 14) provide useful data to set research performance expectations for the University, Faculties, Schools and Research Centres? If not what other framework should be used?
CRICOS Provider No. 00103D Page 26 of 40
CRICOS Provider No. 00103D Page 27 of 40
3.4.4 Excellence in Research Australia (ERA)
In the 2012 ERA, UB was assessed as having two areas of research ranked at world
standard performance: Human Movement and Sport Science and Clinical Sciences. The
following figure plots the number of disciplines at ERA level 3 and above (y-axis) versus the
Research Block Grant (RBG) funding ($) per university (x-axis). Based on the levels of RGB
funding received at the time of the last ERA assessment, UB should have had between 3 to
4 disciplines ranked at world standard or better. Based on the expected research income
projections in section 3.4.3, FedUni should expect to obtain 5 or 6 disciplines ranked at
world standard or better in ERA 2015.
The graphs below plots the research efficiency (defined in The Australian as the number of
research disciplines above world class standard divided by the number of disciplines that
recorded research activity) with the number of research-only and the proportion of research-
only staff for each university in the sector. The analysis can also be repeated using the total
research investment divided by the EFT staff - the results are very similar but not presented
here.
What is interesting in these data is the strong performance of Southern Cross University
(SCU). SCU’s “efficiency” was just above 30%. On this measure, SCU was ranked 10th
highest most research efficient university in the sector. Based on this measure, SCU
CRICOS Provider No. 00103D Page 28 of 40
experienced the most improvement in research between the 2010 and 2012 ERA. SCU
achieved this result with an annual research investment of ~$12M p.a. and with 65 research-
only staff which is about 7% of the SCU workforce.
SCU was evaluated in only 10 specific fields of research (4 digit field of research codes) and
in 6 of these they were rated at world standard or above. SCU was only one of five
universities to achieve ERA 5 rating for geochemistry (0402) from the total of 10 assessed;
CRICOS Provider No. 00103D Page 29 of 40
one of three universities to achieve 5 for zoology (0608) from the total of 21 assessed; one
of two universities to achieve 5 for crop and pasture production (0703) from the total of 11
assessed; one of three universities to achieve 5 for the field of forestry sciences (0705) from
the total of six assessed; one of five universities to achieve 5 for earth sciences (04) from the
19 assessed and one of four universities to achieve the highest rating in the broad field of
agriculture and veterinary sciences (07) from the 23 assessed.
Whilst many may question the validity and assumptions of university ranking models, it is
clear regardless of the model that SCU’s strategy to focus, invest and build a small number
of high quality research teams has paid dividends.
SCU has very similar demographics, size and characteristics to FedUni. In comparison
FedUni should be able to achieve comparable results to SCU.
Question 6: Given the current profile of the University, how many ERA disciplines could/should FedUni aspire to have at or better than world standard for the 2018 ERA? How many of these should we aspire to have at a ranking ERA 5? What is the maximum number of disciplines FedUni could realistically and feasibly support with our current profile?
3.5 Future Research Centres
FedUni, only being newly formed, is in a rapid period of change and evolution. For this
reason, decisions regarding the establishment of new and the dis-establishment of existing
research centres, have been put on hold until a comprehensive review of all research
structures can be undertaken by external consultants scheduled after the completion of ERA
2015. This review will commence late in 2015 and expect to be concluded early in 2016.
Until that time the University has agreed to establish a limited number of interim research
centres.
