Fostering Adoption of More Participatory Approaches to Extension – Some Tales from the West Theo Nabben, Kathryn Egerton Warburton, and Jan Paul van Moort Agriculture Western Australia, PO Box 1231, Bunbury, WA Email: tnabben@agric.wa.gov.au

Abstract

This paper provides a broad context for changes to extension within Agriculture Western Australia, raising some specific questions for discussion, focussing on the difficulty in using participatory extension approaches under Funder, Purchaser, Provider models, the challenges in reconciling group-based approaches with the recognition that individuals take action and the trends towards client focus and regionalism. It then examines three case studies with varying degrees of participatory involvement to issues related to the planning of extension as an integral part of a research design, outcomes versus expectations and evaluation as a dynamic part of extension programs. The paper discusses the implications of these approaches for Australian public and private sector extension in the future. It foresees a growing need for extension to support social learning and embrace participatory approaches more fully and draws some tentative conclusions for further action.

Introduction

The 1990s were a time of flux for extension in Agriculture Western Australia (AGWEST), where movement in government policy, trends in the agricultural industry and evolution of extension methods created numerous challenges, threats and changes. Over the past few decades, agricultural extension has evolved through four significant stages: production, economic, ecological and the institutional stage (Rhoades 1989). We are at currently in the institutional stage, developing new partnership policies and philosophies for change.

The strong movement in government for accountability, economic rationalism and privatisation while seeking to allocate resources more effectively and efficiently and rein in costs was on going. As governments pursued "user pays", "cost recovery" and outsourcing of services there was considerable debate about the future role of public extension. Two key concepts within this were the public and private benefit of extension; and market failure of agricultural information (which services were (not) being provided by the private sector) to determine the role and focus of public sector extension (Coutts, 1997; Dart, et al., 1998; Marsh and Pannell, 1999).

At the same time agriculture, while still significant, was decreasing as a proportion of gross State product and total export, in the face of declining terms of trade, increased farm costs and limited increases in the value of farm output (Hussey, 1994). In 1996, the Western Australian State agricultural portfolio, like many around the nation and world, was restructured to provide a more accountable, participatory and business approach focused on the market and customers.

Key features of the new structure are:

• "Funder-Purchaser-Provider model" (FPP) for resource allocation;
• Partnership Groups (comprising industry representatives) to advise programs;
• customer focus;
• the adoption of programs using outcome-oriented project management and strategic planning;
• regionalisation; and
• "provider" groups based on discipline, such as the Extension Services Unit. In 2000, "provider" groups were altered to be based on geographic location.

This created a challenging environment, where extensionists were being asked to substantiate benefits in the face of recurring costs. Difficulties in managing change are not unusual. What is unusual is the pressure extension is under to show accountability at a time when it is also developing new and innovative methodologies to deal with increasingly complex problems and operating environments.
 

Three Rs for extension: Reality – Rigour – Responsiveness

This section outlines some general issues relating to extension within AGWEST as perceived by the authors. We do not see it as a comprehensive list but rather a list of some key issues, which require further discussion.

Reality ? Difficulties in using participatory approaches under existing frameworks

The reality of modern project funding cycles, is the need to "guarantee" outcomes and outputs relevant to industry needs or the Agency’s vision and make evaluation easier. The timeframe for the outputs and outcomes often tends to be short-term (1-3 years). This favours extension approaches such as transfer of technology (ToT) which lead to quantitative evaluation, rather than participative approaches which are harder to evaluate and which may take longer to bear fruit. Additionally ToT approaches are seen as more likely to guarantee pre-determined outputs. However, they cannot guarantee outcomes, particularly for more complex issues, such as behavioural change in natural resource management.

This perspective is also supported by some extension literature (Black, 2000 and Murray, 2000). Murray outlines the incompatibility and risks involved in attempting to undertake participative evaluations, or evaluations suited to participatory extension, within a framework dominated by traditional ToT, R&D or outcome focused approaches. Basically extension programs with a process orientation that does not rigidly define outcomes and goals at the outset are difficult to evaluate. There is also a fear that the current emphasis on achieving pre-determined outcomes may make it difficult for process-orientated extension activities to receive funding in the future. Dart et al.(1998) argue that:

"…projects without clearly defined outcomes and built-in evaluation strategies will not obtain funding."

