Home page Help page Preferences page
Search for specific termsBrowse by ThemesBrowse by Geographical RegionBrowse by MOST DocumentsBrowse by How ToBrowse alphabetical list of titles

Replicating Social Programmes - Discussion paper n° 18
Open this page in a new windowDon't highlight search terms

Management of Social Transformations - MOST

Discussion Paper Series - No. 18



Nico van Oudenhoven
Rekha Wazir








This paper reviews the key issues and methodologies involved in the replication of social programmes. The related, but more general, processes of knowledge transfer and dissemination as well as the more specific strategies involved in replication and going-to-scale are examined. (1)

Replication has been extensively debated in the non-profit sector in the United States. The other major arena for such discussion is the sphere of development where also there has been an increasing concern with widening coverage of programmes and interventions. This issue has universal relevance and appears high on the agenda of donors and implementors in both developed and developing settings. Unfortunately, these discourses often remain compartmentalized with little or no acknowledgment, cross-referencing, cross-fertilization or exchange. It is remarkable that even the literature emanating from each side shows little knowledge of or interest in the main thinking and trends of the other. This paper explores the linkages between these parallel, but insulated, discourses.


The study of knowledge transfer and replication dates back to nineteenth century anthropology. After World War II, the debate has been pursued in a wide range of fields such as education, planning, sociology, medical practice, commercial and social marketing, and agricultural extension. During the last fifteen years, it has also gained ground in the social sector. A variety of labels, having their origins in distinct disciplines and social practices, are used to describe the phenomena but there is little uniformity or consistency in usage in the research literature and in development language. (2)

In the 1970s and 1980s research on dissemination and related activities reached its peak and culminated in a number of seminal publications. The work of Glaser et. al. (1983) and Rogers (1983) falls in this category and remains mandatory reading on this subject. Rogers, drawing on agricultural extension work, made a significant contribution to the formulation of key concepts and the theory of diffusion, while Glaser offered a detailed overview of research and practice in the field of dissemination and a taxonomy of what to do, in which context, and under what circumstances. (3)

The principles underlying knowledge use and transfer can be best understood through a consideration of parallel discourses on this subject in other disciplines and areas of activity. Illustrative comparisons can be drawn with the fields of medicine, agricultural technology, sociology of education and consumer behaviour. In each of these, transfer, diffusion and replication can be viewed as processes linking the origin of an innovation, idea or product on the one hand, to a universe of potential users, clients or beneficiaries on the other. This separation between the source and the recipient of the impulse is central to this concept. There are, however, other latent premises which need to be made explicit.

First, the subjects (or the universe of potential beneficiaries) are assumed to be unable to generate the required change or transformation by themselves. This inability could stem from a wide range of factors. Second, it is implicitly assumed that the source (donor/supplier/innovator) has the capacity to accurately recognise and prioritise the needs of the recipient. Third, the product or innovation is expected to satisfy the needs of the recipients. Finally, there is the underlying assumption of the universality of needs within the universe of recipients.

Viewed in these abstract terms, the process of going-to-scale is characterised by three components: identification and recognition of a particular need in the target population; a system whereby a product can be developed externally for meeting this need; and finally, a mechanism for the effective delivery of this product from the producer or source to the user or recipient. This process is fairly successful when applied to uni-dimensional single-effect products which cater to relatively homogeneous populations: mass immunization programmes, for example. (4) It is less successful in the case of the so-called green revolution which appears, at first sight, to be similar. However, decades of experience reveal that such thinking might have contained much that was simplistic. (5)

The social sector forms a third domain in which diffusion and replicative processes have been applied. If some of the assumptions which held validity in the case of medicine tended to crumble in the interactive socio-economic arena of agricultural technology transfer, the position is far worse when social-sector interventions are considered. Two examples are provided by anti-poverty programmes that attracted attention in the 60s and 70s - the Community Action Program (CAP) in the USA and the Educational Priority Area (EPA) in the United Kingdom. While both were motivated by social needs which are widely recognized, the simple model for mass transfer and replication did not meet original expectations (Higgins 1978). There was a tendency to overlook the social heterogeneity of the population, and hence the diversity of their needs; there was an over-privileging of the external agency and undervaluing of the voiceless within the recipient population. The multi-dimensionality of the product, of the recipients, as well as of the context, got inadequate recognition.

II.1 Rationale for Dissemination and Replication in the Social Sector

Arguments for increasing the coverage and impact of programmes are readily available. Practitioners, policy makers, researchers, and funding agencies would agree that there is sufficient knowledge and experience to address most problems (6). As so many people are still not reached, there is an obligation to extend, disseminate, or replicate this information so that more can benefit. The assignment, therefore, is not so much to improve the "state-of-the-art", but rather to lift up the "state-of-practice" so that an ever increasing number can benefit. (7)

