Yearly Archives: 2008

137 – Engaging with researchers: tips for policy makers

Practically, what can policy makers do to enhance the prospects of having their policies founded on sound research?

Quite a lot has been written giving advice to researchers about how to engage with policy makers, but I’ve never read anything directed at policy makers giving them advice on how to engage with researchers. This is a shame, because they need just as much help. Here are my thoughts on the subject.

Understand the researcher’s perspective. There are many different types of researchers, of course, but as a generalisation, they are more interested in innovative ideas and being at the cutting edge of their discipline than they are in the information needs of policy makers. Their main goal is probably to have innovative research published internationally. Often, the real information needs of policy are pretty simple, so the research needed to provide that information may be seen as pedestrian by researchers. Just because some information is important, it doesn’t mean it will be publishable in good journals.

Consequently, there is not a lot of incentive for scientists to fill all the information gaps in an area. They will tend to tackle those aspects that are the most interesting or publishable, and then move on. Depending on the subject area, this may mean that there are a lot of basic details unresearched, even when researchers have been active in that area. At least some of the key information needs of policy are quite likely to be in those unresearched details.

So you shouldn’t necessarily ask researchers to provide information about the “latest” research – it may not be all that useful. If your information needs are for basic data, rather than cutting-edge knowledge, then you will probably need to pay for its collection. You can’t necessarily expect to find it sitting on the shelf somewhere.

Even if research is selected by a government research funding organisation, you cannot necessarily expect that it will meet the needs of policy. The decision makers in that funding organisation are likely to be from research backgrounds themselves, and to prioritise innovation ahead of policy usefulness (or to fail to recognise policy usefulness).

Actively engage on research questions and priorities. Most researchers have little or no idea about which information in their field could be most useful to policy makers. Their research questions usually are not posed from a problem-solving, decision-making perspective. Just because a researcher is doing research on a certain issue, it does not follow that the research will provide the information needed to manage that issue.

Work out your information needs and communicate them clearly. When I have heard policy makers attempting to express their research needs, mostly they talk in terms that are much too general – they want research about X. The problem is that there is any amount of research about X that could be done, and if you leave it to researchers to choose which bits they’ll tackle, the chances are that it will not meet policy needs. So there is an onus on policy makers to get it much clearer in their own minds what their specific information needs are, and then to explain them clearly to researchers. Using an investment framework to select priorities for public investment can help to flush out and prioritise information gaps.

Researchers rely on written communication to a greater extent than do policy makers. Brief communication is likely to be effective for busy researchers just as it is for policy makers, but the requirement for brevity is not so extreme for researchers.

Timing issues. The rushed time frames for policy development mean that there is never time to conduct new research to meet current pressing knowledge needs. But this doesn’t mean policy makers should not invest in research. There should be an ongoing targeted investment in research to inform the next phase of policy development. Unfortunately, policy makers rarely seem to invest in research with foresight. In my experience, they only think about their information needs at the moment they need it, and don’t follow up to ensure that key gaps are filled.

Develop relationships with researchers. Attempt to establish a high level of mutual understanding and trust. “Emphasizing from the beginning an expectation that information will flow both from the researcher to the decision-maker and back to the researcher may allow for a more constructive approach” (Jacobs, 2002, p.10).

Distinguish between knowledge and values. As a policy maker, what you need from researchers is knowledge, not an expression of their personal values – you want a scientist to be an independent arbiter on the facts, or an honest broker, rather than an issue advocate (see PD#110). In theory, most research focuses on knowledge and should put personal values aside, but in practice, researchers are human and their values creep into their interpretation of results, and especially their judgments about the policy implications of research. Inevitably, there is something of a trend for those scientists who are most interested in policy to be those who most seriously tangle up their knowledge with their values, particularly among environmental researchers. Obviously, a scientist’s values are no more important than anyone else’s, so you need to get at the knowledge behind a scientist’s recommendation, rather than taking the recommendation at face value.

Don’t rely solely on local scientists who are known to you. Some of your information needs may be better met by, say, national experts who reside elsewhere.

Be wary of “common knowledge” about scientific questions. Just because “everybody knows” something doesn’t make it true. For example, when I started working on dryland salinity, the common knowledge was that all farmers in a catchment need to cooperate to get control of saline groundwaters. This turned out to be untrue in many situations (Pannell et al. 2001). Ask experts to assess such matters of common knowledge and advise you about the ifs and buts.