This raises the prospect that open-ended participatory methodologies may become in short supply under the present "purchaser provider" models over time. Allen cited in Black (2000) notes that "One of the greatest challenges is to build mechanisms into this process [process of evaluation in a FPP model] to allow for learning, correction and adjustment by all parties concerned".

Rigour: Challenge in using group-based approaches which foster individuals taking action.

There has been a move away from linear, "top down", "technology transfer" towards extension methodologies that emphasise information flows, adult learning principles and participation by stakeholders (Marsh and Pannell, 1999). A key element has been the move to working with groups. However, amongst the wide spectrum of private and government people involved in extension, the role of group-based extension is poorly understood and often juxtaposed against other methodologies.

A good example of this is an on-going debate about the need to reconcile the idea of group-based approaches among more traditional extension staff who are used to one to one extension; the demands from many farmers for one to one service; and the stated agency desire to use group-based approaches for both efficiency and group-based learning reasons.

In this debate it is often incorrectly assumed that the shift to group based delivery means that extension efforts are solely directed at the group: group planning; group decisions; group facilitation and maintenance. Recognising that individuals take action, and that groups act solely because individuals within them take action (Clark and Timms, 1999) should focus our attention to extension approaches and techniques which recognise the importance of individuals and the use of relevant information and knowledge systems. Frost and Metcalfe (1999) note the limitations of group-based approaches to individual landholders making changes to farm management and propose a model called ITA (Individuals Taking Action) which can be used to enhance group-based approaches so that individuals' needs are also met.

Responsiveness: trends to client focus and regionalism

AGWEST is also moving towards a client focus and recognising the need for a more sophisticated understanding of its market segments as it evolves from the previously dominant ToT extension paradigm. Examples of this shift include the involvement of primary producers and agribusiness on Partnership Groups which set the strategic direction of Agency programs; the development of branded extension campaigns (e.g. Time to Lime, TopCrop); and a plethora of surveys and rapid appraisals to keep in touch with, and meet, clients’ needs. In the Narrogin agricultural district alone there are approximately 23 "branded extension products". In part this reflects the characteristics of a highly dissected market place with high levels of competition for client "buy in" within the Agency and with the private sector. It may also be a symptom of the ToT supply push – "if we package our good products better the farmers will buy in". One of the challenges facing the Agency is to integrate our extension services, both among programs within the Agency and with the private sector. Another challenge is to develop systems which are flexible enough to meet rapidly changing client needs under an FPP model which favours the predetermination of outcomes and outputs, and the need to be flexible as an Agency (e.g. dealing with the massive reallocation of resources to deal with a locust plague or unseasonal conditions).

One element to foster integration has been that of regionalisation, where regional managers act as providers under the FPP model to ensure integration across Agency programs and negotiate with clients on service provision. However, one consequence of regionalisation has been the disbandment of the Extension Services Unit (which was able to focus on training and management of extension staff, developing and refining extension practice and standards irrespective of which Industry Program staff worked under). Whilst extension will now be delivered at a more local level and be more responsive to local needs there is some uncertainty as to how extension standards will continue to be developed and maintained across the Agency.
 

Currents for Change –Case Studies from Western Australia

The 1990s can be viewed as an era when extension diversified to a discipline with a body of literature, methodologies, methods and processes that are capable of delivering outcomes meeting diverse client needs. It moved beyond technology transfer (telling farmers what to do) beyond farming systems research (defining priorities and working to meet these) and even beyond participatory extension (working with farmers to meet their needs locally) (Fell et al., 1998). Extension is now a holistic systems discipline which van Beek (1997) categorised into 4 main streams:

• transfer of technology and information dissemination;
• problem-solving and decision-making;
• education, training and learning;
• participatory models/human development.

