Science bureaucracy is hugely inefficient: fixing it should be a funding priority

By Prof Adrian Barnett


The government has opened a conversation about Australia’s Science and Research Priorities. Many scientific groups won’t be interested in a two-way conversation, instead they will tell the government that their field urgently needs lots more money. As the priorities are designed to “inform investments”, I expect this lobbying will be depressingly common.

The most obvious field of science that truly needs investment is meta-science, but I would say that because I’m a meta-scientist. Meta-science is the field of science that studies science, and the consensus in our field is that Australian science – whilst often wonderful – could be massively improved.

The scientific system in Australia (and globally) has become hyper-competitive, with many researchers focused on the quantity of their published papers rather than the quality of their work. Current systems of funding and promotion reward those who publish first, but rushed publications often contain errors, making them of no value for society or the economy. The number of published papers being retracted – often because the results or data are unreliable – is growing, including in the world’s most high-profile journals, such as Nature and the Lancet.

There are huge inefficiencies in the bureaucracy of science. For example, Australian researchers spend many months every year applying for funding schemes that have very low chances of success, creating huge opportunity costs. And whilst there may be a global “big data” revolution, in Australia we’ve opted for a “slow data” or even “no data” revolution, as researchers struggle through sludgy approval systems that suck time and money. As an example, a national study of vaccine safety took five years of paperwork to get any data.

Every other industry but science spends a significant amount of time and money on quality improvement. For example, agriculture constantly tests processes and products, and tests innovations to improve quality and efficiency. Science spends virtually nothing on quality improvement, and is still using processes of peer review and publishing designed a century ago. Traditional peer review is struggling to cope with the huge increase in journal submissions. Systems that supposedly measure quality, such as university league tables and the Excellence in Research for Australia exercise, do not capture quality but instead reinforce the incentive for scientists to prioritise quantity over quality.

The growing field of meta-science provides an opportunity to improve science through careful study. Examples include developing more efficient systems of funding and hiring that reward quality over quantity; policy innovations to increase transparency and data sharing; and designing processes which facilitate dynamic, diverse, and inclusive research teams.

Other countries have encouraged meta-science by establishing independent centres and programmes. For example, the US National Science Foundation’s ‘Science of Science’ programme; the UK Research and Innovation Caucus on Innovation and Research Funding Policy; the Volkswagen Foundation’s ‘Researching Research’ programme; and the multi-national Research on Research Institute with headquarters in the UK. These centres are generating, testing, and translating practical solutions to improve science.

The government’s conversation piece wants to hear about the greatest current national challenges and opportunities. The greatest current challenge for science is to improve itself. This is a big call considering the existential crisis of our dramatically changing climate, but the current problems in science are widespread and costly. While Australia must fund environmental sciences, we also need funding to create a better environment for science.

Prof Adrian Barnett, Strength Lead – Statistics and Data Analysis at the Australian Centre for Health Services Innovation, is the current president of AIMOS (Association for Interdisciplinary Meta-Research and Open Science), a society of inter-disciplinary researchers that use science and research to improve science and research.

This article was originally published in Campus Morning Mail, 28 March 2023