Understand that researchers don’t like to compromise on knowledge. As a policy maker, you are used to seeking compromise on policy issues. However, such compromise is simply not relevant to questions about the accuracy of scientific knowledge, which should depend purely on evidence. In areas of scientific controversy, there may be value in policy makers cultivating debate about the evidence. This might later lead to development of some sort of consensus scientific opinion, but generally the debate phase cannot be skipped.

Support efforts to get researchers and users working together. This will help to ensure that the solution being proposed is in fact practical and sufficiently simple. It will help to make sure that the users’ issue of concern is addressed in a way that is relevant to them.

Work with a boundary organisation. Boundary organisations attempt to mediate between the institutions of ‘science’ and ‘politics’. They can help to facilitate collaboration between scientists and non-scientists. In Australia, some Cooperative Research Centres play this role quite well.

If policy is to achieve outcomes, science is probably not an optional extra. In particular, policy that neglects scientific evidence about the link between actions and outcomes is likely to miss the mark.

David Pannell, The University of Western Australia

Further Reading

Pannell, D.J. and Roberts, A.M. (2008). Conducting and delivering integrated research to influence land-use policy: an Australian case study, INFFER Working Paper 0803, University of Western Australia, Perth (submitted to Environmental Science and Policy). Full paper (70K).

Pannell, D.J., McFarlane, D.J. and Ferdowsian, R. (2001). Rethinking the externality issue for dryland salinity in Western Australia, Australian Journal of Agricultural and Resource Economics 45(3): 459-475. Full paper (114K) • brief version (11K) • newspaper article based on the paper (6K) • Full journal paper (164K pdf file) also available via the Journal homepage:

136 – Engaging with policy: tips for researchers

Practically, what can researchers do to enhance the prospects of having their research adopted and used by policy makers?

There is no easy recipe, but there are many potential elements of a strategy. My advice is to do as many of the following as you can manage.

Understand the policy maker’s perspective. Understand the motivations and constraints of the policy makers you wish to influence. They will only adopt your advice if it is consistent with their objectives and perceptions. For example, if a proposed change disadvantages a group that the minister wishes to favour, or requires complex and time-consuming processes to implement, adoption is unlikely. Policy makers face numerous demands and responsibilities and invariably have tight timelines. They are likely to have expertise in areas other than your area of science. They may have different criteria for policy success than you expect.

It helps to align the research results with existing stated policy objectives if possible. Even if the results seem to conflict with current policy, attempt to express them as advancing an existing policy objective.

Science is not enough. Appreciate that “Good science itself is needed but is insufficient to drive informed decision-making. We have to translate it into a form for others to use and to improve decision-making” (King, 2004, p. 190). In considering policy options, policy makers will probably be more concerned with social, economic, political or administrative aspects than with science.

Practice excellent communication. In communications, recognise the lack of time that policy makers have. Be very brief, focus on clear messages, use simple language, free of jargon, using a mixture of approaches. In our experience, written material is useful but is not sufficient. Even more important is effective verbal communication. Communications should always present simplified/synthesised information rather than material suitable for scientists.

Develop relationships with policy makers. Attempt to establish a high level of mutual understanding and trust. “Emphasizing from the beginning an expectation that information will flow both from the researcher to the decision-maker and back to the researcher may allow for a more constructive approach” (Jacobs, 2002, p.10).

Be solution-oriented. It is relatively easy to point out problems with policy, but this may not generate an appropriate response, or any response at all. Do not expect policy makers or managers to see what may seem obvious implications of your research. When offering criticisms of current policy, also offer a practical solution to the problem.

Simplicity is essential. This is true both in one’s communications, and also in the solutions that one offers. In SIF3, based on earlier experience with SIF1/2, we gave a high priority to making the tool as simple to use as possible. It embeds a great deal of past research, but operates as a set of simple decision tables. As far as possible, the solutions one offers need to be simple, transparent and understandable. Policy makers are likely to be suspicious of solutions that rely on complex and opaque computer models.