In preparation for the formal review one possible scenario for the University to consider is a
3-tier structure as outlined below:
CRICOS Provider No. 00103D Page 30 of 40
Tier A Designated University Research Institute (DURI)
Possible characteristics of a DURI might include:
• Predominantly consists of highly-active researchers and ECRs
• Can demonstrate that the Institute is substantially and predominately
contributing to a least one research discipline that is ranked as ERA 5
• At least 50% of core members currently hold ARC and/or NHMRC grants and
75% of academics are highly research active
• Can demonstrate high levels of research performance through research
training outcomes and outputs, publications, grants, esteem factors and
impact
Tier B Designated University Research Centres
• Predominately consists of active researchers and ECRs
• Can demonstrate that the Centre substantially and predominately contributes
to a least one research discipline that is ranked as ERA 3 or above
• At least 50% of core members currently hold significant research grants and
all members are research active
• Can demonstrate high levels of research performance through publications,
grants, esteem factors and impact
Tier C Research Groups
• New and emerging research area/s
• Can demonstrate alignment with University mission and Research Plan
• Can demonstrate critical mass of researchers, excellent potential for
• new research training opportunities, and
• contribution to research quality outcomes
CRICOS Provider No. 00103D Page 31 of 40
All future research institutes, centres and groups should sit within and receive support from
Faculties and Schools. The number that can be supported will depend on the University’s
commitment to support them – There is no point in establishing research institutes, centres
and groups unless there is a significant financial commitment to support them for a period of
time. Comparisons with other RUN universities would suggest that Tier A Institutes (or
equivalent structures) require at least $500k to $1M p.a. for at least 5 years. Tier B Centres
(or similar) require at least $200k to $250k p.a. for 5 years and research groups require $20k
to $100k p.a. for 2 or 3 years. Whatever the final agreed framework to support research, all
formal research structures must develop strong business plans and be required to meet
stringent agreed performance KPIs.
.
Question 7: (a) Is the 3-tier approach to organising University research structures useful? (b) Should these structures sit in Schools or Faculties? (c) Should cross-Schools and cross-Faculties research structures be encouraged and if so how will they be managed and funded? (d) Are the indicative funding levels appropriate? (e) Given the current profile of the University, how many research structures could FedUni feasibly and realistically support in the near future?
3.6 Future Research Workforce
The following illustrates the growth in research-only positions for the entire sector. The
proportion of research-only positions in universities has grown from about 11% of total
workforce in 1998 to a little less than 15% today.
CRICOS Provider No. 00103D Page 32 of 40
The RUN sector average proportion of research-only staff has grown from 2.7% of total
workforce in 2009 to 4% today. Much of the growth in RUN can be attributed to the
Collaborative Research Network (CRN) project. The lowest reported proportion of research-
only staff in RUN was USQ at 1.7%. The highest proportion of research-only staff was
reported by SCU at 7.4%. SCU has experienced the largest growth in research-only staff
from 1.7% in 2008 to 7.4%. Most other RUN Universities have also increased their research-
only staff during this period except for USQ and FedUni (former UB only data). FedUni (UB)
dropped from 3.2% in 2008 to 2.8% (excluding recent departures of CRN staff).
3.6.1 Robert HT Smith Research Professors
The Robert HT Smith Research Professor scheme has proved to be a highly effective
strategy to incubate and concentrate research activity in strategically aligned research
areas. Currently there are 4.4 RHT Smith Professors at Ballarat and 2 at Gippsland. This
number should be increased to a total of 10 at any one time. These professors should be
closely aligned to research disciplines that are operating at or near ERA world standard.
CRICOS Provider No. 00103D Page 33 of 40
3.6.2 Federation University Research Fellows (FURFs)
Due to financial contingencies the Federation University Research Fellowship scheme was
put on hold. The scheme should be introduced as soon as possible. The original proposal
was to appoint up to 10 fellows initially for periods of one, two or three years. These fellows
should be closely aligned to the research activities of the Robert HT Smith Research
Professors and in strategically important research areas.
All Robert HT Smith Research Professors and Federation University Research Fellows
should be appointed to Schools and Faculties. However, Robert HT Smith Research
Professors should jointly report to the DVCRI and relevant Executive Dean. Federation
University Research Fellows should report to a research-active professor or Robert HT
Smith Research Professor.
The complement of 10 Robert HT Smith Research Professors and 10 Federation University
Research Fellows will raise the research-only workforce ratio to about 5%.
Question 8: (a) What percentage of staff on research-only contracts should FedUni aspire to have in the immediate future? Is 5% reasonable? (b) Should FedUni increase the number of Robert HT Smith Professors from its current level of 6.4 to 10 EFT? (c) Should the University proceed with the FURF Scheme?