This diversity is reflected in the way extension operates within AGWEST. We have chosen to illustrate some extension approaches that fit more neatly into the participatory models. This is partly because of the apparent concern that existing institutional support for the FPP model is inconsistent with participatory/human development models and partly because we had some first hand knowledge of these projects. Two of them were undertaken as joint Agency/Research Corporation projects and illustrate some approaches and insights typical to the issues either raised above or the following issues:

• the planning of extension as an integral part of a research design;
• extension outcomes versus extension expectations;
• evaluation as a dynamic part of extension programs; and
• the implications of these approaches for Australian public and private sector extension in the next ten years.

Woolpro – A Changing Extension Approach

Woolpro is a process designed to facilitate improved technology adoption by woolgrowers throughout the woolbelt of Western Australia and is supported by AGWEST and The Woolmark Company. By encouraging farmer evaluations and providing "focus sites" (where feed profiles and pasture growth rates are measured on selected paddocks), Woolpro aims to develop participants’ animal and pasture assessment skills; provide information on tools/‘tactics’ to increase productivity and profitability; and enable farmers to evaluate performance of the new technology on farms. Since it began in 1996, Woolpro has evolved from targeting and working with individuals to targeting and working with groups. Woolpro further evolved towards a more participatory approach as it reached the end of its first funding phase and looked to the future.

Extension as an integral part of research design

Woolpro arose initially in response to the feeling that wool producers in WA had the opportunity to increase productivity and hence profitability through the adoption of technologies, much of which was believed to be available "off the shelf".

As Woolpro has progressed, results of other research have been "funneled" into the Woolpro system e.g. : fibre profile work; worm control; TIMERITE^TM (spring spraying for redlegged earth mite); new pasture species. The perceived success of Woolpro has made it a target for other wool program research activities as a means of conveying messages to farmer groups (through a TOT model). More recent developments in Woolpro relating to participatory approaches will result in Woolpro participants being involved in the identification of research and development priorities (rather than extension planning being an integral part of research design).

Extension outcomes versus extension expectations

Initial targets for Woolpro were based on support being provided to individual wool producers. The likely impact in relation to project objectives (30% of wool producers increasing their productivity by 30%) was not great. Targets were then revised to incorporate farmer groups (and hence an exponential increase in the number of farmers who would participate), together with an increased capacity to provide the service (more staff in key locations). Concerns were still expressed that outcomes related purely to quantity and not quality of service. In a large number of cases, Woolpro had some limited information from the on-farm activities of participants, but not enough to enable an assessment against the project objective.

In 1998, with an injection of new thinking into the activity, the demonstration of the achievement of objectives became increasingly important. This led the project team to consider two issues: the time to achieve change (were we expecting it to happen too quickly), and what were the prerequisites for change. As a result of the Woolpro Beyond 2000 Survey in early 1999, we discovered that the longer participants had been involved in Woolpro, the greater the impact of their participation. Though there were no surprises in this - it reinforced the fact that we expecting significant changes to occur too rapidly.

The Woolpro Beyond 2000 Survey was based on Bennett’s Hierarchy as a program logic i.e. for practice change to occur, then a change in knowledge, attitudes, skills and aspirations was a necessary precursor. Using the hierarchy as a basis, we were able to discover that participation in Woolpro had had significant impacts on KASA, as judged by participants. This survey, while useful in discovering level of impact, was a retrospective evaluation. The project did not have good "baseline" data for participants in Woolpro – something which needs improvement. The evaluation was relying on participant’s memories in regard to attribution of impact (questions framed in terms of "As a result of your participation in Woolpro,…")

Evaluation as a dynamic part of extension programs

Evaluation activities and needs within the Woolpro project have changed dramatically since Woolpro started. Initially focussed on low-level evaluations (inputs, activities) and ultimate outcomes (at the other extreme), the focus in recent times has shifted to responses to activities and changes in KASA. The basis of assessment of ultimate impact was an April 1997 collation of initial benchmarks for wool production on farm in target areas (shire data; selected farms within focus shires). This was to be followed each December by the same data collection process. Other measures included participation, and number of, and attendance at, activities such as field days. Skills baseline monitoring was attempted in early 1997, but few replies were received.