Work with intended users. This will help to ensure that the solution being proposed is in fact practical and sufficiently simple. It will help to make sure that their issue of concern is addressed in a way that is relevant to them. When attempting to convince policy makers, it helps to be able to demonstrate that the solutions being proposed are already in use in the real-world. Our success in applying SIF3 with two regional environmental management bodies was very helpful in enhancing our credibility with policy makers. The fact that the work involved two regional bodies in different states was also an advantage, particularly at national level.

Distinguish between knowledge and values. Be clear that the values that policy attempts to enhance are based on the desires of the community, not science. It is acceptable for research to deal with values (e.g. studies of the non-market values of environmental outcomes, or studies like ours that integrate values with knowledge) but it is essential to be clear that policy makers will have their own views about the values. Traditionally, science deals primarily with knowledge rather than values. In SIF3, the values of natural resource assets were estimated through various means (see Roberts and Pannell, 2008) as part of the investment framework, but they were clearly distinguished from scientific information about degradation threats and cause-and-effect relationships.

Be pragmatic. One has to accept compromise, and it may be necessary to make conscious decisions about where you can and cannot afford to compromise.

Be patient and persistent. My attempts to influence natural resource policy began in 2000 and are ongoing. Establish networks and build support for your ideas over time. Repetition is essential, even to people who are already on you side. “Preaching to the converted, far from being a superfluous activity, is vital. Preachers do it every Sunday. The strengthening of the commitment, intellectual performance and morale of those already on your side is an essential task, both in order to bind them more securely to the cause and to make them more effective exponents of it.” (Harries, 2002).

Be resilient. Numerous problems, frustration and setbacks will arise. I have at times found myself and my work to be the subjects of ill-informed and unreasonable criticism, almost never to my face. In particular, people with vested interests in the status quo will actively resist proposals for change. These people may be insiders to the policy organisation and so have better access to decision makers than outside researchers do. Being part of a team helps when dealing with the various challenges that arise.

Timeliness is important. Be prepared to respond quickly to requests for information. Policy makers cannot wait for additional research. Attempt to respond to requests immediately if possible. It may pay to follow up with additional information gathering/research in certain important cases.

Find a champion. If possible, identify and cultivate a champion for your work within the policy organisation.

Work as a multidisciplinary team. The need for research that integrates biophysical and socioeconomic aspects of complex natural resource problems is well recogised. We have found that having expertise from a range of disciplines within the team has been very helpful when dealing with various policy officers with particular disciplinary backgrounds or interests. The team needs to include “integrators”, who can span the disciplines and draw them together in a way that is relevant to policy.

Our team includes researchers employed by a university and by a government department. This had advantages in that the university researcher was less constrained by government sensitivities, while the government researcher had easier access to policy makers within the same organisation and in a sister organisation.

Avoid any appearance of vested interest. In particular, do not present findings and seek funds at the same time.

Work with or within a boundary organisation. Boundary organisations attempt to mediate between the institutions of ‘science’ and ‘politics’. They can help to facilitate collaboration between scientists and non-scientists. The SIF3 project was conducted within the Cooperative Research Centre (CRC) for Plant-Based Management of Dryland Salinity, which included as partners a number government agencies that we wished to influence. The Chief Executive Officer assisted with our policy engagement in a variety of ways.

A number of the above insights are reflected in the following quote from Australian scientist Richard Stirzaker. In particular, it highlights the need for transparency, simplicity, and mutual understanding.

“To bring together the knowledge and aspirations of managers and researchers, the decision making procedure needs to be:

  • a system of clarity and transparency so participants can understand each other’s knowledge domains (most important);
  •  a system that makes everyone’s understanding of the problem explicit (no black boxes or complicated models);
  • a system that shows how a decision was arrived at (even if it is wrong);
  • a system that can be changed or added to as experience grows; and
  • a system that gives the correct answer (least important)” (Richard Stirzaker, pers. comm., 2007).

David Pannell, The University of Western Australia

Further Reading

Pannell, D.J. and Roberts, A.M. (2008). Conducting and delivering integrated research to influence land-use policy: an Australian case study, INFFER Working Paper 0803, University of Western Australia, Perth (submitted to Environmental Science and Policy). Full paper (70K).

Harries, O., 2002. How to win arguments and influence debate. The Australian Financial Review, Executive Highlights, No. 72, 1 February 2002.

Jacobs, K (2002). Connecting science, policy and decision-making: a handbook for researchers and science agencies. NOAA Office of Global Programs, Boulder, Colorado.

