RAAKSmanual2013 10 2107 47 18UTC

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/254762043
Facilitating innovation for development : a RAAKS resource box
Article · January 1997
CITATIONS READS
129 2,726
2 authors:
Paul Engel Monique Salomon

Knowledge, Perspectives and Innovation University of KwaZulu-Natal
55 PUBLICATIONS 948 CITATIONS 20 PUBLICATIONS 255 CITATIONS
SEE PROFILE SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Keeping cattle in a changing rural landscape View project
Small Business Advisory Services View project
All content following this page was uploaded by Monique Salomon on 02 December 2015.
The user has requested enhancement of the downloaded file.

WINDOWS
RAAKS' WINDOWS A
Materials in Each of the following cards contains one RAAKS ‘window’, or perspective, to B
manual and cards use in better understanding the effect of social organization on networks and
may be freely the use and spread of innovation. At the top of each card, you will find the
reproduced in reference number and name of the window, followed by a summary of what it
limited numbers covers and a list of tool(s) that will help in collecting the needed information.
for training and Each window is labelled A, B or C, indicating the phase to which it is most
other educational relevant. Windows are numbered for convenience, although in Phase B a team
purposes. The is encouraged to chose among the windows, or use them in another order, to fit
following your situation. The reference numbers also connect the window to one or more
statement must tools. Figure 2 in Chapter 2 of Networking for innovation provides a way to
appear on all visualize the use of windows in relation to the study as a whole.
copies: Beside the box describing each window you will find a brief description of the C
design, validity, use and applicability of that particular window. Scientific
Reproduced from
‘validity’ generally refers to whether something does what we expect it to do:
the RAAKS
a valid window is one that helps teams focus on practices and patterns of social
resource box,
organization that are relevant to innovation. These windows have been used in
Paul G.T. Engel
a variety of circumstances (see The social organization of innovation); these
and M.L.
field experiences have shown all of the windows to be valid in this sense. On
Salomon. Royal
the cards, the section describing the validity of a window suggests the aspects
Tropical
it covers. By ‘applicability’ we mean the suitability of a window for
Institute, 1997
participatory inquiry: an applicable window is one that helps participants
For large scale construct images of the system to stimulate their interactive learning. This
reproduction or section of the card suggests how the particular window does this.
commercial use, Overall, it is important to remember that windows do not provide ‘recipes’ –
written instead, they suggest general ideas for ways of looking at the analysis. Please
permission from see Chapter 2 in Networking for innovation for more information! It is also
KIT Press, Royal useful, in learning about the windows, to read them in combination with their
Tropical related tools (see the following cards), as well as the glossary (Appendix 4).
Institute, is
required.
file:///D|/projects/RAAKS/temp/New/Raaks-in-Adobe/Windows.html (1 of 2) [4/5/02 6:12:44 PM]

WINDOWS
WINDOWS TABLE OF CONTENTS
A A1 Defining or re-defining the objective of the diagnosis

A2 Identifying relevant actors
A3 Tracing diversity in mission statements
A4 Environmental diagnosis
A5 Clarifying the problem situation
B B1 Impact analysis
B2 Actor analysis
B3 Knowledge network analysis
B4 Integration analysis
B5 Task analysis
B6 Coordination analysis
B7 Communication analysis
B8 Understanding the social organization of innovation: summing up
C C1 Knowledge management analysis

C2 Actor potential analysis
C3 Strategic commitments to an action plan
file:///D|/projects/RAAKS/temp/New/Raaks-in-Adobe/Windows.html (2 of 2) [4/5/02 6:12:44 PM]

Networking for innovation
file:///D|/projects/RAAKS/temp/New/Raaks-in-Adobe/raaks.html (1 of 169) [4/5/02 6:13:21 PM]

A participatory actor-oriented methodology
M.L. Salomon
P.G.H. Engel

Royal Tropical Institute
The Netherlands
Royal Tropical Institute Royal Tropical Institute (KIT) interest in the RAAKS
KIT Press methodology stems from a long-time commitment
P.O. Box 95001 to the development of ways to increase the real
1090 HA Amsterdam participation of those who are the focal point of
The Netherlands rural development – typically farmers and other
Telephone: +31 (0)20-5688272 villagers – and to increase the potential for others,
Telefax: +31 (0)20-5688286 including governments (from local to national
e-mail: kitpress@kit.nl level), donors and the broad variety of other
‘development workers’ to make a meaningful
contribution.
The materials in the resource box have been revised
and edited for the current edition with the
assistance of Rosemary W. Gunn.

ACP-EU Technical Centre for Agricultural and CTA was established in 1983 under the Lomé
Technical Cooperation (CTA) Convention between the African, Caribbean and
Agro Business Park 2 Pacific States and the European Union Member
6708 PW Wageningen States. CTA's tasks are to develop and provide
The Netherlands services that improve access to information for
Telephone: +31 (0)317-467100 agricultural and rural development, and to
Telefax: +31 (0)317-460067 strengthen the capacity of ACP countries to
e-mail: cta@cta.nl produce, acquire, exchange and utilize information
in these areas. CTA's programmes are organized
around three principal themes: strengthening ACP
information capabilities, promoting contact and
exchange of information among partner
organizations and providing information on
demand.
Stoas Stoas is a Dutch research and consulting agency for
P.O. Box 78 which ‘innovation’ is a key word. Stoas focuses on
6700 AB Wageningen human resource development, strengthening
The Netherlands institutions, and rural development related to the
Telephone: +31 (0)317-472711 agricultural sector worldwide. To support our
Telefax: +31 (0)317-424770 human resource development functions, Stoas also
e-mail: LaV@stoas.nl offers packages of multimedia teaching materials.
© 1997 Royal Tropical Institute – Amsterdam

Design: Basislijn/Hennie van der Zande –
Amsterdam
Printing: SSN – Nijmegen
ISBN 90 6832 109 9
NUGI 651
Printed and bound in the Netherlands
Table of Contents
Acknowledgements

Introduction
1. Changing approaches to knowledge and information
2. Designing RAAKS action research
3. Preparing and carrying out RAAKS action research
Appendix
1: Training for RAAKS
2: Exercises
3: References and other resources
4: Glossary of key concepts
About the authors
Evaluation form
Materials in the manual and cards may be freely reproduced in limited numbers for
training and other educational purposes. The following statement must appear on all copies:
Reproduced from the RAAKS resource box, Paul G.T. Engel and M.L. Salomon.
Royal Tropical Institute, 1997
For large scale reproduction or commercial use, written permission from KIT Press,
Royal Tropical Institute, is required.
Acknowledgements
Many people have contributed to the development of the RAAKS methodology. Over time, a real
‘RAAKS knowledge network’ has formed. Within it, a large number of participating colleagues
have applied RAAKS, or similar approaches, in their respective countries in Europe, Latin
America, Africa and Asia. We hope this manual can be seen as a reflection of the ‘state of the art’
within this network.

As in all professional networks, there is a nucleus of ‘die-hards’ who have contributed especially
significantly to the development of the methodology. These include the co-authors of version 3 of
the manual, Ab de Groot and Stephan Seegers; and version 5, Maria E. Fernandez and Niki
Dieckmann. Others in this group are Niels Röling, André Boon, Willem van Weperen, Annemarie
Groot, Cees Leeuwis, Johan den Bakker, Michael Velders, Luc Adolfse, Lambertus Vogelzang,
Maria Koelen, Lenneke Vaandrager, Bertus Haverkort, Janice Jiggins, Peter van Beek, Emilia
Solis, Augustino Moreno, Mans Lanting and Ricardo Ramirez. We believe our networking has
been intensive and fruitful. A special thanks goes to our colleagues in the Editorial Committees
(English and French), experienced trainers/consultants all of them, who read and re-read earlier
versions of the manual, providing us with valuable comments and suggestions time and time again:
Yanick Lasica, Nour-Eddina Sellamna, Rita Joldersma, Diek van Groen, Ruud Ludeman and
Annemarie Groot.
Many organizations have enabled us to put RAAKS to the test. It is possible to name only a few
here: the Netherlands Ministry of Agriculture, Nature and Fisheries in the Hague (LNV); the
International MSc programme, Management of Agricultural Knowledge Systems (MAKS), at
Wageningen Agricultural University, Wageningen, the Netherlands; the Netherlands Organization
for Energy and Environment, in the Hague (NOVEM); the Development of Bioethanol from
Agricultural Crops (OBL), in Breda; the Information and Knowledge Reference Centre on Tree
Cultivation (IKC-Boomteelt), in Boskoop; the Northern Netherlands Agricultural and Horticultural
Organization (NLTO), in Assen; the Netherlands Association of Rural Women (NBvP), in the
Hague; the International Institute for Irrigation Management (IIMI) Pakistan, in Lahore; and the
Institute for Ecologically Informed Agriculture (ILEIA), in Leusden.
We would also like to thank the organizations that have supported the preparation of this and/or
earlier versions, including the Department of Communication and Innovation Studies of the
Agricultural University of Wageningen, the Centre for Agricultural and Technical Cooperation in
Wageningen (CTA), the Royal Tropical Institute in Amsterdam (KIT), and (all in Wageningen) the
International Agricultural Centre (IAC), International Associates in Rural Extension and
Development (IARED), Perspectives Associates for Innovation and STOAS. The interpretation and
presentation of the materials of course remain entirely our responsibility.
Monique L. Salomon
Paul G.H. Engel

Introduction
Many major issues today – whether sustainable development, waste disposal, negotiations about resource use,
rural conflict resolution, natural resource management, clean water supply, food distribution or others – have
neither clear problem definitions nor readily available solutions. Each issue affects innumerable competing
interest groups, who may or may not even recognize each other, much less work together. Each calls for
innovative solutions; yesterday's approaches no longer seem to work. In agriculture for example, extension
specialists have typically acted as ‘brokers’: go-betweens who transmitted information between members of a
target group and specialists or researchers. The focus has been primarily on one interface, that of the
farmer/specialist – or patient/health worker, public/environmental agency, and so forth. Today, however,
specialist/end-user interfaces are no more relevant than those among other stakeholders, such as national
policymakers, traders, industrialists and retailers, local organizations, non-governmental organizations, action
groups or municipal governments, or among the members within any of these categories. New technologies also
do not seem to be the answer; even existing technologies may not be well used.
It looks as if we need ways to understand and facilitate simultaneous multiple-interface interactions. That is, we
need new ways of working together – new concepts of social organization – far more than we need new
technologies! How can images, knowledge and information be efficiently exchanged among a variety of
parties? How can the widely diverse stakeholders involved join together in seeking solutions to complex social
issues? And, once there is a decision to approach each other, how can different actors communicate, much less
organize themselves to make the best use of the information available, learn new practices, and improve their
capacity to innovate? And how can innovation processes be aligned with broader, even more complex societal
objectives?
RAAKS is a participatory action-research methodology that attempts to provide ways for those involved in such
complex situations to begin to find answers for themselves. It does not give any answers directly, but does
furnish an approach to forming a team and beginning to examine the social organization of the system in which
you find yourselves. It is meant for use in a situation where working together can be expected to promote
positive change, and offers instruments (windows and tools) for gathering, organizing and interpreting
information in a participatory manner: specialists working with others as co-researchers and facilitators. That is,
RAAKS can be used to focus on the present and potential social organization of actors (groups or individuals)
in a situation where innovation is desirable. It encourages the team to think about the system you are in, what
you want from it, what it achieves and does not, and what is needed for improvement. The interplay among
actors with different world views, combined with the varied analytical perspectives stimulated by RAAKS, acts
as a motor for the learning process as a team seeks an understanding of their problem that is inclusive enough to
deal adequately with its many facets.
Using this method increases awareness and understanding and helps to develop a shared sense of purpose. The
interactions involved in carrying out joint inquiry help to transform a diverse set of people and organizations

with an ill-defined sense of purpose into a group with a shared perspective, whose members have agreed upon a
number of tasks and responsibilities and have learned to respect each other. That is, RAAKS promotes the
development of a shared conceptual framework that can facilitate the exchange of ideas, experiences, and
knowledge. Moreover, this establishes a basis for implementation: because people work together in the process,
they tend to emerge with a joint commitment to change. They have identified their shared concerns and the
networks that are most relevant, and can explore possible new alliances and begin to formulate action proposals
that are realistic and implementable.
The RAAKS resource box
The RAAKS resource box provides theory, an introduction to the method, and tools and ‘windows’ (analytical
perspectives) for team use. The book, The social organization of innovation, details RAAKS’ foundation in
scientific theory and development practice, detailed case studies, and an account of its development. This
manual addresses the method itself and gives a brief introduction to building a RAAKS team. The laminated
cards in the binder include a series of ‘windows’ – guides that literally help to ‘open up’ new perspectives on
the analysis – and ‘tools’ to help in gathering and processing information. Together, the manual and cards
provide a group of field-tested methodological elements (windows, step-wise phases, tools, and exercises) for
teams to use in their explorations.
It is not necessary to read any of the elements in the RAAKS resource box from A to Z, and they do not contain
any commandments. Rather, these materials are a sort of ‘menu’ from which users can choose what seems most
relevant to their own particular situation. It contains suggestions, it offers criteria to use in weighing alternative
options, and it provides examples. RAAKS is the product of practical experience, gained in numerous studies of
networking among farmers and others involved in agriculture: how they cooperate and communicate, and what
helps them to learn or evolve new practices faster. Instead of singling out one way to explore such questions,
this approach encourages a team that is implementing RAAKS to seize the opportunity to construct their own
study. That is, it provides a starting point consisting of a number of elements that have proved useful in the past,
so that a team need not start ‘from scratch’.
The first chapter of this manual introduces the reader briefly to the problems addressed, the theoretical approach
chosen and the key concepts used. Chapter 2 covers the design of a RAAKS study and gives examples of the
use of the process, in the form of elements from two case studies. Chapter 3 presents a number of factors
involved in getting organized and carrying out a study. Appendix 1 outlines workshops that can be used in
preparatory training and team building, while Appendix 2 consists of exercises a team can use to help in
organizing its work and carrying out an analysis. References and other resources are listed in Appendix 3, and
Appendix 4 provides a glossary of key concepts.
The manual thus focuses on the steps you can take to design a participatory, action-oriented analysis of the way
people communicate and organize themselves to share information and promote innovation. Reading it will also
increase your awareness of ways to develop criteria for evaluating innovative performance in a particular

situation.
We very much hope you will be an ‘active reader’, making your own plan for becoming familiar with RAAKS.
The first three chapters of this manual provide a brief overview, but working your way through the windows
and tools on the cards is a necessary part of learning how the method works. Windows and tools also help to
make concepts and definitions clearer; the tools in particular also contain additional definitions and
explanations. The glossary too can be used early on to become familiar with concepts as well as for later
reference. You may find it worthwhile to skim quickly through the glossary and manual, read the cards, and
then re-read the manual; to get into the details, reading a window together with its associated tools can help to
understand the perspectives it offers, and to think about whether additional tools are needed for your study.
The social organization of innovation provides more detailed explanations of the concepts and their theoretical
underpinning, as well as the development of the methodology and greater detail on many of the case studies
mentioned. In the book, look to Chapter 9 for a quick overview of the concepts, and Chapters 1 and 6 for more
detailed theoretical background. Chapters 2 and 3 in particular include case study material, but also see
Chapters 4, 5 and 8.
1 Changing approaches to knowledge and

information
Extension systems are under pressure. Those involved – the individuals (such as a farmer or extension agent),
organizations (a farmers' union or fertilizer company) or institutions (an extension service or research institute)
– are becoming more diverse. Current economic trends including liberalization and privatization are stimulating
the development of new industries and the participation of new actors, such as NGOs and private firms, in rural
development. All of this means that the roles played by extension and research in agricultural development are
changing very fast. We are not always able to gauge exactly how these changes take place, nor to know how we
should assess the effectiveness of our responses. As a consequence, all whose interests are involved (including
extension and research services, but certainly not limited to these organizations) need to review the scope of
their activities, reconsider their policies, and build new partnerships. For example, to ensure both sustainability
and food security, extension and research will increasingly need to rely on locally available human and natural
resources. As discussed below, they will need to create effective linkages with others. Moreover, in designing
new strategies and approaches, it will be necessary to thoroughly understand past experience, to judge this in
the light of new requirements and developments, and to use this understanding as a foundation for the new
design.
During the past two decades, many efforts have been directed towards improving the impact of extension and

rural development programmes. The training and visit system (T&V) supported by the World Bank has been
used extensively in many countries. In recent years, the T&V system has been critically assessed by national
extension institutions (as well as by the World Bank itself), and meanwhile other extension approaches have
been developed by governmental and non-governmental organizations (NGOs). It has become apparent that a
uniform extension system, ‘fed’ regularly with pre-packaged technical information, cannot respond to the
demands of richly differentiated target groups of women and men farmers operating under highly variable
agroecological and market conditions.
RAAKS, the methodology described in this manual, grew out of attempts to evolve a new vision of agricultural
extension. However, the rapidly changing environment and need for innovation described above characterizes
many fields today. Thus RAAKS has also begun to be used in other areas.
Innovativeness as a social competence
The focus in this manual is on ‘innovation’, broadly defined. While many of the examples are related to
agricultural extension, they are put in a different perspective. Farmers and others are seen as actively seeking
relationships that will allow them to learn and to make changes in their practices. This is what we call
networking. Such networking may result in either developing new methods and materials themselves, or
adapting ideas, practices and things developed by others. Taking this viewpoint shifts the focus of extension
from ‘transferring technology to a passive farmer’ to a new concept: extension as facilitating networking for
innovation. The key to innovation – including the dissemination and use of innovations originated by others –
lies in the quality of the interaction among farmers, businesses, donors and governments. That is, widespread
innovation is not brought about by one or two stakeholders in agricultural development. Instead, there are
multiple stakeholders (all whose interests are involved), who are mutually interdependent. Each stakeholder
may have his or her particular view of what agricultural development should look like, but none of them could
‘go it alone’. In most cases, before innovation can actually become widespread, the current practices of many
actors must change. The introduction of chemical fertilizer was one such instance (see example).
As a result, innovativeness cannot be seen as an individual competence, or even as the sum of a number of
individual competencies. Instead, it must be seen as a social competence – something shared among all those
individuals, institutions and organizations that have a stake in rural development. In other words, we see
innovation as a social process, rather than simply discovery and dissemination of new technical possibilities by
individuals. The challenge is to learn to create the conditions needed for such innovation to occur – conditions
that enable people not only to develop new ideas, but also to learn and make use of each other's ideas.
Transformation of many practices – not just farming – was required to introduce chemical

fertilizers. Regardless of one's attitude to their use, the worldwide introduction of chemical
fertilizers in agriculture over the last five decades has clearly illustrated this point: successfully
introducing change requires a considerable transformation in a large variety of practices. This
went far beyond farmers' learning how to buy and spread handfuls of chemicals. In the end,
achieving widespread use of fertilizers depended upon enormous shifts in agricultural,
commercial, industrial, financial and political practices. Infrastructural and marketing
arrangements were needed as well, to provide reliable and timely deliveries of fertilizer. Where
the volume available was limited and/or when subsidies were involved, reaching this point may
have called for dramatic shifts in practices related to deliveries. Further, the market had to adapt
to channel the increased production volume. In general, seasonal credits were needed to allow the
use of fertilizer by smaller producers. Banking practices thus had to change, for example by
introducing special interest rates. Since fertilizer use had to correspond to local conditions,
technical recommendations also had to be adapted. Consequently, national and international
research programmes had to be financed, or even established, to produce such recommendations.
Sometimes farmers had to learn to adjust long-held beliefs that had guided them through difficult
times for many years, such as the ‘law of the limited good’, which suggests that structural
increases in production by one farmer will cause suffering and decrease the production of others.
To introduce fertilizer use into farming practice, enormous investments in extension, as well as
more efficient extension practices, were required. Moreover, politicians had to accommodate to
the fact that, as a result of introducing fertilizer use, a large part of the country's foreign exchange
was to be dedicated to buying this input on the international market every year, even though the
expected agricultural products were not, or not yet, intended for export. This list of
transformations could be extended almost indefinitely. Apparently the availability of technical
innovations at farm level is a necessary, but not sufficient, condition for widespread innovation
processes to occur (Engel, 1996).
Understanding the social organization of innovation
If we accept that innovation is the outcome of social interaction among many stakeholders – who are
interdependent, and yet pursue their own strategic objectives – it becomes clear that it is not a straightforward,
technical process. Rather, it is a diffuse, social process, involving both individual and collective searches for
ideas, information and options for decision making. The social organization of innovation may then be
characterized as the way in which actors organize themselves to carry out this search.
But what do actors (individuals, groups, organizations or networks) actually do in searching for ideas and
information to change their practices? They experiment of course, but how do they obtain new ideas, new
options? What they do in practice is what we call networking. To gain access to a range of options and insights,
actors actively engage in building and managing interactive relationships with others – people they consider
relevant to their concerns. In the book accompanying this manual (The social organization of innovation) this
networking is characterized by, first, the creation of joint learning opportunities; second, mutual probing of
relevant ideas, options and contexts; and finally, by a pooling of energies and often other resources, to

implement particular innovation strategies.
How do the great variety of actors involved ‘get their act together’? How do they get organized? Many things
they do are not planned or even premeditated. This is seen in everyday experience: we all know how important
informal contacts, happy coincidences and unexpected events can be in any kind of interaction. On the other
hand, social organization intended to lead to agricultural innovation may be planned and premeditated in the
extreme, as for example with the T&V system for extension. In looking at a knowledge and information system,
the challenge is therefore to take both informal and formal elements into account when attempting to describe
the process of organization for innovation. On the basis of extensive case study evidence, The social
organization of innovation suggests this can be done by looking at:
● relevant actors and their practices; and

● the patterns of social organization, or ‘social forms’, that emerge as a result of networking among these
actors.
The relevance of an actor (researcher or research institute, extension staff, farmer, NGO, trader, and so forth) or
a practice (research, extension, farming, trade, etc.), however, is in the eyes of the beholder. Actors build
networks with those they consider relevant to their cause: for example, with respect to innovation, those they
think for one reason or other may help them to learn something new and useful. Therefore, to select relevant
actors and practices for study, our first question is who the various stakeholders themselves consider relevant to
achieving what is, in their view, a desirable agricultural development. From a relative outsiders' point of view,
we can also ask who we, as students of innovation, think might be relevant, and why.
After a preliminary assessment of the relative relevance of the actors involved, we can address the question of
the degree and kind of social organization that exists (or could exist) among them. The social organization of
innovation suggests that networking tends to lead to the emergence of three social forms: convergences,
resource coalitions and communication networks. These merit particular attention when studying innovation;
they represent macrostructures, with dynamics of their own. ‘Macrostructures are in part the result of the
unintended consequences of numerous social acts and interactions, which … become the enabling and
constraining conditions of social action…’ (Long, 1989: 229).
Convergences
Convergences emerge as actors narrow down the scope of their searches, along with the range of issues and
alternatives they consider relevant to innovation. That is, as some actors make similar choices, coming to define
problems and solutions in similar ways, a ‘school of thought’ may emerge. Even so, as mentioned above, full
consensus among all stakeholders involved in any type of development is very rare. In practical situations, a
number of schools of thought normally coexist, such as ‘low-external-input’ versus ‘highexternal- input’
thinkers with respect to agriculture. When they hold very different views of what agriculture and agricultural

development are and should be, actors from different convergences may find it difficult to cooperate or even
communicate effectively. However, strong convergence may also be a disadvantage: when actors agree on
many issues, it becomes easier to ignore useful information that points in another direction.
Resource coalitions
When actors decide to pool their resources for a joint performance, a resource coalition emerges. Leading actors
may enrol others in their ‘projects’. As a result, effective leadership patterns may develop – but constraints to
coordination and widespread searches for information can also occur. More than one coalition may arise, or
cooperation in some but not all areas. Classical examples of the latter can be found in countries where donor
agencies compete with each other for the use of local staff, and separately approach target groups in the same
area with similar programmes. Studying coalitions will require an analysis of the instruments of power and
influence used to forge alliances and/or to impose one's approach upon others.
Communication networks
These emerge as a direct consequence of actors' decisions to make use of available information (newspapers,
journals, extension information and so forth) or to exchange ideas, experiences, knowledge and information
among themselves. As a result, particular communication practices are adopted (such as meetings, newsletters,
workshops, journals, conferences or courses). A diverse range of communication media may be used, ranging
from interpersonal contacts to mass media, and from formal to informal channels. In studying innovation, the
role of informal channels of communication must not be underestimated, as has often been the case in
traditional studies in this area. Communications networks among farmers – especially women farmers – are
often very informal, yet quite effective.
Innovation configurations
Actors in agricultural innovation have long recognized the need to divide tasks among themselves. As a result,
various types of organizations have been set up: advisory and extension agencies, research institutions,
policymaking bureaus, research and development units, auctions, agricultural information bureaus, liaison units,
farmer organizations and the like. Over time, networking or the lack of it among these units may lead to a
pattern of more or less durable relationships.

We call the particular combinations of convergences, communication networks and resource coalitions that
result from networking ‘innovation configurations’. Innovation configurations encompass the views, procedures
and ground rules for collective behaviour with respect to innovation (or perhaps a particular type of innovation)
that are accepted within a network. Success in innovation requires a degree of strategic consensus. If a network
is to succeed, its participants must be aligned to a considerable degree; this is reflected in the configuration – the
convergences, resource coalitions and communication networks – that we observe. At the roots of such
consensus lies a common concern and strategy shared among a number of relevant and powerful social and/or
institutional actors.
Generally, a lot of casual, unintentional problem solving (as well as ‘divergent’ institutional behaviour – actions
not obviously related to the apparent objectives) remains, even within a strong and successful configuration. In
fact, many authors argue that ‘opportunity grasping’ and divergent behaviour may be important elements in
explaining successful innovation. Also, some actors within a configuration may not want to express their
agreement with others explicitly. For example, a Colombian-Netherlands agricultural development project had
to navigate carefully so as not to be seen as ‘fully absorbed’ into the World Bank sponsored national rural
development programme. Its own identity had to be stated clearly and repeatedly for the other parties involved,
to ensure the continuation of independent funding (Engel, 1996).
The AKIS perspective
No one group, whether farmers, scientists or technicians, is alone responsible for innovation in agriculture.
Rather, the interdependent activities of a whole set of actors – including farmers, school teachers, traders,
producers of inputs and services, processing industries, policymakers, planners, researchers, technicians and
extension services – stimulate or, alternatively, frustrate innovation processes. Looked at in this way, innovation
can be seen as the outcome of a mutual learning process taking place among a large number of autonomous
actors.
The agricultural knowledge and information systems (AKIS) perspective (Röling and Engel, 1992; for a
comprehensive introduction see Chapter 1 in The social organization of innovation) provides a way of looking
at the social organization of innovation. A pivotal idea in the AKIS perspective is its recognition of the level of
mutual interdependence among the actors on the agricultural development scene. This idea can be made visible
by drawing a picture representing actors and their practices. Begin with the agricultural practices of the farmer,
and then add other actors whose practices are or might be linked to this in the situation that concerns you.
Research, extension and commercial practices are apt to be there, of course, but also policymaking and
industrial practices, and educational practices may be related to all the rest. Figure 1 is an example of an attempt
to picture all of these practices. In reality, each picture represents a variety of actors. For example, the farmer
stands for individual women and men who work their land, but also for farm families, farmer organizations and
farm labourers. The arrows suggest what building a knowledge system – a collective competence for innovation
– can mean. Each of the separate figures shows a person or organization that is learning; however, we are often
absorbed in the practices and priorities of our own tasks or discipline. When the arrows appear, learning and

practices become linked, so that we can begin to understand and learn from each other; when practices become
interconnected, it becomes possible to make decisions on a broader, more realistic basis. Thus, the more arrows
the more tightly the actors are joined together in a learning system.
Figure 1
Agricultural knowledge and information systems

In fact, we can define any knowledge and information system (KIS) as a linked set of actors – individuals,
organizations or institutions – that emerges (or is strengthened) as a result of networking for innovation. Such a
set of actors (or network) can be expected to work synergistically to effectively support learning for innovation;
indeed, a network can be managed – in a sense – to achieve this aim (Engel and van den Bor, 1995; see also
Windows C1, C2 and C3, and the related tools).
Using an AKIS perspective can help people like extension staff and researchers to focus on those actors within
an agricultural system whose knowledge and information contribute the most to agricultural innovation. The
AKIS perspective assumes that the more effective and efficient networking is among the stakeholders in
agricultural development, the better the chances for innovation. However, to look at networking, and to take
into account the many types of actors who may be relevant to the network, it is necessary to look beyond the
boundaries of conventional extension. The AKIS perspective can help to carry out such a comprehensive
analysis, and to provide a practical contribution to knowledge management and policy (Röling and Engel,
1992). Research carried out over the past few years indicates that the performance of the system involved in an
AKIS depends heavily on:
● cooperation among the various actors involved

● effective internal and external communications
● transparency and agreement among different actors with respect to interests and objectives
● the degree to which tasks are divided and coordinated within the system, so that relevant knowledge
networks are activated and people acquire a shared sense of direction.
Key concepts in knowledge and information systems
When we use the AKIS perspective, a number of concepts play a pivotal role in the analysis. In the first place,
systems are seen as constructs. That is, they are products of our imagination, but we can use them to understand
the world and to act in it. A knowledge and information ‘system’ does not exist: it is just a way of thinking that
helps us understand the social organization of innovation in agriculture better. What does exist is a large variety
of people (as seen in the example on the greenhouse sector) who are stakeholders in agricultural development,
such as farmers, policymakers, traders, researchers, extension staff, teachers or people who sell inputs.
The Dutch greenhouse sector, seen as a knowledge and information system, consists of a variety
of different actors: extension agents, managers of various extension agencies, private consultants,
commercial firms, agricultural schools, growers' associations, rural women's groups, banks and
research stations.
All of these actors manage, generate, transform, transmit, store, retrieve, integrate, diffuse and use knowledge
and information within a specific area of human activity: horticulture. They are all stakeholders in agricultural

development – therefore we consider them part of an agricultural knowledge and information system. The
knowledge systems perspective can also be applied to other areas, such as health care, traffic and transport,
education, and so forth.
The boundary of an AKIS – the definition of the problem and the line between actors who are seen as part of
the system and those who are not – is established arbitrarily. In using the RAAKS methodology, it is the task of
the team, together with some or all of the stakeholders, to determine which actors are relevant to innovation in a
particular situation (this decision is then periodically re-evaluated). Boundaries may be based on geographical
or physical aspects, or defined on the basis of a problem. Thus examples of boundaries include actors ‘in the
horticultural sector’, or involved with issues related to a particular watershed, or problems of declining soil
fertility. Some actors may be quite relevant even though this is not obvious at first sight; therefore, careful
thought is needed before deciding to exclude groups or individuals. Diversity is an important breeding ground
for innovation, so a team should not shy away from including new or hitherto unrecognized actors. For
example, a failure to seriously consider environmental activists as actors in the Dutch agricultural knowledge
system has cost the sector dearly. However, practical considerations are always involved. Criteria to use in
establishing the boundaries of an AKIS may be derived from the purpose of the diagnosis, the time available,
and the relative importance of the actors vis-à-vis the type of agricultural development that is intended.
Linkages among actors, and the related linkage mechanisms, are a quite significant part of a knowledge and
information system: they show how actors communicate and work together. Linkages enable actors to exchange
resources such as information, money, labour and other materials; or immaterial assets, such as power, status
and ‘goodwill’. Linkage mechanisms are arrangements that facilitate communication (such as meetings among
farmers or with extension staff, or liaison offices), coordination (e.g. mutual adjustment of activities, or water
distribution) or resource transfers (perhaps credit, salary payments, or sharing labour). Some linkages are of a
more formal character, such as administrative links between a project and its donor(s). Others are more
informal, as the links that often exist between extension staff and researchers at a personal level.
Within an AKIS, actors often perform specialized tasks: policymakers formulate agricultural policies and plans,
researchers do research, teachers educate, farmers farm and donors finance programmes. On the other hand, it is
important not to overemphasize specialization. Conflict may arise with respect to agricultural development, for
example, when researchers define ‘research’ as limited to analysis based on formal, academic procedures. If
they then see themselves as the only ones qualified to carry out such work, it becomes easy to forget that
farmers have been innovators for centuries, based on their own on-farm experimentation. Acknowledging
farmers' expertise, involving them in setting the research agenda and/or as partners in research, can lead to
additional forms of learning and innovation. This is true overall: different actors have access to different ideas,
knowledge, information and experiences. As a result they learn different things, but also they are apt to learn in
different ways. This makes some form of coordination essential.
The importance of linkages and coordination does not mean there are no conflicts among system actors. On the
contrary, consensus is the exception. More often, because actors may perceive agricultural development in
many different ways, and because they each play their own roles in this development, consensus is hard to
achieve. Nevertheless, for the smooth ‘performance’ of an AKIS, it is necessary to build a degree of strategic
consensus among at least the main actors who are involved. External factors (and actors) are influential as well.