Whilst evaluation activities were planned for Woolpro at the beginning, information gathered was not systematically used for improvement. More recently, evaluation has been explicitly designed as an integral part of project development and refinement. Lessons learned from (lack of and difficulties with) evaluation in the early part of Woolpro have been taken into account in evaluation plans for Woolpro phase 2. Elements of the evaluation plan will include:

• activity based ® evaluation aimed at delivering an improved service in the short term;
• assessment of long-term change ® assessment of impact
• participant stories ® based on significant changes, and extent of adoption.

The implications of these approaches for Australian public and private sector extension in the next ten years

Woolpro has undertaken a shift from an individual ToT focus to a group-based approach where presumably the needs of individuals have been satisfied. We expect there will be disquiet among some project staff over this change in approach and believe it may be useful to evaluate what factors support and which hinder a shift in attitudes, values and behaviours as participatory methodologies are introduced to people with a more traditional research-extension paradigm. Many extension program are aiming to develop similar skills and approaches (using different mechanisms sometimes) but delivery is in an enterprise context, so the application of those skills doesn’t cross enterprise boundaries or account for system interactions. As a result, these skills and approaches appear to be "trapped within" the branded products (i.e. TOPCROP, Woolpro), Industry Program or farm enterprise. The successes of Woolpro need to be shared beyond the staff, offices and Wool Program directly involved in the Woolpro project. Structures and processes are needed to ensure that they are shared across Agency Industry Programs and different farm enterprises. AGWEST is currently planning to integrate Woolpro groups with other branded extension / research project groups such as TOPCROP (a cropping version of Woolpro) and Better Business (Property Management Planning).

Wiley et al. (2000) in their assessment of three Woolpro evaluations raise questions about the roles of various players in the research, development and extension process. Participatory action models (PAM) such as Woolpro allow those experiencing the problems to define them, and develop appropriate strategies to address complex whole-farm issues. This raises questions for traditional agricultural research under participatory approaches where farmers are more likely to be directly involved in on-farm research. On the basis of their evaluations, Wiley et al. (2000) suggest that researchers can:

• contribute to the development of techniques and benchmarks for use in monitoring;
• generate information which challenges accepted wisdom;
• collect information farmers require but cannot easily generate themselves;
• assist in the identification of appropriate strategies; and
• assist in the interpretation of farmers observations.

Developing Farming Systems with Lower Recharge for Western Australia - Attempts at Integration

Developing Farming Systems with Lower Recharge for Western Australia is a joint AGWEST/Grains Research Development Corporation (GRDC) project with a total budget of $2,500,000 from 1999 to 2004. The project aims through participative R&D, to develop, test and validate viable new cropping systems with lower groundwater recharge than current systems. The expectation is that farmers will have the basis for future profitable broadacre farming systems with lower groundwater recharge than existing systems. It is anticipated that these systems will be viewed as credible because they will have been developed in full partnership with farmers; trialed and adopted at a commercial scale by some farmers; and based on rigorous research and analysis, involving experts who will have identified the major opportunities for and threats to the new systems. This project aims to reduce the spread of salinity in Western Australia by developing management options that maintain a profitable, productive grains industry.

The planning of extension as an integral part of a research design

The project is essentially focussed on research built around a multi-disciplinary team with a central focus on participatory, on-farm R&D, as the primary vehicle for development and extension of the new farming systems. Extension is recognised as an integral part of the project alongside on-farm R&D, training, agronomic aspects, farming systems, recharge effects, marketing and economics.

It is envisaged that the R&D would be conducted with farmers operating at various levels. In particular a group of ‘primary farmers’, people interested in making a transition from current farming systems, will work in equal partnership with researchers to define the problem and develop the research and design aspects of the project. A full-time extension specialist has responsibility at this early stage in supporting the primary farmers’ input and ensuring the project develops in a participatory manner. He is responsible for planning, implementing, monitoring and evaluating the extension campaign to achieve the target levels of knowledge and adoption. He will also be liaising with regional "High Water Use" extension workers conducting extension via group processes with local catchment / TOPCROP type groups. A training component for key extensionists to support farmers in the implementation of the new farming systems will be linked with existing training programs.