CRICOS Provider No. 00103D Page 34 of 40
3.7 Future Research Development
Academic research support is currently driven through Research Services. Many universities
have different organisational structures for research services and research development.
Should FedUni create a Centre for Academic Research Development (CARD)?
CARD would be responsible to the University for a range of activities including but not
limited to
ECR training
Graduate Centre programs
Supervisor training
Emerging research leaders
Research mentoring programs
Grant, publication writing and development
Research commercialisation support, etc
CARD would require a full-time Director and a complement of staff including research
training and commercial development officers. Some officers maybe embedded in Schools
and Faculties. It would be expected that the Associate Deans Research, the Professoriate
and highly research active academics be significantly engaged in CARD activities.
Question 9: (a) Should FedUni establish a Centre for Academic Research Development (CARD)? (b) In addition to activities listed above, what other activities should CARD undertake? (c) How should CARD interface and collaborate with Faculties and Schools?
CRICOS Provider No. 00103D Page 35 of 40
4. Feedback
Question 10: Any other comments on how FedUni can improve its research performance?
Please forward your comments and suggestions to
This document and excel spreadsheets containing the data in the Appendix and in Tables 1
and 2 can be found on the research services website.
Closing date for discussion: 31 October 2014.
Shortly after the conclusion of the discussion (31 October 2014), comments will be collated
and summarised. The comments and this paper will inform the final draft of the University
Research Plan and become the basis for a further consultation paper.
CRICOS Provider No. 00103D Page 36 of 40
APPENDIX A: 2012 ERA Sector Average Data
CRICOS Provider No. 00103D Page 37 of 40
CRICOS Provider No. 00103D Page 38 of 40
CRICOS Provider No. 00103D Page 39 of 40
APPENDIX B: HERDC Definitions
Research
Research is defined as the creation of new knowledge and/or the use of existing knowledge in a new and creative way so as to generate new concepts, methodologies and understandings. This could include synthesis and analysis of previous research to the extent that it leads to new and creative outcomes.
This definition of research is consistent with a broad notion of research and experimental development (R&D) as comprising of creative work undertaken on a systematic basis in order to increase the stock of knowledge, including knowledge of humanity, culture and society, and the use of this stock of knowledge to devise new applications. This definition of research encompasses pure and strategic basic research, applied research and experimental development. Applied research is original investigation undertaken to acquire new knowledge but directed towards a specific, practical aim or objective (including a client-driven purpose).
Category 1 Income: Australian competitive grants
Category 1 consists only of income from those research schemes and programs listed on the 2013 ACGR:http://www.innovation.gov.au/Research/ResearchBlockGrants/Pages/AustralianCompetitiveGrantsRegister.aspx. Grants received from the Australian Government and not included in the 2013 ACGR may be eligible for inclusion in Category 2: Other public sector research income.
Category 2 Income: Other public sector research income
Category 2: Other public sector research income includes:
Australian government – Non Category 1:
This is any other income for the purposes of conducting research received from the Australian Government, whether via programs, grants or contracts, that is not eligible for inclusion as Category 1 research income.
State or Territory government, Local government
This is income for the conduct of research received from state or territory government departments or agencies, whether via programs, grants or contracts.
Government business enterprises:
This is income for the conduct of research received from enterprises that are wholly or partly owned or funded by Commonwealth, state or territory, or local governments; have a board; and operate on a profit or cost-recovery basis.
Cooperative Research Centres:
This is research income from CRCs in which the reporting HEP was not defined within the
CRICOS Provider No. 00103D Page 40 of 40
Commonwealth Agreement as “The Researcher” or a “Participant”.
Category 3 Income: Industry and other research income
Category 3: Industry and other research income must be categorised in the following subcategories:
Australian: Contracts, Grants, Donations, bequests and foundations, HDR fees for domestic students, and
International A: Competitive, Peer-reviewed research grant income
International B: Other income
International C: HDR fees for international students
Category 4 Income: CRC Research income