A related angle is provided by the assertion that it makes sense on pragmatic and economic grounds to replicate what has proven to be working rather than reinventing the wheel. This argument is increasingly gaining ground with donors and policy makers alike. In an environment of shrinking resources for the social sector, both funding agencies and governments are coming under increasing pressure to show "results". They are also accused of spending the bulk of their resources on innovation and project identification rather than on replication. It is suggested that replicating good practice is a cost-effective means of utilizing scarce resources. The assertion is that money would be saved if project experience could just be transferred to other sites. (8) Consequently, funding and implementing organizations are under pressure to focus on bringing to scale existing programmes rather than supporting yet more "experimental", "pilot", "innovative", "alternative", or "trial" projects. (9)

Recently, a more developmental rationale has been put forward. (10) The spreading of good practice is viewed not merely as trying to persuade others to mount identical programmes, but rather as an opportunity for mutual learning and sharing of experience. A positive outcome of exchanging experience, according to this view, is that it allows networks of people and groups to develop. These can, in turn, grow into coalitions that can demand more political attention and appeal for larger allocations of means and also evolve into institutional vehicles for internal problem-solving.

In practice, a combination of economic and developmental motives may well provide the justification for dissemination and scaling-up. However, it is important to make a clear distinction to help bring order to the discourse on the subject.

II.2 Strategies for Increasing Coverage in the Social Sector

In many human development and research circles concerned with developing innovative and effective social programmes, dissemination was rarely an issue for deliberate reflection at the start of a project. It was more or less assumed that once a pilot project had been successfully completed, replication would follow as a matter of course. At most, a report would be written and a set of recommendations formulated "for further action". This further action was then considered to be the task of others. As a rule, no information was provided on who the others were, or only in general terms such as "practitioners", the "government", or the "NGO community". Neither was it made clear how these others should go about spreading (or receiving) the good news.

In response to rising pressure to look beyond the pilot phase of a project and to assume active responsibility for following up on project outcomes, many donor agencies made the inclusion of dissemination a mandatory objective for providing funding. Even when formally stated as one of the objectives, project designers tended not to look beyond the boundaries of their present work or think about its wider implications. Many donor agencies that carry the pursuit of replication in their banner tend not to move beyond rhetoric. The majority do not provide long-term support for replication work, resulting in the creation of "white elephants" and dependency on external funding. (11)

In recent years there has been a renewed interest in applying the principles of knowledge transfer in social programmes. (12) Donor agencies, governments and the international development community are all expressing a concern for making use of existing, well-tested experience. This has provided an impetus to documenting and broadcasting illustrative cases and there is now a steady flow of descriptions of commendable projects, models, and of approaches "that work". This information about good programmes is expected to assist others in developing their own work. (13) Attempts have also been made to highlight the ingredients or key features that make for success and to provide practical guidelines and strategies for dissemination. (14)

The following distinct paths to replication may be distinguished in the social sector: (15)

    Franchise Approach: Also known as the "cookie-cutter" approach, it is closest to the private sector in its policy and practice. It assumes that there is a product - in this case a programme - that can be replicated. The components of this prototype programme and performance standards are largely inviolable. There is a central agency, usually the franchiser, which provides technical assistance, marketing, training and other services.

    Mandated Replication: This approach is usually, though not necessarily, sponsored by government and occurs when a parent body wants to disseminate a prototype programme through the organizations under its jurisdiction. Mandated replication is always top-down and there is usually no element of choice involved.

    Staged Replication: This is the most structured approach to replication and takes place in three stages. The first is the pilot stage where the viability of the programme concept is tested; followed by the demonstration stage where the programme is implemented in a variety of sites. This stage is usually closely monitored and rigorously evaluated and successful demonstration is followed by replication. The analogy is drawn with prototype testing and development in the private sector and the need for an independent replicating agency is stressed.

    Concept Replication: In this approach the focus is not on the universal and specific elements of the prototype programme but rather on general components and principles which can be transported to other sites. Unlike the approaches mentioned earlier, strict adherence to the strategies and the model of the prototype are not required and success is measured in terms of adaptation and sensitivity to each unique local context. There is no accountability for how components are transferred and used at each local site.

    Spontaneous or Endogenous Replication: The essential difference here is that the demand for information comes from below. It is need based and is characterized by spontaneous and informal contacts between like-minded individuals. Additionally, the communication flow is not one-way - from recognized model to recipient - but is rather a two-way process of convergence where participants "create and share information".

In reality, dissemination efforts usually combine features of the approaches listed above. However, for the sake of clarity, two contrasting approaches are identified in this paper. The first will be called universalist - broadly speaking, proponents of this view share a belief in universal principles which can be applicable to a very wide band of practices and situations. The dissemination effort is supply-determined. The second approach is termed contextual - the emphasis here is on local practice, local initiative, spontaneity, mutual learning and problem solving. The dissemination effort is demand-driven. While both approaches are acceptable in principle, emphasizing one or the other would determine the choice and adoption of sharply contrasting strategies. There are sufficient commonalities in the first three forms of replication to warrant grouping them under the universalist label, while the fourth and fifth could be termed contextualist. It may be worth noting here that advocates of the universalist and contextual approaches largely debate their separate viewpoints in insulated groups and there is little evidence of learning from each other.