Extension outcomes versus extension expectations

It is not possible to assess extension outcomes versus expectations at this early stage of the project. However, project staff believe there is a high probability of success, partly because there are a number of possible components of lower recharge systems (e.g. lucerne, warm-season crops). Typically ambitious adoption targets for the duration of the project have been set including:

• 80% of farmers have a positive view of the need to introduce lower recharge farming systems and believe they will introduce such systems within a decade;
• 30% of farmers have a good understanding of the new farming systems developed in this project, believe they have the characteristics described above, and that it will provide the basis for their adoption of lower recharge farming systems;
• 5% of farmers have trialled the farming system in an area large enough to provide a model to make decisions about it becoming a significant part of their farm operations.

Initially, it was anticipated that full adoption of the systems would occur by 2014. However, this target is currently being reviewed in the light of project development to reflect more realistic outcomes. The review recognises the incongruency in sticking to predetermined targets within a participatory project which also involves other stakeholders.

Evaluation as a dynamic part of extension programs

Though evaluation is seen as an important part of the project, evaluation tools have not contributed to the project design process. Initial work on a baseline study is planned one year into the project. Though not ideal, it is at least early in the project cycle, before any major outputs have occurred. Overall, evaluation will be used to change project direction if required (it will feed into process and research improvement), as well as reporting to funders and other stakeholders. The performance of the project will be evaluated and reported on using a number of approaches based on Bennett’s hierarchy. Planned evaluation activities include measuring changes in practices, changes in KASA, reactions of participants to activities, description and enumeration of activities and inputs.

The implications of these approaches for Australian public and private sector extension in the next ten years

The project works on the assumption that agribusiness consultants will be a significant group in extending research results to farmers, and ultimately, a major group in ensuring adoption of profitable low recharge farming systems. To this end, agribusiness consultants have been invited to participate in early project design activities with researchers and primary farmers. However, for various reasons there has been little direct agribusiness involvement to date; their involvement will increase as business opportunities or useful information for their clients becomes available. The project will collaborate with TopCrop and a range of programs and agencies involved in addressing the salinity problem. The team leader and extension officer have key roles in gaining industry-wide ownership of the project.
 

Working the Networks - explicitly flexible and participatory

In another example, AGWEST funded a consultancy to undertake an innovative project called "Working the Networks" in November 1999. This aims to improve the flow of ideas and information in agricultural networks. The projects’ specific goals are to build the skills of agricultural service providers in working in networks, support innovation in creating and working in networks and to build and hold knowledge about networks.

While networks and networking have always been a part of agriculture and rural communities, skill, innovation and knowledge in networking are more and more critical to the effectiveness of individuals, or enterprises when there is rapid change in the agricultural information and advice industry. Initial work with agricultural service providers identified a number of weaknesses in relation to AGWEST’s extension in selected parts of the wheatbelt. There were weak links with farm consultants and also weak links between industry programs and the land management section of AGWEST. AGWEST’s networking capacity needed to be enhanced since the old networks, based on long personal relationships within discipline groups and in communities, have weakened in the target communities.

The planning of extension as an integral part of a research design

The project is essentially an extension research project; planning and delivery, using the Action Learning Method, creating and facilitating action learning groups around the topic of networking. The project uses action learning as its core process because this methodology simultaneously creates new knowledge, new practices and new networks. The primary research task is not storage of data, but creating relationships between action learning groups and building the capacity of participants in networking and social enquiry. A website and manual on Best Practice in Networking will be developed by early 2001.

Extension outcomes versus extension expectations

Initial project expectations, initiated by senior AGWEST managers and based on a supporting body of literature, were to develop action learning groups, including both public and private extension agents to support on-farm innovation in two selected wheatbelt communities.

The initial research phase of the project revealed there was no support for networks around on-farm innovation among local extension agents. In the second phase the participants of the learning groups were less interested in making changes to network linkages in a local area, preferring to concentrate on building their own network building competencies as opposed to changing a particular network. There are now four action learning groups dispersed across the State.

To a large extent this shift occurred because local extension agents saw themselves as less skilled and ill equipped to be involved in supporting on-farm innovation or focusing on local networks; they were more interested in developing their own capacity to work as network practitioners.

