What’s the “paradox of interdiscipliarity”? It’s the difference between what institutions say and how they act. Universities promote interdisciplinary research, yet they rarely provide practical support to researchers forging connections across disciplines. And although funding agencies say they support interdisciplinary research, gaining funding for projects that span academic fields is much harder than for those that sit neatly within a single discipline.
But could this just be sour grapes? After all, funding rates are depressingly low for fundamental research, with many funding schemes having success rates of 10 per cent to 20 per cent at best. It’s not unusual for researchers to feel that their great ideas are not meeting with the approval and support they deserve. Could it be that interdisciplinary researchers are blaming their discipline-breaking applications, when it’s simply that they weren’t good enough to be funded?
To answer this question, we analysed applications for the Australian Research Council’s “Discovery” scheme, an ideal test case for several reasons. It is the main research funding scheme for curiosity-driven research in Australia, covering all fields except medical research. Every proposal nominates one or more research fields and specifies the weighting between them, providing a tractable basis for evaluating disciplinary spread.
- Where the Nile’s waters meet: creating value through interdisciplinarity and collaboration
- To get started with interdisciplinarity, ask: what’s the problem?
- Juggling competing priorities in research collaborations
Most importantly, the head of the research council was keen to ask whether interdisciplinary proposals were at a disadvantage, so he made the bold decision to share de-identified application data with us. This was critical, because it gave us access to information not only for successful grants, but also the unsuccessful applications, essential for testing if success rates were biased against interdisciplinary proposals.
To compare degrees of interdisciplinarity between proposals, we adapted metrics from conservation biology to reflect not only the number of disciplines identified in each proposal, but also their relative weightings and the degree of disparity between them. For example, a proposal combining mathematics, music and biology would get a greater interdisciplinary score than one that combined physiology, anatomy and genetics, and a proposal with even weighting between disciplines would be considered more interdisciplinary than a proposal dominated by one field with much lower representation of the others.
We found that, across all areas of research, increasing interdisciplinarity was associated with decreased chance of funding. Anecdotal evidence suggests that a similar pattern occurs in many national funding schemes. Application of this analysis to other schemes would answer the question, but it would require cooperation from funding bodies to provide anonymised data on all applications.
Why is this happening?
Although we showed bias against interdisciplinary proposals, we couldn’t test why success rates decreased with increasing degree of interdisciplinarity. But, based on experience with interdisciplinary funding proposals, we can suggest some possible explanations.
High on the list of suspects is the difficulty of pleasing reviewers drawn from different fields. Academic disciplines differ in their concepts of what good research looks like, what are valuable questions to ask and what constitutes a valid way to answer them. It’s much easier to write a proposal that targets one discipline because it can be aligned with the expectations of that field.
This disciplinary difference of values extends to ranking researchers. For a scientist to look funding-worthy, they need lots of publications, preferably in high-impact journals and often in team-based research where they are one author among many. Philosophers, on the other hand, typically publish fewer papers but more books, and they are often sole author. But even the best philosophy journals have lower impact factors than most science journals.
These cultural differences between fields won’t matter for disciplinary proposals evaluated by peers, but a proposal that spans multiple fields, such as philosophy of biology, can lead to the same researcher being considered to have both an excellent track record (by their own peers) and a terrible track record (by researchers from a different field). This places a penalty on researchers with ongoing engagement with interdisciplinary research because they may struggle to compete on measures of success in any one of the component fields.
For example, a researcher with an excellent output of high-profile papers in computational linguistics may be regarded unfavourably both by scientists (not enough papers and grants) and linguists (too many multi-author papers, no monographs). Outputs of a successful art-science collaboration, such as artworks or exhibitions, might not provide any CV points for the collaborating scientists, who are evaluated largely by number of peer-reviewed publications.
How do we solve these problems?
The Australian Research Council responded to our findings by asking applicants to identify whether their proposals are interdisciplinary – and now two-thirds of all applications tick “yes”. Interdisciplinarity looks different from within and without. Researchers examining the effect of temperature on metabolism in a lizard species might identify their proposal as interdisciplinary because it includes both ecology and physiology, but to anyone outside the field, it’s all biology. In any case, it’s not clear if proposals self-identified as interdisciplinary are treated any differently.
An alternative solution could be for funding bodies to create dedicated panels to assess interdisciplinary proposals, with panel members from different disciplinary backgrounds and with experience of conducting interdisciplinary research. Researchers could choose if they wish to be evaluated by the interdisciplinary panel or a discipline-focused panel. Would the researchers on an eco-physiology project want their research evaluated by a panel of biologists or a panel consisting of artists, mathematicians, social scientists and physicists?
The advantage of a special panel for self-identified interdisciplinary proposals is that researchers would know who they are writing for and could focus on explaining the value of their research to a broad audience. Applications to the interdisciplinary panel would be aware they should put their track record, experience and outputs in context, demonstrating productivity and value beyond accepted disciplinary norms. A further benefit would be to allow funding bodies to walk the talk, put their money where their mouth is, and dedicate funding for interdisciplinary projects.
Lindell Bromham is a professor at the Australian National University.
If you would like advice and insight from academics and university staff delivered direct to your inbox each week, sign up for the Campus newsletter.
comment