Unstable market prices for their products may cause actors to become reluctant to invest in innovation. Or low
interest rates or favourable policies may stimulate more systematic investment in the development of knowledge
and know-how. Moreover, access (or lack of it) to international donors can play an important role. Influential
external actors or combinations of actors may create specific types of problems, constraints and opportunities
for the performance of the system (see ‘Basic configurations’ – Tool B6, and Chapter 4 in The social
organization of innovation.)
Key concepts are summed up in the Glossary (Appendix 4) for quick reference as you work through the
RAAKS materials.
RAAKS: Participatory action research
RAAKS is a methodology that has been designed and tested to help stakeholders gain a better understanding of
their performance as innovators. The acronym stands for rapid (or relaxed) appraisal of agricultural knowledge
systems. RAAKS provides a way to improve the generation, exchange and utilization of knowledge and
information for innovation. Men or women villagers, researchers, policymakers, extension workers, consumers,
producers of inputs or services, industrialists and/or traders, guided by a team of specialists, can all be involved
in a RAAKS study.
Central elements of RAAKS are team work, focused collection of information, qualitative analysis, and
strategic decision making. RAAKS uses a variety of windows, as mentioned in the Introduction, to achieve a
fundamental analysis, a transparent problem definition and recommendations for action. The important issues
addressed in RAAKS include forms of cooperation between actors, actors' objectives and their conflicting
and/or shared interests, integration and coordination of activities, relevant knowledge and information networks,
and the division of tasks (research, experimentation, training, farming and so forth).
Although RAAKS has been developed for use in agricultural and rural development situations, it has been
applied to non-rural problem situations as well, such as the management of services for the elderly and the
introduction of solar energy in the Netherlands. In what the West tends to call ‘developing countries’ its
application has generally been directed at bottom-up policy formulation and planning, and at the organization
and evaluation of agricultural research and extension. In this manual and the accompanying materials, we
present examples from a variety of different situations. We have included some from the Netherlands; these
show that even though there are differences, the basic principles are the same in all countries. As a result, the
method can be widely used to describe situations and generate the recommendations needed for planning
purposes.
RAAKS, women and other stakeholders

As a participatory methodology, the use of RAAKS immediately suggests the inclusion of women and other
groups of stakeholders who are often forgotten – depending on the local situation, this might also include newer
immigrants, young people, the elderly, or landless farmers. While not every RAAKS tool includes a reminder
about these groups, they definitely need to be considered. In many knowledge and information systems it is
important not only to understand their role as actors, but also to seek ways to build them into the information
system – supplying as well as receiving information. If they have been defined as target groups, this is essential!
Many of the readings in Appendix 3 can be useful here. See in particular the sections on gender analysis and on
participatory methodologies.
For whom is RAAKS useful?
RAAKS is first of all useful to organizations or institutions who feel pressed to improve their performance with
respect to innovation. As innovation processes grow and change more and more quickly, the need to align them
with broader, more complex societal objectives becomes more urgent. This increases the pressure on
governmental, nongovernmental and private organizations – such as extension, research or development
agencies, exporters of agricultural products, agro-industries or national or international policymaking bodies –
to improve their own capacity to innovate and to make use of innovations. RAAKS provides a way to evaluate
the functioning of their knowledge and information systems, and to see ways to make improvements. Therefore
RAAKS may help such organizations to develop a more adequate response to the pressures they feel.
Second, some organizations or institutions take it upon themselves to intervene on behalf of particular
developments and hence to guide innovation in a particular direction. This is the case, for example, when a
government decides to subsidize research and development programmes, the use of clean energy sources, low-
external-input agriculture, sustainable farm practices, and so forth. Assuming such a leadership role requires an
understanding of the impact of technology upon development, but also an appreciation of the role the
organizations, institutions and companies involved play in continuously obtaining, developing, redeveloping
and putting to use relevant knowledge, information and technologies. Such actors may find that RAAKS can
help them to develop a thorough understanding of the social and organizational issues involved in innovation, to
formulate concrete proposals for action in a participatory manner, and to assure that they understand the
positions of others before moving ahead.
Typically, managers of agricultural development programmes, extension managers, researchers or, more
generally, development professionals are the ones who implement RAAKS. They may find RAAKS useful as a
participatory approach: it is a way of facilitating inquiry into the constraints and opportunities that affect
networking, cooperation and communication for innovation; moreover, this approach can aid in formulating
strategies to overcome problems that have been identified in a specific situation. Some of the ways RAAKS can
be used are summarized in Box 1.

Box 1
Ways to use RAAKS
Field workers As a field worker, you can use this method to chart out the knowledge and information networks
in which you operate, and design ways to improve your performance.
Trainers As a trainer, you can use RAAKS to encourage trainees to take a comprehensive, critical look at
their performance as facilitators of innovation in agriculture.
Managers As a manager, you can encourage teamwork, self-monitoring and the generation of ideas on how
to improve collective performance related to innovation, with built-in feedback and followup.
Researchers As a researcher, you can use this method to develop an understanding of the social organization
of innovation, based on input from those who are involved; this can be used as a basis for
proposals for action and/or further analysis.
Consultants As a consultant, you can use RAAKS to facilitate a shared understanding among stakeholders,
to define what can be done and by whom, and to improve the way stakeholders function
together.
2 Designing RAAKS action research
Because RAAKS is a participatory approach, much thought must be given to the role of relevant stakeholders
(farmers, extension staff, researchers, policymakers, traders and so forth). The degree of participation of
stakeholders in the RAAKS study, and in the formulation of solutions and recommendations, are critical
elements. In fact, decisions in this area are probably the most important (and have the most implications) of any
that must be taken by a RAAKS team when designing the particular form in which they will use the
methodology.
This chapter should allow you to become familiar with the issues and steps involved in designing such action
research. The intentions involved in the use of RAAKS, its underlying principles and its procedural and
analytical design are covered. Again, RAAKS is not a ready-made solution; rather it offers a ‘menu’ made up of
a number of field-tested methodological elements (windows, step-wise phases, tools, and exercises). A RAAKS
team can choose among these, deciding to leave out certain ingredients or add new ones, fitting their approach
to local circumstance or to the problem situation at hand.
Appendix 3 lists further readings on related topics including research, RAAKS and participatory methodologies.

What is RAAKS?
RAAKS is based upon what has been called a ‘soft systems’ methodology (Checkland and Scholes, 1990). If
innovative performance is to improve, the various stakeholders themselves have to decide to do something
about it. Therefore, at the core of RAAKS lie the appreciations – the perceptions, preoccupations, judgements
and understanding – of the principal stakeholders. That is, they (like all of us) filter incoming information and
construct a reality that makes sense to them. (See Chapters 1 and 6 in The social organization of innovation.)
The RAAKS process helps to make these appreciations explicit by encouraging stakeholders to assess and re-
assess their understanding of the problem situation and their own role in it. It also offers ways to address
specific issues in a transparent manner, using methods that have been validated and can be agreed upon in
advance by those who are taking part. This method thus makes it easier for people with diverse interests to
begin to work together. It also helps identify possible constraints and opportunities, and allows the participants
to design strategies to define measures to improve their current interaction. They can also make adjustments to
better fit trends and changes in their environment – for instance, increasingly stringent regulations concerning
the preservation of natural resources, changing developmental objectives, or new developments in the
international market. Directly involving those additional actors responsible for policy formulation and
implementation helps to move beyond the identification of opportunities, constraints and clues for action, and
toward finding a point of departure for future cooperation and change.
RAAKS and other participatory methods
We see RAAKS as part of an emerging family of ‘alternative systems of inquiry’. All of these have several
points in common (Pretty, 1994; see Chapter 8 in The social organization of innovation):
● a defined methodology and built-in learning process,

● the use of multiple perspectives,
● an insistence upon group inquiry,
● context-specific methodological design,
● facilitation of participation by both experts and other stakeholders, and
● a focus on designing and implementing sustained action.
Well-known examples are PRA (participatory rapid appraisal – Chambers, 1992) and PTD (participatory
technology development – Jiggins and de Zeeuw, 1992). RAAKS is distinguished by its focus on the social
organization of innovation: the way actors (individuals and organizations) build and maintain relationships with
each other to foster innovation. The ways actors organize themselves to learn, how they network, cooperate and
communicate for innovation, what hampers their capacity to learn and what helps them to learn new practices

faster, are central research questions. PRA focuses more on analysing local farming and livelihood systems and
general conditions enabling and/or constraining their development; PTD helps to create a process of creative
interaction between local community members and outside facilitators, to experiment with and develop
technologies for improving the local agroecological system and to increase the capacity of the local community
to sustain the technology development process. All three methods use techniques (visualization, for example)
that are especially useful in working participatively with people who cannot read; these techniques also tend to
stimulate action. Thus RAAKS complements both PRA and PTD. In practice, RAAKS teams often combine
techniques from these and/or other participatory approaches. (See also Exercise 5 in Appendix 2.)
Objectives
Generally, a RAAKS study has three objectives, as shown in Box 2. The second objective is particularly
important. Careful preparations are needed to guarantee close collaboration. The RAAKS team may include
representatives of actors familiar with the situation as well as members familiar with the RAAKS methodology.
As noted earlier in this chapter, decisions about stakeholder participation in a RAAKS study are vital ones.
Relationships with stakeholders (individuals and representatives) and their participation must be carefully
prepared and managed. A preparatory workshop may be held so that team members can get to know each other
and become familiar with the methodology. (See also the section on ‘Preparation’ in Chapter 3, and the
exercises and workshop planning materials in Appendixes 1 and 2.)
Box 2
Objectives in using RAAKS
● To identify opportunities to improve a knowledge and information system – that is, to better the
organization, decision making and exchange of information among actors, with the aim of improving
the potential for learning and innovation.
● To create awareness among relevant actors (such as target groups or constituencies, managers,
policymakers, producers, traders, researchers and extension workers) with respect to the opportunities
and constraints that affect their performance as innovators.
● To identify actors and potential actors who do or could act effectively to remove constraints and take
advantage of opportunities to improve innovative performance and to encourage their commitment to
such changes.
Designing a RAAKS study

RAAKS action research may be characterized along three dimensions. The first is concerned with intentions,
with what one hopes to achieve when using this method. Why is the study thought to be necessary? And what
implications does this have for the design and implementation of the study? For example, the problem may
seem to be that farmers do not participate actively enough in extension programmes; the intention is therefore to
suggest ways to increase their participation. Or the problem seems to be of a more institutional sort, such as a
lack of coordination between research and extension; this could make improved cooperation desirable, to avoid
duplication of effort and a waste of government funds. Questioning the initial wording of the statement of the
problem is a part of the process, which helps to refine the statement and make it more useful. Nevertheless, the
initial intentions of the various actors provide a starting point.
Analytical concerns come next. What issues should be addressed to clarify the problem, identify potential
solutions, and create commitment among actors to implement solutions? What aspects of the knowledge system
are we to study? How can we reach an understanding of a complex situation without ‘running around in circles’
all the time? Answering these and similar questions is not easy. The choices made must be relevant to the
intentions, and practical within the scope of the resources allocated to the team. It will not be possible to study
everything; the team will have to limit itself to the most relevant issues and variables within the knowledge
system. However, the ‘windows’ provided by the RAAKS method itself are helpful, as will be discussed below.
Also, RAAKS allows for making changes and focusing in more closely in the course of the study, as the
problem situation becomes clearer.
Third, an agreed procedure is needed. What will the team do first, what next? How intensively will each of the
stakeholders be encouraged to take part in the study? How many individual or group interviews will be done,
and during which phases of RAAKS? And how many workshops will there be where stakeholders will meet?
The answers to such questions are related to the choices made about intentions. If more participation from
farmers in extension is desired, farmers should take a very active part in the RAAKS process so that their views
and knowledge are adequately represented in discussions and in decision making. If, on the other hand, the
relationship or lack of a relationship between researchers and extension managers is seen as the problem, our
first concern might be with getting those parties actively involved.
Designing a RAAKS study requires taking decisions related to all three of these dimensions: intentions must be
clarified, a design for the analysis made, and procedures agreed upon. Together these make up the design for the
RAAKS study. While in reality these three dimensions are not entirely separate, in the following paragraphs we
elaborate on them and present a number of related issues that you may want to consider when designing your
own RAAKS study.
Intentions and underlying principles

RAAKS directs attention first to helping actors to study the ways they have organized themselves for
innovation, rather than immediately focusing on specific solutions. Strategic diagnosis – an appraisal of
constraints and opportunities, leading to a joint definition of useful strategies for improvement – is emphasized.
A very important characteristic of the approach is that men and women farmers, other rural people, consumers
or other target groups are considered stakeholders and sources of relevant knowledge and information. As
suggested by the objectives listed above, RAAKS aims at three types of outcomes: a more comprehensive
understanding of the social organization of innovation in a particular situation, suggestions for improvement,
and interest on the part of some or all of the relevant actors in implementing these. To produce these outcomes,
three different yet intertwined learning processes (see Figure 4, Chapter 8, in The social organization of
innovation) are combined:
● joint inquiry, involving both the team and other stakeholders in exploring a shared concern with respect
to innovation-related performance;
● contrasting results obtained by using different ‘windows’; this creates a useful tension among different,
but equally relevant and valid, interpretations of the same situation;
● a task-oriented path that leads participants from analysis and interpretation of the problematical situation
to designing and committing themselves to potentially useful actions.
Joint inquiry into the social organization of innovation
RAAKS focuses on the performance of actors, as those who are responsible for making innovations in their own
practices. RAAKS helps actors to work together to identify relevant networks – and appreciate the role and
relevance of these networks with respect to effective innovation in their own situations. It helps to draw
attention to relevant issues and shared concerns within the knowledge system, offering instruments (windows
and tools) for gathering, organizing and interpreting information on relevant ideas and events as well as
exercises that provide guidelines and techniques to support interactive learning. Using different windows and
tools and contrasting the results makes a profound analysis possible, while the use of exercises helps team
members to design their own learning path and implement the analysis in a participatory manner.
Task-oriented learning: organizing for effective innovation
RAAKS revolves around understanding and diagnosing the problem at hand. This makes it possible to reach the
point of suggesting measures to improve the social organization of innovation. That is, studying the social
organization of innovation is not enough: something has to be done about it. This lies behind the design of
RAAKS as a participatory action-research methodology. The interaction between stakeholders and team
members has the explicit purpose of arriving at suggestions for improvement. Establishing an agreed procedure
among team members helps to assure that such suggestions will have been thoroughly discussed beforehand.
The task-oriented design of RAAKS is aimed at creating consensus, whether general or partial, among relevant
actors. This increases the potential for taking decisions, as well as the potential for commitment to
implementation of recommendations. This line of inquiry is necessarily more conclusive than the other two.

Therefore, RAAKS offers specific windows (A5, B8 and C3; windows are discussed in more detail in the last
section of this chapter) to help achieve a synthesis of the problem situation and of the existing social
organization of innovation.
It is important to note that in many development situations it takes time for actors to come to understand each
other and the mutuality of their interests, work together, agree, arrive at sound plans, consult others within their
several organizations and commit themselves to action. ‘Taking time,’ however, generally pays off nicely in the
end!
Design and analysis using multiple perspectives: a choice of windows and tools
In reading some of the questions listed earlier, like ‘what issues should be addressed’ or ‘how can we reach an
understanding of a complex situation without running around in circles’ you may have wondered how teams can
come up with answers. The windows and tools included on the laminated cards in this resource box can provide
part of the answer; building a good team – and hard work! – will do the rest. The social organization of
innovation can be studied in many ways. Looking at a problem from different angles, or perspectives, can help.
The windows are meant to ‘open up’ these possibilities, guiding a RAAKS team to seek information effectively
and make sense of their observations. Figure 2 gives an overview of RAAKS windows. Each window draws
attention to particular issues and relationships, and all have been developed and tested in field research, by
others and/or ourselves. We have chosen and designed or redesigned them in the form presented here to fit an
action-oriented methodology. As has been mentioned earlier, the windows, tools and exercises offered here are
by no means the only ones possible. On the contrary, since this is a participatory action-research methodology,
the team serve not only as researchers, but also as facilitators. They encourage actors to participate in producing
modified or additional windows and designing new tools. However, extensive field research suggests these
particular windows do provide a RAAKS team with a useful starting point.
Figure 2
Juggling with perspectives

A window helps the RAAKS team focus attention on particular issues that have proved to be relevant in other
studies. Using a window brings particular ideas, actors, activities and relationships forward. Each window
therefore provides a different view of the situation we are studying. This does not mean that all windows are
mutually exclusive; overlap has been carefully built in. Within a RAAKS study, it is not necessarily inefficient
to arrive at similar insights via different routes. For example, the barriers that generally prevent the sharing of
knowledge between peasant farmers and industrial producers will become apparent whether we observe their
linkages, the types of knowledge they exchange, or the way they communicate. However, each of these angles
may tell us something different about why such a situation has developed and how we might be able to do
something about it. Contrasting the results of various windows often provides additional important information.
When the intention for a RAAKS study is a comprehensive analysis of a situation, a new team should use all of
the windows in the first and last phases (A and C, described below). During the second phase, the actual study
of the social organization of innovation, a selection can be made. The choice of windows and tools depends on
the situation, the capacities of the team and the issues and problems identified in the first phase. The windows
may be used as they are, or teams may decide to adapt and/or extend them to address particularly relevant issues

in their situation.
Even though field testing has helped to assure that all windows can be considered valid and applicable, this
alone cannot guarantee the quality of RAAKS action research. The windows help, but they cannot replace
individual and group skills. Applying RAAKS and its windows requires analytical as well as communications
skills on the part of both team and stakeholders. As noted, the appendixes give some suggestions on how a team
can build such skills. However, the use of these windows also requires creativity. None of the windows provides
an unequivocal recipe for analysis. On the contrary, each must be studied, discussed and, if necessary, re-
designed before it can be used in the field. Moreover, specific plans must be made on how to collect,
consolidate and interpret relevant information, and how to make cross references between different windows.
To provide a quick reference when choosing the windows to be used in a RAAKS study, all windows are listed
in Table 1, where they are related to the main issues discussed in the first chapter (see the section on
‘Understanding the social organization of innovation’). In reading the table, look for example at Window A3.
This window explores – analyses – the degree of convergence (or divergence) among stakeholders with respect
to innovation. Therefore an ‘A’ is seen in the ‘convergence’ column. Window B3 on the other hand addresses
the analysis of networking practices with emphasis on thematic convergences, communication networks and
innovation configurations; Window B8 facilitates the synthesis of relevant issues of social organization and the
analysis of relevant action proposals; and so on. (Detailed RAAKS windows are included as one section among
the laminated cards in the binder.)
Table 1
Issues of social organization for innovation addressed by each RAAKS windowa

a Each letter in the table relates to one of these activities:

A = analysis/appraisal
S = synthesis
D = design/choice
Each of the windows is equipped with one or more tools. Windows can be seen as opening up a vista, a way of
looking at a situation, while tools supply practical ways to begin gathering and organizing the relevant
information. Tools provide a means for the actors involved in RAAKS to do a ‘quick scan’ from various angles.
A tool helps the team to gather information systematically and to process it. In a way, a RAAKS tool represents
a straightforward way to address all or part of a particular window. However, because it involves a choice with
respect to which questions will be addressed and how, a tool is generally more limited than the corresponding
window.
Studies of the social organization of innovation have produced a number of specific tools: we have collected 23
of them here. Each of these tools has proved itself in various RAAKS studies. They are not the only useful
possibilities, but they provide a good start for a RAAKS team that is designing a study. However, we expect

teams to critically assess the strengths and limitations of each tool in view of their own purposes, and to adapt
or even replace them before going into the field. In practice, this means that RAAKS teams often start by using
the tools provided; but after they become more familiar with the approach and methodology, they start adapting
the tools or developing some of their own, adjusting them more closely to the situation they are studying.
(RAAKS tools are included as one section among the laminated cards in the binder. Tools bear the number of
the related window. If a tool is numbered, for example, ‘A5/B6’ it applies to both windows – A5 and B6.)
Even though you may not use all of the windows and tools at once, it is a good idea to read through all of them
in combination with these chapters of the manual. You might for example take a quick look now, a closer look
after reading the following material on the three phases of RAAKS action research, and then read and consider
them in more detail after you have completed Chapters 1–3. This may also be a good point for a quick review of
the Glossary.
A step-wise design for procedures
After a preparatory phase – team building, getting acquainted with the methodology and building relationships
with relevant stakeholders – a RAAKS study consists of three phases:
● Phase A: defining the relevant system and its problems;

● Phase B: analysing constraints and opportunities;
● Phase C: articulating policy and strategy/planning for action.
Specific research objectives are pursued during each phase; these correspond roughly to the operational
objectives of RAAKS. To reach these objectives, the windows are used to help in gathering, organizing and
interpreting information and in the presentation of results. As mentioned above, in Phase B the combination of
windows used may be different for each RAAKS study. Teams can decide to modify windows or construct their
own, to adapt the analysis to the specific issues and the situation at hand. At the end of each phase, a workshop
is organized with stakeholders and/or their representatives. The purpose of the workshops is to feed information
back to the stakeholders. Both the process of producing the reports and the resulting discussions enable people
to debate, complement and/or amend the research findings and to participate actively in the learning process.
Field reports are made as a way of sharing and discuss ideas and information among team members; discussion
reports are made to share ideas with stakeholders during the workshops. These are not intended to be lengthy
descriptive reports – instead, for example, use can be made of drawings, tables and brief statements.
(Workshops and reports are discussed further in Chapter 3.)
The phases cannot be strictly separated. Sometimes elements are brought forward, and considered in an earlier
phase; sometimes an analysis from an earlier phase is repeated in greater detail in the next. The phases are
meant to be taken as a point of departure, not as a blueprint. Nevertheless, we will briefly sketch the general
contents of each of the three phases below. They have some factors in common, in that the following tasks need
to be planned and implemented during each one:

● debate, specify and agree upon the research objectives;

● debate and choose relevant windows, further operationalize them to make them fit the situation (that is,
state specifically what aspects are most relevant, what information will be needed and how to obtain it)
and/or add new windows;
● use these windows to gather relevant information through individual or group interviews and from
secondary sources;
● produce field reports;
● use group discussions to integrate ideas, issues and information;
● produce a discussion report that reflects important issues, information gaps and tentative conclusions.
Figure 2, above, is a graphic reminder that teams make a choice of windows to be used; it also emphasizes that
RAAKS is not a process that takes place in a straight line. Instead, it is interactive, bouncing forward and
backward. It evolves as teams work to achieve their goals: teams choose the sequence in which they will use the
tools. Often they are used in 1, 2, 3 order, but this is not required.
Phase A: Defining the relevant system and its problems
The broad objective of the first phase of the diagnosis is to identify opportunities to improve a knowledge and
information system. Therefore, one of the tasks in this phase is to establish the boundaries of the system that is
to be studied – including a definition of the problem, and the stakeholders to be included in the study.
Secondary information, brainstorming and field data can all be used to define and further specify the problem.
The fact that different stakeholders will have different ideas about the problem definition is a key element in a
RAAKS study. The team's picture of the underlying problem may change over time. Various windows can be
used to make a preliminary inventory of relevant actors, to study the problem definitions embraced by different
actors at the start of the process, to review different actors' views about what is desirable, or to analyse
environmental constraints. At the end of this phase, the result is a first approximation to a picture of the
knowledge system and the ways relevant actors currently act and interact in their search for ideas and
information that can lead to innovation in their practices. The discussion on the use of bio-ethanol as transport
fuel provides an example. Laminated cards for the windows in this phase are numbered A1 to A5. When a
comprehensive analysis is desired, it is recommended that all of these windows be used. They provide a means
for the team to identify the elements in the knowledge and information system that are relevant to coping with
their problem situation.
Facilitators often work through Phase A in the form of a desk study, to get a quick overview of the knowledge
system and make a first appraisal of the problems at hand. They read relevant documents and literature and talk
with a resource person who knows the study area or topic. The results from the desk study are then used to
formulate a proposal for involving a team in RAAKS action research.
Phase A results – reopening the discussion on the use of bio-ethanol as transport fuel: In the
Netherlands, a complex discussion on the advantages and disadvantages of the use of bio-ethanol

has stagnated and become polarized. The Scientific Council for Government Policy produced a
balanced report on the situation, recommending an experimental project on a practical scale. Then
nothing happened for some time. One of the industrial stakeholders decided to ask Wageningen
Agricultural University to apply the RAAKS methodology, to try to reopen the discussion and
develop ideas for specific followup activities. Rather than addressing the national level debate, a
choice was made to focus upon the north of the country, where the problem of developing
alternative farming activities is most severely felt. The research team included an actor of
regional importance: the Van Hall Institute for higher agricultural and environmental education,
in Groningen. The team focused on the discussion process. In Phase A, key actors were
interviewed and hot topics, communication practices and relationships were identified. Over 20
stakeholders participated in the first workshop (following Phase A), including farmers,
industrialists, environmental activists and specialists from various government departments. In
this workshop, the team proposed that the search for viable alternatives like bio-ethanol had
stagnated because of first, the lack of communication between three different sets of actors, each
focusing on different priorities; second, the fact that the discussion had been dominated by
experts, rather than social stakeholders; and third, a certain competitiveness among actors, each
trying to convince others that their own solution is best. The workshop recognized the worth of
these reasons, and suggested additional political and technical causes that also help to explain the
situation. It became apparent at this workshop that support could be mustered to initiate action
that would move towards establishing a collaborative experimental project for testing and
evaluating the production and use of bio-ethanol on a practical scale. Further action research as to
the most effective way to do this was proposed (van Weperen et al., 1994).
Phase B: Analysis of constraints and opportunities
During the second phase, team members go into the field to more systematically gather information on the
social organization of innovation. A RAAKS study generally means interviewing ‘key informants’ (people
chosen for interviews because their views are apt to reflect those of relevant actors). Windows are available to
study impact, actor characteristics and linkages, knowledge networks, task performance, and coordination and
communication among actors. Windows for Phase B (also to be found on the laminated cards) are numbered B1
to B8. The team's choice of windows depends upon the problem situation and the priorities expressed during the
first phase. Phase B results in a more detailed picture of how different networks of actors interact, the issues that
dominate their debates, and the way they coordinate their activities or fail to do so. This can be seen in the
example from the horse husbandry sector.
Phase B results – the horse husbandry sector: The first comprehensive application of RAAKS was
in 1990, and was related to the horse husbandry sector in the Netherlands. The then recently
formed Knowledge and Information Reference Centre of the Ministry of Agriculture, Nature and
Fisheries was interested in a diagnosis covering the way in which knowledge and information are
generated, exchanged and used in the horse sector. The aim was to be better able to define or
redefine the role of the Centre in facilitating effective knowledge management. A general

conclusion, recognized by all actors interviewed in Phase B, was that in fact no ‘horse husbandry
sector’ exists in the Netherlands. There are three quite separate clusters of actors: one cluster is
dedicated to horse racing, one to horse breeding for export and one to facilitating recreational
horseback riding. For obvious reasons, actors in these three clusters formulate the mission
statements for their innovation activities quite differently. Interconnectedness, sharing of
knowledge and information, and other communications are much stronger within clusters than
between them. It became clear that the recreational cluster, though it had the greatest number of
entrepreneurs and beneficiaries, had only a marginal influence on the agendas related to breeding,
research and extension; it could benefit from organizing itself better and building stronger
alliances. On the other hand, those involved in breeding had very strong linkages with research,
but could benefit from stronger ties with both extension and education. Knowledge management
within the sector was found to be weak, particularly with regard to farm management economics
and marketing. Even though a great deal of knowledge was available within the sector (held
mostly by the horse racing and export actors), other entrepreneurs had difficulty gaining access to
it. It was recommended that the Reference Centre should adopt the improvement of this situation
as a goal. Another observation was that, due to the lack of a direct relationship between research
and the various groups of clients and/or beneficiaries of the horse sector, economic support for
and feedback to research was weak (Engel et al., 1990).
Phase C: Articulating policy and strategy/planning for action
During the third phase, propositions for policy and strategy, and/or plans for action are formulated and debated.
Three windows provide support in analysing the potential of different actors to carry out specific strategies.
Here too, a comprehensive analysis can best be carried out by using all three of the Phase C windows, which
appear on cards C1 to C3. The constraints and opportunities identified in earlier phases provide a basis for
preparing realistic action proposals. The involvement of some or all stakeholders or their representatives in all
of the phases of the study helps to stimulate their commitment to putting recommendations into practice. Phase
C results are illustrated by the example on basic grain production in Central America.
Phase C results – basic grain production in Central America: The European Unionsponsored
Basic Grains Programme (PRIAG), covering six countries of Central America, is intended to
improve the relevance and impact of research and extension, especially for small grain producers.
RAAKS was used as an instrument to generate recommendations for action. Teams were selected
and trained to carry out RAAKS in selected grain growing regions of their own countries. In the
analysis, the importance of understanding the diversity of social organization for innovation was
clear. Often two entirely different subsystems were seen to exist: one whose beneficiaries are
subsistence farmers (we will call this group A), and one (group B) whose beneficiaries are
farmers who, in addition to producing for their families, also sell part of their produce. The
problems faced by these two subcategories of producers are very different, as are the ways they
acquire and share their agricultural knowledge. In fact, group A have not generally been
considered to be beneficiaries of research and extension programmes at all. As a consequence,

technical packages have never been developed to fit their needs; most often information reaches
them indirectly, through contacts with B farmers or sometimes via local traders. Traders and
group B farmers in turn receive most of their information through the representatives of private
multinational companies who sell inputs and/or services. Public institutions often play a
secondary role in providing technical recommendations to grain farmers. Particularly in category
A, farmers' basic knowledge of fundamental issues is relatively poor – for example, regarding
improved varieties and their adaptation, integrated pest management and cultivation methods.
Similarly, they are not very familiar with the practices and circumstances of researchers and
extension staff. However, category B farmers also have problems. Their adoption of improved
technologies is often partial, due to lack of credit, difficulties in access to marketing channels, and
so forth. Another feature frequently noticed by the teams, and recognized by the various
stakeholders, was the lack of coordination or even disarticulation: in attending to the information
needs of grain producers, public, private and non-governmental institutions were not sufficiently
coordinated. Recommendations stemming from the RAAKS study ranged from re-orientating
research and extension policies, to establishing documentation and information centres, to
activating or re-activating a number of inter-institutional coordinating mechanisms and
stimulating the organization of farmer study clubs. A notable feature of the studies in the six
countries was that each analysis exhibited specific characteristics related to the social situation in
the particular region and country: for example, the active involvement of elected municipal
authorities in Nicaragua, the recommendation to reinforce institutional intervention and
coordination in Honduras, and a discussion of the possibilities for agro-tourism in Costa Rica
(PRIAG MSICA Workshop II, September-October 1992).
RAAKS followup
RAAKS emphasizes developing a shared understanding and appreciation of the problem, carrying out a
diagnosis, and articulating strategies and strategic commitments. Thus RAAKS can encourage change
processes. And, while at present the methodology does not include preparation and implementation of activities
to follow up on the commitments made, it does help to establish a sound foundation. Further, some of the same
techniques can be applied. RAAKS teams are increasingly being called in for followup sessions, to stimulate
and monitor this part of the process; these experiences are being used to develop the methodology in this area.
Implementing RAAKS: examples
To give a practical example of the use of windows and tools, two cases are given here. You may want to return
to these examples again later, after reading through the cards that describe the windows and tools.