King, L., 2004. Impacting policy through science and education. Preventative Veterinary Medicine 62, 185-192.

Acknowledgement: Thanks to Sue Briggs for her writings in this area and for sharing other relevant papers.

135 – Differences between the worlds of research and policy

Continuing the theme of research-policy links, this week’s PD outlines some of the key differences between researchers and their world on the one hand, and policy makers and their world on the other.

One of the interesting things as a researcher attempting to influence policy is observing the stark differences between the two worlds. There are major differences in the cultures, the values and the way things operate. Here are a few of the more striking differences, as well as some similarities or parallels.

Outcomes valued. Scientists place a high value on knowledge and innovation while policy officers seek to advance the public interest and to please their political masters, who are primarily concerned about being re-elected. Both seek to capture resources, although in different ways and on different scales.

Source of recognition. The sources of recognition for practitioners are different: from an administrative or political master in the case of policy, and from peers in the case of science.

Achievements rewarded. Policy officers tend to be promoted based on their ability to successfully implement desired policy programs and processes, while scientists are promoted according to their productivity of scientific outputs, especially of those judged to be high in scientific quality. Many scientists do not place a high value on research being practically useful. Comparably, some policy officers don’t seem to place a high value on policy achieving real outcomes, as long as political outcomes are achieved.

Controversy vs compromise. Policy officers aim to resolve management or political problems with minimal controversy, making pragmatic compromises wherever necessary, whereas a healthy scientific discipline thrives on debate, and should not compromise the truth. (Of course unhealthy areas of science can be identified, often when the science gets politicised.)

Communication. Scientists and policy makers speak different languages, with different acronyms and jargon and different hidden assumptions. Scientific communication can be hard to understand even between different scientific disciplines. Policy officers deal mainly with very brief, simply written and highly interpreted/synthesised material conveying only essentials with a focus on practical implications and recommendations. On the other hand, even brief scientific writing is considerably more detailed and qualified and much science appears to deal with practical implications as an afterthought, if at all.

Time frame. The time frame for policy development is usually short and the process places participants under great pressure, with little time for careful consideration or analysis. Science is generally slow and unresponsive to urgent policy needs, although it can be responsive in the longer term. Science usually takes a conservative approach to new knowledge, only accepting a change when the evidence is compelling. Policy can also be slow to change in response to new knowledge, especially if the redundant knowledge has strongly influenced the design or implementation of a current program.

Supply vs demand. Policy usually addresses a problem identified by someone else (in a sense, it is demand-driven), while the directions of science are usually selected by scientists (supply-driven).

Complexity vs simplicity. Policy officers prefer simple, straightforward advice with few, if any, caveats, whereas scientists tend to enjoy unraveling the full complexity of an issue, with all caveats and knowledge gaps highlighted.

Specialisation vs breadth. University training and the academic reward system encourages narrow specialisation, whereas policy officers need to consider a broad range of factors.

People focus. Policy involves intensive interaction among diverse groups of people, requiring highly developed social skills, while for some scientists working with lay people is not comfortable. They may prefer to work individually, or at least in groups of like-minded scientists.

David Pannell, The University of Western Australia

Further Reading

Pannell, D.J. and Roberts, A.M. (2008). Conducting and delivering integrated research to influence land-use policy: an Australian case study, INFFER Working Paper 0803, University of Western Australia, Perth (submitted to Environmental Science and Policy). Full paper (70K).

Acknowledgement: The article draws on various published papers, particularly those by Sue Briggs, Anne Tomes and Sandra Nutley. See the paper by Pannell and Roberts (2008) for full details.


134 – Why is it hard for research to influence policy?

Some research, probably a minority, is conducted with the serious intent of supporting or influencing policy. Within that subset of research, only a minority succeeds in its policy intent. Why is that?

I spend a fair bit of time talking to people involved in government policy at different levels, trying to encourage the use of approaches to decision making that will address questions such as: Will this policy really achieve the desired outcomes? Will the desired behaviour change really occur in response to the planned policy mechanisms? Of the things that money could be spent on in a particular program, which will give the most worthwhile results for the dollars spent?

They seem like common-sense questions that should be of interest to anybody involved, but it often seems extremely difficult to have them taken seriously. Why is that?