Some of the reasons that management has accepted the rapid change in project approaches are:

• the philosophy of the project from the start has been that the extension outcomes cannot be predicted and uses the action learning methodology to develop and test the hypothesis as the project evolves;
• project managers/funders were prepared to take risks in developing this project because they had the been wanting to foster a more robust and flexible extension methodology for some time;
• regular communication on changes with management/funders meant they were aware of the real situation (limited capacity among extension staff) and that there were no surprises; the proposed changes fell into a logical consequence;
• the consultant has a proven track record with action learning groups and has credibility with the funders; and
• the process allowed rapid and responsive changes to focus strategy and outputs whilst ensuring broad outcomes were still met.

Evaluation as a dynamic part of extension programs

The action learning process is an adaptive model for project planning, fostering dynamic evaluation within the action learning cycle of plan, act, reflect. The evaluation takes account of both processes used within the project, individuals’ own reflections and also across the project strategies and focus. This has allowed the project to respond to participants' needs as opposed to funders’ initial expectations of needs.

As another example, two groups challenged some planned mid-term progress reports and reshaped these into peer review and normal action learning sessions. This recommendation was accepted by the consultant because the participants’ wishes were aligned with his past experience that the groups may need more time (at least three meetings) to be comfortable enough as a learning group to report on progress to other groups.

The implications of these approaches for Australian public and private sector extension in the next ten years

Assuming the pace of change in the Agency roles and functions, turnover of staff, issues facing rural communities and the need to change agricultural systems is maintained at the same pace, we will need an adaptive and flexible extension process. In this regard the concept of an action learning set offers us a social learning process fostering the ability to learn from, adapt to, and manage changing situations in a flexible and rapid manner. We contend that this is a preferable option to traditional extension approaches that define outcomes up front and do not, or have trouble, adjusting these to changing real world conditions.

The process for setting up and running an action learning group can be readily applied to other aspects of extension practice. Action Learning Sets among peer groups of private service providers may have some appeal as agribusiness seeks to continuously improve its effectiveness and efficiency. For the public sector, the limiting factors will be agency sponsorship of the process, the ability of participants to meet competing demands on their time, and the level of facilitation support for the particular requirements of action learning. Successful implementation will also provide a framework within which public extension agents can discuss and critically review a cornerstone of capacity building among rural communities – developing and maintaining networks as a way of managing the explosion of information rural communities need to manage.
 

Comments on the future directions for extension

The Woolpro and Lower Recharge Farming Systems projects typify the general relationship between public and private sector extension providers and extension issues under an FPP model. Marsh and Pannell’s (1999) study on public sector extension shows a focus on the public benefit/market failure end of the extension-agricultural research spectrum, brokers' involvement in research and projects that have a broader, more integrated applicability and actively seek partnerships. They also demonstrate that some Research Corporations will allow projects to shift to a more participatory approach and that those with in-built participatory methods may also be funded. We believe this occurred in the Woolpro example because some project successes were already evident and because of the interest of key extension staff in participatory methodologies and evaluation. This was partly stimulated by training in evaluation and participatory extension approaches. Furthermore, the Woolmark Company were open to more participatory approaches, illustrated by their decision to fund the Bestprac program for rangeland wool growers and extension agents (see also the work of Clark and Timms (1999) which Woolmark has supported).

The case studies also raise questions and provide some insights into the following:

• the extent that participatory approaches to extension are made more difficult under FPP models;
• the need to extend extension learnings and practice beyond the narrow confines of particular projects, industry programs or farming enterprises in which they generated;
• the integration of research and extension and the roles for traditional researchers in more participatory extension approaches; and
• the difficulties in predetermining outputs and extension objectives such as adoption rates within extension projects.