Case 1
Joint reflection on irrigation practices in Senegal: Ile à Morphile
This first case presents results from a full-fledged participatory RAAKS study in Senegal.1 It
shows the design and procedures used in this real-life study, and is particularly interesting
because of the involvement of a large number of farmers as a part of the team.
The context
Traditional floodland agriculture is dominant in Ile à Morphile. Irrigation was introduced 17 years
ago. At village level, small-scale irrigation projects were created where rice, sorghum and maize
are grown. Each plot is managed by a group of villagers. These village groups are organized as a
union, which is part of a federation. Farmer's groups are supported by technical advisors,
animators and extension workers from the project. A Mechanics Section is responsible for
maintaining the irrigation pumps. The Training Section has been charged with organizing lectures
and giving instruction.
At first, project activities focused on food security and income generation. Later, the mission
shifted towards organizing farmers and improving the position of women. In the near future,
however, the Senegalese government and the Dutch donor will withdraw their support. Thus,
improving self-management of the irrigation system is the project's current major concern. It is
necessary to formulate a new project to guide the process of phasing out external support.
Farmer's organizations will play a pivotal role in the new project. Federation, union and village
groups will become responsible for building and managing irrigation plots, supply of inputs,
credit and marketing. All units will come under the authority of the Federation, which is now part
of the interim management, along with the Senegalese and Dutch project staff. In the future,
maintenance of irrigation plots will be privatized. Marketing tasks have to be transferred from the
Union to a special unit. To be effective, the new project will also have to take other organizations
that work in the area into account.
The RAAKS team
Two Dutch consultants were asked to prepare a more detailed plan for the new project. They
decided to use RAAKS. From September to December 1994 the relevant individuals and
organizations were involved in a joint learning process. They reflected together on the
performance of the irrigation system at that time and decided how tasks, responsibilities and
institutional arrangements should be redesigned.
In the future, trainers and technical advisors will have to play an important role in strengthening
farmer's organizations. Therefore they were invited to facilitate the learning process, together
with the consultants. In a five-day preparatory workshop these facilitators were introduced to
RAAKS, made a work plan for the coming period and identified relevant parties to be involved
during the process.

The procedural design
Project staff, farmers, input suppliers, researchers, policymakers, NGOs and merchants were
invited to a subsequent workshop for ‘actors’. During the five-day workshop, participants
discussed the workplan and its objectives. They prepared and carried out field studies in which
they evaluated the performance of the then-current irrigation practices. A second workshop was
then organized for actors. Over four days participants elaborated, presented and discussed the
preliminary field results. They wrote action plans for the various sections of the project. A third
workshop was organized, at which the different action plans were presented and analysed. All
parties contributed in one way or the other to all or parts of the action plans. A number of
consultant missions were planned for 1995.
A ‘feedback’ day was later held for people who did not have the opportunity to participate in the
workshops, plus relevant governmental and non-governmental organizations from Dakar. During
this meeting, those who had participated earlier were encouraged to mobilize the support and
commitment of the ‘newcomers’ for the action plans that had been made.
The analytical design
During the field studies much attention was paid to present and future tasks, linkages, knowledge
and skills. This process showed that shifts in tasks and responsibilities would need to be discussed
and clarified among the parties involved. Further, new tasks would require new knowledge and
skills. Extension workers, animators and technical advisors needed to be able to support farmers
in their decision making on organizational and technical issues: thus they would have to change
from acting as advisors and teachers, and become facilitators. Too, expertise was needed to build
a credit and savings system to enable farmers to pay for maintenance of irrigation canals. The
studies carried out also made clear that linkages needed to be improved. Farmers, village groups,
union and federation were core actors, so strong linkages among them would become important.
The role of extension was also discussed and clarified by village groups and farmers. External
linkages needed to be strengthened and collaboration sought with both governmental and non-
governmental organizations.
The action plans
During the second workshop participants expressed a need for training to help in carrying out
their new tasks. The training section made an inventory of learning needs and started formulating
training programmes. Trainers themselves acknowledged a need for training in participatory
methodologies and formulated a request for external expertise. To help in synchronizing
development approaches and activities, a mechanism was installed to link the project to other
NGOs working in the area. Staff from the training section and the governmental research
department formulated an action plan for collaboration on applied research and development. A
request was also made for an external expert who could help establish a credit and savings
systems.

In February 1995, one of the consultants returned to Ile à Morphile for 8 weeks. Her task was
threefold: 1) to support the work of an external expert on management and technical matters (for
the Mechanics Section), and an expert on credit, 2) to organize training in participatory methods
and 3) to monitor the action plans formulated during the workshops. A lot remains to be done, but
the people of Ile à Morphile are well under way towards preparing for their future responsibilities.
Case 2
The coconut system in Tanzania: examples from a RAAKS seminar
This second case is from a short classroom exercise carried out as part of a seminar at
Wageningen Agricultural University, to allow students to gain experience.2 A limited number of
windows and tools were used. This example shows the results of their work in some key areas.
The case began as follows: coconut palms are widely grown in Tanzania. In the 1960s the country
produced enough coconuts to satisfy both its fresh nut and copra markets. By the late 1970s, the
coconut industry was in very serious decline. It was generally agreed that the decline of the
industry was due to a number of factors: an overage palm population, a lethal disease of palms (in
some areas), lack of improved planting materials, poor crop husbandry, poor pricing, lack of
research efforts, ineffective marketing channels, and lack of a government unit responsible for
development and policy. The Tanzanian government decided to take steps to rectify the situation.
In 1979 the National Coconut Development Programme (NCDP) was created, based on
cooperation between the Tanzanian Ministry of Agriculture and Livestock Development
(MALD), the Coconut Extension Service (CES), the International Development Agency of the
World Bank (IDA) and the German Technical Development Cooperation (GTZ). This established
a new linkage mechanism for research and extension in the coconut sector.
Diagnostic objectives
(Window A1/Tool A1)
In this situation, the following alternatives were identified (using Tool A1) by the student team as
possible diagnostic objectives:
● To identify the factors limiting coconut production in Tanzania.

● To formulate recommendations intended to speed up the development of improved technologies to be
used by coconut farmers.
● To discover what constraints limit the ability of resource-poor farmers to grow coconuts, and what can
be done about these constraints.
In practice, the students focused on the second objective: increasing coconut production by
developing improved technologies, beginning with the identification of actors involved in
research and development.

Actor objectives
(Windows A2 and A3/Tools A2 and A3)
Several key actors, including farmers, were identified (Tool A2). They had somewhat different
objectives (Tool A3):
to provide financial means that would allow the government of Tanzania to stimulate economic
IDA
development
GTZ to provide financial and technical support to the government of Tanzania for development
MALD to re-evaluate the agricultural sector, revive the coconut sector and increase coconut production
to increase coconut production; to increase the number of coconut farmers; and to provide farmers
NCDP
with improved technologies
CES to increase the number of coconut farmers and increase their adoption of improved technologies
Farmers to generate income by growing coconuts.
Approximation exercise I
(Window B8/Tool A5/B8)
Drawing a figure (an ‘approximation model’) such as the one below is a way for a team to work
as a group to sum up the way they visualize the system they are working to understand: the tasks
being carried out, the hierarchy within which this happens and the ‘power relationships’ within
the system. This makes it easier to decide what elements of the current situation make a positive
contribution to meeting the objectives of the system, and which do not. Drawing such a picture
can make a real contribution to the team's shared understanding of the problem situation. Getting
away from words and looking for a visual metaphor moves away from a focus on ‘what are the
problems’; it can stimulate thinking in terms of the connections between these problems. Thus it
can lead the team toward a more adequate definition of the problem situation. Drawings can also
be used as input in actor workshops (see Chapter 3).
The figure below pictures the actors in the coconut knowledge and information system (KIS) and
their relationships. The coconuts are there, along with straight lines indicating linkages. (The
thickness of these lines shows the strength of the linkages; a wavy line means a weak linkage; and
the ? after ‘market’ indicates the lack of a linkage between the coconut KIS and the market: the
focus is on production, without too much thought of the market.) The picture also illustrates the
fact that with respect to power relationships, this is a top-down system. Money and messages too
flow from top to bottom – little or no information flows from bottom to top, so it is quite possible
that decision makers will know little about the realities of the situation at the ‘bottom’.

‘Demonstration farmers’ are reached by the project, but in general citrus farmers are not: the
falling beetles and hooks (see following text) represent the only aspect of the project, the hooks,
that reaches them. Subsistence farmers are not reached at all.
Task analysis
(Window B5/Tool B5)
In carrying out a task analysis (Tool B5) the student team found that the coconut knowledge and
information system was made up of a central core of key players – NCDP, MALD, GTZ and
IDA. There was also a peripheral ring of institutions – the Tanzania Agricultural Research
Organization (TARO), Tropical Pesticide Research Institute (TPRI), Tanga Integrated Rural
Development Project (TIRDEP) and the Cashew Nut Improvement Project (CIP). A clear
division of tasks could be distinguished among the core institutions. IDA and GTZ provided
institutional linkages for financing and managing the programme. GTZ also financed
consultancies for evaluation. MALD provided and managed financial, research and extension
manpower to NCDP and CES. NCDP, as the implementing agency, was responsible for the day-
to-day implementation of the programme. CES was in charge of carrying out field extension.
Further, the activities of the collaborative National Coconut Development Programme were
directed at improving tree husbandry, controlling plant pests and diseases and increasing
production of good quality seedlings. NCDP had developed and improved a package of
technologies which was being transferred to farmers through the CES extension workers. This
package included messages about weeding, mulching, spacing (in intercropping) and fire
prevention. Nitrogen and phosphorus fertilizers were provided, as well as small hooks to be used
to combat the rhinoceros beetle, and hybrid mawa nuts. These technologies were explained to
farmers during field days for farmers. Reference farmers were selected on a competitive basis to
run demonstration plots. A system of incentives was used to stimulate extension workers to
promote the six-point package. (The matrix that was used to sum up this part of the exercise is
shown on the card for Tool B5, Task analysis.)
By 1989 about 54,000 beetle hooks had been fabricated and 50,000 had been distributed to
farmers. Between 1986 and 1988, more than 200,000 hybrid mawa nuts had been planted in
NCDP nurseries and more than 400,000 were distributed. Small entrepreneurs selling coconut
seedlings had appeared in the city of Dar es Salaam, which was evidence of a growing market for
coconut production. Apart from technology related to fertilizer application, the available
technologies in production were quite relevant and affordable even for resource-poor farmers.
Nevertheless, constraints remained:
● a lack of adequate feeder roads, limiting access to some villages

● an inefficient marketing system, which increased prices and affected the availability of some inputs,
such as fertilizers and pesticides, and

● the attitude of peasant farmers, who believed that full-grown coconut palms do not need to be tended.
Technology development, production and delivery were well implemented. But the consolidation
and evaluation of technology left much to be desired. With regard to the use of the hooks, hybrid
mawa, spacing and application of fertilizer, testing was done well. Other consolidation activities,
however, such as verification tests in farmers' fields, large-scale pilot testing, and tests for
profitability and farmer acceptance varied from technology to technology. Assessment of the
adoption rates for the technologies, categorization of users of technologies (by, for example,
income, education, and size of the farm), was yet to be done. An attempt was therefore made to
get feedback from farmers about the NCDP in general, rather than in relation to specific
technologies.
Several external factors influenced the performance of the NCDP. These included a national policy
related to the revival of agriculture; a World Bank reorganization of research and extension in
Tanzania; the result-orientated stance of GTZ and IDA; the rapid growth in demand of elite farmers for
advice and inputs; the diversity of ethnic groups and systems of land ownership and tenure in the
country; and the complexity and diversity of the environment.
The case study on which this exercise was based included many uncertainties concerning the influence
of external actors and factors. This included the restructuring of research and extension by the World
Bank and the possibility of future withdrawal by the North American and German donors. Also, the
sustainability of the technologies offered by NCDP could not be clarified by the exercise; this would
only become clear in the future.
Coordination analysis
(Window B6/Tool B6)
An analysis of coordination within the KIS identified the dominant basic configurations (Tool B6
explains this term). Figure 3 shows the system model drawn by the team. This model too
identifies the actors, their relationships and influence, but puts the emphasis on leadership and
coordination. The coconut knowledge and information system is seen as a policy and donor
driven configuration, in which the government of Tanzania and the foreign donors GTZ and IDA
are dominant actors. They exercise direct supervision of the other actors through MALD and
NCDP. Again, it is clear that only the demonstration farmers are fully part of the system; others
may benefit to some extent by purchasing beetle hooks, but subsistence farmers – and people
related to the market as well – are completely unrepresented within the system.
Figure 3

The coconut KIS in Tanzania
Source: Drawn by RAAKS seminar participants, Wageningen Agricultural University, 1993
Conclusion
This RAAKS exercise explicitly demonstrates the necessity of identifying the varied groups

involved. Clearly, in this situation it would be a great mistake to address coconut farmers as one
homogeneous group. Such an exercise could serve as background for a RAAKS study; it could
also lead to a second exercise, focusing on the newly identified target groups and the networks in
which they play a part.
3 Preparing and carrying out RAAKS action

research
In a participatory action-research methodology like RAAKS, the quality of communication, team work, actor
participation and the consolidation of results assume great importance. Managing these areas is never easy;
since RAAKS brings together a very heterogeneous group of actors, often from more than one culture,
thoughtful preparations are especially necessary. This chapter provides suggestions to help in dealing with these
issues.
Use of communication skills is an essential part of a RAAKS study. Ways to build in communication both
within the team and involving other actors will therefore be found throughout the rest of this chapter. (See also
Appendix 3.) Using RAAKS involves actors in working together to make sense out of their collective learning
experiences. This makes good communication essential: teamwork, focused collection of information,
qualitative analysis, reaching agreement on strategic decision making and following through on
recommendations all play a central role.
A very important characteristic of RAAKS is the active participation of all actors involved. A great many
different actors may take part in diagnosis, problem definition and action planning. Men and women farmers,
other rural people and consumers are involved as sources of relevant knowledge. In traditional methodologies,
these target groups have often been seen as users or receivers of knowledge; this leaves out the very important
role of local knowledge networks in innovation. And, for example, if rural banking, commerce and industry are
key actors or stakeholders, they too ought to be part of the process. The need to incorporate this broad variety of
actors means that the team must play the roles of both researcher – defining the problem and gathering,
analysing and synthesizing information – and facilitator – helping people to listen to each other, to work
together, and to make the most of their learning experiences. This puts high demands on communication skills!
Preparation

The team and supporting groups
A RAAKS study can be facilitated by an individual or a team, by insiders and/or outsiders, by professionals or
students, or any combination of these categories. Different objectives may require different expertise on the part
of facilitators. Facilitating a process of participatory planning requires conflict handling skills; these are not
likely to be required in carrying out a desk study to design a communication strategy for a firm. When RAAKS
is used for training purposes, a need for expertise on learning styles and creating open learning situations is
evident. (Appendix 3 lists relevant literature to help in gaining these and other needed skills.)
In the irrigation project in Senegal presented as Case 1 in Chapter 2, the RAAKS team was made up of 30
people. Because farmers were expected to take over general responsibility for irrigation management, a number
of farmers (plus project staff and trainers) made up the team. In other cases, the team has consisted of three
researchers, a steering committee and a consultation board (see below). In general, however, RAAKS has been
carried out by an interdisciplinary team, generally made up of not more than four or five people. Ideally, such a
team will include both actors familiar with the area and the situation, plus members who have had experience in
applying RAAKS. Achieving a degree of balance among team members is an important consideration: not only
should the mixture of disciplines included match the potentially relevant aspects of the situation, but also it is
often extremely important to include stakeholder representatives, women, people who can speak local
languages, and so forth.
As a support to the team, it is often helpful to establish a steering committee made up of relevant stakeholders.
In some cases a ‘consultation board’ has been installed as well, to provide comments on the actors' proposed
actions and their likely effect on the outcomes of the study. Such supporting groups help to ensure the active
participation of relevant actors and increase the chances of achieving some convergence in their viewpoints.
(The selection of the members for such a committee is of course a very sensitive matter.) While this can take
place in various ways, in a RAAKS study planning for the inclusion and meaningful participation of farmers or
others who have typically been called ‘beneficiaries’ or ‘target groups’ is essential.
Preparing the team
Team activities generate a great deal of information – together with opinions and observations. Aided by the
materials given in this manual (including the windows and tools), plus issues and questions that the team itself
develops, its members interview key actors, observe and discuss events, and develop an idea of what the
situation is with respect to linkages, interaction and so forth among the actors concerned. Careful preparations
are needed to assure close collaboration within the team carrying out the study. The process that will be used to
involve relevant actors also needs to be carefully discussed and planned. Team members are not ‘objective

outsiders’. Their task is not only to facilitate the learning process, but also to be actively involved and to
participate in it. Therefore team quality and interaction become very important. Team members must work
together effectively and efficiently. They must be aware of each others' background and capabilities, be able to
work as a team, and come to share a general understanding of RAAKS. Moreover, they must be able to
establish acceptable work procedures, and to argue for their preferences or choices without becoming mired
down in endless procedural debates.
A preparatory workshop should be held to allow team members to get to know each other and become familiar
with the methodology. (One possible outline for a workshop is given in Appendix 1. Also see the readings on
group dynamics in Appendix 3; for example, Pretty et al. offer information and exercises on this and on team
building.) Such a preparatory workshop has multiple objectives. These include getting to know each other,
getting to know the methodology, designing a work plan and time schedule, and designing a framework for
monitoring and evaluating teamwork, as well as preliminary and final results.
When teams study the social organization of innovation in agriculture, differences in points of view (whether
among team members or between team members and other stakeholders) are frequent. This should not be seen
as negative: differences can be essential to a community's capacity to achieve innovation in its practices, and
they may provide clues to essential points a teams and other actors need to understand the network. Rather than
being suppressed, therefore, differences should be recognized and put to use in the debate on useful
interpretations, interventions and accommodations among stakeholders. The windows and tools in this resource
box are intended to encourage recognizing the broad variety of opinions held by relevant stakeholders, and
constructively exploring their differences.
In designing a work plan and a time schedule the following factors (at a minimum) should be taken into
account:
● the scope and transparency – the clarity of the process and the field of inquiry – of the work defined for
and given to the team as a starting point (perhaps in the form of terms of reference for the group);
● the extent to which team members have had previous experience with knowledge systems research, as
well as the degree of ‘inside knowledge’ they have about the system to be studied and its context;
● the need to build up the team and to develop working relationships;
● the extent to which secondary data is available for use;
● the travel and other time that will be involved in gaining access to key actors, as well as other aspects of
gathering information in the field.
In addition, the team must decide how to consolidate its own learning process, how to share the main findings
with key actors during interviews and, even more importantly, how to share the main findings with these key
actors during the workshops or seminars to be held at the conclusion of each phase. If team members keep a
diary or journal of their reflections, this may be helpful in guiding the research process. Further, flip charts can
be used to summarize the preliminary results. When hung on the walls of the workroom, these also help to make
the process visible and to guide the team in its research. Drawings, tables and matrixes may be used to sum up
the preliminary and final results and to stimulate debate.

The material in the following sections of the manual can be used to aid in the development of the team,
including team building, communication and joint learning among team members, and in planning and
managing workshops. Sometimes prospective team members and stakeholders can be trained as a group. In
some current projects, training workshops in RAAKS have been facilitated by two consultants. In ten days, 20
to 30 local people are introduced to the methodology and carry out brief work in the field. These trainees then
work on RAAKS activities for two to three months. The results are presented and discussed in meetings co-
facilitated by the RAAKS consultants.
The interview as a method of inquiry
An interview is a social event with a purpose: a person or a team carries out an interview to obtain relatively
specific information from another person or group. Interviews are never isolated events; although the interview
is a key element in the process of gathering information, it does not stand alone. Both formal and informal
conversations with key actors, personal observations, secondary sources (such as a description of the region),
and participation in social events (including markets, church activities and festivals) can provide important
information.
Before beginning an interview, it is important to have a good idea of what information is really needed and
whether an interview is the best way to obtain it. Often the information needed can be gathered without taking
up someone's time – a precious resource for each of us, and for some villagers, one of the most limited. When
an interview is needed, do not try for perfection; there is no such thing as the perfect interview, nor the perfect
technique. Designing an interview requires making choices. The setting in which the interview will take place
can make a difference: a comfortable and secure environment will increase the chance of success. The questions
asked must be clear to the person being interviewed, and should not be ‘slanted’ to suggest a particular answer.
Any more sensitive questions should be asked toward the end of the interview, when contact has been well
established. It is wise to have a ‘trial run’, with team members interviewing one or two actors and then
discussing the results. Clearly interviewing is not just a matter of asking the right questions in a friendly voice.
Interviewing is about communicating, and becoming partners in a conversation. For the interviewer, this
requires skills in listening and asking followup questions when more detail is needed, rather than speaking.
The visualization techniques (see Appendix 3) used in participatory rapid appraisal have proved quite useful for
use with both individuals and groups. For instance, a seasonby- season analysis can elicit information about the
crops grown by individual farmers. Alternately, ranking can be done during a community meeting to identify
and discuss priority problems.
Box 3 summarizes some of the choices to be made in planning interviews. The place and timing may depend
not only on the comfort of the interviewee(s), but also their time constraints, the relevance of place and time to
the discussion and so forth. ‘Why’ and ‘what’ refer to decisions about the reasons and the information to be
collected. ‘Who’ decisions are not only about the people to be interviewed, but also about how they will be

selected – whether this will be done randomly or not, and what the criteria for choice will be. (If all or most
individual actors are known, random sampling can be a way to try and assure that all sorts of farmers are
represented; however, in many cases it is necessary to select key informants, to assure representation and
participation, including that of the often forgotten groups discussed in Chapter 1.) Finally, the mechanics of the
interview must be decided, including whether it will be structured (that is, based on a written list of questions,
decided in advance and used similarly by all interviewers) or semi-structured (based primarily on a list of
questions).
Box 3
Six decisions in preparing an interview
Where: Home/field/office/other
When: Time of day/season of the year
Why: Explore special knowledge of the interviewee; test hypotheses and inquire about topics of interest
What: Life histories
Representative information
Community interactions
Who: Selection (specific/random)
Key informants (individual/group)
Criteria (sex, age, wealth, occupation, etc.)
How: Semi-structured/structured
Appointment/chance encounter
Tape recording/notes/both
RAAKS interviews are often semi-structured. It is important to have a list of topics that need to be covered, and
to assure that the interview does not take too long (one to one and a half hours should be the maximum). It is
not practical to carry out separate interviews for each of the tools. Instead, the team needs to decide what
information it needs before carrying out the interviews. Combined interviews might be carried out for Phases A
and B, or these may be separated. In the latter case, the same people are generally interviewed in each phase;
however, before the second round of interviews the team should decide if additional people need to be included
(or if some do need not be interviewed a second time). To prepare a list of topics, the team needs to get an
overview of the information needed, rather than trying to ask separate questions for each tool. Russell Bernard
and Pretty et al. (see Appendix 3) give more information on interviewing.
During the study
Reports

Taking a ‘pause’ at the end of each of the phases in a RAAKS study is an essential element. This can stimulate
reflection within the team as well as among other key actors or their representatives; it can also provide a
crucial cross check of the ideas the team, working with these actors, has developed. This process helps to
sharpen the focus of the process and paves the way for the next phase. (Tools A5, B8/b, and C3/a suggest ways
to consider and summarize the material collected at the end of the successive phases; the material below on field
reports and actor workshops can also contribute.) This is the time to look at the hypotheses of the team and ask
if they hold up, making use of the contrasting information provided by different windows and tools. Sometimes
you may discover that vital information is missing! At that point the team must decide what to do about the
gaps. Can you do without it, making estimates based on what you do have? Should additional windows and/or
tools belonging to the just-completed phase be used? Or can the missing information be collected in another
way (perhaps in discussions during a workshop, or an extra interview)? In addition to looking back, the team
also looks forward. Do the plans made for the following phase need to be amended? And, for example at the
end of Phase B, it is not too early to think about preliminary ideas for recommendations based on the RAAKS
study. This process of summing up is good background for the actor workshops (described below). Preparing a
report is an important part of the process: it can be seen as a tentative attempt at making sense out of the
information, opinions and observations gathered so far. (An outline for a report can be found in Appendix 1.)
Field reports are made as a way of sharing, discussing and debating ideas and information among team
members. The field report – after it has been discussed with other key actors who are not a part of the team, and
if necessary after it has been modified – also provides a basis for preparing a discussion report. Discussion
reports are used to share ideas with stakeholders at subsequent actor workshops (see below). They are not meant
to be lengthy descriptive reports – instead, for example, use can be made of drawings, tables and brief
statements. At the end of a RAAKS process, the team produces a synthesis report, describing the process and
outcomes of the action research. This provides a basis for implementation, monitoring and evaluation.
System drawings
System drawings are a graphic way to develop a better understanding of the interactions among actors. They
provide a sort of model – a way of representing social organization on paper. Since social organization is
typically complex, many different models may fit to some extent, but also different actors may perceive the
situation very differently. Figure 4 suggests some of the basic building blocks for a system drawing: types of
actors and the various sorts of linkages among them. (In this case, three types of institutional leadership co-exist
in one system.) The different types of actors relevant to the system are shown by using cards or drawing images
to represent them. Such a drawing makes it possible to create a visual impression of the dominant patterns of
interaction among the actors. These models provide a visible object for debate with and among the actors. Of
course, the team will have to carefully define the meaning it attaches to the various categories of actors and the
linkages among them. More importantly, the actors need to have a chance to say whether the results represent
their reality, and to begin to think about the steps to take next. System drawings encourage actors to consider

and discuss how they feel the system is working, and how they feel it should work in the future.
Figure 4
A system drawing can make relationship patterns and dominant issues visible Rural population
Such ‘linkage models’ are not the only type of models of the system that can be created during a RAAKS study.

In fact, each window represents a different way of modelling the system: for example, by looking at the actors
from an actor perspective (actor analysis, Window B2), from a network perspective (integration analysis,
Window B4), from a functional perspective (task analysis, Window B5), or from an organizational perspective
(coordination analysis, Window B6), to name just a few. (See the window cards in the binder.) However, the
modelling of linkages suggested in Figure 4 has been chosen as a central axis for RAAKS. It has proven its
usefulness in many studies. Moreover, it provides an overall picture that connects directly with the main issues
that come up in studies of this type: who is important to the functioning of the system? Are they internal or
external actors? How important are they? Which relationships are important, very important, or vital? Which
patterns of relationships are predominant among the relevant actors? Can we speak of one ‘system’ or do, in
fact, various fundamentally different systems or subsystems coexist in the situation we are studying? These are
the kinds of questions addressed by the windows and tools.
Actor workshops
Question: Why do people have two ears and only one mouth?
Answer: Because listening is so much more difficult than talking.
This old Asian riddle touches on the essence of communication. Listening requires sitting back in a sense, but
still actively listening to and understanding what is being said. It is about stimulating people to exchange views
by asking appropriate questions and refraining from immediately giving our own opinions. Using traditional
methods, the people interviewed are often seen as passive informants. In RAAKS field interviews, however, the
active involvement of key actors and perhaps others – not just providing information, but really being involved
in the process – is essential. This can be organized in different ways, depending on the local situation.
Workshops are one way of inviting key actors to ‘sit at the table’ together with team members to discuss the
problem situation. Although some actors may have met before, listening to and especially learning from each
other may be a new experience. Keeping people interested and involved may require creativity on the part of the
team members, but more often – when the topic is clearly relevant to them and they know their opinion is
valued – people are quite willing to spend time in discussion. A striking instance of creative action, which
comes from a gender workshop organized by an Indian NGO, is given in the example.
…There was one man in the group who strongly dominated the discussions and yet insisted that
women had an equal say in decision taking. Finally (and on the spur of the moment) one female
participant said ‘for the next five minutes, I will play the wife and you be my husband’. She
developed a situation in which the ‘wife’ had sold off a bag of grain to get herself a nose-ring that
she had been asking her husband to buy for her for the last two years. Within five minutes the
situation culminated in the ‘wife’ being beaten by the ‘husband’! The action was spontaneous;
another person in the group almost joined in, and the others who were watching agreed that such
reactions are not uncommon in the community (MYRADA PRA-PALM Series 7, 1993).

The fact that workshops bring actors together to discuss what the team has observed and to negotiate about what
the problems are (and according to whom), and who can do what and how, is essential. They are designed to
adjust and refine the ‘images’ constructed based on material collected in the field, and to check again for
missing or misunderstood information. With each subsequent workshop, the ideas and analysis expressed in the
field report become clearer. The conclusions, reflected in a modified report prepared at the end of each
successive workshop, serve as a starting point for the following phase. The quality of participation in these
workshops has a great impact on the outcome of the RAAKS process. Seeing how key actors react gives the
team a fresh look at the situation, plus a chance to discuss and adjust the results. This helps to assure that the
work of the team will be relevant to the situation; it also greatly increases the chance that key actors will support
the results, and that action will result after the RAAKS study is over.
As a part of any RAAKS study, at least one actor workshop involving actors or key actors should be organized;
preferably, however, there would be three – one at the conclusion of each of the RAAKS phases described
above. The number of workshops depends on the objectives set out for a particular study, the level of analysis
required to identify opportunities for action, and the time period and person hours available. The systems
drawings described above, or the models suggested by several of the tools (see the example in Case 2 at the end
of Chapter 2, under ‘Approximation exercise’ can be used as input into actor workshops. This is one way of
presenting the material from the team's field reports (see above) while maintaining a lively atmosphere. The
techniques mentioned in Case 2, in which the team presents the pictures it has drawn and asks for discussion,
can be an extremely fruitful part of a workshop: the team asks to what extent is this picture an accurate
portrayal? Can participants suggest changes or other metaphors that would be more accurate? Care should be
taken to avoid suggesting that the drawings or other models show ‘how the situation is’; if instead the team
suggests ‘this is how we see it – what do you think?’ the models become a useful way of bringing out the
various viewpoints of the actors present.
The team will need group dynamics skills to handle such occasions adequately, and will need to spend a
considerable amount of time and energy preparing and implementing the seminars in a way that helps all
relevant actors to participate actively. Especially when there are traditional barriers to participation in meetings
– as is often the case for women, for example – team members will have to put all their skills to work.
(Participatory learning and action, Pretty et al., provides a great deal of relevant information; see Appendix 3.)
It can also be hard to create common ground and room for joint action among actors who simply don't like each
other, have hidden agendas or who feel too threatened to say what is in their hearts. Therefore, in facilitating
actor workshops, establishing a positive atmosphere, seating arrangements, listening and encouraging others to
listen, trust, valuing the contributions of participants, looking for common ground, decision making, and the
role of the facilitator are all relevant issues.
The facilitator
Some of the requirements for those playing a facilitator role include:
● being aware of and able to work within his/her own limitations;

● having the ability to encourage actors to communicate explicitly but in socially acceptable ways;
● knowing how to act as an intermediary, or broker, helping those with different views to see their

common interests;
● having no need to be the ‘big chief’ who provides solutions, and instead stimulating others to find their
own;
● helping participants to define possible ways to end problems and conflicts;
● group dynamics skills – understanding how groups work, how to be sure that all have a chance to speak
(without offending those who are very talkative), being able to sum up the attitude of the group and
check this with participants, and so forth;
● awareness of his/her management and communication styles;
● an ability to experience problems as challenges for learning;
● an ability to be a part of the learning process, and to stimulate others to do the same.
The resource list in Appendix 3 includes a section on facilitation skills; the readings listed cover both acquiring
these skills and putting them into practice.
Appendix 1 Training for RAAKS
The success of RAAKS very much depends upon the skills of the team. Therefore a great deal of emphasis
needs to be given to preparation and training. Such training, of course, should be designed in accordance with
both the personal and professional qualities of the team members and the situation at hand. This chapter
describes a number of training modules that could serve as a starting point for designing RAAKS training.
However, these examples (and the material in Chapter 3, including the section on team composition and the
characteristics needed by a facilitator) can be incorporated as you see fit: they illustrate some possibilities, but
are not intended to prescribe the way training must occur. The modules given here all make the assumption that
participants will have had a chance to read the manual and cards ahead of time, so that the RAAKS vocabulary
is not new to them. (More information on specific skills and techniques, such as those listed under ‘reading’ in
Table A1, can be found in the materials listed in Appendix 3.)
Table A1
Workshop programme
Day Topics Readinga

Introduction
Introduction to RAAKS
1 Exercises 2 and 5 (Appendix 2)
Knowledge and information systems
Division into groups
Preparation for field exercise Guidelines
2, 3
Contents/teamwork Working/learning as a group

Field exercise and presentation Terms of reference for field exercise,

Interviews documentation and information
4, 5, 6, 7, 8 Group discussions
Report writing
Presentation/discussion
Evaluation of teamwork
Innovation
9, 10 Introduction to RAAKS:
Facilitation
relevance and applicability
Alternative systems of inquiry Criticism of
11 Quality criteria for ‘alternative systems of inquiry’
RAAKS
Terms of reference for field exercise,
12, 13, 14 Preparation for field exercise
documentation and information
a Sources of background reading include the earlier chapters of this manual, plus the example of a field exercise below and the
glossary (Appendix 4); The social organization of innovation (Chapter 9, for an overview; the Introduction, on changing
approaches; Chapter 1, on innovation, knowledge and information systems, ‘soft-systems’ thinking and related criticism; and
Chapter 8, on alternative systems of inquiry); and the readings listed in Appendix 3 (in particular, see Pretty et al. on adult
learning, facilitation, group dynamics and teambuilding, participatory learning and action, organizing workshops, and for more
exercises and games). For teams focusing on related issues, Röling 1995 is quite relevant to the changing role of agricultural
extension.
RAAKS training workshop
The outlines below can be used as part of a general capacity building workshop for employees of governmental
and non-governmental development organizations, and/or for students or a team preparing for a RAAKS study.
Some of the elements found here, such as the field report form, are also useful in the study itself. Specific
objectives include:
● familiarizing participants with the RAAKS methodology

● carrying out a first network analysis in the field
● formulating action plans for RAAKS studies.
As a result of such a workshop, participants can be expected to know the principles and framework of RAAKS;
to be able to undertake a RAAKS study; and to be aware of the constraints and opportunities offered by
working with RAAKS. The products might include a report on constraints and opportunities within a given
problem situation that has been examined during the workshop, with suggestions for improvement; and action
plans for carrying out RAAKS studies.