There is actually a fair bit of commentary in the research literature about the difficulties that researchers have in influencing policy. A broad range of reasons for poor uptake of research results by policy makers have been identified, including the following.

Non-scientific considerations matter. Sometimes scientific information may be known to policy makers, but the decision reached may still appear to be inconsistent with the science. This may or may not be a concern. It may simply reflect that policy makers and managers consider additional factors, such as legal mandates, societal desires, economic benefits and costs, rights, distributional equity and procedural fairness.

Hidden agendas. There may be political or bureaucratic objectives unrelated to the public interest, so that research that seeks to advance the public interest is not wanted.

Policy fashions and crises. Policy attention tends to be directed to certain issues with high currency, often issues where there is a perceived crisis (e.g. see PD#123), and this may leave little scope for research (or any other input) to influence policy in a policy area that is not currently high on the agenda. We observed this occurring in Australia around 2000 with land and water degradation due to salinity. Now salinity is no longer perceived as a crisis, and policy attention has shifted to another area of perceived crisis: climate change. The fall of salinity as a policy issue is remarkable. At one point it dominated all discussions about the environment in Australia, and now it seems to be hardly mentioned outside Western Australia. Even when an issue is perceived to be in crisis, it doesn’t necessarily follow that research results will be noticed and acted on. For example, research that challenges preconceptions will tend to be dismissed, especially when there is a lot of momentum behind a particular view of the issue.

Timing issues. “Information must be timely to be useful” (Jacobs, 2002, p.9). “Policy generally moves faster than science, and the capacity of science to provide information may require more time than policy makers are willing to accept, especially for politically hot issues” (Clark et al., 1998, p.9).

Difficulty getting access. Policy makers often rely primarily on locally based and trusted experts with whom they are familiar. This is understandable, given the flood of information that policy makers can face on some issues, but it does not ensure that the most appropriate information is used, and it can make it very hard for new researchers or outside researchers to be heard.

Distrust. There may be suspicion about motivations of scientists, so that they are treated as just another interest or lobby group. The growth of public advocacy by some high-profile scientists feeds these suspicions. Even without overt advocacy, some environmental scientists tend to intertwine facts and values, and this affects the perceived independence of their scientific advice.

Incentives facing researchers. It may be that scientists are not conducting research that is relevant to policy, or not making efforts to make their science known to policy makers. “University reward systems rarely recognize inter-disciplinary work, outreach efforts, and publications outside of academic journals, which limits the incentives for academics to participate in real-world problem solving and collaborative efforts” (Jacobs, 2002, p.14). This probably helps to explain the observed lack of so-called “boundary spanners”, these being individuals who can link the worlds of science and management and translate the concerns of one to members of the other (Clark et al., 1998).

Communication problems. Research findings may be communicated in ways that policy makers cannot understand, using jargon, technical language, or mathematics. Conversely, policy makers also have a comprehensive set of jargon, and may have trouble expressing their information needs in ways that researchers can respond to.

Lack of expertise. In some situations there is rapid turnover or movement of staff in government policy agencies, leading to lack of expertise by responsible staff and lack of knowledge of relevant science and scientists. A culture may develop in government agencies that detailed subject knowledge is not necessary.

I believe that understanding of the adoption of new policy ideas by policy makers can be enhanced by a knowledge of the research literature on adoption of innovations (e.g. Pannell et al. 2006). Here are some generalisations from that literature that are relevant to the policy sphere:

  1. Most potential adopters considering an innovation are sensibly cautious. They do not rush in, but seek information to improve their eventual decision about the innovation.
  2. Where decision makers do not have personal experience with an innovation, they rely to some extent on external sources of information. As decision makers gain personal experience, this tends to have a dominant influence on their perceptions and their actual behaviour.
  3. External sources of information are given more or less weight depending on factors such the expertise and credibility of the information source, the relevance of the external information to the decision maker’s circumstances, and the number of external sources reinforcing the message with similar information.
  4. Apparently misguided decisions to adopt or not adopt an innovation can often be easily understood and seen as reasonable if one makes the effort to learn about the objectives and perceptions of the individual decision makers involved.
  5. Many factors influence the speed of adoption of an innovation. Key ones include: the extent to which adopting the innovation is superior to maintaining existing practice; the ease with which the innovation can be observed and evaluated; the number of other potential adopters who have already adopted it; and the intensity and quality of efforts to promote the innovation.

David Pannell, The University of Western Australia

Further Reading

Pannell, D.J. and Roberts, A.M. (2008). Conducting and delivering integrated research to influence land-use policy: an Australian case study, INFFER Working Paper 0803, University of Western Australia, Perth (submitted to Environmental Science and Policy). Full paper (70K).

Pannell, D.J., Marshall, G.R., Barr, N., Curtis, A., Vanclay, F. and Wilkinson, R. (2006). Understanding and promoting adoption of conservation practices by rural landholders. Australian Journal of Experimental Agriculture 46(11): 1407-1424. If you or your organisation subscribes to the Australian Journal of Experimental Agriculture you can access the paper at: (or non-subscribers can buy a copy on-line for A$25). Otherwise, email to ask for a copy.

Clark, R.N.; Meidinger, E.E., Miller, G., Rayner, J., Layseca, M., Monreal, S., Fernandez, J., and Shannon, M.A., 1998. Integrating science and policy in natural resource management: lessons and opportunities from North America. Gen. Tech. Rep. PNW-GTR-441. Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station. 22 p. Here.

Jacobs, K (2002). Connecting science, policy and decision-making: a handbook for researchers and science agencies. NOAA Office of Global Programs, Boulder, Colorado. Here.

Acknowledgement: The article draws on various published papers, particularly those by Sue Briggs, Michael Carolan, Lonnie King and those by Clark et al. and Jacobs cited above. See the paper by Pannell and Roberts (2008) for full details.


133 – Environmental budgets: two distinct questions

Environmental advocates damage their own cause by confusing two distinct questions about the public environmental budget.

In recent years I’ve put quite a bit of effort into trying to improve the environmental and natural resource outcomes from publicly funded programs. One of the challenges in this is convincing some people that it’s better to accept that some environmental assets should not be targets for public funds. By not attempting to save those assets that would be the most intractable and expensive to save, we can increase the chances of saving those assets that are a better bet. In taking this position, I’m focusing on a particular question:

1. How should the available budget be spent to get the highest-value environmental outcomes?

Some people get quite upset at the very idea that we should write off some parts of the environment. For example, among the public responses to the Victorian government’s recent Green Paper on biodiversity, there were many outraged objections to the very sensible statements in the document recognising that not everything can be saved. In taking this position, people are focusing on a different question:

2. How big should the environmental budget be?

Both questions are relevant, but they should not be mixed up. A classic example of mixing them up is the environmental advocate Stuart Pimm, who argues that concepts like triage are defeatist and unduly pessimistic and offer politicians and managers an easy way out.

It is quite possible to believe that the environmental budget should be much bigger (question 2) but to also believe that we should not attempt to save every threatened environmental asset (question 1). Even if you held the extremist position that the budget should be big enough to save absolutely every threatened environmental asset, it would still be sensible to accept that, for as long as it is not that big, hard decisions about how to spend the environmental budget cannot be avoided. (Incidentally, a view that we should save everything is really side-stepping another set of hard decisions, about how public resources should be allocated between competing issues such as environment, health, education, …).

The problem is that by objecting to hard-nosed decision making about the environment (on the grounds that the budget should be big enough that hard-nosed decision making is unnecessary), environmental advocates run the risk of reducing the environmental outcomes achieved by policy. Whatever one would like, one can be certain that the environmental budget will never be so big that hard decisions are unnecessary.

I have no problem with people advocating for a bigger budget for environmental programs, but I wish they would not compromise those programs by arguing against systematic, rigorous and tough decision making. In my view, if they do argue this way, and their arguments are accepted, the result will be bad for the environment: the loss of environmental outcomes from spreading resources thinly across most or all assets is likely to be greater than the environmental benefits from any extra resources that result from this advocacy.

A version of this problem arises in the way threatened species are handled in Australia. When a species or ecological community is officially recognised as threatened with extinction, the government is legally required to look like it is doing something to save it. A recovery plan is prepared and some money is spent on it, but the system does not take seriously the question of how public resources should be allocated among species. In theory, at least, all species are viewed as equally deserving, no matter whether they would be easy to save, or difficult to the point of impossibility. Inevitably, the system results in more extinctions than one that took a more hard-nosed approach.

David Pannell, The University of Western Australia