Frost (2000), Muktasam and Chamala (2000) and Black (2000) comment on the shift from previous instructional paradigms of extension (i.e. ToT approaches) to more participatory approaches, in particular, ones where a social learning is recognised as important. Frost (2000) believes that "the likelihood that knowledge systems [are] key to addressing NRM (Natural Resource Management) problems [and one could add general agricultural development] are those generated in the communities, in other words holistically and involving a range of perspectives, is generally not well appreciated nor supported in agriculture". Accordingly, acknowledging the importance of knowledge systems, particularly those of local communities, presents the following challenges for extension. It requires development of a systematic approach with both common language and goals. A challenge is to engage the NRM community more fully. This represents a shift from land managers as recipients of ‘good’ scientific research to fostering an environment of shared critical thinking with land managers. Such a scenario would involve questioning the approaches taken to address NRM issues as well as the physical research.

Frost (2000) sees a future where extension actively engages the community (land managers and researchers) in social learning. In such a situation, solutions to problems are allowed to emerge and be accepted by the community. This in turn requires extensionists and the community to be involved in self-evaluation on outcomes and reasons behind outcomes. "In other words, in addition to the communication and facilitation skills of extension practitioners, extension processes must challenge assumptions, belief systems and values. This is difficult".

The discipline of extension must become more accountable, becoming part of research and development programs rather than an adjunct to them. Therefore it must develop new relationships with researchers, shifting to a learning focus and not just a communication focus. Extension must shift from ‘single loop thinking’ to ‘double loop learning’ and have the capacity to mix extension methods as appropriate. The challenge for extension is to devise methodologies and techniques that will allow us to adapt and improve on the delivery of our participatory approaches within an FPP model. We see action/social learning as a potentially useful methodology which can assist extension agents develop and manage highly flexible projects capable of dealing with changing conditions.

If social learning is to become the key for the future of a relevant extension approach, then the question is how will extension encourage and sustain this environment, especially when the recognition that knowledge systems may be the key to addressing complex agricultural and NRM issues has not yet permeated the mindsets of many Agency staff. The challenge is to find a place and a way that social learning/action learning processes can be funded and supported in a routine manner within an FPP model. We do not have any easy answers for this, but, the Working the Networks project is premised on the importance of knowledge systems and is actively involved in developing a working model that may offer some valuable lessons.

There is a need to develop and test approaches, which will ensure the philosophy of participation and social learning is supported, particularly in times of rapid change and for complex systems. The Woolpro and Lower Recharge Farming Systems case studies demonstrate that participatory approaches can survive under the present FPP framework and that extension methodologies can be shaped to meet changing needs. We propose some tentative plans for action as a result of our preliminary analysis about how participatory methodologies can be fostered. Specifically we suggest three broad areas for discussion and action:

• More rigorous evaluation of, and reporting on, the unintended consequences /benefits of participatory approaches.
  The Woolpro example demonstrates that participatory approaches are working well, even when they have been adapted to a previously ToT approach. As a discipline, extension needs to examine what has worked well and assisted in delivery; extension staff must consciously document successes and the lessons from our activities. Evaluations should be designed which have a broader perspective than reporting against project outcomes, outputs and milestones. However, the push for impact/outcomes evaluation and traditional monitoring focussing on inputs and outputs to satisfy policy makers or treasury/funders can overshadow evaluation efforts suitable for more open and participatory extension approaches. In short "there is a danger in programs with a strong focus on outcome evaluation that the indicators become the mission"(Dart et al. 1998). Nabben (2000) argues that it is particularly dangerous for public agencies with a strong process orientation to ignore the opportunities offered by broader evaluations. By restricting themselves to traditional evaluation approaches they are doomed to failure or at the best underestimate the unintended impacts they may achieve. This under-reporting of outcomes does not allow for an upward flow of information to policy makers or funders about achievements or new issues and as such stifles the direction of new research by its very omission. Future evaluation frameworks should support the capturing of unintended consequences of our interventions if we are to promote the benefits of participatory approaches to others.
• That APEN, in partnership with AGWEST extension staff, investigate the factors supporting the adoption of participatory and social learning approaches in Western Australia.
  Our limited work suggests that the areas of training, supportive champions, the ability to redesign projects, a clarification of, and focus on ‘real’ outcomes, fostering a peer support network within a social learning framework and a degree of institutional support are some useful points for discussion.
• Development of structures and processes to extend learnings from extension efforts into the broader discipline of extension.
  One of the limitations of workplace-based learning is that it lifts the skills of practitioners without necessarily improving the organisation’s learning. Extension has a responsibility to critically examine what has worked and what needs improvement. Broader discussion should occur around extension learning so that project-level lessons are shared amongst more staff within the Agency. This suggests a changed emphasis for extension from "doing for/doing to "external clients to one which also accommodates a reflective aspect aimed at building capacity in social learning. At the present moment, processes and structures to support this are variable across the Agency. Such sharing is opportunistic, ad hoc and largely based on the transmission of ideas along personal networks. Though still in the early stages of development, a network of Discipline Leaders who have agency-wide responsibility for fostering excellence among all staff who require access to the techniques, practices and processes in specific areas of expertise has been initiated within AGWEST. A challenge for the extension discipline within this network is to foster the capacity for social learning and continuous improvement.