Such a workshop gives participants a framework for RAAKS or for working together in other ways, using
exercises, a short field exercise and lectures. The emphasis is on learning by doing, reflection, and discussion of
the possibilities and constraints of RAAKS and/or the situation at hand. The proposed workshop programme,
which as given in Table A1 covers two weeks, is flexible; in practice, the programme used must be in line with
participants' needs.
Planning a RAAKS workshop including a field exercise
A three week RAAKS workshop including a more detailed field exercise, to help participants learn more about
working with RAAKS, is outlined below. A sample schedule (Table A2) and related materials follow, including
some to help in preparation for the field exercise. The objectives and expected results are the same as those
outlined for the RAAKS workshop, above. Sources of background reading can be found in Table A1, note a.
Table A2
Sample schedule for workshop including a field exercise
Monday Tuesday Wednesday Thursday Friday
Week I
The changing role of

Keynote address Teambuilding Teambuilding
extension
Getting to know Agricultural knowledge How to use the information
Teambuilding Teambuilding
each other Systems collected
Week II
Preparation for field exercise Field exercise Field exercise Field exercise Field exercise
Field exercise Field exercise Field exercise Field exercise Feedback to the field
Week III

Individual Process of
Identification of windows New perspectives Identification of
problem information
to be used on analysis actors and networks
definition gathering
Continue
Choice of Field reports Identification of Presentation of
problem
windows/design new ones revisited windows action plans
Definition
Learning objectives
In planning this schedule, it is useful to outline the specific learning objectives per day and per week, using a
format such as the one shown in Table A3. This also helps participants to think about what they can expect to
get out of the programme. Learning objectives can be outlined by a facilitator, or participants can outline their
own objectives. These can also be used in evaluating the results of the workshop.
Table A3
Format for outline of learning objectives
Week I
Week II
Week III

Plan for group exercises
A simple format such as that shown in Table A4 can be used to make a plan for the group exercises to be used
during the workshop to achieve the stated objectives.
Table A4
Format for planning group exercises
Week I
Week II
Week III

Example of a field exercise
Any field exercise, even though it is done for practice, requires specific preparation. The case below includes
examples of a problem statement, terms of reference (TOR) and division into working groups, which suggest
some of the preparation involved in an ICRE (International Course for Rural Extension) classroom exercise at
the International Agricultural Centre in Wageningen, the Netherlands. The objectives that evolved are also
included. This example also demonstrates the use of a real-life current problem in a classroom exercise.
Problem statement
In Limburg, the Netherlands, considerable efforts were being undertaken in 1994 by the government, farmers'
organizations, research establishments, extension services and other relevant parties to reduce the impact of
agro-chemicals on the environment. Policy measures – such as a multi-annual crop protection plan and a law
regulating the use of manure – had imposed restrictions on previous practices. In many cases, farmers saw
themselves as being confronted with a need to make new investments and to adjust their farming systems.
A proposed new law intended to reduce the pollution of surface water was much discussed, particularly among
producers and related parties in the greenhouse sector in which such pollution is a common problem. By the
year 2000, according to this law, greenhouse growers would be required to take measures allowing them to
produce almost exclusively within ‘closed systems’ (recirculation of water, water applied in carefully calculated
dosages, minimization of run-off water from e.g. cleaning activities, reuse – insofar possible – of agro-
chemicals, and so forth).
Two farming systems could be distinguished: a) culture on soil and b) culture on a substrate (artificial soil). The
authorities' aim was to achieve a situation in which the substrate method becomes the predominant method by
the year 2000. In such systems, flows of water and chemicals can be much better managed than in method (a).
Terms of reference
The agencies within the AKIS network in the example were doing their utmost to create programmes of
activities that were expected to enable farmers to come to grips with the new regulations and their
consequences. The Limburg Farmers' and Growers' Organization (LLTB), an agency representing the interests
of its farmer members, found itself in the forefront of these efforts. This organization agreed to the use of their
problem situation in a RAAKS study concerning the AKIS network, to be carried by 1994 ICRE participants,
and was interested in knowing the results.

The terms of reference given to the ICRE students included the background information given in the example,
and a requirement that the study should result in one or more recommendations (an ‘advice’) as to how actors
within the AKIS network could best approach the problems of keeping farmers and growers informed about the
rapidly changing situation, keeping in touch with farmers' information needs, and establishing linkages that will
make a positive contribution in these areas.
Among the questions in the terms of reference were:
● To what extent are the problems related to the proposed law on surface water pollution known in the
region? How great are the differences among the various actors within the AKIS (agencies, farmers and
other relevant actors) with respect to their degree of knowledge? What are their opinions and attitudes
toward the proposed measures? How do they see the implementability and consequences of the law?
● Which actors are involved in producing and distributing information about the proposed law? What
types of information are available?
● How is information being exchanged among the various actors involved (quality, quantity, methods,
frequency etc.)?
● Given the information obtained in answer to the above questions, can measures be suggested that might
contribute to improvements in the knowledge system? If so, which measures? Which actors (research,
extension, LLTB, farmers, clubs (including study clubs) and so forth, would need to be involved in
which measures? Who is apt to benefit from such improvements?
● Can recommendations be formulated specifically for the benefit of the LLTB, which would enable the
organization to optimally carry out its role as a representative of farmers and growers?
Objectives
Based on the terms of reference above, the following objectives were defined:
● to gain insight into the use of a methodology for analysing an agricultural knowledge and information
system
● to create a background of common experience and a common frame of reference
● to make a constraints-opportunities analysis of innovation processes within an AKIS
● to study the possible role of a specific organization in improving the performance of an AKIS
● to reflect on the usefulness of the methodology.
Division into groups during field exercise
To optimize exchange of information during the ICRE field exercise, each of the participants took part in two
different subgroups: one subgroup for interviewing (interview groups) and one for analysis of the information
that was collected (window groups). Table A5 gives an example.
Table A5
Example of division into groups
Window groups Interview groups

Names Windows Names

John W1, W2, W8 John
Lena Maria
Bernard Sonia
Maria W3, W4, W6 Lena
Ahmed Ahmed
Pedro Arif
Sonia W5, W10 Bernard
Arif Pedro
Myra Myra
Field report
When the field exercise has been completed, the group prepares a report. One possible outline to use in
assembling this material is given below. (See also the section on ‘Field and synthesis reports’ in Chapter 3,
regarding the purposes and use of this report.)
Summary An overview of the contents of the report. This should include a short description of the problem
studied, the methods used by the field team during its work, and the major findings.
Introduction A short description of the problem studied, why it is important for the farmers who are affected
by it and the approach used in collecting, processing and analysing data.
The agricultural knowledge system and related information
The AKIS should be described, including the people and organizations who play important roles
in the problem; the activities of these people and groups; and their objectives, opinions, attitudes
and interests. Further, the information and knowledge they produce or have access to should be
covered, along with their view of the problem and suggestions for solving it.
Annexes
Maps
Graphs
Terms of reference
Teambuilding workshop (two days)
The teambuilding workshop is meant as additional preparation for a RAAKS study. It follows on from a
RAAKS training workshop, as further preparation for a team. Alternately, such a workshop can serve as a
‘crash course’ before beginning action research. Its elements focus on getting to know each other, setting

working rules and designing a work plan. Such a workshop can be facilitated by either an external RAAKS
consultant or by the team itself. In the latter case, at least one person should have basic knowledge of RAAKS
and experience in using the method.
Objective
To establish comfortable working relationships, so that team members can work together intensively during the
diagnosis.
Expected results
● Team members with insight into each others' backgrounds, work experience and opinions.
● A shared idea of how knowledge systems can be examined, and how to recognize the potential for
change in a system and its actors.
● A choice of methods for working together, plus agreements on how to do this (how to divide tasks, how
to process results, how to discuss and analyse the results together).
● A choice of methods for collecting information, plus agreements on how to do this (how to interview,
how to collect secondary data).
● A choice of methods to stimulate the active participation of the actors who will be involved, plus
agreements on how to do this.
● A work plan and time schedule for RAAKS.
Programme
Day I
9.00 Introduction
9.15 Getting to know each other. Each team member takes a few minutes to draw an image of her/himself,
including characteristics that seem important to share with the other members (background, education,
work experience, hobbies etc.). The images are then presented and explained.
10.15 Break
10.45 Knowledge networks in our daily work. Team members work together in pairs to identify the knowledge
networks that led to their participation in the diagnosis, using Exercise 1 from the following appendix.
Each pair of team members summarizes the results of their discussion on a sheet.
11.45 Presentation of the summary sheets and discussion of different types of knowledge, sources of
knowledge, ‘reliability’ of sources and relevant topics.
12.30 Lunch

13.30 The RAAKS team as a knowledge network. Working in pairs and using Exercise 3, team members
identify the key actors of their own networks, the individual objectives and expectations of these actors,
their tasks and relationships, and the driving and constraining factors that affect the performance of the
networks.
14.45 Break
15.00 Presentation and discussion of the results. Attempt to formulate clues related to getting the greatest
possible benefit out of a RAAKS exercise.
15.45 Teamwork: learning styles and team contract. Roles and learning styles that team members may use
during the diagnosis should be identified, as discussed in Exercise 4 in the following chapter.
17.00 End of session.
Day II
9.00 Methods of collecting and processing information (open or structured interviews, use of
literature/reports/censuses/maps/aerial photographs etc., transect walks, joining meetings/special events)
– a brief introduction. Discussion of methods most appropriate for the diagnosis in the situation to be
studied. (See Appendix 3 for related readings, particularly Russell Bernard and Pretty et al.)
12.30 Lunch
13.30 Communication with actors – role play: facilitating an actor workshop. Divide the group into teams of
four to six people. Each team prepares a role play of an actor workshop, each with one or more
facilitators, team members and other actors (the ‘audience’). The theme of the workshop may be
assigned or decided by the teams. Role plays are presented to the group as a whole. After each one,
those watching write each of their observations on a separate card. At the end of the plenary session,
cards are pinned on a board, discussed and systematized. This will bring out many different points,
which can be used to formulate guidelines for real-life actor workshops.
15.00 Break
15.15 Work plan. The team summarizes the concepts, methods and communication techniques discussed, and
uses these to produce an outline of a workplan.
16.30 Final agreements regarding the way the team will proceed.
17.00 End of workshop.
Appendix 2 Exercises
This appendix presents a few exercises that RAAKS teams and trainers can use for team building and as a first
introduction to RAAKS. Exercises 1, 3 and 4 support team building, and are meant to be used before starting to
work in the field. Exercise 4 can also be used later, to monitor and evaluate team interactions. Exercises 1 and 3
can be integrated into workshops such as those described in the previous chapter. It is assumed that participants
will read the manual and cards before beginning these exercises, so that the RAAKS vocabulary is a bit
familiar.

The materials needed for the exercises include flip charts, marking pens and masking tape (or an overhead
projector and transparent sheets to write on), small cards in different colours; if a video is to be used (Exercises
2 and 5), obviously a videorecorder will also be required.
Exercise 1 Knowledge networks in our daily work
The assignment in this exercise can be used to introduce workshop participants to RAAKS concepts, and to help
in preparing a new team. The team gains insight into RAAKS by applying these questions to itself. In the
process, team members come to understand more about networks in general, and specifically about the
networks in which each team member plays a part, and the differing perceptions of RAAKS among team
members.
Further, the importance of making explicit choices as part of a RAAKS study becomes evident. Participants
may, for example, come up with different ideas about types of knowledge, and begin discussing which one is
correct. However, in RAAKS the question is not which is ‘right’ or ‘wrong’. The point is to become aware of
the differences, and to use this awareness in deciding among the options for the analysis.
In using the questions below, participants should first answer for themselves, then interview each other and sum
up the answers as a group, putting the summaries on paper. At least one or two networks should be worked out
as examples. See Tool B3/c (the Source-intermediary-user sheet) for more information on types and sources of
knowledge.
Questions
What types of knowledge and information related to RAAKS have you encountered?
What sources of knowledge and information have you used (or are you using) to learn more about
RAAKS?
Can you give these sources of knowledge and information a name – that is, assign them to categories? If
not, why not?
Which of these sources do you consider to be the most reliable?
Who or what first directed your attention to RAAKS?
What themes in the situation to be studied most appeal to your imagination?
Exercise 2 Identifying a knowledge system

This exercise can be used to practise applying the knowledge systems perspective to a specific problem
situation. The time required for this exercise is approximately 120 minutes.
Participants are divided into small groups and asked to divide the key questions (below) among their groups. A
case study/problem situation is then provided. This could be on paper, or if possible as a video.1 Alternately,
one of the participants could describe a problem situation, or participants could act out a situation (role play).
Each group is asked to divide the key questions (below) among themselves. Information can be summarized as
outlined in Tool A5, ‘Summarizing Phase A’.
Answers to the questions are then presented (using for example a flip chart) and discussed, comparing the
similarities and differences in the answers of the small groups. If the participants will go on to Phase B, they
will need to agree on one problem situation to be used as a starting point for the Phase B analysis.
Questions and activities
What actors (individuals, groups and organizations) are involved?

What are their objectives?
Is there a shared objective?
What problems do you think each of these actors perceives?
External factors: what and/or who are driving/constraining forces?
Make a drawing of the problem situation.
Exercise 3 Knowledge networks and the team
This is a further exercise that can be used by a RAAKS team or by participants in a training workshop. Team
members are asked to apply the knowledge systems perspective to their coming RAAKS study. They look at
themselves and the team as participants in one or more knowledge networks, and how these are relevant to the
study. How does the knowledge network that is being examined perform, and how this will affect the
performance of the team and the success of the study?
In working to understand knowledge networks, it is very helpful to use graphic methods, drawing one or more
pictures of the situation. (See ‘Systems drawings’ in Chapter 3.)
Questions

Who are the actors in the knowledge network? Who can be seen as the key actors, and who should not?
What other actors would it be necessary to talk to, to understand the network? Why?
What information will be needed from each key actor, to understand the network?
What individual objectives may each of the different actors have?
What results are expected from the knowledge network, and by whom?
What/who are driving/constraining forces behind the functioning of the network?
What tasks have to be performed by whom to achieve an optimum result?
What contacts already exist between the various actors?
What relevant knowledge networks do the actors already utilize? In what areas? (See also Exercise 1,
‘Knowledge networks in our daily work’.)
Exercise 4 Teamwork
This exercise can be used not only at the beginning of the first phase of RAAKS, but also as an interim
evaluation of the ongoing teamwork during the process. The main idea is to explore the way the team will work
together, and to make agreements related to this. As preparation for a RAAKS study, you need to discuss how
to work together as a team – do's and don'ts, how to solve conflicts, division of tasks, and so forth. The
agreements that come out of this discussion can be formalized in a sort of ‘contract’.
The following questions can be used to structure discussions; they also provide an outline for the team's
contract. The questions and information gained can be used to divide up tasks within the team, and in
monitoring and evaluating team performance.
Questions
Does the team find it necessary to divide tasks?

How is the exchange of information between team members organized?
How are decisions taken (discussion and giving arguments; forced by time pressure; focus on results;
consensus; dominant team members, etc.)?
How are disputes settled (consensus, voting, etc.)?
Can different roles be seen among the team members?
What learning styles are seen among team members? Are the roles being played appropriate to the learning
styles of the ‘player’?
Styles of learning

When asked to reflect on how they managed their teamwork during a RAAKS seminar, one group
of students used Kolb's styles of learning (1986) as a framework to discuss and make use of their
experiences. Each team member was asked to think about thetasks he or she had carried out
during the seminar: what did this suggest about their ways of working? Finally, each member
decided which part of the diagram best described his or her learning style.
The diagram suggests two axes, along which you can locate yourself (see Figure A1). It describes
elements of personality or character that affect the way you learn. One axis ranges from concrete
to abstract: do you see yourself as generally thinking in more specific, real-world terms, or more
abstractly? The other axis goes from practical to theoretical. The more practical thinker ‘learns by
doing’ – first acting, then using experience to help in grasping the underlying concepts. A more
theoretical person prefers to understand concepts before acting. This leads to the four
combinations shown in the diagram. For example, in this system someone oriented both to
concrete thinking and to practice is called a ‘practitioner’, while someone who combines
abstraction and theory is called a ‘thinker’.
Figure A1
Learning styles
Source: Adapted from Kolb, 1986
Focusing on learning styles can help both individuals and the team to gain insight into individual
styles. This provides clues that can help the team to cooperate better: learning styles affect the
way a team performs. Talking about learning styles helps to avoid the tendency of groups to
cover up differences to prevent conflict – but then the differences can stand in the way of
cooperation. The ‘thinkers’ in the group may be seen by the ‘practitioners’ as ‘not wanting to do
anything but talk’; they want to ‘get things moving.’ On the other hand, thinkers may feel
practitioners ‘act too hastily, without any plan of action’. In this situation, everyone can easily
loose sight of their common goals in a battle to support their own points of view. If instead team
members acknowledge their different styles and see these differences as a strength, they can
achieve real ‘synergies’ – bringing together their different approaches, to make use of the
strengths of all and achieve results that exceed what could have been expected from any one
alone.
Exercise 5 Exploring a problem situation

The objectives of this exercise are: to define the boundaries of a problem situation and analyse it from a
knowledge systems perspective; to operationalize Phase A (that is, to translate the results into actions) and to
apply one window from Phase B; and to integrate participatory methodologies in the work of the team. As in
Exercise 2, this exercise can be used to practice applying the knowledge systems perspective to a specific
problem situation. Here, a ranking technique from PRA (participatory rural appraisal) is added to the analysis.
The time required for this exercise is approximately 120 minutes.
To carry out this exercise, participants divide themselves into three diverse groups – that is, within each small
group people should be different with respect to work and educational backgrounds, sectors they represent,
gender, age, ethnicity and so forth. People who often work together should be separated. Choose a problem
situation from practice or use a video (see Exercise 2). After becoming familiar with the problem situation,
groups answer the questions and then carry out the activities, using the cards mentioned below. The cards that
remain after the ranking can be put on the ground or pinned up in such a way that the actors, their objectives,
the problems they see and the external factors that influence them become visible.
Questions
What actors (individuals or groups) are involved? (Participants make an inventory of the actors, writing the
name of each actor on a card)
What are the actors' objectives? What do they see as the problem(s)? How successful are the actors in
solving these problems? (Participants write each objective and each problem on a separate card.)
What are the external factors? (Participants write each external factor on a separate card.)
Activities
● Identify the most important problems, actors and external factors, using steps 1, 2 and 3, below.
● Using the results of steps 1, 2 and 3, look again as a group at the problem situation. Does the group still
accept the problem statement, or do they want to re-define the problem situation?
● Make a drawing of the problem situation.
● Choose one of the ‘B’ windows and apply it to the case. Make the results graphically visible in some
way.
● Present the results.
Group ranking of problems, actors and external factors
Step 1: Problem ranking

Spread out the cards that show the problems identified. Each participant then receives three

beans. (Alternatively: women and men receive beans of different types or colours.) Each person
decides which three problems seem the most important to them, and puts their beans on those
problem. (You can put all three beans on one problem if you like.)
The number of beans received by the various problems shows which ones the group sees as most
important. If men and women have used different beans, any differences in their voting will now
be visible. The most important problems are separated and spread out on the ground. (Option: if
for example there are not clear differences in the number of beans received by the problems
listed, you can carry out another round of prioritizing with beans, first setting aside the problems
that were clearly less important).
Note: during step one, each of the problems should be written down, and a record should be kept
of the ‘scores’ – the number of beans received by each problem.
Step 2: Ranking actors

In step 2, for each of the high-priority problems, participants are asked which actors are the most
important and who should take the lead in solving the problem. Again, for each problem, after the
group has listed the actors each participant is given three beans to indicate which ones they see as
most important.
Step 3: Ranking external factors

Participants are then asked which external factors have the strongest influence, and receive three
beans to indicate the most important factors.
Step 4: Results
The results of the ranking process can be made visible by making a table, as seen in Table A6.
Table A6
Summary of a group ranking
External factors Actors
Problemsa
Policy Drought Internal Local Community NGO
factor government
x x 1. Food x x x
x 2. Roads x
x 3. Education x
x x 4. Income x x
aNumber indicates the priority assigned by the group External factors Problemsa Actors

Appendix 3
References and other resources
References
Beek, P. van, ‘Managing knowledge systems involving QDPI’. (MSc thesis, University of Queensland,
Australia, 1989.)
Beek, P. van, ‘The Queensland dairy AKIS: a systems approach to the management of research and extension’.
In: D. Kuiper and N.G. Röling (eds.), The edited proceedings of the European seminar on knowledge
management and information technology. Wageningen, Wageningen Agricultural University, 1991.
Chambers, R., Rural appraisal: rapid, relaxed and participatory. Brighton, IDS, 1992.
Checkland, P. and J. Scholes, Soft systems methodology in action. Chichester, John Wiley, 1990.
Dijk, T. van, P.G.H. Engel and C. Leeuwis, ‘Evaluatie AGROCOM proefproject’. Wageningen, Vakgroep
Voorlichtingskunde LUW, 1991.
Engel, P.G.H., ‘The impact of improved institutional coordination on agricultural performance: the case of the
Nariño Highlands in Colombia’. The Hague, ISNAR, 1990.
Engel, P.G.H., Facilitating innovation. An action-oriented approach and participatory methodology to improve
innovative social practice in agriculture. Wageningen, Wageningen Agricultural University, 1995.
(Dissertation.)

Engel, P.G.H. and W. van den Bor, ‘Agricultural education from a knowledge systems perspective: from
teaching to facilitating joint inquiry and learning’. European Journal of Agricultural Education and Extension,
vol. 1, no. 4 (1995), pp. 1–23.
Engel, P.G.H., A. Groot, A. Meyering and T. Elema, Case study kennissysteem paardenhouderij. Wageningen,
Wageningen Agricultural University, 1990.
Engel, P.G.H. and S. Seegers, ‘Towards a design and management of effective linkage strategies: a diagnostic
tool’. ISNAR R/TTL synthesis working paper. The Hague, ISNAR, 1992.
Frempong, C.A., ‘A systems approach to the evaluation of research-extension interface. The case of cocoa and
maize in Ghana’. (MSc thesis, Wageningen Agricultural University, 1988.)
Groot, A. and S. Bakker, Renforcement du cadres institutionnel autour des PIV: RAAKS (rapid appraisal of
agricultural knowledge systems) au Senegal. Wageningen, Wageningen Agricultural University, 1994.
Kolb, D.A., Experimental learning: experience as a source of learning and development. New Jersey, Prentice
Hall, 1986.
Lindblom, C., Inquiry and change. New Haven, Yale University Press, 1990.
Long, N. (ed.), ‘Encounters at the interface: a perspective on social discontinuties in rural development’.
Wageningen Sociological Studies no. 27. Wageningen, 1989.
Lupanga, I.J., ‘The Tanzania case study’. Paper prepared for the Research-technology transfer linkages case
studies workshop. February 28-March 3, 1989a. The Hague, ISNAR, 1989a.

Lupanga, I.J., ‘The national coconut development. Subsystem Draft Report’. Paper presented at workshop on
research technology transfer linkages, the Hague, the Netherlands, 25–30 September, 1989. The Hague, ISNAR,
1989b.
ONVD, ‘The work of ONVD’. Plenary presentation at the international workshop on extension, Yaoundé, 1994.
Pretty, J.N., Alternative systems of inquiry for a sustainable agriculture. Wageningen, ICRA, 1994.
Röling, N.G. and P.G.H. Engel, ‘The development of the concept of agricultural knowledge and information
systems (AKIS): implications for extension’. In: W.M. Rivera and D.J. Gustafson, (eds.), Agricultural
extension: worldwide institutional evolution and forces for change, Amsterdam, Elsevier, 1992, pp. 125–137.
Röling, N.G., Agricultural extension in Africa. Proceedings of an international workshop. Yaoundé, Cameroon,
January 1994. Wageningen, Technical Centre for Agricultural and Rural Cooperation, 1995.
Russell Bernard, H., Research methods in cultural anthropology. Qualitative and quantitave approaches.
London, Sage, 1994.
Salomon, M., Weperen, W. van and P.G.H. Engel, Onderzoek naar draagvlak voor gebruik van bio-ethanol als
motorbrandstof in de Noordelijke provincies met behulp van de RAAKS-methodologie. Wageningen,
Wageningen Agricultural University, 1995, 38 p.
Swanson, B., ‘INTERPAKS and the development and the development of its knowledge system model’.
Plenary lecture for the International Course on Rural Extension.
Wageningen, International Agricultural Centre, 1986.

Wageningen Agricultural University MSc students, ‘Report for discussion’, 1993.
Wanga, E.O., ‘Kenya agricultural development and extension perspectives’. In: Agricultural extension in
Africa. Proceedings of an international workshop, Yaoundé, Cameroon, January 1994. Wageningen, Technical
Centre for Agricultural and Rural Cooperation, 1995.
Weeperen, W. van, Salomon, M. and P.G.H. Engel, ‘Samenvatting concept rapport fase 1 bio-energie in
verkeer’. Wageningen, Agricultural University, 1994.
Zeeuw, H. de and Veldhuizen L. van, Learning for participatory technology development. A training guide.
Leusden, ETC, 1992.
Other resources
Communication
Communication for development case studies (CDEECS), Rome, Italy.
This series of case studies can be obtained by writing to CDEECS, FAO, Viale delle Terme di Caracalla, 00100
Rome, Italy. A recent example is Farmer-first approaches to communication. A case study from the Philippines.
Facilitation skills
Heron, J., The facilitators handbook. London, Kogan Page, 1989.

Klebert, K., E. Schrauder and W.G. Straub, Winning group results. Techniques for guiding group thought and
decision-making processes with the moderation method.
Leatherman, D., The training trilogy. Facilitation skills. Massachusetts, Amherst, 1990.
Nolan Davis, C., Planning, conducting and evaluating workshops. A practitioner's guide to adult
education.Mentor, Ohio, Learning Concepts US, 1974.
Pretty, J.N., I. Guijt, J. Thompson and I. Scoones, Participatory learning and action. A trainer's guide. London,
IIED, 1995.
Gender analysis
Feldstein, H.S. and J. Jiggings (eds.), Tools for the field. Methodologies handbook for gender analysis in
agriculture. Connecticut, Kumarian Press, 1994.
Feldstein, H.S. and S.V. Poats (eds.), Working together. Gender analysis in agriculture. Vol. 1: case studies.
Kumarian Press case studies series. West Hartford, Connecticut, Kumarian Press, Inc. 1989.
Feldstein, H.S. and S.V. Poats (eds.), Working together. Gender analysis in agriculture. Vol. 2: teaching notes.
Kumarian Press case studies series. West Hartford, Connecticut, Kumarian Press, Inc. 1989.
Gianotten, V., V. Groverman, E. van Walsum and L. Zuidberg, Assessing the gender impact of development
projects. Case studies from Bolivia, Burkina Faso and India. Amsterdam, Royal Tropical Institute, 1994.

May, N., No short cuts. A starter resource book for women's group field workers. Actbook 1. London, Calvert's
Press, Actbook 1, 1992.
Moffat, L., J. Geadah and R. Stuart, Two halves make a whole. Balancing gender relations in development.
Ottawa, Canadian Council for International Co-operation, 1992 (revised edition).
Ostergaard, L. (ed.), Gender and development. A practical guide. London/New York, Routledge, 1992.
Oxfam, Gender training manual, Oxford, Oxfam, 1995.
Sandhu, R. and J. Sandler, The tech and tools book. A guide to technologies women are using worldwide. New
York, International Women's Tribune Centre and London, Intermediate Technology Publications, 1986.
Group dynamics games/exercises
Forbess-Greene, S., The encyclopedia of icebreakers. Structured activities that warm up, motivate, challenge,
acquaint and energize. San Diego, University of Associates Inc, 1983.
Newstrom, J.W., Games trainers play. Experiential learning exercises. University of Minnesota and Edward
Scannell Arizona State University, New York, McGraw Hill Book Inc., 1980.
Newstrom, J.W., Still more games trainers play. Experiential learning exercises. University of Minnesota and
Edward Scannell Arizona State University, New York, McGraw Hill Book Inc., 1983.
Pretty, J.N. et al., Participatory learning and action. (See above under ‘Facilitation skills’.)

Management
Checkland, P. and J. Scholes, Soft systems methodology in action. Chichester, John Wiley, 1990, 329 p.
Clark, R.F. ‘Management for amateurs. A "total resource" approach’. Manchester training handbooks no. 11.
Manchester, University of Manchester, 1989.
Grimble R. and M.K. Chan, ‘Stakeholder analysis for natural resource management in developing countries:
some practical guidelines for making management more participatory and effective’. Natural Resources Forum,
vol. 19, no. 2 (1995), pp. 113–124.
Labra, I. and I. Labra, PELUM & Community Link Trust, 1996, 29 pp.
The method described can be used to strengthen the capacity of marginal communities to manage their own
development. Address: PELUM Association, Box MP 1059, Mt Pleasant Harare, Zimbabwe. (29 pp.)
Senge, P., R. Boss, B. Smith, C. Roberts and A. Kleiner, The fifth discipline fieldbook. Strategies and tools for
building a learning organization. London, Nicholas Brealy Publishing, 1995.
Participatory methodologies
Thrupp, L.A., New partnerships for sustainable agriculture. Washington DC, World Resources Institute, 1996,
136 pp.
This booklet can be ordered from WRI Publications, P.O. Box 4852, Hampden Station Baltimore, MD 21211,
USA.
Bergdall, T.D., Methods for active participation. Experiences in rural development from East and Central
Africa. Oxford, Oxford University Press, 1993.

Boonekamp, G.M.M., H.W. Vaandrager, M.A. Koelen and L. Kennedy-Haynes, Travelling through health
promotion land. Guidelines for developing and sustaining health promotion programmes, derived from a
European experience. Valencía, Institut Valencía d'Estudis en Salut Pública, 1995.
Burke, S., People first. A guide to self-reliant, participatory rural development. London, Zed Books Ltd, 1993.
Deutsche Stiftung für Internationale Entwickling, Training manual. Extension in rural development. Feldafing,
1996.
Guijt, I., Perspectives on participation, views from Africa. An inventory of rural development institutions and
their uses of participatory methods. London, IIED, 1991.
Guijt, I., Questions of difference. PRA, gender and environment. London, IIED, 1995. A training video.
IIED, PLA Notes, London.

PLA (prior to 1995, RRA) notes are newsletters that cover useful themes related to participatory learning and
action. International Institute for Environment and Development (IIED), 3 Endsleigh Street, London WC1H
0DD, UK. Tel.: +44 171 388 2117. Fax: +44 171 388 2826.
Kerstan, B., Gender-sensitive participatory approaches in technical cooperation. Trainers manual for local
experts. Eschborn, GTZ, 1995.
Leurs, R., A resource manual for trainers and practitioners of participatory rural apraisal. Birmingham,
University of Birmingham, 1993.
Moris, J. and J. Copestake, Qualitative enquiry for rural development. A review. London, ODI, 1993.

