Designing applied research for impact

By Andrew Campbell

Andrew Campbell (biography)

How can we get the three critical groups in transdisciplinary research—researchers, end users of research, and funders of research—to work together in designing applied research for impact? As Roux and colleagues (2010) pointed out:

A key characteristic of transdisciplinary research is that the domains of science, management, planning, policy and practice are interactively involved in issue framing, knowledge production and knowledge application.”

A critical challenge is that each of the three groups is likely to have different perspectives on the goals of a given research project or program and how to achieve them, and therefore likely to define success differently.

In synthesising the lessons from eight years of the Australian National Environmental Science Program (NESP) and its predecessors (which involved more than 500 researchers from more than 50 organisations and a federal government investment exceeding $300 million Australian dollars), colleagues and I brought together the three groups to review accountabilities in transdisciplinary research and reach consensus on the relative responsibilities that funders, researchers and end users have in delivering these accountabilities.

The accountabilities seen as important for all three groups were leadership, engagement and discourse. All participants in collaborative, transdisciplinary research need to demonstrate leadership and to remain engaged and actively communicating throughout the research process.

Funding agencies need to maintain sufficient continuity in staffing to be intelligent purchasers, able to ‘take the long view’ and undertake high quality strategic planning and adaptive management at a research program level, responding to changing circumstances and priorities as necessary, but no more than necessary. Research funders need competent project management systems, extending to management of data, information and the knowledge legacy from past investments. They need sufficient scientific capacity to be able to evaluate research proposals and the track records of research providers, but not to the extent of second-guessing researchers once programs and projects are contracted.


Researchers’ accountabilities emphasise scientific quality, competent project management and willingness to engage in two-way knowledge sharing with end users.


Research end users in turn must engage in the research process to the extent necessary to maximize the chances of research outputs being implemented in their real world. They need sufficient organizational research capacity and scientific competence to engage effectively with researchers in problem definition and/or co-design of the research.


The ultimate performance measure for such research investments is the extent to which program outputs are adopted, and the resulting environmental benefit. The capacity to interrogate, adapt and utilise research outputs, and their ability to engage in adaptive learning and decision-making as new knowledge emerges, are crucial accountabilities for end users.

Continuity is also an important attribute for all three groups. Extended interaction bridges cultural differences between the different worlds of researchers and end users. It helps researchers to better understand the needs of end users, it makes it easier for end users to challenge researchers and to interrogate research findings more freely, and it gives funders more confidence to invest in possibly riskier, less well-defined or more adaptive projects, in a spirit of co-learning. Where research programs are funded for four years or less, as is increasingly common, it is difficult to sustain continuity of personnel and to build social capital (familiarity, respect, trust, reciprocity) between funders, researchers and end users.

Our research showed that there is a high level of consensus among the leaders of multi-institutional, transdisciplinary environmental research programs in Australia that the chances of such research making a difference are maximised when research investments are designed such that funders, end users and researchers have shared goals, sufficient continuity of personnel to build trust and sustain dialogue throughout the research process from issue scoping to application of findings, and sufficient flexibility to be able to adjust and respond to new knowledge, changing circumstances and priorities.

How well does this gel with your experience?

Campbell, C. A., Lefroy, E. C., Caddy-Retalic, S., Bax, N. J., Doherty, P., Douglas, M. M., Johnson, D., Possingham, H. P., Specht, A., Tarte, D. and West, J. (2015). Designing environmental research for impact. Science of The Total Environment, 534: 4–13.

Roux, D. J., Stirzaker, R. J., Breen, C. M., Lefroy, E. C. and Cresswell, H. P. (2010). Framework for participative reflection on the accomplishment of transdisciplinary research programs. Environmental Science and Policy, 13: 733–41.

Biography: Andrew Campbell is Professor and Director of the Research Institute for the Environment and Livelihoods, Director of the Darwin Centre for Bushfires Research and Director of the Centre for Renewable Energy at Charles Darwin University, in Darwin Australia. His research interests span the interactions between sustainable agriculture, food systems, climate, water, and energy, and the interface between science and policy.

1 thought on “Designing applied research for impact”

  1. It is aligned with my experience in that the need for continuity is often shorted by “program-mentality” or the real constraints of organizations. In uncovering some transdisciplinary and transformation traits needed in resolving the wicked problems of landscape sustainability, I “cheated” by stringing together a half-dozen projects that occurred over a decade or so. In a nutshell, I concluded that resolving wicked problems requires a transdisciplinary approach supported by a shared governance model housed within a multi-sided platform.


Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: