The skilled graduate must become the most important outcome of research training in Australia – as it is in many other parts of the world – urges a new report by the Australian Council of Learned Academies (ACOLA).

The findings from the ACOLA review of the Australian research training system suggest that universities need to better serve the needs of doctoral graduates. The report provides recommendations for how to help better connect research students and graduates with industry.

With fewer than 50% of Australian doctoral graduates being employed in the university sector, industry experience and training that better equip graduates for jobs beyond universities are essential.

The report calls for transferable skills to be central to the training of doctoral candidates. Such skills development must be flexible and candidate-directed, and build on the diverse backgrounds and experience of candidates.

Improve industry-university research collaboration

Industry placements are at the heart of this model. Industry is defined broadly as: businesses, governments, government business enterprises, non-government organisations, not-for-profit organisations and community organisations. This needs to include research students in the social sciences and humanities as well as those in science, technology, engineering and mathematics (STEM) disciplines.

Australia has been slow to adopt such initiatives. This failure is likely to have contributed to our lacklustre performance in research collaborations between higher education providers and industry. Our poor performance is illustrated by several indicators.

For example, Australia ranks last out of 30 OECD countries on the proportion of small and medium enterprises collaborating with higher education and public sector institutions on innovation. It is second-last on the same indicator for large businesses.

Encouraging PhD students to form a bridge between academia and industry will bolster collaborations and be a win-win for all.

Improve PhD supervision

The review also calls for more professional supervision through recognising excellence, driving metrics and providing high-quality training and ongoing professional development for supervisors.

While rates of satisfaction with supervisory experience for research students have steadily increased over the past 14 years, surveys reveal that one in five research graduates is dissatisfied with the supervision they received. Negative expectations of cloning (creating graduates in the image of the supervisor) and cheap labour still linger.

Research training needs to be valued as the human capital development required to build our innovation economy as well as for the research that research students produce. Increased emphasis on the skilled graduate as the major outcome of research training will require cultural change in supervisory practice.

What other countries are doing

In 2002 in the UK, the Roberts Review catalysed a policy initiative that provided £120 million in new government funding for skills development of research higher degree candidates and postdoctoral research staff in STEM disciplines.

This program was a major cultural change in the provision of skills and career support for researchers in UK higher education institutions. It championed the value of investing in the development of individual researchers.

A review of the program in 2010 confirmed that it had resulted in major improvement in the way career development and transferable skills training were provided for researchers. It was unable, however, to provide quantitative metrics of success because of the absence of baseline data.

Although no such dedicated funding has been available in Australia, many universities have invested in the delivery of transferable skills training to higher degree research (HDR) candidates. However, transferable skills development is not as strongly embedded in our research training system as in the UK and some other similar countries.

In 1999, Canada established Mitacs as a part of its Network Centre of Excellence Program. Mitacs started as an initiative in mathematics, prompted by the loss of Canadian mathematics PhD graduates to the US.

The program expanded from 17 internships in 2003 to almost 3,200 in 2014-15. By 2020, 10,000 internships are expected to be delivered per year.

Although Australia has several small internship schemes for research students, we estimate that fewer than 120 of more than 20,000 HDR candidates undertook an internship in 2015. A Canadian HDR candidate is 16 times more likely to undertake a placement than their Australian counterpart.

Mitacs has 14 government partners including the federal government, most of the provincial governments and more than 60 universities. In 2014-15, Mitacs collaborated with 1,065 partners in Canada and 23 partners outside Canada; 79% were small to medium enterprises.

In 2014-15, it enabled two million hours of research work by interns for industry and not-for-profit partners, underpinned by an investment of C$22 million.

Mitacs surveys of their partners indicate that 82% have continued collaborations with the academic supervisor, 66% of the products of the collaborations have been commercialised and 47% have hired at least one intern. The majority of participating interns feel more employable as a result of their internship.

The 2015 Review of Research Policies and Funding recommended that Australian government funding of $12.5 million per year be provided to create a program to support universities to increase numbers of industry placements for PhD students.

The ACOLA review recommends that this funding be used to seed a national program of placements for research doctoral candidates, along the lines of Mitacs.

Certainly, some Australian research students already obtain experience by working on externally defined research problems, in non-academic settings with non-academic supervisors, especially as part of the Co-operative Research Centre system.

However, the review did not have enough data to evaluate the significance of such partnerships at a comprehensive national scale. This is an example of the inadequacy of the data available to determine the performance of the research training system and its value to Australia’s economic and social well-being.

The review calls for a longitudinal national data-collection exercise to monitor course satisfaction, course completions and career outcomes for HDR training.

Without such data, it is impossible to determine the return on this almost A$1 billion annual government investment in the development of the human capital required for the innovation economy.

This piece was co-authored by John McGagh, chair of the ACOLA review expert working group. He is an adjunct professor at the University of Queensland.