Pretty, J.N. et al., Participatory learning and action. (See above under ‘Facilitation skills’.)
Srinivasan, L., Tools for community participation. A manual for training trainers in participatory techniques.
New York, UNDP, 1990.
Technical University of Berlin, Developing a participatory extension approach. A design for Siavonga District
Zambia. Berlin, Technical University of Berlin, 1992.
Waters-Bayer, A. and W. Bayer, Planning with pastoralists: PRA and more. Eschborn, GTZ, 1994.
Werner, J., Participatory development of agricultural innovations. Procedures and methods of on-farm
research. Eschborn, GTZ, 1993.
White, S. (ed.), with K. Sadanandan Nair and J. Ashcroft, Participatory communication. Working for change
and development. New Delhi, Sage, 1994.
Wilcox, D., The guide to effective participation. Brighton, Partnership, 1993.
Zeeuw, H. de and L. van Veldhuizen, Learning for participatory technology development. A training guide.
Leusden, ETC, 1992.
RAAKS
Engel, P.G.H. and M. Salomon, ‘RAAKS;, a participatory action-research approach to facilitating social
learning for sustainable development’. In: A. Budelman (ed.), Agricultural R&D at the crossroads. Merging

systems research and social actors approaches. Amsterdam, Royal Tropical Institute, 1996.
Kuiper, D., M.A. Khan, J. van Oostrum, M.R. Khan, R. Roovers and M.ul Hassan, ‘Applying Rapid Appraisal
of Agricultural Knowledge Systems (RAAKS) for building inter-agency collaboration’. Report on three
RAAKS studies in the project "Managing irrigation for enviromentally sustainable agriculture in Pakistan".
Report no. R-15. Lahore, Pakistan, Irrigation Management Institute, 1996.
Salomon, M., ‘Forging partnerships. Looking through RAAKS windows’. ILEIA Newsletter, vol. 13, no. 1
(1997).
The ILEIA newsletter covers low external input and sustainable agriculture and related topics. It is available
free to suscribers in developing countries. Address: ILEIA, c/o ETC, Kastanjelaan 5, 3830 AB Leusden, the
Netherlands.
Salomon, M. and S. Seegers, ‘Rapid appraisal of Agricultural Knowledge Systems (RAAKS) and its use in
irrigation management research. Training workshop report’. Lahore, Pakistan, International Irrigation
Management Institute, 1996.
RAAKS networks and organizations
Perspectives, associates for innovation: a network of consultants using RAAKS as part of their core expertise.
Contact person: Stephan Seegers, Hemonystraat 18-4, 1074 BP Amsterdam, the Netherlands. Tel./fax: +31
(0)20 6700312, E-mail: stephan.seegers@pi.net
STOAS: a consultant agency in rural development and agricultural education offering tailor-made training in
RAAKS.
Contact person: Lambertus Vogelzang. STOAS Head Office. Agrobusiness park 91a, P.O. Box 78, 6700 AB
Wageningen, the Netherlands. Tel.: +31 (0)317 424711. E-mail: lav@stoas.nl, fax: 0317 424770.

RAAKS in irrigated agriculture: a network of RAAKS users in Pakistan. A number of reports are available.
Contact person: Chris de Klein, 12 KM, Multan Road, Chowk Thokhar Niaz Baig, Lahore 53700 Pakistan. Tel.:
+92 (0)42 5410050-53. Fax: +92 (0)42 5410054, E-mail: iimipak@cgnet.com
Research and action research
Budelman, A. (ed.), Agricultural R&D at the crossroads. Merging systems research and social actors
approaches. Amsterdam, Royal Tropical Institute, 1996, 247 pp.
McNiff, J., P. Lomax and J. Whitehead, You and your action research project.Bournemouth, Routledge,
London and Hyde Publications, 1996.
Russell Bernard, H., Research methods in cultural anthropology. Qualitative and quantitave approaches.
London, Sage, 1994.
Sapienza, A.M., Managing scientists. Leadership strategies in research and development. New York, Wiley &
Sons Inc. Publication, 1995.
Visualization techniques
IIED, PLA Notes. (See above under Participatory methodologies.)
Linney, B., Pictures, people and power. People-centred visual aids for development. London, McMillan
Education Ltd., 1995.

UNICEF, VIPP: Visualisation in participatory programmes. A manual for facilitators and trainers involved in
participatory group events. Bangladesh, UNICEF, 1993.
Appendix 4 Glossary of key concepts
Actors/social actors
People are at the core of the knowledge system perspective used in RAAKS – not as passive recipients but as
active, knowledgeable participants who can arrive at decisions. ‘Actor’ as used here may refer to an individual
person or to a group, organization or network: all interact, taking and implementing decisions on the basis of
their own perceptions, interests, agendas, understandings and the opportunities they are able to see. (Also see
‘The AKIS perspective’ in Chapter 1.)
Area of human activity
The area that is the primary concern of the actors in the knowledge system being considered (this could be any
type of activity – providing care for the elderly in place X, or the shrub and tree nursery sector in place Y).
Three dimensions can be used to describe an area of human activity: the particular target groups or
constituencies involved; the problem(s) that concern the actors; and/or where these problems occur (the region,
agroecological zone and so forth). (See Tool A1, the Problem definition exercise.)
The breadth or specificity of the area of activity chosen plays an important role in defining the problem to be
considered by a RAAKS team. (For example, it may be necessary to choose between a focus on ‘agriculture’
and one on ‘rural development’.) Too broad a definition may produce so much information that it becomes
difficult to handle, but a narrow definition may focus the team's attention too early, excluding important
information. (See Problem, below.)
Communication
A process that may take place when people or groups of people exchange information, including symbolic
information, with each other. The information involved in such exchanges can take many different forms – for
example, it may be spoken or written, on video or radio, or electronic. In describing the process, any one or
combination of various aspects may be emphasized, such as the communications media, the participants, the
cultural issues involved or the rules that govern the exchange process.(See ‘The knowledge and information

systems perspective’ in Chapter 1 of the accompanying book, The social organization of innovation.)
Communication networks
Patterns of communication that arise as a direct consequence of actors' decisions to search for, to exchange and
to make use of the experiences, knowledge and ideas of others. This leads to increases in both formal and
informal communication – perhaps through e-mail exchanges, meetings, newsletters, workshops, journals,
conferences or courses. These communication patterns are one of three ‘social forms’ that emerge from
interactions among actors in their search for new ideas. (See ‘Understanding the social organization of
innovation’ in Chapter 1; Window B3, Knowledge network analysis; and Convergence and Resource coalition
below.)
Configuration
A particular arrangement of actors. An innovation configuration appears when actors work together to achieve
innovation. It is made up of the convergences (see below) and networks (for both communication and resource
sharing) that emerge as a part of this process. Mutually accepted views, procedures and ground rules for
collective behaviour with respect to innovation come into being, making joint decision making and coordinated
action – and eventually the steering of innovation processes – possible. (See ‘Understanding the social
organization of innovation’ in Chapter 1 and Innovation below).
Basic configurations are models that can be used to help identify prime movers and dominant relationships
among actors in a situation that is under study. They provide one way of looking at innovation configurations.
(See Window B6, Coordination analysis; Tool B6, Basic configurations; and Chapter 5 in The social
organization of innovation.)
Convergences
When a number of actors begin to share ideas and define relevant problems, alternatives and solutions in similar
ways, their views can be said to be converging. In a team or network, actors whose views converge to some
extent can work together more effectively – for example, it becomes easier to narrow down the scope of their
inquiries and the range of issues and alternatives they see as relevant to innovation. (However, strong
convergence can be a disadvantage, leading actors to ignore important information that does not match their
perceptions!) Convergences are one of the three ‘social forms’ that emerge from interaction among actors in
their search for new ideas. (See ‘Understanding the social organization of innovation’ in Chapter 1 and
Communication networks and Resource coalitions in this glossary.)

Information
For the purposes of this resource box, information will be defined as the explicit part of knowledge, which can
be exchanged among people. It is a pattern imposed on a carrier such as sound, radio waves, paper, diskettes,
electronic cables and so forth – any sort of written or spoken message. Although actors who produce
information generally do so to express a particular meaning, they can never be sure the intended beneficiaries
will attach that same significance! (See ‘The knowledge and information systems perspective’ in Chapter 1 of
The social organization of innovation.)
Innovation
A social process of interactive inquiry that actors carry out in order to construct or reconstruct their practices
(see Social practice, below). The main elements are experimentation and networking, which may result in
developing new methods and materials (technical, social or other) or in adaptation of ideas, practices and other
elements developed by others. Because innovation requires interaction among actors, it can be seen as the
outcome of a process of mutual learning. (See ‘Understanding the social organization of innovation’ in Chapter
1, plus The social organization of innovation, Chapter 5.)
Innovativeness
A social competence. A collective capacity to learn: to generate, identify, obtain, develop and put to use
technologies that are appropriate to specific conditions and societal objectives. The capacity for innovativeness
is embedded in the social relations and interactions of a large number of semi-autonomous actors – individuals,
groups, organizations and institutions. (See ‘Innovativeness as a social competence’ in Chapter 1; more detail
can be found in The social organization of innovation, Chapters 1 and 6.)
Interface
A shared boundary between actors where interactions may occur. One way to increase relevant interactions (and
therefore communication) is to identify strategic interfaces, and encourage or build linkages at these points.
Knowledge
The set of concepts, meanings, skills and routines developed over time by individuals or groups as they process

information. Knowledge is in people, ‘between the ears’. It is intrinsically related to social practice. Actors
generate, transform, integrate, exchange, disseminate and utilize knowledge while going about their daily
business. (See Chapter 6 in The social organization of innovation.)
Knowledge network
Individual people who join together to deliberately generate, share and use ideas, knowledge and information;
each participant is both a source and a user of information. Often existing local knowledge networks can be
identified and encouraged. (See Window B3, Knowledge network analysis; and Tool B3/c, the Source-
intermediary-user sheet.)
Knowledge and information system (KIS)
A linked set of actors – individuals, organizations, institutions and networks. Multiple linkages emerge (or are
strengthened) as a result of their networking in search of innovation. System performance depends heavily on
elements such as cooperation among actors, effective communication, agreement with respect to objectives and
interests, and how well the system defines and coordinates its tasks. (See ‘The AKIS perspective’ in Chapter 1,
Networking below and The social organization of innovation, Chapter 1.) RAAKS involves actors in looking at
their existing system or systems, studying for example communication patterns, convergences, resource
coalitions and configurations to see how the system is put together. This makes it possible to consider what
changes might be useful – perhaps new networks or other types of linkages that could be improved or
established to promote innovation – and to work towards commitments to achieve this.
Knowledge management
This concept suggests that a KIS can be designed or re-designed to improve its performance. That is, when we
understand the way a knowledge and information system functions, we can begin to design interventions to
increase the individual, social and/or institutional benefits it provides. (See Window C1, Knowledge
management analysis; Tool C1, Knowledge management analysis; and ‘An eye on knowledge management’ in
The social organization of innovation, Chapter 1.
Learning
A complex activity that manifests itself in a relatively stable change in behaviour of a person or a group of
persons. Learning is rooted in the human capacity to improve one's understanding and skills on the basis of day-

to-day experience. Such a change is sometimes also observed after a person or group has been involved in an
organized learning activity or has gone through some kind of ‘critical’ experience.
Linkages
Connections between actors that allow the exchange of resources such as information, money, labour and other
material or immaterial assets, such as power, status, or ‘goodwill’. Linkages are a quite significant part of a
knowledge and information system; studying them can show whether (or how tightly) actors are connected to
each other, how actors communicate and work together, and which actors and relationships are most important
to the functioning of the system. Thus when knowledge management is to be used to facilitate change, linkages
are a major focus. (See ‘The AKIS perspective’, in Chapter 1; Cases 1 and 2, in Chapter 2; Window B4,
Integration analysis; Tools B4/a and B4/b; and ‘Drawing attention to integration…’ in The social organization
of innovation, Chapter 1.) Linkages can be either formal or informal; both are important!
Linkage mechanisms
Organizational arrangements (e.g. liaison offices, meetings or administrative relationships) that help to link up
the parts of the system. This facilitates the exchange of resources – perhaps by contributing to communication
(meetings among farmers, with extension workers or liaison offices are some examples), coordination (e.g.
mutual adjustment of activities, such as water distribution or publicity campaigns) or resource transfers (perhaps
credit, salary payments or shared labour).
Mission statement
A short concise statement setting out the essential purpose of an individual or organization; ideally, this
statement specifically describes objectives, relevant constituencies and activities, plus other stakeholders and/or
partners, rather than being ‘beautiful but vague’ (Window A3)! Different actors strive for different kinds of
development: they may have widely differing views of what must be achieved, how and by which actors. What
an actor sees as a mission sets the stage for their activities; it affects their expectations, the ways they search for
information, and their impact. Similarly, the degree of diversity among the mission statements of relevant actors
has a strong influence on the system. (See Windows A3 and C1, and Tool A3, the Actor objective sheet.)
Further, a mission statement operationalizes the concept of volition (discussed in The social organization of
innovation, Chapter 6.)
Networking

An active process of building and maintaining meaningful interactive relationships with relevant actors; one of
the most important ways actors organize themselves to search for the ideas and information they need to change
their practices – that is, to innovate. Within a system, networks link relevant sources and users of knowledge
and information or other resources. Therefore, the more effective and efficient the networking among
stakeholders in development, the better the chance of innovation. The process of networking leads in turn to
more or less stable patterns of relationships: convergences, communication networks and resource coalitions; in
short, innovation configurations. This makes actors increasingly capable of supporting the purposeful
generation, use and transformation of innovations.
Because the composition and integration of a network affect communication and resource linkages, they impact
the availability and relevance of these resources and the extent to which the system achieves its objectives.
Networks are thus an important part of knowledge management strategies. (See ‘Innovativeness as social
competence’ and' Understanding the social organization of innovation' in Chapter 1; Window B4 and Tools
B3/a, B3/b and B3/c – the Info-source-use exercise, Communication network sheet, and Source-intermediary-
user sheet. See The social organization of innovation, Chapter 5, for more discussion of networking, networks
and configurations; and Chapter 6 for more on convergences.)
Participation
The involvement of actors in the process of making decisions that will affect them, including what is to be done
and how. In a RAAKS study, planning for the inclusion and meaningful participation of those who have
typically been called ‘beneficiaries’ or ‘target groups’ is essential. Successful use of participative approaches
requires respect for others' knowledge and experience and willingness to involve them in implementing,
contributing, sharing and evaluating proposed solutions. Participation may involve individuals, groups or their
representatives. The quality of participation has a great impact on the outcome of the process, and can greatly
increase the chance that key actors will support the results – so that action results. Decisions in this area are
probably the most important of any taken by a RAAKS team. (See Chapter 2 and Problem/problem definition
below.)
Problem/problem definition/problem situation
Even when a problem is stated in their assignment, a team must look carefully at the way they will define the
area of human activity (see above) and problem (Window A1) to be considered. The fact that different
stakeholders have different ideas about the definition of any problem is a key element in a RAAKS study,
encouraging stakeholders to assess and reassess their understanding of the problem situation and their own role
in it. The picture of the underlying problem evolves as a variety of actors work together.
These definitions determine the objectives the team will work towards and, critically, the participants in the
study. For example, the problem may seem to be that farmers do not participate actively enough in extension
programmes; or it may seem more institutional, such as a lack of coordination between research and extension.

In either case, the importance of this decision makes clear that the way the problem is stated must be critically
examined! Either of these problems might not be the real issue. (See Windows A1, A4 and A5, Tool A1 and
Area of human activity and Participation above.)
Resource coalitions
Alliances among actors who decide to pool their resources to improve their performance. These alliances are
one of the three social forms (see below) that emerge as a result of interactions among actors as they search for
new ideas. Such alliances can be used as a means of wielding power and influence. (See ‘Understanding the
social organization of innovation’ in Chapter 1 and Convergences and Communication networks here.)
Social forms
Patterns that emerge as a result of networking for innovation. With continuing contact, these may stabilize and
persist. The existence of these social structures influence later behaviour/interactions of the actors involved;
they may have either an enabling or a constraining influence. Engel identifies three such social forms:
convergences, resource coalitions and communication networks. (See ‘Understanding the social organization of
innovation’ in Chapter 1).
Social organization of innovation
This phrase recognizes innovation as taking place within a social context; this context can thus either facilitate
or impede the process. Further, if the context is sufficiently well understood, the actors involved can work
together to organize themselves in ways that are more beneficial to their objectives – that is, they can organize
to search for ideas and information that will allow them to change their practices. (See Chapter 1,
‘Understanding the social organization of innovation’ and the Introduction to The social organization of
innovation.)
Social practice
The activities we all engage in are largely socially defined. That is, the society in which we function – whether
within the family, a particular type of work, or a country, region or tribal group – tends to have rules (written or
unwritten) about what makes a competent ‘X’, where X may be a mother, a doctor, a farmer, a particular sort of
business, a teacher, a policymaker and so forth. Any set of actions can be seen as a ‘social practice’, in the sense
that actors socially define and reinforce rules of competence as they relate to each other. A social practice can

then be defined as a discernible set of actors, who define and uphold performance through some form of rule-
governed social interaction.
Interplay in-and-between diverse social practices is a breeding ground for innovation; networking is a way of
encouraging this interplay among relevant practices. Further, this concept can be useful in understanding the
differences in the rules that govern different practices – for example, farmers in comparison to researchers or
extension workers. A clearer perception of each others' activities as social practices in their own right could
help to improve the interplay between groups and to generate more effective networking practices among them.
Stakeholders
Actors whose interests are affected by a particular area of human activity, whether as victims or beneficiaries.
In a RAAKS study, generally those who are concerned with maintaining or changing performance in this area.
(See ‘The AKIS perspective’ in Chapter 1.) As with actors, stakeholders may be individuals, organizations,
legal entities, etc.
Synergy
An effect arising from the cooperative activity of two agents that, when working together, produce a combined
result greater than either one could have achieved alone. When actors whose resources (see Linkages above)
complement each other work together, the collaboration increases the effectiveness of all. Each actor then
achieves more than they could alone. (See Chapter 1, ‘The AKIS perspective’; and Window C1, Knowledge
management analysis).
Window
An analytical perspective that focuses on particular issues relevant to understanding a problem situation.
Windows provide a framework for the use of a RAAKS team. (See Chapter 2.)
About the authors

Monique Salomon holds an M.A. in Cultural Anthropology. She is currently based at the Department of
Agriculture of the University of Zululand, South Africa as senior lecturer in rural development and extension.
She also works for STOAS as an international consultant on knowledge management, participatory
development, gender issues and RAAKS, and is a member of Perspectives, a network of RAAKS consultants.
Previously she was a lecturer and consultant at the Department of Communication and Innovation Studies,
Wageningen Agricultural University, the Netherlands; she has also had professional experience in social work
and education in the Netherlands.
Paul G.H. Engel has an MSc in irrigation agronomy and a PhD in agricultural and environmental sciences. After
field work with small farmers in Peru, Ghana and Colombia he joined the Department of Communication and
Innovation Studies at Wageningen Agricultural University as a lecturer and researcher. He focuses on action
research as a means of developing methodologies to facilitate stakeholder participation in agricultural
innovation and natural resource management. As a consultant he is primarily engaged with stakeholders who
are re-inventing their informal and institutional networks to strengthen smallholder agriculture. As a researcher
he concentrates on institutional development and knowledge management issues in decentralized, market-
oriented settings. He is now a senior consultant with STOAS International Projects, based in Concepción, Chile,
and is also a senior member of Perspectives, a network of RAAKS consultants.
Evaluation form
RAAKS is still under development; we would very much like to hear your comments and
suggestions. Please make a copy of this form (so that others can use it later) and send the
information to the address given in the box.
1. In which professional field(s) are you active?
………………………………………………………………………………………………………………
………………………………………………………………………………………………………………
2. Had you had any experience with RAAKS before using the Resource box?
………………………………………………………………………………………………………………
………………………………………………………………………………………………………………

3. How applicable is RAAKS to your work?
………………………………………………………………………………………………………………
………………………………………………………………………………………………………………
4. Do you expect to use RAAKS (or aspects of RAAKS) in the future? If yes, in what ways?
………………………………………………………………………………………………………………
………………………………………………………………………………………………………………
5. What suggestions do you have for improving RAAKS or the Resource box?
………………………………………………………………………………………………………………
………………………………………………………………………………………………………………
If you would like to receive information on future RAAKS developments, please fill in your name and
address:
Name
………………………………………………………………………………………………
Organization
………………………………………………………………………………………………
Address
………………………………………………………………………………………………
Postal code/country
………………………………………………………………………………………………

Telephone
………………………………………………………………………………………………
Other, e.g. fax, mobile phone or e-mail
………………………………………………………………………………………………
………………………………………………………………………………………………
Order form
If you would like more copies of the RAAKS resource box or other information
about publications, please fill in a copy of this form, checking one or more items,
and send it to the address in the box.
I would like to order …… (number) additional copies of the RAAKS Resource box at Dfl. 59 (Dutch
guilders) per copy*
Please send by airmail
By surface mail
Please send me (free) the annual KIT publications list (participatory developmentrelated publications).
Please send me (free) information on:

RAAKS-related publications
publications on agriculture
publications on health care.
Name
………………………………………………………………………………………………
Organization

………………………………………………………………………………………………
Address
………………………………………………………………………………………………
Postal code/country
………………………………………………………………………………………………
Signature
………………………………………………………………………………………………
RAAKS' WINDOWS A
Materials in Each of the following cards contains one RAAKS ‘window’, or perspective, to use in B
manual and cards better understanding the effect of social organization on networks and the use and
may be freely spread of innovation. At the top of each card, you will find the reference number and
reproduced in name of the window, followed by a summary of what it covers and a list of tool(s) that
limited numbers will help in collecting the needed information. Each window is labelled A, B or C,
for training and indicating the phase to which it is most relevant. Windows are numbered for
other educational convenience, although in Phase B a team is encouraged to chose among the windows,
purposes. The or use them in another order, to fit your situation. The reference numbers also connect
following the window to one or more tools. Figure 2 in Chapter 2 of Networking for innovation
statement must provides a way to visualize the use of windows in relation to the study as a whole.
appear on all Beside the box describing each window you will find a brief description of the design,
copies: validity, use and applicability of that particular window. Scientific ‘validity’ generally C
refers to whether something does what we expect it to do: a valid window is one that
Reproduced from
helps teams focus on practices and patterns of social organization that are relevant to
the RAAKS
innovation. These windows have been used in a variety of circumstances (see The
resource box,
social organization of innovation); these field experiences have shown all of the
Paul G.T. Engel
windows to be valid in this sense. On the cards, the section describing the validity of a
and M.L.
window suggests the aspects it covers. By ‘applicability’ we mean the suitability of a
Salomon. Royal
window for participatory inquiry: an applicable window is one that helps participants
Tropical
construct images of the system to stimulate their interactive learning. This section of
Institute, 1997
the card suggests how the particular window does this.
For large scale Overall, it is important to remember that windows do not provide ‘recipes’ – instead,
they suggest general ideas for ways of looking at the analysis. Please see Chapter 2 in

reproduction or Networking for innovation for more information! It is also useful, in learning about the
commercial use, windows, to read them in combination with their related tools (see the following
written cards), as well as the glossary (Appendix 4).
permission from
KIT Press, Royal
Tropical
Institute, is
required.
WINDOWS TABLE OF CONTENTS
A A1 Defining or re-defining the objective of the diagnosis

A2 Identifying relevant actors
A3 Tracing diversity in mission statements
A4 Environmental diagnosis
A5 Clarifying the problem situation
B B1 Impact analysis
B2 Actor analysis
B3 Knowledge network analysis
B4 Integration analysis
B5 Task analysis
B6 Coordination analysis
B7 Communication analysis
B8 Understanding the social organization of innovation: summing up
C C1 Knowledge management analysis

C2 Actor potential analysis
C3 Strategic commitments to an action plan

WINDOW A1
Defining or redefining the objective

of the diagnosis
The team that is to carry out the diagnosis usually receives terms of reference from others. However, these may
reflect the views of only some of the relevant actors. These actors may be quite important, but a thorough,
critical assessment is needed to determine whether others need to be included, and whether the problem
definition and objectives are workable for all of the relevant actors. To achieve a definition of the problem and
objectives that will be truly useful, the team must take into account the many types of actors relevant to the
innovation processes to be studied.
TOOLS Problem definition exercise
Design
This window is intended to raise questions such as ‘who stated the problem as it is now? Who thinks it is
important? Who does not? Why? Who holds a key to solving the problems mentioned?’ It requires great skill on
the part of the team. Generally the initiators of a study find it difficult to see that a new formulation of the
problem might be more useful. Nevertheless, the team needs to carefully probe the views of different
stakeholders. The aim should be to generate a ‘rich picture’-one that describes relevant diversity. It is not
necessary to seek consensus in this early stage.
Validity
This window obliges the RAAKS team to make their own objectives clear. Uneasiness on the part of some or all
stakeholders usually reflects a partial understanding of the situation. The window calls for the RAAKS team to
probe views and arguments and to confront actors with each others' views vis-à-vis innovation.

Use and applicability
When other actors believe that those who have declared the problem have only a partial understanding of the
situation, they may not want to cooperate, fearing their views may be misrepresented and/or their efforts
frustrated. Dominant views that have this effect might include ‘farmers always resist change – if only they
would do what we say!’ or ‘extension workers should not bother us with their comments; they should just
implement the recommendations we provide’. Such views may lead to terms of reference that read ‘to design
ways to improve farmers’ response to extension programmes' or ‘to improve the effectiveness of extension
workers’. Such partisan views ‘pass the buck’ to some actors and ignore the relevance of others to the problem.
This makes it difficult – but even more essential – for the RAAKS team to work towards a balanced inquiry.
WINDOW A2
Identifying relevant actors
The actors relevant to the innovation process must be identified; this window follows up on the process begun
in Window A1. Here too, there will probably be different points of view on the relevance of each proposed
actor to the problem at hand. Remember, at this stage inclusive thinking is required – being open to a variety of
possibilities, rather than focusing too narrowly.
TOOLS Actor identification exercise
Design
The inclusion or exclusion of a particular actor is a matter for careful consideration. On the one hand, it is
important to generate a broad list. The decision to exclude certain actors (perhaps groups labelled traditional
farmers, accountants, or traders) may reinforce traditional views with respect to the social organization of
innovation. On the other hand, actors may be included simply because someone suggests that they might be
relevant. Not only does this make the list of actors long and difficult to work with, but also if these actors are
later excluded from the study, those who suggested them may have less interest in participation.

Validity
This window takes on the ‘boundary issue’ – how broad the study will be. It calls for a tentative definition of
who the ‘relevant actors’ are. This forces the team to formulate criteria for assessing the relevance of the
contributions specific actors make or are expected to make to innovation. Here we must deal not only with the
current situations some actors have in mind, but also with the desired situations. It is important to see that the
picture drawn while using this window is never ‘finished’. At any point during the study, an actor may acquire
new relevance in the eyes of the team and/or participants, so that they need to be included; or some may be
excluded as more is learned about the system.
With the help of Tool A2 (the Actor identification exercise), it is very easy to generate a long list of actors that
seem to be related to the issues in one way or another. It is important to list each of the actors by name, not just
as a category. This list may contain not only names of persons but also organizations, units, journals,
committees, etc. Eventually, the team might choose to work at the level of either persons or organizations, not
both. In any case, the team must work with relevant organizations in the difficult task of choosing representative
individuals to participate in the RAAKS study.
WINDOW A3
Tracing diversity in mission

statements
Different actors strive for different kinds of development. Each actor related to the process may have their own
view of what must be achieved, between which actors and how. We will refer to their statement of this view as
their mission statement. The team can use this diversity among relevant actors to explore fundamentally
different or even conflicting objectives. Differences in objectives can serve to indicate the direction in which a
particular actor might look when seeking information relevant to innovation in his/her practices.
TOOLS Actor objective sheet
Design

This window presupposes that different sets of actors generally have different ideas about which developments
are desirable. Consensus is the exception. Therefore, to better understand the system and ways it could change,
the search is for convergences and divergences of opinion between relevant actors. Mission statements provide
a way to summarize the objectives, strategy and beneficiaries (or intended beneficiaries) of a particular actor.
Similarities may suggest convergence, while fundamental differences may indicate divergence. A discussion
among actors about their mission statements may be very revealing!
Validity
Probing the strength of the convictions and/or arguments that lie behind mission statements helps to gain insight
into the possibilities for negotiations among actors with respect to differences in their views. Any inquiry into
the social organization of innovation requires spelling out the preoccupations or ambitions that characterize
relevant actors. One way to do this is to ask for a definition of the actual versus the desired situation in the eyes
of some or all of the actors involved. Another possibility is to trace the motives different actors state as a reason
for taking part in innovation-related activities.
Tracing the motives and intentions of actors can be fun. It gives participants an opportunity to probe their own
views and those of others with respect to innovation. However, it may also be threatening. Beautiful but vague
objectives such as ‘to increase smallholder family income through the introduction of improved farm
technology’ must be questioned, and actors must be gently forced to specify their intentions, favoured
technologies and target groups – and to discuss them with others!
WINDOW A4
Environmental diagosis
This window examines which actors and other factors influence the performance of the system – these make up
the ‘environment’ in which it functions. Which external influences and/or conditions may affect the
performance of the actors involved with respect to innovations? Examples might include natural resource
availability and agroecological constraints, but socioeconomic and cultural factors are also apt to be important.

TOOLS Environmental limits checklist
Design
Some factors are more frequently seen as affecting the dissemination of innovation than others. Articles in the
literature refer to both natural and man-made conditions: agroecological and socioeconomic diversity, available
technologies, external market and/or policy pressures, the availability and need for external resources, and the
adequacy of agricultural services, marketing, inputs and communications infrastructure. In studying the impact
of the socioeconomic and natural environment upon the innovative efforts of actors, these factors (at a
minimum) should be carefully considered.
Validity
To improve networking for innovation, limiting and enabling conditions must be assessed. This window brings
out the wider context relevant to the performance of the system.
This window helps to assure that the team will take relevant environmental conditions into account. This
includes agroecological, climatological and other natural conditions. In practice, however, many ‘environmental
limits’ related to resources and services are man-made; they are created by actors. When a team begins to
identify external factors, it may become clear that some who for one reason or another were earlier considered
to be ‘outsiders’ are actually quite important to the system. This brings us back to the boundary question raised
in Window A2: should these actors be included in the RAAKS study? One frequent example: ‘the market’
might well be called an environmental factor; market prices normally fall outside the control of local actors. But
when particular traders or agro-industries play a role in these prices, and thus in the ‘theatre’ defined for the
RAAKS study, it may be better if they participate actively in the networking related to innovation. On the other
hand, some things may remain as ‘external.’ The team and participants can use this window to think further
about actors, distinguishing those whose involvement is direct enough to be affected by local networking
strategies from those who are considered too distant (physically or in terms of power) to be influenced. During
this process, relevant resources and services become clearer.
WINDOW A5

Clarifying the problem situation
Here the team works to achieve a first approximation of the knowledge system and its performance. Further, the
diagnostic objective is reconsidered. Is the problem situation identified in the process of diagnosing the
objective in line with the results of this window? If not, why not? How can the objective be redefined? This
helps to close the first ‘loop’ of the diagnosis, and is part of considering whether any revisions in procedures are
needed before going on to Phase B.
TOOLS
Prime mover septagram
Approximation exercise II
Design
This window asks the team to draw up a synthesis of the objectives and findings established during Phase A,
creating a tentative image of the social organization of innovation. As part of this process, the area of interest,
the relevant actors, their main concerns, their objectives and target groups are specified. Moreover, the team
discusses the influence of different leading actors, as perceived by other actors. As a result, a first attempt can
be made to trace relevant coalitions related to resources – actors who pool their resources to achieve their aims.
This requires the team to state their own terms of reference more specifically and to declare what it sees as a
workable and widely acceptable definition of the problem situation. In general, this is the point to reflect on
Phase A, before moving on.
Validity
This window stimulates the team to formulate tentative ideas with respect to what is most relevant to the
system, including practices, networking and newly emerging forms of social organization. It does so in a
general way, not in detail. The end result tentatively describes or pictures the team's perception of the way
relevant actors organize themselves to achieve innovation. This picture is to be presented and discussed –
including a discussion of its validity – during the first workshop with the stakeholders.