Acknowledgements

The authors wish to acknowledge the contribution of the following people who provided information or comments on the paper: Bill Porter and Richard Olive (AGWEST), Fionnuala Frost (AgInsight), Ross Colliver (Training and Development Group).
 

References

1. Bennett C, Rockwell (1995) 'Targeting outcomes of programs (TOP): An integrated approach to planning and evaluation.' Draft.
2. Black A (2000) Extension theory and practice: a review. Australian Journal of Experimental Agriculture 40, 493-502.
3. Clark R, Timms J (1999) (Eds.) 'Enabling Continuous Improvement and Innovation: The better practices process.' Rural Extension Centre, University of Queensland.
4. Coutts JA (1997) Changes in extension – an Australian perspective – the coming of age of extension. In 'Proceedings of the 2^nd APEN Conference' Albury New South Wales, pp. 22-32.
5. Dart J, Petheram RJ, Straw W (1998) 'Review of evaluation in agricultural extension.' Rural Industries Research and Development Corporation Publication 98/136, Canberra, ACT.
6. Fell R, Clark R, Easdown W, Mobbs R (1998) Extension, dealing with diversity, coping with change. In 'Proceedings of Post Harvesting Handling Conference Technology Conference' GRDC, Canberra, ACT.
7. Frost F (2000) Engaging scientists in community action programs, knowledge systems and social learning; new challenges in extension. ExtensionNet 8, 10-11.
8. Frost F, Metcalfe P (1999) Landcare in the future: using landscaping now to design our futures. In Fenner Conference - Visions of Future Landscapes, Canberra, ACT.
9. Hussey R (1994) 'Report of the agricultural portfolio review to the Hon. Monty House MLA Minister for Primary Industry.' Government of Western Australia, Perth, WA.
10. Marsh S, Pannell D (1999) 'Agricultural extension – a decade of change.' Rural Industries Research and Development Corporation Short Report No. 66. Canberra, ACT.
11. Muktasam A, Chamala S (2000) Group approach in Indonesian agricultural extension: A need for paradigm shift In 'Sustaining rural livelihoods: Building communities, protecting resources, fostering human development. Proceedings of the tenth world congress of rural sociology, CDRom,July- August 2000, Rio de Janerio, Brazil, (International Rural Sociology Association)
12. Murray P (2000) Evaluating participatory extension programs: challenges and problems. Australian Journal of Experimental Agriculture 40, 519-526.
13. Nabben T (1999) Using evaluation tools to clarify outcomes and promote project synergy in the South West Sustainable Rural Development (SRD) region. In  'Ten examples of workplace learning projects from December 1998 Evaluation of Programs & Projects course'. (Ed. J Crisp J) (Agriculture Western Australia, NHT Landcare Training Project), unpublished.
14. Rhoades RE (1989) 'Evolution of agricultural research and development since 1950: Toward an integrated framework.' Gatekeeper Series No SA12, Sustainable Agriculture Program, International Institute for Environment and Development, London.
15. van Beek P (1997) Beyond Technology Transfer. European Journal of Agricultural Education and Extension 4, 3, 183-192.
16. Wiley T, Egerton-Warburton KL, Evans E (2000) Evaluating Woolpro - Lessons for Research and Extension. Asian-Australian Journal of Animal Science 13 (Supplement)B, 83 - 86.