The interactions between relevant actors, as well as with their respective constituencies, can be made visible by
constructing pictures of ‘soft systems.’ You can either draw very general pictures or use cards or papers in
different shapes to represent different actors, drawing one or more circles around groups of actors who appear to
converge on the same or a very similar objectives – that is, who seem to form coalitions (see Tool A5/B8,
Approximation exercise I). Discussion can be stimulated by seeking (whether this is achieved or not) to define
precise characterizations of actors and their influence on the innovation process. Tool A5, Approximation
exercise II, helps to sum up and to prepare for Phase B.
WINDOW B1
Impact analysis
Does the system or subsystem succeed in achieving the several desired objectives of its actors – that is, what
impact does it have? What desired or undesired side effects does this have? Answering these questions requires
reflection on the degree to which the actors involved recognize a joint objective or purpose, and whether they
feel this is achieved by the current social organization. Rather than attempting to reach full consensus and
define a joint mission, the objective of this window is to understand the ways different actors make sense out of
their own individual performance, and what they expect from a joint effort. The outcomes desired by one may
of course be the undesired consequences another hopes to avoid.
TOOLS Impact analysis sheet

Source–intermediary–use exercise
Design
This window assumes that actors who define objectives differently will also judge impact differently. Also, they
often recognize the need to improve communication and/or cooperation with relevant others, but fail to specify
their expectations clearly enough. Too, they may not know exactly what others expect of them. Impact analysis
helps to make these expectations – which influence the criteria used by actors in judging impact – more explicit,

so that they can become the subject of discussion and debate among relevant actors.
Validity
This window permits the team to probe somewhat deeper into the practical consequences of the views stated by
various actors. A central question is what criteria are used by actors to judge their joint operations. The purpose
of the window is to identify the criteria used by actors in defining what level of impact, effectiveness and
efficiency is sufficient, rather than to measure it in an ‘objective’ manner.
When there is agreement on a joint purpose, it may be a relatively straightforward operation to assess the
knowledge, skills and technologies relevant to achieving the objectives of different sets of actors, and the extent
to which these are available to everyone concerned. If there is no convergence, probing for information must go
much deeper. For example, in southern Chile small producers were hesitant to cooperate with milk processing
plants: they expected the companies would take advantage of them. The plants, on the other hand, complained
that small producers were unreliable and always threatened to sell to their competitor. Still, they recognized the
need to cooperate to strengthen milk production in the region. In such a case, getting realistic answers about
possible convergences may require clarifying biases on both sides.
WINDOW B2
Actor analysis
Who are the most important actors and what are their characteristics? This window assumes that some actors
are more relevant to successful innovation than others; it works towards learning who they are, why they are
more relevant, and what types of innovation they favour.

TOOLS Actor analysis checklist
Design
This window continues the work begun with the help of Window A2. It brings into focus the views and role,
characteristics, and contributions to agricultural innovation of each individual actor. Validity Looking at actors
individually makes it possible to identify and describe relevant practices, trace convergences and resource
coalitions, and assess the actor's strengths and weaknesses with respect to stimulating innovative performance in
a particular direction. The effectiveness of leadership (including institutional leadership) within the system can
also be assessed.
Actors who are networking for innovation can be looked at and compared on the basis of many different
characteristics. The RAAKS team chooses characteristics that describe the role of actors in their situation; these
can vary widely from one situation or one team to another. For example, actors could be compared with respect
to views and strategies, and their power to influence events in the ‘theatre’ under consideration. Or official
mandates could be compared to actual practice. Clearly, the use of actor analysis goes hand in hand with the use
of other windows. For example, the results of this window are often combined with those of Integration analysis
(B4), to provide an analysis of relationship patterns in the context of the importance of particular actors to the
system. Combination with the results of Coordination analysis (B6) has proved powerful. Looking at
interactions among the results of B2, B3 and B5 has also been fruitful.
Carrying out such an actor analysis requires careful thinking. Actors such as farmers or villagers, who are
sometimes seen as ‘beneficiaries’ or ‘target groups’, may be essential to the functioning of the system. If this is
the case, representatives of the group need to be included in the study and perhaps on the team. If they are
clearly peripheral to the system under study, but will be expected to be influenced by its results, an analysis of
these categories, their views and practices can be very relevant. Also, if farmers for example are not organized
to act collectively, consideration should be given to ways to meet their participation needs and be sure that they
are adequately covered in interviews and feedback sessions. Some teams have opted for drawing a random
sample, but most have selected key informants, hoping to guarantee farmer participation in the inquiry.
WINDOW B3

Knowledge network analysis
This window goes beyond the question of the needs of individual actors, to that of the larger whole. What types
of knowledge are important for the successful performance of the system? Who are the sources and users of
these types of knowledge and information? Who or what are the intermediaries – the actors, printed materials or
other media that move knowledge and information among actors? How effective are the existing
communication networks in linking relevant sources, intermediaries and users of knowledge and information?
TOOLS
Info–source–use exercise
Communication network sheet
Source–intermediary–user sheet
Design
This window assumes that actors deliberately exchange information on relevant topics or concerning particular
types of knowledge: they talk to those they consider knowledgeable, read papers they consider well informed,
listen to interesting radio programmes, and so forth. During this process, relatively stable patterns of interactive
relationships evolve in which information is produced, exchanged and used. One characteristic of these
networks is that every participant is at the same time a source as well as a user of information. Some may be
more knowledgeable on one issue, and others on another. In the exchange with others, each actor adds value to
the network by transforming his or her ideas, experiences and information into intelligible information.
Validity
This window helps the team study networking practices and the generation, exchange and practical use of
knowledge, and to appraise communicative interactions relevant to agricultural innovation.

The main difficulty in the application of this window is to avoid traditional conceptions of knowledge – taking
it as a static and/or technical entity – and to begin to recognize the extent to which knowledge is ‘socially
constructed’: a product of our interactions with each other, which changes over time and with new interactions.
Team discussions might cover the relevance of different types of sources to particular sets of actors, the lack of
access to relevant knowledge and information of particular actors or constituencies, or the speed of exchange of
knowledge and transfer of information. In each case, the relevance of each of these issues to the innovative
performance of the whole should be considered.
WINDOW B4
Integration analysis
This window is about linkages. Who has contact with whom, why, and how intensively? Can clusters of actors
be distinguished – for example, around key actors? What characterizes these clusters?
TOOLS Linkage matrix

Linkage mechanism checklist
Design
The starting point of this window differs from earlier ones: the focus is on whether actors are connected to each
other. If they are, do their linkages imply communication alone, or also control? First, the types of links to be
included in the analysis are defined. Normally these include resource linkages plus administrative and
communication linkages. Sometimes more detailed information is recorded here, such as the characteristics of
the linkage or the frequency of contacts. The results of integration analysis can be presented as a drawing or as a
‘linkage matrix’ (see Tool B4/a, Linkage matrix): a table of any size, with the same components listed on each
axis. The information given in each of the cells of the matrix relates to a particular linkage (see Tool B4/a for an
example). As in Actor analysis (Window B2), team discussions are needed: what is the most relevant
information to collect to characterize and describe the linkages in your situation? Later, the team can consider
the relevance of particular links and their impact on innovative performance, as well as the relevance of the

resources a cluster of actors can pool together. To what extent can pooling resources change the course of
agricultural innovation?
Validity
Determining the linkages between actors makes it possible to recognize resource coalitions and communication
networks. If necessary, a detailed analysis can be made of one or more specific linkage mechanisms, to
determine the role a mechanism plays in enhancing coordination of tasks among actors (Tool B4/b, the Linkage
mechanism checklist). In the process, indications of innovation configurations may also appear. (These
combinations of convergences, resource coalitions and communications networks are discussed in Chapter 1 of
the manual Networking for innovation.)
This window is one of the instruments most frequently used in exploring the social organization of innovation.
It permits the elaboration of a relatively comprehensive picture of relevant coalitions and networks in a brief
period of time. However, it tends to overemphasize more structural, formal contacts at the expense of informal
ones. Initially, researchers often assumed that each linkage was equally relevant to innovation. However, as van
Beek (1991) demonstrates, managers for example attach different priorities to different links. His suggestion is
to include perceived importance as a characteristic of linkages when carrying out an analysis.
WINDOW B5
Task analysis
Who does what in the system? Are there functional connections between system actors? Do these function
adequately? These are the questions dealt with by this window. Practices relevant to innovation, such as
farming, research, trade or quality control are identified, along with the actors in charge. Gaps or overlapping in
the performance of tasks become more evident.

TOOLS Task analysis sheet
Design
This window focuses on the role and functions of the actors in the system, including the way tasks are divided
among them. As a first step, the team and participants define which functions need to be performed to achieve
innovation in agricultural practices. Traditionally, such questions were referred to research and extension and
perhaps to trainers working in farming and education. However, experience and field studies have demonstrated
the relevance of other actors – policymakers, veterinary services, input suppliers, agro-industries, banks,
certification committees, traders and others. Therefore, the relevant practices in a particular situation must be
sought out, using intensive probing and debate. Following this, a team can ask which actors are involved in each
of the practices listed.
Validity
A task analysis helps shed light on relevant practices of actors, on overlapping or missing functions, and (in
combination with the results of Tool B4/a, the Linkage matrix) on the adequacy of any social structures such as
convergences, resource coalitions and communication networks that are present as well (see Chapter 1 and
Windows B4, B6 and B7).
The importance of a thorough discussion prior to a declaration of ‘relevant practices’ is illustrated by a RAAKS
seminar in Costa Rica. Five groups elaborated a relatively predictable list of relevant practices including
policymaking; fundamental, applied and adaptive research; transformation; dissemination; and use. However,
the Nicaraguan group, very conscious of the role of the free market in their economy, added quality control as a
function. They argued that the actors who control quality standards for agricultural inputs and/or produce are
extremely relevant in deciding the course of agricultural innovation. This confirmed earlier observations by for
example Swanson (1986) with respect to the role of certification procedures. It also shows the importance of
input related to the specific local situation.
WINDOW B6

Coordination analysis
Who takes the most important decisions? Who ‘pulls the strings’ and sets the agenda? How is influence
exerted? Who involves others in their ‘projects’? Who has the means to implement important decisions? This
window helps to identify leadership and coordination efforts made by actors and directed at innovation. It also
looks at the ‘basic configurations,’ or patterns that may be seen within the system.
TOOLS Basic configurations

Design
This window assumes an organizational perspective, with a focus on leadership and coordination for
agricultural innovation. It helps identify leading actors and the means by which they create and strengthen
coordination among relevant actors in the ‘theatre’. This leads first of all to a characterization of the way tasks
are coordinated among relevant actors (if at all). Second, by combining the findings from this window with
those of B1, B3 and others, the team can better understand the issues the various actors consider in assessing the
impact of each of the leading actors – their influence on agricultural innovation and its direction.
Validity
Using this window helps bring out dominant resource coalitions – coalitions that have control over resources
and can thus dominate the situation – and the one or more configurations that may be the result. To achieve a
more comprehensive interpretation of configurations and emerging networks, it must be combined with
Windows A3, B2, B4 and B7.
There are different types of leadership. Some give political or financial leadership, or both. Leaders set

priorities, provide financing and impose administrative or other regulations. Others acquire technical leadership
on the basis of knowhow and experience. Still others represent relevant constituencies or markets. Looking at
the basic configurations (see Tool B6, Basic configurations) has proved an interesting way to highlight
leadership issues and thus to encourage debate. Consequently, this helps to study how different types of
leadership impair and/or enhance innovation. Such a debate may however be threatening to certain stakeholders.
In addition to analytical skill, this method therefore requires considerable skill in group dynamics and
communications.
WINDOW B7
Communication analysis
Do people speak the same ‘language’ in a figurative sense? Even when people use the same word, do they mean
the same thing? Is effective communication among system actors possible? This window helps to study the
effectiveness of communication among actors, and thus to understand whether, if the varied actors involved
meet each other, a fruitful dialogue will be possible.
TOOLS Communication analysis exercise
Design
The point of departure of this window is an assumption that innovation is contingent upon effective
communication among relevant actors. This window has been chosen as a way to focus on cultural barriers that
may obstruct effective communication between social groups, not to discuss networking practices as such (that
has been done in B3).
Validity
This window focuses the attention of the research team and participants on constraints on communication
implied by culture and use of language. This is extremely important for RAAKS: these same constraints can be

expected to have a direct influence on the outcome of the team's work.
The sorts of problems that can arise between indigenous communities and an extension worker who speaks only
the official language are well known. But even when extension workers or researchers have the same mother
tongue as the community, the cultural differences created by upbringing and education may create formidable
communication barriers. For example, peasants in southern Colombia were puzzled at times by the Spanish
word selección, used by extension workers to refer to the selection of potatoes to use in planting. For them the
selección was the village soccer team! After some explanation, of course, it became clear that the same process
of ‘selecting the ones that will do the best’ was the basis for both usages, but the object of the selection differed.
In this same region, a detailed study by a team of communication specialists, including a local anthropologist,
found over 175 words in common ‘extension language’ that were of low or doubtful comprehensibility, even
though farmers were native Spanish speakers and all extension workers were sons or daughters of local farmers.
When actors still farther from the community are included, and when the concepts in question are farther from
daily experience, it becomes even more important to check the effectiveness of communication.
WINDOW B8
Understanding the social organization

of innovation: summing up
This window recalls the insights the team has gained. These are summarized in the form of a report and
presentation to be used in the next workshop with other participants. The central questions to be answered are:
what are the major convergences, the resource coalitions and the communication networks within the ‘theatre’?
What are the main impairments to innovation? What opportunities are there that could be used to improve the
way actors interact, and to encourage innovation?
TOOLS
Window reporting sheet
Understanding the social organization of
innovation
Approximation exercise II

Design
This window suggests a continuation of the discussions initiated in Tools A5/B8 and A5 (Approximation
exercises I and II), and an integration of the team's results into a more detailed and clearer picture of the way
actors interact for innovation.
Validity
This window stimulates the team to draw conclusions as to the convergences, resource coalitions and
communication networks that characterize the social organization of innovation. Are the missions and joint
missions of actors clear? Is the leadership of some or all key actors well established? Does this contribute to
achieving the agreed mission(s)? What stands in the way of adequate performance with respect to innovation?
Can configurations be identified? Are new networks emerging? What opportunities can be identified for
improving the performance of the system? How would this work?
The results of this window are as varied as the situations teams encounter. The most difficult aspect may be the
need to achieve a synthesis, given the richness of materials collected in a short time. In preparing for a
workshop, discussions within the team of alternative propositions, arguments, and ways of presentation must be
open-ended and inclusive at the start – yet towards the end they must become selective and decision oriented.
This requires skilful management of group dynamics and communication. Successfully representing the
eventual results in the form of drawings and a synthesis report requires a careful choice of figures and texts, and
if possible some professional editing. The presentation to the workshop participants needs to be complete but
concise, so that as much time as possible will be available for discussion.
WINDOW C1
Knowledge management analysis

What can be done to enhance performance related to innovation? This is the question addressed by the
knowledge management window. Performance may be seen from different angles, as it is by different actors –
therefore there may be more than one answer. Using this window, the team and actors can design positive
changes and/or actions for each of the objectives or mission statements identified earlier.
TOOLS Knowledge management analysis exercise

Defining possible actions
Design
As a first step, this window suggests that the team should make a basic decision with respect to the situation:
could the problems possibly be overcome by improving current networking practices among actors? Where this
is the case – that is, where there appear to be no structural impediments – the team has the option of choosing to
design a network improvement strategy. If, on the contrary, structural impediments have been uncovered (that
is, they cannot be overcome without a fundamental change in current networking patterns), the team may decide
to work towards a more demanding network re-configuration strategy. Such a strategy is intended to improve
networking efficiency and effectiveness. It implies the re-working of present social structures, including
configurations, convergences, coalitions, and communication networks. For example, a strategy with the aim of
privatizing extension and research would suggest a structural intervention. Such strategies originate from a
realization on the part of leading actors that existing configurations are unable to cope with new demands for
knowledge within the system, whether from farmers, extension workers, private institutional actors or others.
Improving networking strategies can be seen as a kind of knowledge management. While the usual sort of
management is generally impossible with respect to a knowledge system, a team can attempt to improve its
‘synergy’. (Synergistic relationships are those in which collaboration increases the effectiveness of both
partners – working together, each achieves more than they could alone.) This can improve the interactions and
contribute to the operation of the system as a whole.
The mission held by actors sets the stage for their activities. If there is agreement among actors, a single mission
may be taken as the point of departure. This provides a standard against which performance related to
innovation can be judged. Otherwise, the separate missions of different subsets of actors must be recognized
and treated in a parallel manner.
Validity
The concept of managing the knowledge within a given system suggests that a KIS can be designed or re-
designed to improve its performance. When a team begins to look at the system in this light, the information

collected earlier may take on new meaning. This window helps to make the transition from understanding the
system to making plans for the future.
Few simple recipes can be given to would-be knowledge managers. However, when actors work together to
understand their knowledge system, the recommendations generally refer to improved cooperation and/or
communication strategies. This makes wide participation and consensus among relevant actors overridingly
important to such management – otherwise, no matter how carefully the recommendations are formulated, they
may not be implemented.
WINDOW C2
Actor potential analysis
Who has the mandate to cooperate in making the changes seen as necessary for the successful performance of
the system? Do they also have the means to do this? Who is most interested in making these changes? The use
of this window acknowledges that no single person or actor directs complex social innovation processes.
TOOLS Actor potential checklist
Design
If the application of Window C1 has produced specific suggestions to improve innovative performance, C2
helps the team to review the support you can expect such innovations to receive from the actors, and to assess
the relevance of specific actors to successfully carrying out the changes. On the basis of their analysis the team
can propose particular new or revived linkages or coalitions, or measures to stimulate such relationships, among
relevant actors, joint projects or activities. This also requires looking at each actor's capacity to influence the
way innovation is socially organized.

Validity
An analysis of actor potential should pave the way for negotiations among actors. These may lead to building or
strengthening linkages and coalitions that can probe and decide on new missions and alternative options
(including technical options). In addition, such actors can look at and interpret the environment of the system –
the external factors that influence it – in ways that may enhance innovation. The validity of this window lies not
in its focus on any one aspect of the social organization of innovation, but in its contribution to the process of
exploring possible linkages and coalitions among stakeholders, pooling of resources, and so forth, to improve
performance related to innovation.
Active participation of stakeholders is a fundamental condition for the successful use and application of this
window. The window can only be used if a number of actors are willing to meet and assess their possibilities for
collectively improving the situation. Further, any joint action must fall within both the mission assigned to the
team and the mandates of the actors who are represented on it. If this is not the case, or if (even when the
RAAKS study has almost been completed) such willingness does not exist (or if no agreement can be reached),
the team may make suggestions; but in this case the actors must be left to draw individual conclusions and, if
they wish, to act upon these.
WINDOW C3
Strategic commitments to an action plan
What practical recommendations can be made to key actors and accepted by them? What will lead them to
contribute to the improved performance of the knowledge and information system? When the team has
discussed this, recommendations for interventions and strategies for cooperation and/or communication can be
drafted; their implementation can then be negotiated among key actors or selected key actors. As a part of the
strategy, the actors who took part in the analysis of actor potential may be asked to make a further commitment
to participate in these negotiations.
TOOLS Defining possible actions

Strategic commitments

Design
As the RAAKS cycle is completed, negotiations among actors are needed to reach at least partial agreement on
specific joint interventions and/or actions. Specific project proposals are necessary at this stage, with an
assessment of the resources that will be required to carry these out. Generally, taking decisions on such
proposals is not within the mandate of the RAAKS team members and participants. However, proposals can be
prepared, ready to put through the proper channels. Further, commitments can be sought from important actors
(not necessarily all relevant actors!) to follow up on the project proposals.
Validity
The validity of this window lies in its contribution to achieving verbal commitments on the part of relevant
actors to implement or participate in recommended actions. This is vital to the RAAKS study as a whole.
In the design of the RAAKS methodology, diagnosis is emphasized rather than project planning and
implementation. The need for tangible results is recognized, but accomplishing the concrete activities needed to
follow up on recommendations is left to the actors. The applicability of this window could be further enhanced
by designing specific proposals that facilitate the detailed design and implementation of the followup.
WINDOW
RAAKS' TOOLS
A
Materials in Tools help a RAAKS team to gather information systematically and to process it. To
manual and cards some extent, each tool represents a practical way to address a particular window: it
WINDOW
may be freely provides specific questions and working procedures. Although some tools are more
reproduced in encompassing than others, a tool generally limits the analysis to a narrower range of
limited numbers issues than the corresponding window. There is also some overlap among them. In B
for training and addition to being used in the analysis, the information collected can also become input

other educational for a focused discussion with actors, giving a clear view of ‘how actors see this
purposes. The element of the system.’RAAKS teams often begin by using the tools provided here. As
following you become more familiar with the approach and methodology, however, you may see
statement must aspects of the situation that you would like to explore in more detail. In this case,
appear on all teams should feel free to adapt the tools (perhaps by combining or extending them) or
copies: to develop new tools of your own.
Reading through the tools is also a good way to begin becoming familiar with
Reproduced from RAAKS. Each tool begins with expected outputs: what can we expect it to produce?
the RAAKS What is it supposed to add to our analysis? The relevant questions suggest topics for
resource box, the team to work through: areas to think about and types of information to collect to
Paul G.T. Engel help in understanding the social organization of the system. Working procedures
and M.L. provide more details; in some cases examples of the use of the tool or its results are
Salomon. Royal also given. Some Tools (A5, Approximation exercise II; B8/b, Understanding the
Tropical
social organization of innovation; C3/a, Defining possible actions; and C3/b, Strategic WINDOW
Institute, 1997
commitments) can be used to sum up the preceding phase or phases and prepare to
For large scale move on. Each tool bears the number of one or more related windows. If a tool is
reproduction or numbered, for example, ‘A5/B6’ it is relevant to both Windows – A5 and B6. The C
commercial use, Glossary mentioned in the cards is Appendix 4 in the manual Networking for
written innovation.
permission from
KIT Press, Royal
Tropical
Institute, is
required.
TOOLS TABLE OF CONTENTS
WINDOW Problem definition exercise Window: A1

Actor identification exercise Window: A2
A Actor objective sheet Window: A3
Environmental limits checklist Window: A4
Prime mover septagram Window: A5/B6
Approximation exercise I Window: A5/B8
Approximation exercise II Window: A5
Impact analysis sheet Window: B1

WINDOW
Actor analysis checklist Window: B2
Info-source-use exercise Window: B3/a

Communication network sheet Window: B3/b

B Source-intermediary-user sheet Window: B3/c
Linkage matrix Window: B4/a
Linkage mechanism checklist Window: B4/b
Task analysis sheet Window: B5
Basic configurations Window: B6
Communication analysis exercise Window: B7
Window reporting sheet Window: B8/a
Understanding the social organization Window: B8/b
of innovation
WINDOW Knowledge management analysis Window: C1

exercise
C Actor potential checklist Window: C2
Defining possible actions Window: C3/a
Strategic commitments Window: C3/b
WINDOW
TOOL A1
Problem definition exercise
Expected outputs
● A concise description of the area of human activity (see Glossary) the team is studying.
● A first listing of relevant or possibly relevant actors (see Glossary) and/or target groups.
● A specification of the problems as perceived by these actors and target groups, and identification of
preliminary problems and objectives for RAAKS.
Relevant questions
Which area of human activity are you looking at?
How diverse is this area (with respect to farming systems, agroecological zones, social groups, etc.)?
What actors are involved in this area?
What general problem or problems can you identify?
Who thinks there is a problem? What characterizes these actors?
Who is affected by the problem? What characterizes these actors/constituencies?
What is the history of the problem?
What are possible causes of the problem?
What problems or aspects are seen as urgent? By whom?
Do you think there is a problem?
Working procedure
The team works through the list of questions above, beginning with identifying the areas of human activity
involved, relevant actors, and problems or possible problems. One way of generating a list of problems is for
the team to have a ‘brainstorming’ session. The rules for such a session are simple: all ideas are acceptable; and,
during brainstorming, participants are not allowed to discuss or criticize each other's ideas. The idea is to
generate a list with lots of ideas, on a flip chart or overhead sheet. Next, the team narrows the list by discussing
the importance of the problems and the extent to which they can be grouped – for example, several items on the
list may all be the result of one underlying problem.
Each of the sets of problems selected then needs to be explored by the team in more detail, using the questions
on the above list. A separate flip chart page or overhead sheet should be used for each problem, to keep a record
of the results. If necessary, easy-to-read summary sheets should be prepared for later use.
At this time, it is less important to focus on finding the best possible problem statement or diagnostic objective
than to think about how to fit together the elements of the problem in one statement. An important point to
remember is that this definition of problems is preliminary – the team's view may change over time, especially
when discussions with actors begin! However, this tool gives a starting point and a means to explore team
members' ideas.
Among the considerations in choosing the problem on which the team will focus are the terms of reference of
the team, the interests of the team members, and of course feasibility – including the time and resources you

have available. A broad problem statement requires involving more actors (which takes more time for
interviews and analysis); also, team members may find it hard to relate a broad statement to their own interests.
On the other hand, it may be hard to motivate people to work on a problem that is too narrow. Further,
narrowing the problem too early makes it easy to miss vital information: it is important to make sure women,
landless farmers, recent immigrants and other sometimes forgotten groups are considered. The team must also
decide whether to accept the problem as stated (or implied) in its terms of reference, or to redefine the problem.
Either decision may create a necessity to negotiate with sponsors of the RAAKS effort: for example, a problem
statement based on a very specific TOR may not provide enough context to reach a real answer.
Some problem statements are listed in the example. Which of these do you think would provide the best basis
for a useful study? Why? (Additional examples of problem statements can be found in Case 2 in Chapter 2 of
Networking for innovation.)
Sample diagnostic objectives
In a study related to dairy farming, the following possible objectives for the RAAKS diagnosis
were identified:
● To identify the factors limiting the success of the dairy knowledge system.
● To formulate recommendations to increase the development and use of ecologically sound technologies
by small dairy farmers in the south of the country concerned.
● To identify factors that hold the national dairy knowledge system back from developing adequate
solutions to environmental problems in the intensive dairy industry, and what can be done to remedy this
situation.
WINDOW
TOOL A2
Actor identification exercise

Expected outputs
● A general list of system actors and possible actors, which is as complete as possible given the information
available at this stage.
● A first general diagnosis of the relative importance of each actor for the functioning of the system.
● A list of arguments to support this diagnosis.
Relevant questions
For the area of human activity defined by the team (Tool A1), which actors play a role? Try to be as
specific as possible. ‘Farmers’ for example are often not a homogenous category but need to be specified
as small/large scale, female/male, young/old, remote or not, or by ethnic group, etc.
Which actors play a significant role in technological innovation, policymaking, research, or exchange or
utilization of new or existing knowledge? Why and how? Be sure to include gender aspects.
Who else could make an important contribution? Why and how?
What do the various actors contribute? Is there a difference between mandatory and ‘de facto’
contributions? Why?
Which actors can be seen as key actors? Why?
What subsets of actors can be distinguished?
Working procedure
The actor identification sheet (see below) can be drawn on a flip chart or on overhead sheets, so that the team
can keep the information. The team then has a brainstorming session (see Tool A1) to generate a list of actors in
the system – again, these might be either individuals or organizations. After making a list of ideas, the team
discusses which of the actors listed are truly relevant, and which are to be considered key actors. Reasons for
the choices regarding key actors should be written on the sheet.
In Phase B the team may find it necessary to go into more detail on actors (the Actor analysis checklist, Tool
B2, will help at that time). This is relevant, for example, if the system needs to adjust to changing conditions,
and actors are seeking new partnerships.
Actor idenitification sheet

System actors Do you see this person/organization as a key actor? (yes/no) Why or why not?

1.
2.
3.
.
n
Resettlement project for youngsters in the Congo
The National Volunteers Organization for Development (ONVD) was founded in Congo in 1987.
The objective was to offer unemployed urban youngsters new possibilities in rural areas. Starting
capital was provided by UNDP, a private French institute, and the Dutch Ministry for
Development Cooperation. In 1994 this Congolese NGO provided housing, education and work
in agriculture for thirty male and three female youngsters. During a RAAKS study, the listing of
key actors given below was made by the team.
Key actors – Congo-Brazzaville project

Actors Key actors? In what way?
ONVD yes Provides housing, training and work for youngsters
Village Chief yes Provides access to village life
Youngsters
female yes
male yes
Parents of youngsters
from the village yes Provide social network for youngsters
from elsewhere no In effect, absent: group is small in number
Landowners
parents yes Provide access to land
state farms yes Determines availability of land
Consumers yes Provide market opportunities
Village farmers yes Compete with new farmers
Transporters yes Provide distribution to market
UNDP yes Provided initial funding
French private institute yes Provided initial funding
Dutch development cooperation yes Provided initial funding
Banks yes Provide loans for future investment
Congo Ministry of Agriculture yes Designs national agricultural policy

Congo Ministry of Cooperatives yes Designs national policy on coöperatives

Traders yes Determine market opportunities
Bakers yes Customer of bread
Subdistrict administration yes Implements government policy
Research yes Provide production technology for manioc,
bananas, beans
Source: Congo case discussion at the International Workshop on Agricultural Extension in Africa, Yaoundé,
Cameroon 24–28 January 1994
WINDOW
TOOL A3
Actor objective sheet
Expected outputs
● Statements of the objectives/mission statements of a number of actors, which will be used and further
developed during the RAAKS study.
● An overview of the arguments actors put forward and the criteria they apply in rating the degree of
importance of each of their objectives with respect to the direction agricultural development should take.
● A first assessment of the driving force(s) and actor(s) behind each of the different objectives.
Relevant questions
What do the relevant actors see as their objectives?
How does each of these actors perceive their contribution to the development of the process?
Who are the real beneficiaries of each of these objectives?
What technologies and/or activities are being developed or implemented as a result of each objective?

Which actors are crucial to implementing each objective?
Is there a shared objective?
Working procedure
Each team member will be able to think of several possible actor objectives, based on their own perception of
the system, and/or the information they have gathered. Sources might include documents (annual reports,
articles, speeches) as well as interviews with relevant actors. Generally no one objective can adequately reflect
the visions of the many different actors involved in the system. Therefore the team creates a large variety of
possible objectives, even including ‘unrealistic’ ones, to explore the boundaries of the system. These objectives
are discussed, compared, and evaluated by the team. One or several actor objectives are then selected as the
most relevant to use in beginning the RAAKS study. In the course of the diagnosis, the team will develop a
clearer picture of shared or conflicting objectives which may stimulate or hamper system performance.
Sample mission statements
The following examples are from (or suggested by) discussions with different parts of the horse
husbandry sector in the Netherlands. (See Chapter 8 in The social organization of innovation for
more information on this example.)
Those with different client groups may perceive themselves as having different missions:
● To meet the knowledge and information needs of professional horsekeepers.

● To meet the knowledge and information needs of professional and semi-professional horsekeepers.
● To meet the knowledge and information needs of professional, semi-professional and recreational
horsekeepers.
The area of concern can also lead to differences in mission statements:
● To provide professional horsekeepers with technical knowledge and information.

● To provide professional horsekeepers with knowledge and information concerning legislation relevant to
rural areas.
● To provide professional horsekeepers with knowledge and information on the horseback riding market
and the related consumer preferences.
Also, the problem addressed may vary:
● To prevent maltreatment of horses by amateur horsekeepers, by providing knowledge and information

on handling and treatment.

● To inform the public about the recreational possibilities of riding.
● To extend the infrastructure for horseback riding by informing municipalities of the number of their
citizens who ride, to encourage provision of facilities.
Each of these statements suggests a particular perception and emphasis. Each has specific
implications, in terms of the actors who are to be considered a part of the system, the roles they
might play, the knowledge products they might develop and the linkages they will find important.
Clearly some statements are compatible, and might be combined; others may even be mutually
exclusive!
WINDOW
TOOL A4
Environmental limits checklist
Expected outputs
● More clearly distinguished internal and external actors and factors; identification of external
agroecological and socioeconomic environmental factors – that is, factors in the ‘environment’ of the
system – that influence the problem situation.
● An inventory of resources and services external to the situation.
● An improved understanding of which environmental factors the actors in the system can or cannot
influence.
Relevant questions
Which factors (agroecological, cultural, socioeconomic, political, etc. – see the checklist below) play an
important role in determining the pre-conditions for developing technologies and/or services?
Which actors can influence such factors?
Which external actors put direct pressure – either positive or negative – on the actors within the system
with respect to developing technology and/or supplying services? How?
What relationships with the environment are seen as crucial to adequate performance of the system (or
those that need to be created)?

How complex is the environment within which the system must perform? Is it subject to rapid change?
Working procedure
The team brainstorms (see Tool A1) regarding the questions above. Answers are written on flip charts or
overhead sheets. If the team wants to go into detail on the first question, all or part of the checklist below can be
used. Throughout this tool, look for ways to make the results of team discussion visible – e.g. by making
drawings that show the relationships being discussed.
External factor checklist
Environmental complexity
● Degree of agroecological and production system diversity the system must deal with.
● Degree of cultural and socioeconomic diversity among the technology users included in the system.
● How stable or subject to change is the environment?
Available technology
● Degree to which the information needed within the system is already held by some of its actors. Does
their knowledge stay up to date? Do they generate new knowledge?
● The availability and accessibility of technologies that are relevant and can be adapted or modified for use
within the system.
External pressures
● Dominant types of policy pressures exerted upon the system.

● Degree of direct exposure to market demands.
External resource base
● Types of external resources used by the system and how these are acquired; degree of dependence upon
external resources.
Adequacy of agricultural services, marketing, inputs and communications infrastructure
● To what extent do agricultural services, marketing, inputs and communications infrastructure serve the
needs of the system?

Source: Adapted from Engel and Seegers, Towards a design and management of effective linkage strategies: a
diagnostic tool.
ISNAR R/TTL synthesis working paper. The Hague, ISNAR, 1992
WINDOW
TOOL A5/B6
Expected outputs
● Identification, based on actors' perceptions, of the ‘prime movers’ – those who give the leadership and
have the most influence on what happens within the system.
● A picture, in the form of several septagrams (see example), of the influence and/or leadership of each of
the prime movers as seen by different subgroups/actors.
Relevant questions
Who do different actors see as the prime movers in the system?
Which of these prime movers exert the strongest influence?
Who could change the situation and would be interested in doing so? Why?
Working procedure
Actors influence interactions within the system in different ways. For example, policymakers design and
implement policies and regulations, market actors influence prices, donors finance certain programmes,
research stations offer certain technological solutions, consumers choose certain products, agro-industries
favour relationships with particular producers and producers may favour specific techniques. Each actor
therefore has their own influence on the social interactions within the system. However, some actors may exert

more influence than others, so that coalitions appear around these ‘prime movers’. They may exert strong
leadership on the way the knowledge system functions, and hence on the type of outputs and impact the system
achieves. This tool focuses on identifying these prime movers and the degree to which they effectively ‘steer’
the system in a given direction.
To use this tool, the team asks each actor or group of actors to say how strong an influence each different type
of actor (internal or external) exerts upon the functioning of the knowledge system. This discussion can be made
visible by asking the interviewee to fill in a blank ‘septagram sheet’ (see example) consisting of a circle and one
line for each type of actor in the system. (There may be fewer lines than in the example – or more.) Each type of
actor is assigned a line; the group or individual being questioned is asked about each actor separately. They
decide where to place a sticker on the line representing this particular type of actors. The stronger (the more
‘controlling’) the influence of this type of actor, the further away from the centre the sticker is placed. The
weaker (the more ‘following’) the influence, the closer it is put to the centre. There may be more than one prime
mover in the centre. The use of a septagram in an interview is a good way of discussing and coming to
understand the perceptions of the person being interviewed.
Septagrams help to give the team a coherent picture of the system: actors may have similar or very different
perceptions (as shown in their drawings), which can have major implications for the knowledge system. The
team and other actors can work to put together a single picture that applies to the system as a whole. Further, the
team's attempt to sum all of this information up in one septagram is a good discussion starter for a workshop!
Septagram of a policy and donor-driven knowledge system

Scoring
5 = 100% controlling
1 = 100% following
WINDOW
TOOL A5/B8
Approximation exercise I: Leadership and coordination
Expected outputs
● A tentative system model, graphically displayed.

Relevant questions
Are all key actors involved in the RAAKS study? (Results from the actor identification sheet, Tool A2, and
the Actor objective sheet, Tool A3, can be used.)
Which ‘prime movers’ can be identified? ‘Prime movers’ are internal and external actors who take the
decisions that are most important to the system, and whose influence is important to other actors. Prime
movers may include investors, policymakers, suppliers of services and technology, users of technology and
services and consumers of final products. (Use the results from Tool A5/B6, the Prime mover septagram.)
What actors desire change similar to that desired by the prime movers?
If the system is to function well, what should be its objective? (Use the results from the Actor objective
sheet, Tool A3.)
Is one basic configuration prominent? Otherwise, which combination of basic configurations best describes
the coordination of the system? Use the results from Tool A5/B6, the Prime mover septagram, and the
definitions or results from Basic configurations, Tool B6.)
Working procedure
Drawing pictures – ‘approximation models’ – such as the one below (and the coconut tree in Chapter 2 of
Networking for innovation) is a way for a team to work as a group to sum up the way they visualize the system
they are working to understand. The results from the questions and tools mentioned above can be used to
produce several possible approximations. This makes it easier to decide what elements of the current situation
make a positive contribution to meeting the objectives of the system, and which do not. In Phase A this can be
very general, even metaphorical: in a study of the dairy industry, perhaps the industry will be represented by a
cow, surrounded by combatants labelled ‘extension’, ‘research’, ‘feed suppliers’ and so forth! The important
thing is to try and include the important actors (heavy arrows, stars or other marks could indicate which are
most important, or maybe each prime mover will wear a crown or cowboy hat…). Critical external factors need
to be in the picture too. What really determines the performance of the system?
Each model is discussed by the team to see how well it explains the phenomena you have observed. The model
or models that most closely resemble team perceptions are chosen. They can be used as a starting point for
Phase B of the RAAKS study (see Chapter 2 in Networking for innovation); also, they are useful in workshops
to get responses and see if the audience agrees with the team. In choosing one or more models, the team does
not have to achieve a consensus. It does, however, have to achieve clarity about the nature of the differences
between the models.
Finally the models that appear most useful are compared to the objective of the diagnosis: what sort of

performance can be expected from a system that conforms to the one or more tentative models? How does this
compare to the ‘desired’ state of affairs suggested by the objective? This discussion can yield important insights
that help to establish study priorities for Phase B, and ‘set the tone’ for the choice of windows in that phase.
Phase B covers more specific aspects of the KIS. In Approximation exercise I, the system may be pictured in
much more detail (depending on the windows addressed) than in Phase A. The aim is to make the tasks, the
hierarchy within which they are carried out and the ‘power relationships’ in the system more visible, so these
can be compared to the objectives of the system. One way to do this is to make cards of different shapes for the
actors and main external influences in the system. These can be stuck or pinned to a board (in such a way that
they can be moved easily during the discussion). Using the name cards, team members again construct several
possible models of the system, taking into account the actor objectives, their assessment of external factors and
actors that make up the environment, and the importance of each of these to system performance. As above,
discussion then centres on comparing the models, asking how well they fit the system and what they imply
about its functioning.
The example below shows a first approximation to a drawing of one subsector (the breeding subsystem) in
Phase B, using cards. (This study of the horse husbandry sector in the Netherlands is discussed in Chapter 8 of
The social organization of innovation.) One important feature of knowledge systems that comes out in this
phase are the linkages among actors. Linkages – especially the most dominant, important, and problematic ones
– are drawn here as lines. Different sorts of lines are used to show different types of linkages, while the squares,
circles and so forth indicate the type of actor. The important role of the pedigree organization (KWPN) shows
up clearly, as does the limited role played by extension and education. The landbouwschap (an organization that
protects the interests of farmers) has an important policymaking role, since it acts as an intermediary between
the breeding subsystem and the policymaking bodies of the state.
An approximation model: The horse husbandry sector

WINDOW
TOOL A5
Approximation exercise II: Problems and diversities
Expected outputs
● Synthesis of results from windows addressed in Phase A.

● A drawing of the knowledge system you are studying.
Relevant questions
Findings based on analysis of the material collected using Phase A windows

How would you formulate the diagnostic objective? Think especially about the area of human activity that
is being studied: how would you describe it?
What are the objectives of the most important actors? Is there some agreement among them that could be
seen as a shared objective? If not, why not? What key words or issues characterize actors' primary long-
term objectives? Why are these useful descriptions? Give arguments!
Make a drawing of the knowledge system as you see it; use for example Approximation exercise I (Tool
A5/B8) and/or the Basic configurations (Tool B6). Give arguments for the connections you draw.
Preliminary conclusions regarding the knowledge system as a whole
What are, as far as you can see now, three important problems that actors will have to deal with before you
can speak of the knowledge system as one that is functioning optimally? Why or in what way are these
particularly important?
Implications for the RAAKS study: do the results so far suggest a need for revisions in procedures?
Is the team getting the information it needs? Does this include information from and about women and
other groups that are sometimes forgotten?
What windows should be chosen for use in Phase B? Why?
Should any additional actors be included in the team, or in the interviews?

Exercise 4 in Appendix 2 of Networking for innovation can be used to check whether the team is making
the best possible use of its members' individual styles of learning.
How does the problem look in comparison to the team's terms of reference (TOR)? Does the team need to
attempt to negotiate amendments?
Working procedure
The team analyses the material collected from the literature and the interviews carried out using the windows
chosen for the first phase, with the questions above as a starting point. Results are discussed and written down
in a brief report.
WINDOW

TOOL B1
Impact analysis sheet
Expected outputs
● An inventory of outputs that different actors or clusters of actors expect to be available from the system.
● An assessment of the impact of the current system on the outputs listed.
● Identification of actors and factors that promote or hold back the achievement of the desired impact.
Relevant questions
How similar are the views of different actors or clusters of actors regarding what the system must achieve?
Are they very different or even incompatible? (See also Window/Tool A3, on mission statements, and B2,
Actor analysis.)
How effectively is the system organized to respond to the demands of various actors or clusters of actors?
How well do the outputs of the system meet the expectations of the different actors or clusters?
Is some form of evaluation built into the system?
Are there particular client groups whose needs do not receive attention?
Does the system generate any negative socioeconomic or ecological effects?

Is the system flexible enough to respond to changes in its environment, including the market, policy
targets, users and/or consumer demands?
Could the system be reduced (in size or in resources) and still retain its impact – that is, still meet the
objectives of the actors?
Working procedure

First, the team compares the actor objectives identified in Window/Tool A3 on tracing mission statements: what
are the differences among the actors? Do their objectives overlap? Do actors cluster around overlapping or even
similar objectives? Second, the team addresses the questions above. This provides an assessment of the outputs
expected by the actors or clusters of actors, as well as the actual impact of the system. Answers can be summed
up on an Impact analysis sheet.
Impact analysis sheet

Actors' objectives Clusters of actors Expected outputs Impact
To assess impact, the team needs to decide what standard (what ‘measuring stick’) will be used in defining
‘high’ or ‘low’ impact. In the example below, the degree to which knowledge is generated and exchanged
within the system is the standard. Actors' ratings of impact of a particular sort might be another possible
standard; so would for example ‘the degree to which the system supports ecologically sound agriculture’.
Note: in the case below, Tool B3/c, the Source–intermediary–user sheet, would be used before applying the
Impact analysis Tool (B1).
The horse husbandry sector in the Netherlands
Within the horse breeding sector in the Netherlands, seven client groups and ten knowledge types
were distinguished during the RAAKS study. A question mark after an item on the list indicates a
type of knowledge that seems to be needed, but appears not to be offered by any of the actors – it
is not currently available within the system. The types of knowledge identified were:
1a Basic knowledge related to veterinary care for horses
1b Advanced knowledge related to veterinary care for horses
2 Knowledge of horseback riding
3 Knowledge of horse breeding
4 Knowledge of training
5 Knowledge of entrepreneurship
6 Knowledge of consumer preferences (?)

7 Knowledge of knowledge transfer mechanisms (?)
8 Knowledge of public relations
9 Knowledge of regulations
10 Knowledge of legislation
The knowledge and information needs of the various client groups are shown in the following
table. For example, horse breeders need knowledge of types 1a, 1b, 3, 9 and 10. The extent to
which the needed types of knowledge are available within the system is shown for each type of
knowledge; an indication of the extent to which each client group is reached by the system is also
given.
Types of knowledge needed versus those available

Client groups 1 2 3 4 5 6 7 8 9 10 Client groups reached by system
Knowledge types a b
Horse breeders * * * * * ±
Horse trainers * * * * +
Riding school owners * * * * * * * * * +
Horse traders * * * * * * * −
Horsemen/horsewomen * * * * −
Stud owners * * * * * * +
Trotting and racing * * * * * * * * * * ±
Sportsmen
Knowledge types offered by system + + + + ± − − ± ± + +
This table suggests the conclusion that, in the horse breeding sector studied, there is a discrepancy
between the types of knowledge needed by several client groups and the knowledge available
within the system. In this example, the most important ones not available are knowledge
concerning entrepreneurship and knowledge with respect to market/consumer preferences. That
is, the knowledge system does not make available all of the types of knowledge that are relevant
to its client groups.
Further, the knowledge system does not make knowledge available to every client group to the
same extent. In particular, semi-professional (e.g. breeders and traders) and recreational client
groups (recreational horsemen and -women) are less well served by the knowledge system. This
also is an indication that the system is not very responsive to its resource-poor clients. (More
information on this example can be found in The social organization of innovation, Chapter 8.)

WINDOW
TOOL B2
Actor analysis checklist
Expected outputs
● Descriptions of the relevant actors (individuals or organizations), summarizing their most important
features in relation to the way they function within the knowledge system – in short, why they are
relevant. See the working procedure below regarding who is ‘relevant’.
● A more profound understanding of the functioning of particular actors, from their own point of view.
Relevant questions
Primary activities
What primary activities are carried out by individual key actors? Include both knowledgerelated activities
and other activities.
What resources are controlled by each key actor? This might include human resources – (wo)manpower
and qualifications; market share; budget; capital and/or land.
What is the knowledge base of each of the key actors? Include experience within the knowledge system
being covered, qualified (wo)manpower, staff skills, knowledge and information about the problems to be
addressed, contacts with complementary sources of knowledge and information.
How would you describe each key actor in terms of organizational capacity? This refers for example to
efficiency, effectiveness, flexibility, task orientation, bureaucracy, and transparency of policies/mandates.
Positioning within the knowledge system

How does each key actor define their primary function and mandate? If this actor has executives and field
staff, how do they perceive the function and mandate?
To what extent do the key actors consider themselves part of a ‘system’? Do they recognize that they and
other actors are mutually dependent? With whom?
In general terms, how does each key actor perceive the objective of the system?
Does the actor have policies related to their role in the knowledge system? Are these formally established
and agreed upon? With whom?
To what extent do the key actors know what other actors have to offer with respect to knowledge, skills
and technology/resources?
Impact on system performance
How relevant is the knowledge base of each actor to meeting the needs of the system's various client
groups?
How would you describe each key actor in terms of their status, influence, and power within the
knowledge system?
In general, what would you list as strong and weak points of each actor in terms of their contribution to
achieving each objective of the knowledge system?
Is any key actor capable of exerting effective leadership in the system? Does this occur at present?
What type of innovation is favoured by each key actor?
Working procedure
This tool helps in assessing which actors are most important to the system, describing them and their place in
the system in some detail. Note that often those like farmers or villagers – sometimes seen as simply
‘beneficiaries’ or ‘target groups’ rather than actors – may be highly relevant to the functioning of the system;
therefore they must be included in the assessment. See the section on ‘The team and supporting groups’ in
Chapter 3 of Networking for innovation.
First, the team should attempt to answer the questions above, based on the information that has already been
collected. If additional information is needed, can it be collected from the sources you have already found, or is
a further search needed? For some questions, quantitative data can be collected, e.g. numbers of professional
and technical staff, operational and personnel budgets, and financial allocations to relevant lines of research.
Other questions, such as the capability of actors for effective leadership, require that the team and the actors
involved use their best judgement to provide answers. A summary sheet such as the one below can be used to
record your results.

Actors: Summary sheet

Position in the Impact on system
Actor Primary activity
knowledge system performance
WINDOW
TOOL B3/a
Info–source–use exercise
Expected outputs
● Matrices showing most-used information sources, for each type of information and actor; these suggest
the comparative relevance of these sources in providing information to clients.
● An indication of the relevance of the information available.
● An indication of the extent to which each source reaches relevant users.
Relevant questions
What sources of information (actors, products, media, etc.) do the various actors use regularly?
What types of information (e.g. strategic, operational, technical, policy or market information) do they
obtain from these sources?
In what ways do they put it to use (such as shaping their corporate image, determination of client's needs,
comparison of alternative possibilities, and/or implementation of solutions)?
For each type of information, what are the ‘top five’ most important sources used by each actor?

Working procedure
In using this tool, the team identifies the value actors place on the services of those who provide knowledge
within the system. This process shows which sources are the most frequently used, and which are considered
the most important. The outcome is often very revealing, showing differences among the actors involved,
whether the services provided reach the intended target groups, the extent to which these target value the
services, and what improvements are needed.
Agricultural producers in Aste, the Netherlands
Tables such as those below can be used to summarize the information collected. The categories
used in both of these tables can of course be revised by the team to fit your own situation. These
examples come from a case in Asten, the Netherlands, involving the agricultural communication
network (see Case B, Chapter 2 in The social organization of innovation for more information).
The first table shows that these farmers use a widespread, diverse network of sources to get the
information they need to manage the farm. Actors engage in different types of relationships, each
offering distinct types of support, including information. The table gives the frequency with
which farmers use priority sources, for several types of information. Strategic information has
direct relevance to questions or issues of a strategic or tactical nature; operational information has
immediate relevance to the daily management of the farm business; technical information relates
to technical and scientific developments; policy information has to do with rules and regulations
affecting farm operations; and market information concerns the development of prices, consumer
demand, and national and international developments.
Frequency of use of sources, by type of information

Source: van Dijk, Engel and Leeuwis, Evaluatie AGROCOM proefproject. Wageningen, Vakgroep Voorlichtingskunde LUW, 1991
The second table shows how farmers use information from the sources identified in the table
above. One remarkable feature to be observed in this table is the broad influence of product and
service-related advice. These advisers seem to be tied into all of the major decision making
processes at farm level. Colleagues and independent extension advisers, as well as the accounting
bureau and the manufacturers of farm technology are consulted in a more limited, more specific
manner. Again, the role of the farm journals is evident. Although they do not appear to play a role
in other forms of information use, their role in general opinion formation, in maintaining a vision
of what is going on, is pivotal. Farmers rely on these journals, with their broad coverage of issues,
to keep up with current developments that may affect their operations.
The four categories that appear across the top of the second table relate to this particular RAAKS
study: information used in ‘image formation’ helps the farmer to keep up with what others are
doing; determination of client's needs deals with the identification of problems or need for
specific interventions; information in the third category is used for comparison of alternative
solutions or interventions; finally there is information used for implementation of a selected
solution or intervention.

Use made of most important information sources
Source: van Dijk, Engel and Leeuwis, 1991
WINDOW
TOOL B3/b
Communication network sheet
Expected outputs
● A communication network (see Glossary) for each actor.

● Insight into the positions actors have within different communication networks.
● Increased understanding of the relevance of actors to each other.

Relevant questions
For the actors and information services identified in Tool B3/a, how close is the information source to the
user?
What value does the recipient put on the information provided by the source?
If providing information is the mission of the source, is this being achieved?
Working procedure
A communication network sheet – the concentric circles shown in the examples – is used to make actors'
relevance to each other visible. The information gathered can be used to construct a picture of the
communications network of a given actor or type of actor. It should be filled in on circles drawn on a flip chart
or on overhead sheets, so that it can be kept. These sheets show how actors position other actors in their
networks, what types of information are available to whom and which actors are prominent in all networks. The
type of actor being interviewed is put in the centre; the team decides where to place other actors in the diagram,
depending on their relationship to the actor in the middle; other aspects can be shown by the area an actor
occupies and so forth.
This tool can be used in various ways; it could also be used more than one way for a given actor. In the second
drawing below, the positions of actors in networks is being explored using questions about who provides
advice, information via various media, or ‘documentation’ on specialized topics (such as technical innovations,
rules and regulations, or financing). More information about Example 1 can be found in Case B, Chapter 2, in
The social organization of innovation. (The other examples are simply for purposes of illustration, to show
various ways to use this sheet.) In Example 2, information about actors relates to their ‘closeness’. For example,
for farmers, does a particular actor visit farms, or not? Example 3 on the other hand is based on the degree of
relevance: one actor may be quite relevant (or not) to another, without being ‘close.’ If such alternatives are
used, the team needs to have a clear definition for itself of words like ‘closeness’ or ‘relevance,’ such as those
given in the examples.

Example 1 Positions of actors drawn to show the type of information they provide to the actor interviewed
Example 2Positions of actors drawn to show their closeness to the actor interviewed

Example 3Positions of actors drawn to show their relevance/importance and closeness to the actor interviewed
(Figure shows the agricultural product communication network in Asten, the Netherlands)
Closeness can be defined in various ways – for example, frequency of visits or other interactions:
● inner circle = frequent contact

● outer circle = few or no contacts
Relevance/importance is indicated by size of the shaded space:
● big = very relevant/important

● small = not so relevant/important
● closeness to actor interviewed is indicated by position, as in Example 2
WINDOW
TOOL B3/c

Expected outputs
● One set of actors for each type of knowledge that has been defined.
● Insight into gaps and overlapping among knowledge networks.
● Insight into the quality of the networks supporting the generation, use, and transformation of particular
types of knowledge and information within the system.
● Insights into reasons that the system serves some client groups more adequately than others.
● Observations concerning the ways scientific, technical, and practical knowledge can be integrated to
produce workable solutions.
● Observations concerning the emphasis particular types of knowledge and information receive within the
system.
Relevant questions
How does the fact that particular products are developed by particular actors affect the availability and
relevance of these knowledge products for specific client groups?
Which actors could be integrated in the network to improve linkages and thus increase availability and
relevance for a particular client group?
Does the composition and integration of the networks contribute to over-or underrating the importance of
some types of knowledge? Does this work against achieving one or more of the objectives of the system?
Working procedure
The name of this tool emphasizes that farmers and other actors may be not only users of knowledge; they may
at the same time be sources of knowledge or intermediaries – actors who channel knowledge or knowledge
products from one or several actors, so that they reach other actors. Making use of the information gathered to
answer the questions above, the team fills in a Source–intermediary–user sheet. For each type of knowledge,
this identifies a set of actors related to its production, transformation, and use: the knowledge network for this
particular type of knowledge. ‘Sources’ of information are filled in, along with their ‘knowledge products’;
intermediaries, with the information (knowledge products) they pass along; and the users of this information.

The next step is to compare these networks, to see if there are gaps between the information offered and the
information needed. On the other hand, there may be bridges between the networks – actors who participate in
more than one network (‘gatekeepers’). Some networks may overlap to such a degree that they can be
combined. The transformations that occur within a network – for example, making it possible to put more
theoretical knowledge to practical use – can be studied by comparing the products actors receive with those they
pass on to others.
Knowledge Sources Knowledge Inter- Knowledge User
type product mediaries product
Looking at these networks also may provide some insight as to reasons some client groups are served better
than others by the system. Comparisons with information gathered in an Actor analysis (Tool B2) may provide
insights into the limitations (whether inherent or self-imposed) actors deal with while performing their roles in
the system. Furthermore, in cases like the one below, information gathered about networks can be used for
Impact analysis (Window and Tool B1).
A knowledge network in the Netherlands horse husbandry sector
The table shows the information that might be filled in on the Source–intermediary–user sheet for
just one of the knowledge types (knowledge of breeding) identified in the example given for Tool
B1, Impact analysis – the horse husbandry sector in the Netherlands (see Chapter 8 in The social
organization of innovation). This gives the team a way to analyse the knowledge network. As can
be seen in the table, the pedigree organizations (especially the largest ones) play an important role
in the breeding knowledge network. They are active not only as a source of knowledge but also
through their magazines, which are another example of an intermediary. These magazines seem
to have a great deal of importance or potential importance, because they are often the only
communication channel between breeders and institutional actors. Also, the magazines reach
large numbers of people in a sector that is characterized by its diversity.
Breeders' networks related to breeding: A knowledge network analysis

Knowledge Intermediary
Sources Products Products User
type /ies
Knowledge - Pedigree Management Breeding recom- -Members of
- Pedigree org.
of breeding organization info. system mendations of breeding org.

Contributions - Members
- Pedigree org.
Practical in meetings, - Breeders
- Breeders - Breeders
experience evaluations, - Breeders
- Extension
etc. visited.
Visits, written
- Universities Research results - Extension
articles, etc.
- Pedigree org. - Members
- Research Research results, Articles, other
- Research - Horse
station publications publications
station keepers
Integration of
- Landbouw-
knowledge on - Members of
- Landbouw- schapa schap Visits, articles
e.g. venereal Pedigree org.
- Pedigree org.
diseases
a An organization that works of protects the interests of farmers
WINDOW
TOOL B4/a
Linkage matrix1
Expected outputs
● A linkage matrix, summarizing information on the existence, frequency and intensity of contacts between
pairs of system actors.
● Visible clusters of actors – very intensely linked to each other (system segmentation).
● Insights into the degree to which system segmentation stands in the way of optimum performance of the
system.
Relevant questions
What contacts are there among actors?
Who initiated these contacts?

How relevant are they (frequency, intensity, value)?
Is the linkage one-way or two-way?
Are contacts formal or informal?
What resources are available to maintain these contacts?

What clusters can be identified?
Do these clusters coincide with those seen while using other windows and tools? (Such as Actor objectives,
Window/Tool A3; Network analysis, Window B3 and Tools B3/b and B3/c; and the Prime mover
septagram, Tool A5/B6.)
Working procedure
The linkage matrix helps in making a systematic description of the linkages or interfaces that can be seen
between pairs of actors. Differences between actors with respect to their linkages can be described with respect
to intensity (tightness of the connection) or frequency of contacts or other factors. This focuses attention on the
degree of integration within the system: how much do the actors have to do with each other? The main aims in
using the linkage matrix are to identify clusters of actors, missing linkages (what additional linkages would
change the way the system functions, and in what way?) and overlapping.
The matrixes are a way of summarizing and analysing information collected during interviews with actors. You
will not be able to predict exactly which actors will be listed in the matrix beforehand. Instead, ‘who has contact
with whom’ comes out in the interviews. If you learn of additional important but unexpected contacts, more
interviews may be necessary.
Two examples illustrate this process. One shows the use of a linkage matrix in considering a quite varied group
of actors; the other is specific to linkages among institutions. In the example from Kenya, the team simply filled
in an X wherever those associated with an actor reported in interviews that they had contact with another actor
(for example, financial institutions reported contact with coops, chemical companies, farmers and extension
workers). If possible, you can ‘score’ these responses, as in the example from Columbia. Even a simple scoring
system like this makes it possible to summarize more information in the matrix, so that it is more useful. The
matrix in the second example also reports on two time periods, so they can be easily compared. If a scoring
system is to be used with this tool, the team must define the scores. Often a scale of 1 to 5 is used, with 1
indicating for example ‘very frequent, formal and informal contacts sought’ and 5, ‘hardly any contacts sought
– neither informal nor formal’. Or, if scores are to be based on the importance of linkages, scores could range
from 1 (most important) to 5 (least important).

Flower farming for the expot market in Kenya
In preparation for the International Workshop on Agricultural Extension in Africa (24–28 January
1994), a survey related to the innovation process in flower farming was carried out in Kiambu
district, Kenya. Flowers were a newly introduced crop here. They require sophisticated new
methods of production with which farmers were not familiar. The survey covered contacts among
the actors; these were summarized in the linkage matrix shown below.
In this case, the matrix does not indicate the frequency of contacts. However, we can see that:
● farmers and extension workers report contacts with most other actors;
● there are few contacts among other actors. In other words, there is little integration among actors;
● both points suggest the conclusion that farmers and extension workers may be actively seeking
information on flower farming. This is indeed the case. Farmers took the initiative to begin the
development of flower farming. Together with agricultural extension officers, they identified problems
and needs. Other actors such as the local research station, private horticultural specialists, Hoechst, farm
chemical companies and the agricultural finance corporation became involved later.
Source: E.O. Wanga (1994) Kenyan case study. Performance of the agricultural knowledge and information
system in Kiabu district. Paper presented at the International Workshop on Agricultural Extension in Africa,
Yaoundé, Cameroon 24–28 January 1994
Matrix 1
Identification of actors and linkages
Rural development institutions i Nariño, Colombia

Institutional integration among core institutions within the agricultural knowledge system was
studied in the periods 1975–1978 and 1978–1985 in the Nariño Highlands of Colombia. Matrix 2
shows two linkage matrixes – one for each period.
The matrix on the left relates to the situation before an integrated rural development project was
undertaken in the Nariño Highlands (1975–1978), while that on the right covers the situation after
the intervention (1978–1985). As can be seen in these matrixes, integration between the
institutional actors increased during the project. During this period of higher institutional
integration, other results (not shown) indicate that yields of wheat, guinea pigs (curies) and dairy
cattle (key project activities) increased, while those of two other activities (maize and beans)
stagnated. In the three successful activities, a number of specific linkage mechanisms (such as
collaborative task groups, subject matter specialists and on-farm trials) were observed to be
operating: these were absent with respect to maize and beans.
Here again, the matrix is a way of summing up the statements made by the actors in interviews. A
simple scoring system was used: if actors reported strong formal and informal inter-institutional
linkages, a ‘1’ was entered in the matrix; if there were only informal linkages, a ‘2’ was used.
Looking across the matrix from left to right, the number refers to what the actor on the left said
about contacts with other actors. Thus above the diagonal in the matrix for 1975–1978, we see
that extension did not mention contacts with research; under the diagonal, however, research is
seen to report limited contacts. In 1978–1985 limited contacts were reported by both. Overall, the
matrixes suggest that with a change in mandate, activities and thus contacts also changed.
Source: Engel, P.G.H. The impact of improved coordination on agricultural performance. The case of the Nariño
Highlands in Columbia. The Hague, ISNAR, 1990
Matrix 2
Integration of core institutions – Nariño

WINDOW
TOOL B4/b
Linkage mechanism checklist
Expected outputs
● Descriptions of individual linkage mechanisms in terms of a list of attributes.

● Comparisons of linkage mechanisms with respect to these attributes.
Relevant questions
A linkage mechanism checklist
Origin of the mechanism: was it initiated by a particular actor or cluster of actors?
Who are the participants in the mechanism?

Describe their activities.
When was the linkage used? What task did it carry out?
Specific outputs?
Impact of the mechanism on performance?
Is it a facilitative or a control mechanism?
Does it have an officially mandated purpose? What?

Degree of formality?
Type and quantity of resources exchanged among actors, if any?
Type, quantity and source of resources required for the linkage mechanism to function?
Who provides/controls these resources?
What is the relative importance of the resources exchanged?
What is the frequency, duration and permanence of the exchange?
Administrative level at which the mechanism operates?
Evidence and perceptions about the competence of participants in fulfilling their functions?
What, if any, decision making powers are attached to the linkage mechanism under examination?
Working procedure
Linkage mechanisms are arrangements that help to link up the parts of the system. They may do this by
contributing to communication (meetings among farmers, with extension workers or liaison offices would be
examples of such mechanisms), coordination (e.g. mutual adjustment of activities, or water distribution) or
resource transfers (perhaps credit, salary payments, or sharing labour). The team can begin with the information
from Tool B4/a – what sorts of mechanisms support the linkages identified there? The checklist above should
be filled in for each linkage mechanism. Comparisons can then be made among the mechanisms. In both cases,
team discussion should be used to help clarify the mechanisms; they can then be summarized for future use.
WINDOW
TOOL B5

Task analysis sheet
Expected outputs
● A task diagram for the knowledge system under consideration.

● Insight into gaps and overlapping in the functions of actors in the system.
● Greater understanding of linkages between actors who perform different but complementary tasks.
● Insight related to the effect of system segmentation (clustering of actors) on the implementation of tasks
and functions in the system (in combination with the Linkage matrix, Tool B4/a).
● Increased clarity on the quality of knowledge and information generation, transformation, and use by
various parts of the system (in combination with Impact analysis – Window and Tool B1).
Relevant questions
Which tasks/functions are performed by which actors?
What activities do the actors carry out in the process of performing these tasks? How effective is this?
What gaps are there between tasks? Is there some overlapping?
Do the tasks/functions that are carried out match actors' expectations for the system?
Do actors who are within the same system or subsystem see themselves as interdependent? As
complementary?
Is there a coordinated effort among the most relevant actors to integrate their tasks?
Which subsystems, networks or clusters exhibit high performance in terms of matching and integrating
tasks among actors?
What factors within the system have a positive or negative influence on task performance? In the context
where the system is found, what factors help or hinder optimum performance of tasks?
Working procedure

This tool helps a team to identify and visualize the tasks carried out within the system. The coordinating
linkages can then be seen in the patterns of contacts. In the Coconut KIS example here, the tasks explored are
the knowledge tasks of the continuum that connects policy–science–practice. During interviews, the team writes
down the tasks that actors say they carry out. These are then summarized in a matrix: the procedure is to list
tasks on the X-axis, and the actors who perform these tasks on the Y-axis. The tasks carried out by particular
actors can then be plotted. (See the example; the coconut system case in Chapter 2 of Networking for innovation
gives more background information.)
The resulting matrix produces information about gaps and overlapping among essential tasks various actors say
they carry out. Here for example GTZ reported tasks related to finance, programme management and
evaluation. (See Tool B4/a, for more information on reading a matrix.) The task matrix does not, however, give
information about the performance of tasks or how well they are carried out. To obtain this sort of information,
use the Impact analysis sheet (Tool B1): it provides an analysis of the performance of the system. The results of
the analysis from the Source–intermediary–user sheet, Tool B3/c, add information about the functional linkages
among tasks. Actor analysis (Tool B2) may provide additional information that makes it possible to deduce the
tasks of specific actors.
A task analysis can be used in discussing which additional tasks will be needed to have the system function
better. It can also be used to sum up this information. Further, a task analysis can also be used in interviews or
in workshops, to check whether various actors can identify tasks that have been missed.
Coconut knowledge and information system in Tanzaniaa
aThis example was produced by students at Wageningen Agricultural University, the Netherlands, during a course on RAAKS in
1993. Abbreviations:
IDA International Development Agency

GTZ German Technical Development Cooperation
MALD Ministry of Agriculture and Livestock Development
NCDP National Coconut Development Programme

CES Coconut Extension Service
Sources: Lupanga, I.J., ‘The Tanzania case study’. Paper prepared for the research-technology transfer linkages case
studies workshop. February 28–March 3. The Hague, ISNAR, 1989a; Lupanga, I.J., ‘The national coconut development.
Subsystem Draft Report’. Paper presented at workshop on research technology transfer linkages, the Hague, the
Netherlands, 25–30 September, 1989. The Hague, ISNAR, 1989b
WINDOW
TOOL B6
Basic configurations
Expected outputs
● A multiple-system model, reflecting the coordinating mechanisms that are dominant in the system and the
influence of different types of prime movers (see the Prime mover septagram, Tool A5/B6); together,
these suggest the basic configurations that are present.
● Possibilities for intervention by the actors involved in the way coordination is achieved within the system.
These one or more potential interventions are intended to improve performance with respect to their most
important objectives.
Relevant questions
What patterns of relationships can be seen among the actors, and especially the prime movers, in the
system?
Which actors are dominant?
What sorts of coordination occur now? What gaps and overlapping can be seen with respect to
coordination?
Working procedure

The team uses cards or drawings such as those made during Approximation exercise I (Tool A5/B8), plus the
septagram sheets (Tool A5/B6) and/or the ‘tree’ shown in Case 2 (Chapter 2 of Networking for innovation) to
create a picture of the system being studied. This identifies the prime movers and dominant relationships. For
each system, subsystem or network, the team may also choose to make a separate analysis.
There may be several types of prime movers – for example:
● donors
● policymakers
● marketing or processing industries
● international research
● users (for example, farmers).
When prime movers have been identified, it becomes possible to show their interrelationships graphically. Such
a picture helps to identify the relatively stable patterns of institutional relationships among actors and the
balance of power among them. These patterns are related to the ways the system is coordinated – how the actors
coordinate among themselves to reach their objectives. Some of the coordination mechanisms found in
organizations are outlined below.
Coordination mechanisms foun in organizations
Direct supervision
A boss or supervisor takes responsibility for coordinating the work of others.
Standardization of skills
The need for coordination is reduced by specifying the skills needed to perform the work, either
by training or by hiring workers who have these skills. (Example: nurses in a hospital.)
Standardization of work processes
The need for coordination is decreased by specifying ahead of time the activities that are to be
performed. Requires little or no communication, but is useful only for routine tasks.
Standardization of outputs
The need for coordination is decreased by specifying the results that must be achieved (setting

standards), leaving the method up to the worker.
Mutual adjustment
Coordination through simple informal observation or communication; appropriate to small groups

of people working side by side. However, when the situation is complex and the tasks at hand
cannot be standardized, mutual adjustment again becomes the preferred method of coordination.
Perhaps you have experienced this in your team!
Standardization of norms and/or behaviour
Coordination is achieved as a result of a shared organizational ideology.
Source: Adapted from Mintzberg, H., Structure in fives. Englewood Cliffs, New Jersey, USA, Prentice-Hall, Inc.,
1983
Basic configurations
The basic configurations shown below will help as you look for a model to describe the situation you are
studying. These configurations are patterns of institutional relationships. They are ‘pure’ forms, each with one
dominant actor and one dominant coordination mechanism, as stated in the table and illustrated in the figures. In
reality, you will almost always find a mixture of several basic configurations and several types of coordination
mechanisms, with much less distinct boundaries. Furthermore, different subsystems may be coordinated in
different ways. This often gives rise to a multiple system, as shown. There is no need to attempt to find one
configuration that fits the system exactly; instead, the question is which configuration or configurations are
dominant, and help the most in explaining the behaviour of the system. (Configurations are discussed and basic
configurations are described further in Chapter 4 of The social organization of innovation.)
Basic configurations and dominant coordination mechanisms

Industry-driven knowledge system Standardization of outputs/work processes
Central policy-driven knowledge system Direct supervision/standardization of norms/behaviour
User-driven knowledge system Ideological adjustment/standardization of norms
Technology-driven knowledge system Standardization of work processes/skills
Donor-driven knowledge system Standardization of outputs/mutual adjustment
Basic configurations in agricultural technology systems

Source: Engel, P.G.H. and S. Seegers, ‘Basic configurations in agricultural technology systems.’ Summary. In:
Proceedings of the international workshop on agricultural knowledge systems and the role of extension. Bad
Boll, Germany: 21–24th May 1991, pp. 353–358.

There is also no ideal type of configuration for a given situation. The configurations you see in practice have
historical roots in the countries and organizations where they have developed. When a system does not work
well under the present circumstances, this should be taken as a challenge to modify, rather than replace, the
existing coordination mechanisms. The aim is to find a solution that achieves coordination in a manner more
appropriate to the objective(s) of the system. For example, the design or re-design of a project intervention
could encourage the use of direct supervision as a coordinating mechanism (as in early T&V project designs).
On the other hand, as seen from the list in the table, there are other options. Coordination could also be achieved
by putting more emphasis on the standardization of outputs (e.g. by introducing packaging technologies) or
through particular work processes (such as FSR/E – farming systems research/extension), or by encouraging
mutual adjustment between professionals and farmers (using techniques such as functional groups or other
people-centred approaches).
Figures such as those shown can be used to picture and discuss many factors. For example, in the industry-
driven model, key relationships are commercial farmers–consumer markets and commercial farmers–agro-
industries. Key actors are these farmers, marketing agencies and agroindustries. The farmers are also the
primary client group; standardization of work processes/outputs serves as the coordination mechanism. In this
case, the extension approach favoured is specific to relevant commodities.
The team can begin its discussions by attempting to apply one of the basic configuration models to the system
being studied, then evaluating the pros and cons of this kind of simplification – does it help to explain how the
system functions? Often you will find that different subsystems are coordinated in different ways. Such
discussions allow the team to work towards a multiple image of the knowledge system. This is more apt to
resemble the much ‘fuzzier’ picture in Figure 4 of Networking for innovation than the neat models here! The
result is a picture that reflects the team's perception of how coordination actually takes place within the system.
In addition, the team tries to formulate ways to improve coordination: how can a closer fit between the current
situation and the coordination needed to improve system performance with respect to its different actor
objectives be achieved? It is useful to state action possibilities in terms of strengthening one or more
coordinating mechanisms in relation to others.
WINDOW
TOOL B7
Communication analysis exercise

Expected outputs
● Identification of social and cultural differences among actors, as well as differences in their knowledge
and perceptions.
● Identification of constraints and opportunities related to communication among the actors.
Relevant questions
Do the actors have similar or different opinions on the nature of the problem, and on the objective to be
achieved by the knowledge system?
Do the world views, ways of reasoning, social circumstances, languages and/or ideologies of the actors
differ?
How do these differences influence communication among actors and/or between actors and prime
movers?
Working procedure
The team tries to assess whether there are fundamental differences of opinion among actors or groups of actors
regarding the nature of the problem situation and the objective to be achieved by the knowledge system, or if
there are differences in their world views (or in other ways listed above). In this case, the analysis focuses
particularly on differences among the actors in their assumptions about the nature of development and the
problems of the knowledge system. Can those with very different backgrounds or social positions understand
each other sufficiently to communicate? If there are problems, are they technical or social? Even if no
fundamental differences are identified, discrepancies in perception and reasoning between system actors are
worth exploring. What themes do they consider important – for example, is ecological sustainability seen as
more essential than safeguarding income, or the reverse? How do they view the need to maintain a position in
the international market? Such considerations should include the specific meaning assigned to key words by
particular subgroups, even when both parties speak the same language. Finally, the third question regarding the
impact of these factors on communication within the system can be addressed.
Actor differences
Doing collaborative research with farmers requires that researchers and farmers learn to
understand each other! Not only language but also ways of reasoning are important. Nitsch, for
example, contrasts the adaptive rationality of farmers with the formal rationality often

recommended to them by scientists. In an adaptive rationality, decisions ‘grow out of a synthesis

of information and the owner-manager's own experiences and aspirations’.a This can be
contrasted with the formal decision making or problem-solving process that is often the basis for
the decision support systems developed by scientists.
Less obvious, but potentially no less damaging, are the differences in ‘culture’ between ‘blue
collar’ field staff and ‘white collar’ management, between women and men, researchers and
extension workers, or government and non-government workers.
WINDOW
TOOL B8/a
Window reporting sheet
Expected outputs
● An overview of the contribution of the windows used, and their results, to understanding the performance
of the system.
● Identification of constraints and opportunities met in applying the windows.
Relevant questions
Relevance: why was this window chosen? What is its relevance or expected relevance to deepening our
understanding of the problem situation?
Main issues: what main issues were addressed? Or: how was the window designed or redesigned for use in
this particular situation?
Information gathered: most important findings or information obtained.
Preliminary conclusions: what conclusions has the team or subteam suggested, based on its analysis from
the point of view of this window? This might for example include:
– conclusions on the network in general;
– conclusions on the problem situation, constraints and opportunities for action.

Evaluation: did the analysis yield interesting ideas or insights? Were these in accord with the results
expected beforehand? Did it yield any unexpected insights? What sorts of insights did it help to generate
that differ from ideas generated by using other windows?
Working procedure
The team evaluates each window used during Phases A and B of the RAAKS study. This can be done by
discussing the questions and/or writing a brief report (maximum two pages). If a report is written, it can become
a part of the team's final report.
WINDOW
TOOL B8/b
Understanding the social organization of innovation: Constraints

and opportunities
Expected outputs
● Synthesis of results from windows addressed in Phases A and B.

● Identification of the constraints and opportunities of the knowledge system.
● Preliminary identification of possible solutions.
Relevant questions
Findings based on analysis (e.g. from Tool B8/a, the Window reporting sheet) of each window used

What convergences, resource coalitions and communication networks can be identified? Include often-
forgotten groups e.g. women's networks. (See also ‘Understanding the social organization of innovation’ in
Chapter 1 of Networking for innovation.)
What are some possible constraints to the optimal functioning of the knowledge system, as seen from the
standpoint of different windows? Why are these important? Give arguments!
Preliminary conclusions regarding the system as a whole
What are the objectives of the most important actors in the knowledge system? Is there some agreement on
a shared objective? Are there marked differences in the objectives of women and men, or between other
often-forgotten groups and other stakeholders?
What are the three most important problems the actors will have to deal with before it will be possible to
speak of an optimally functioning knowledge system? Why?
Looking back at the figures in Tool B6, do you see a ‘pure’ basic configuration in operation?
What possibilities do you see for improving the functioning of the knowledge system? Why would this
help?
What can be done to achieve such improvements? By whom?
Implications for the RAAKS study: as you look forward to Phase C, where you will work on making
recommendations and achieving commitment to a plan for the future, do the results so far suggest a need for
revisions in procedures, or a need for further work before moving on?
Is the team getting the information it needs, including information about ‘forgotten’ stakeholders?
Should any additional actors be included in the interviews, workshops and so forth?
Is there a need to work through additional Phase B windows or tools?
Is any other preparation needed before moving on to Phase C? Why?
Working procedure
The team analyses the material collected from the literature and the interviews carried out using the windows
chosen for the first phase, with the questions as a starting point. Results are discussed and written down in a
brief report. Note: if you feel that vital information is missing, consider using additional windows/tools; also see
the section on workshops in Chapter 3 of Networking for innovation.

Complemenarity among tools
In analysing your material and answering the questions above, look for ways to compare, combine and contrast
the information you have gathered by using various windows and tools. This can help the team to see how the
system is connected, and whether any new structural forms, such as convergences, coalitions, linkages and so
forth, are emerging – or missing! For example, the results from the linkage matrix can be used in combination
with results from the Task matrix/task analysis sheet, Tool B5 and the knowledge network analysis achieved by
using the Source–intermediary–user sheet, Tool B3/c. Also, combining the linkage matrix with the results of
task analysis allows an evaluation of the linkages among the essential tasks in the system. Using the linkage
matrix in combination with a knowledge network analysis, clusters of contacts can be compared with the
networks of actors supporting the generation, transformation and use of certain types of knowledge. This allows
the team to ask whether the knowledge networks do in fact show integrated clusters – or whether there are
significant barriers within the networks identified earlier as being important to the performance of the system
with respect to a specific type of knowledge.
WINDOW
TOOL C1
Knowledge management analysis exercies
Expected outputs
● Explicit identification of opportunities ‘knowledge managers’ can use to improve system performance.
Relevant questions
Which persons (whether they are actors themselves, or a part of an institutional actor in the system) carry
out knowledge management tasks (e.g. those of policymakers, project designers, managers or liaison
officers)?
Which knowledge management tasks do they fulfil? (See checklist below.)
What is/are the explicit or implicit aim(s) of these knowledge management tasks?

Do these knowledge management tasks influence the interactions within the system in a way that
maximizes the performance of the system and/or changes its direction?
What opportunities do the ‘knowledge managers’ have to improve the generation, exchange and utilization
of knowledge and information in the system?
Working procedure
The process of analysis includes at least two steps: first, with an eye on the objective of diagnosis and further
insights gained during the diagnosis, the team tentatively selects the individuals/units who hold key positions in
the system with respect to knowledge management. Second, the team works further on defining potentially
positive interventions by these actors, keeping the constraints and opportunities formulated previously in mind.
One aspect that is particularly relevant here is the possibility that different actors will define the objective of the
system differently. From the point of view of knowledge management, this may or may not be a problem. The
team will need to consider this point, and, if necessary, outline a path that will encourage negotiations among
actors, aimed at increasing the convergences between them.
Complementariy with other tools
The basic configurations in Tool B6 can be helpful in discussing the feasibility of different possible actions. The
dominance of a particular basic configuration within the system implies that particular coordination
mechanisms are also dominant. Proposing changes in knowledge management may imply fundamental shifts in
the balance of power and influence in the system. If, for instance, interventions are intended to shift an
‘industrial’ agricultural knowledge and information system towards target groups that are relatively ‘resource
poor’, such as farmers, a shift from a marketdriven system towards one that is more policy-driven will be
required. Such a shift can only be realized if and when policymakers are committed to the change and have the
resources available to actually make it happen.
A knowledge management checklist
When the team wants to go into the details of existing or potential knowledge management in the system, the
following checklist can be used. It can suggest questions to ask to bring out the positive and negative aspects of
the way knowledge is handled by the system. It can also beused in thinking of recommendations toward the end
of a RAAKS study, to suggest possibilities for new elements that are needed. The questions – or
recommendations for developing these elements – can be applied to either subsystems or to the system as a

whole.
● Does the system or subsystem have a shared culture, policy and purpose?
● Is there a shared language, to facilitate exchanging experiences and information?
● Does the system or subsystem have ways to minimize blocks to formal and informal communication,
such as large social differences?
● Are linkages at strategic interfaces between organizations encouraged?
● Is the use of informal networks encouraged (by any means from social events to availability of electronic
media)?
● Are the existing information systems within the AKIS linked (e.g. by computer systems or other data
sharing)?
● Is strategic cooperation – that related to achieving the aims of the system – encouraged among actors?
● Are mechanisms to enhance user control operating (e.g. political, market or technical means that help to
coordinate user needs with information delivery)?
● Are there mechanisms to strengthen the responsiveness of the system to the objectives of the society,
changes in the environment of the system, and market opportunities?
● Does the system avoid becoming isolated from external sources of knowledge and information,
including that from non-traditional target groups?
● Does the system make strategic investments in physical infrastructure and human resources? Is there
flexibility in funding?
● Have key parts of the system been identified? Does resource allocation among them match their
requirements?
● Has a KIS management information system been developed?
WINDOW
TOOL C2
Actor Potential checklist
Expected outputs
● An assessment of the potential of internal and external actors for a positive effect on the performance of
all or part of the system.
● Identification of the resources various actors have that could help to achieve positive changes.
Relevant questions

Do the proposed interventions fall within the scope of the current activities of the actors?
Does a relative consensus exist concerning the need to intervene and the need to carry out the interventions
proposed by the team and the participating actors?
Do the relevant key actors see the interventions as being in their interest?
Whose perceived interests might be negatively affected by the intervention(s)?

Which of the key actors share a need for the perceived interventions? If they work alone, could they make
changes?
Do the relevant key actors have the resources needed to implement the proposed interventions (financial,
human, knowledge base, leadership, organizational capacities)?
Which of the key actors has actually expressed willingness to implement some or all of the proposed
interventions?
Working procedure
This tool can be used within the team and/or in a workshop with other actors, for example, in discussing the
potential for change and who can do what to accomplish it.
It is important to realize that such meetings do not imply pushing all actors into one mould or ‘unified system’.
Successful knowledge and information systems are often diverse and segmented – there may not be much
contact among different clusters of actors. In fact, successful systems generally need both this diversity and a
certain degree of redundancy to perform well.
When subsystems have fundamentally different objectives, conditions, or requirements, ‘knowledge

management’ should be approached separately within each subsystem. For example, the team will need to look
at the objectives set, and decide whether commercial farmers and peasant farmers will be best served by
designing a single knowledge system, or by subsystems with fundamentally different objectives, prime movers
and coordinating mechanisms.
WINDOW
TOOL C3/a
Defining possible actions: Moving towards improvement

Expected outputs
● Synthesis of results from windows addressed in Phases A, B and C.

● Identification of the constraints and opportunities within the knowledge system.
● Background material to use in preparing a draft report.
Relevant questions
Findings related to problem definition or redefinition in the earlier phases
What are the most important problems to solve, to achieve optimum functioning of the knowledge system?
Why do you see these as the most important? Give arguments!
Compare these conclusions with those reached using Tools A5, Approximation exercise II and B8/b,
Understanding the social organization of innovation.
Findings related to the analysis of constraints and opportunities in the second phase
What windows did you use and which gave you the most insight?
What opportunities or potential do you see for improving the functioning of the knowledge system? In
what way?
What constraints need to be taken into account in preparing interventions? Why?
Are any of the constraints/opportunities related to the way the system takes (or does not take) account of
information related to groups such as women, who are sometimes ignored?
What windows would you choose if you were to start over again? Why?
Possibilities for improving the innovation process (third phase)
Who could do something that would help to realize the possibilities mentioned in the second ‘Expected
output’ above?
What could he/she/they do?
Who should take the lead or the initiative?

What potential obstacles or problems do you foresee?

Can you see any side effects (positive or negative) the proposed intervention might have on the system as a
whole? On related systems? On specific groups (women, older people, farmers etc.) within the system?
Working procedure
The team analyses the material from literature and interviews collected in the first, second and third phases,
using the questions above as a starting point. They discuss the results and prepare a report. See the working
procedure and the section on ‘Complementarity among tools’ in Tool B8/b for more information.
WINDOW
TOOL C3/b
Strategic commitments: Action planning
Expected outputs
● A forum where key actors can discuss and negotiate alternatives, reaching agreements concerning the
actions to be taken.
● An assessment of the resources needed to carry out the proposed actions.
● An action plan/strategy for achieving a solution to the problem.
● Strategic commitments among the actors who are essential to realizing the objectives of the action plan.
Relevant questions
What? Defining the preconditions for formulation and implementation of an action plan
What are the specific activities that will be needed to formulate a plan?

What actions are needed?
What are the objectives that an action plan must achieve?

What broad general strategies will be followed?
What are the main areas to be included in the action plan?
Who and how?
What actors will need to be involved in the action plan?
Who will write a draft work plan?
Who will write each of the separate parts of the action plan?
Who will carry out each of the activities needed for implementation?
When?
What are the deadlines for completion of the final action plan?
When will the action outlined in the plan take place (deadlines for implementation)?
At what points will there be monitoring and evaluation of the implementation?
Working procedure
To the extent that the team and the key actors involved have been able create a forum for discussion and
negotiation, you have already achieved a great deal. Window C3 and its tools move further along this path. At
this time, it is necessary to reach more formal agreements, and to get a plan – including agreements about
implementation – down on paper.
It may be useful to begin with a brief review of previous work, including for example the agreed-upon
objectives of the system, and the work done related to needed coordinating mechanisms. Strategies for action
plan interventions that stimulate actors to work more closely together should certainly be considered. Teams can
also move beyond this, seeking strategies to broaden the system – to involve new actors who are essential to the
desired impact. In either case, it may be necessary to improve the mechanisms that encourage consultation
within the system. Policy measures, such as regulations and subsidies, project activities that facilitate desired

changes, or improvements to information infrastructure are additional possibilities, as are collaborative

activities such as training or lobbying to influence external actors or factors.
The questions above outline an approach to planning action and achieving strategic commitments. However, in
reality the ‘what’, ‘who’ and ‘when’ questions cannot be quite so easily separated. Moreover, the constraints
and opportunities identified during the entire RAAKS study serve as a basis for later strategic commitments
among key actors. Designing an action plan together builds and solidifies commitment. To assure smooth
implementation, the team can think throughout the study, and during action planning in particular, about the
activities that are apt to be needed, and whether additional actors will be needed to carry out these actions.
These actors or their representatives need to be included in the process. Further, anyone who is involved needs
clear permission from their parent organization to devote time, attention or other resources to the planning
process (and later to the activities specified in the action plan).
The results of a RAAKS study depend heavily on the willingness of autonomous actors to arrive at practical
promises. The results of the process should either be practical enough for quick implementation, or transparent
enough that they can become an input into a continuation of the process, in the form of negotiations among key
actors. During action planning, the number of meetings and their content depend on the degree of agreement
within the group. One or more preliminary meetings may be necessary to discuss the outlines of a plan and to
identify other important actors who should be involved in planning and implementation. On the other hand, if
actors are already committed, it may be possible to discuss the topics and activities to be included in an action
plan briefly. Next, group members can be assigned to write up each topic for the plan, and activity groups can
be assigned to discuss each of the needed activities and write up that element of the draft work plan. In either
case, the table below provides a framework for the discussions; the example outlines the outcome of one such
process. Another example is that of action planning among actors in the Basic Grains programme in Central
America, discussed in detail in The social organization of innovation, Chapter 8 (see ‘Strengthening agricultural
institutions…’).
Action planning sheet
What? Who? When?

(action needed) (who will participate) (timing for this action)
An action plan for the perennial plant sector n the Netherlands

A RAAKS exercise was carried out covering the perennial plant sector in the Netherlands. Two
subsystems were identified: one non-governmental and one governmental. These were in effect
separated by a barrier that hampered the capacity of actors (including government, research,
extension and growers) in both subsystems to adjust to emerging environmental problems and
changes in the market. External pressures were making it necessary to produce in a more
sustainable way, so that the two subsystems needed to link up. The team suggested that linkage
mechanisms should be installed and/or strengthened between research and growers, growers'
organizations and policymakers, and growers and extension workers. They also proposed the
formation of a regional environmental forum. A meeting was organized with key actors, in which
the findings and recommendations were presented and discussed. The team's proposal for an
environmental forum was very much in line with ideas already developing among some actors.
Also, the Community Board for Horticultural Products had planned to implement a system for
monitoring plant production on environmental aspects, which could feed information into the
system. The Agricultural Extension Service, suffering from its image as the bringer of ‘bad news
from the government’, took the results of the meeting to heart. As a result, they designed a special
unit of extension workers to provide strategic advice to growers to help them cope with the rapid
changing environmental regulations. The approach used in this solution might be called a system
management strategy, in which various key actors planned to act in ways that would facilitate
linkages.
Source: RAAKS seminar 1993
Footnote
1. This complete RAAKS analysis was facilitated by Annemarie Groot and Sjoerd Bakker, Department of
Communication and Innovation Studies, Wageningen Agricultural University, Wageningen, the Netherlands.
2. Lupanga, 1989a, 1989b
1. Two effective video packages are a video about Benin (The system and the soil) or the Netherlands (A day in
the country). Each package includes a film made from the perspective of knowledge systems, a workbook for
facilitators, and a booklet explaining the context of the film. Both films are available in either French of
English, and may be free of charge to governmental and non-governmental organizations in low and middle
income countries: submit a letter explaining how the film will be used and ask if this is possible. (At the time of
publication the price for others, including all commercial organizations, is US$ 150). Contact TVE Training
and Distribution Centre, P.O. Box 7, 3700 AA Zeist, the Netherlands, telephone 31 340 692 0499/fax 31 340
692 2484.
* Please indicate what is most convenient. Do not pay in advance: we will send the book with an invoice for the
box, plus shipping charges. Details regarding payment are given on the invoice; we accept major credit cards.
Orders from countries where we are represented will be forwarded to our local distributor.

1. The linkage matrix was developed as a tool by several knowledge system researchers, including Peter G.H.
van Beek (1989), Charles Frempong (unpublished), and Paul G.H. Engel (1990). This tool resembles the
‘connectedness’ matrix used in social network analysis.
aNitsch, U., ‘Computers and the nature of farm management’. In: D. Kuiper en N.G. Röling (eds.), The edited
proceedings of the European seminar on knowledge management and information technology. Wageningen,
Wageningen Agricultural University, 1991.

file:///D|/projects/RAAKS/temp/New/Raaks-in-Adobe/raaks.html
View publication stats
(169 of 169) [4/5/02 6:13:22 PM]

RAAKSmanual2013 10 2107 47 18UTC

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

RAAKSmanual2013 10 2107 47 18UTC

Uploaded by

Copyright:

Available Formats

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

Facilitating innovation for development : a RAAKS resource box

Article · January 1997

Paul Engel Monique Salomon

SEE PROFILE SEE PROFILE

Keeping cattle in a changing rural landscape View project

Small Business Advisory Services View project

The user has requested enhancement of the downloaded file.

file:///D|/projects/RAAKS/temp/New/Raaks-in-Adobe/Windows.html (1 of 2) [4/5/02 6:12:44 PM]

WINDOWS TABLE OF CONTENTS

A A1 Defining or re-defining the objective of the diagnosis

C C1 Knowledge management analysis

file:///D|/projects/RAAKS/temp/New/Raaks-in-Adobe/Windows.html (2 of 2) [4/5/02 6:12:44 PM]

file:///D|/projects/RAAKS/temp/New/Raaks-in-Adobe/raaks.html (1 of 169) [4/5/02 6:13:21 PM]

Networking for innovation

A participatory actor-oriented methodology

file:///D|/projects/RAAKS/temp/New/Raaks-in-Adobe/raaks.html (2 of 169) [4/5/02 6:13:21 PM]

Royal Tropical Institute

file:///D|/projects/RAAKS/temp/New/Raaks-in-Adobe/raaks.html (3 of 169) [4/5/02 6:13:21 PM]

© 1997 Royal Tropical Institute – Amsterdam

Printed and bound in the Netherlands

file:///D|/projects/RAAKS/temp/New/Raaks-in-Adobe/raaks.html (4 of 169) [4/5/02 6:13:21 PM]

1. Changing approaches to knowledge and information

2. Designing RAAKS action research

3. Preparing and carrying out RAAKS action research

About the authors

file:///D|/projects/RAAKS/temp/New/Raaks-in-Adobe/raaks.html (5 of 169) [4/5/02 6:13:21 PM]

Paul G.H. Engel

file:///D|/projects/RAAKS/temp/New/Raaks-in-Adobe/raaks.html (6 of 169) [4/5/02 6:13:21 PM]

file:///D|/projects/RAAKS/temp/New/Raaks-in-Adobe/raaks.html (7 of 169) [4/5/02 6:13:21 PM]

The RAAKS resource box

file:///D|/projects/RAAKS/temp/New/Raaks-in-Adobe/raaks.html (8 of 169) [4/5/02 6:13:21 PM]

1 Changing approaches to knowledge and

file:///D|/projects/RAAKS/temp/New/Raaks-in-Adobe/raaks.html (9 of 169) [4/5/02 6:13:21 PM]

Innovativeness as a social competence

file:///D|/projects/RAAKS/temp/New/Raaks-in-Adobe/raaks.html (10 of 169) [4/5/02 6:13:21 PM]

Understanding the social organization of innovation

file:///D|/projects/RAAKS/temp/New/Raaks-in-Adobe/raaks.html (11 of 169) [4/5/02 6:13:21 PM]

implement particular innovation strategies.

● relevant actors and their practices; and

file:///D|/projects/RAAKS/temp/New/Raaks-in-Adobe/raaks.html (12 of 169) [4/5/02 6:13:21 PM]

file:///D|/projects/RAAKS/temp/New/Raaks-in-Adobe/raaks.html (13 of 169) [4/5/02 6:13:21 PM]

The AKIS perspective

file:///D|/projects/RAAKS/temp/New/Raaks-in-Adobe/raaks.html (14 of 169) [4/5/02 6:13:21 PM]

file:///D|/projects/RAAKS/temp/New/Raaks-in-Adobe/raaks.html (15 of 169) [4/5/02 6:13:21 PM]

● cooperation among the various actors involved

Key concepts in knowledge and information systems

file:///D|/projects/RAAKS/temp/New/Raaks-in-Adobe/raaks.html (16 of 169) [4/5/02 6:13:21 PM]

file:///D|/projects/RAAKS/temp/New/Raaks-in-Adobe/raaks.html (17 of 169) [4/5/02 6:13:21 PM]

RAAKS: Participatory action research

RAAKS, women and other stakeholders

For whom is RAAKS useful?

file:///D|/projects/RAAKS/temp/New/Raaks-in-Adobe/raaks.html (19 of 169) [4/5/02 6:13:21 PM]

2 Designing RAAKS action research

file:///D|/projects/RAAKS/temp/New/Raaks-in-Adobe/raaks.html (20 of 169) [4/5/02 6:13:21 PM]

RAAKS and other participatory methods

● a defined methodology and built-in learning process,

file:///D|/projects/RAAKS/temp/New/Raaks-in-Adobe/raaks.html (21 of 169) [4/5/02 6:13:21 PM]

Designing a RAAKS study

file:///D|/projects/RAAKS/temp/New/Raaks-in-Adobe/raaks.html (22 of 169) [4/5/02 6:13:21 PM]

Intentions and underlying principles

file:///D|/projects/RAAKS/temp/New/Raaks-in-Adobe/raaks.html (23 of 169) [4/5/02 6:13:21 PM]