II.3 The Universalist Approach: an Appraisal

As stated earlier, the most concerted discussions and reflection on the theme of replication have taken place in the United States. The universalist viewpoint appears to be gaining ground as a potential strategy for extending the scale of effective programmes in the social sector. (16) Two broad trends are discernible in the literature. First, while paying lip service to the validity and importance of concept replication, a strong preference is expressed for developing a more planned, structured and controlled approach to disseminating good practice. (17) This is reflected in a call for adherence to standards and principles; for protecting the identity of the programme that is being replicated; and for charting out admissions requirements for selecting local sites. There is a move towards giving this controlling function to an intermediary organization which acts as the replicating agent and has "final accountability for program performance". A need has also been expressed for a national agency which can develop and promote replication strategies and speak with an authoritative voice on the subject. (18)

A second and related move is towards applying theories and practices developed in the private sector to replication efforts in the non-profit sector. (19) According to RPS (1994:ii), replication in the social sector "... is entrepreneurial, market-driven ... In short it appears to be analogous to our market economy." This conclusion is based on their finding that the most successful replication does not take place on account of deliberate policy but is the result of a private entrepreneurial effort, very similar to starting a new business. Thus, in the social sector as well, replication efforts would require a "champion" or "programme entrepreneur" who has the charismatic and leadership qualities required to design programme strategies, promote its achievements and secure long-term funding. (20) Once conceived, the programme would need to be "marketed" and "promoted" in order to raise its public profile and increase its ability to compete for scarce funding. Concern is expressed for the lack of "incentives", in sharp contrast to the private sector, which would be required to sustain commitment to the programme. Finally, there would be a need for protecting the programme prototype from being cloned or expropriated without due acknowledgement and payment. It is suggested that in order to avoid loss of revenue, protect the reputation of the programme, and prevent misuse of key concepts and strategies, the social sector would in future need to enforce copyrights and patents and levy licensing fees.

In short, in order to be successful, replication strategies should look to the business sector for inspiration, in particular to the field of business franchising. (21) It is suggested that the participants and donors can be assured of certain quality standards in large-scale franchised programmes. The licensing agreements to carry out the programme usually entail an adherence to fixed standards of effectiveness. Training, upgrading and inspection are the responsibility of the franchising agent and there is the stipulation that the licence can be removed if the subcontractor fails to meet certain standards. Funders are more inclined to invest in the expansion of such programmes.

Despite earlier critiques of the universalist approach within various disciplines, it is once again in the foreground of social action, but this time taking its cue from the private sector. Several factors could be held responsible for this trend. The recent resource crunch has implied a move away from government funding of social programmes and a corresponding increase in the importance of the voluntary or non-profit sector for the delivery of such programmes. The government has increasingly incorporated principles of corporate philosophy with respect to the use of its own resources and also for the disbursal of funds to the voluntary sector. At the same time, social problems have not decreased and there is an urgency in the search for successful prototypes. Given this climate, and the parallel rise in corporate philanthropy, techniques developed in the private sector have percolated into the world of social programmes. While it is undeniable that there is a real and immediate need to search for solutions that reach more people, it remains necessary to consider the appropriateness of these techniques and to anticipate and examine their weaknesses.

First, replication is seen as the culmination of a unilinear unfolding of discrete activities starting with the pilot and the demonstration stages. Agencies' annual reports, conference discussion papers and research and evaluation reports present innumerable examples of variations and elaborations of these stages. But the existence of these stages and their sequential order is hardly challenged. A close look at practice reveals that they are often not clearly distinguishable and they, or their elements, may exist simultaneously. Projects never work in total isolation, they have radiation effects and they respond to environmental influences from the onset. This is especially the case when project staff belong to varied networks and are in constant communication with others.

Second, this approach looks on programme replication as an activity to be carried out largely by the sponsor or initiator of the original project. These initiators are generally governmental or parastatal agencies and private or non-governmental organizations as well as so-called "pioneers", "champions", or "charismatic leaders" belonging to these organizations. The designation of the sponsor as the main directing and initiating actor has far-reaching consequences. It immediately builds into the work a "source bias" reflecting the interests, style and values of the sponsor. (22) Going-to-scale becomes a centralized, top-down process with the major decisions made at "headquarters". (23)

Third, there are dangers inherent in transferring strategies developed in the business sector to improve replication practice in the social sector. There can be vast differences in objectives, guiding philosophy, target groups, values and mission. Principles and practices that have been developed to maximise profit might not be applicable, beyond a point, to agencies working for the benefit of disadvantaged sections of society. For example, notions of staff ownership and commitment, and the need for transparency and participatory decision-making are viewed very differently in the two sectors. (24) It would be difficult to juggle the need for local participation, ownership and responsiveness to contextual variables - crucial elements in social programmes - with the requirements of standardization which are central to the franchise approach. It is interesting to note that even franchised operations are increasingly allowing for local input and creativity and imposing standardization only where necessary.

The World Organisation of the Scout Movement/World Scout Foundation aptly illustrates this move towards increased contextualization in the social sector. (25) World-wide the organisation counts over 32 million members, boys and girls. Scout groups differ widely from country to country, and within countries, from place to place. In Indonesia, Scouting is compulsory, and every school-going boy has to join; in Yemen it is very much a an elitist activity; and in the USA, scouting is a strong commercialized movement (26). The activities carried out by scouts cover the range from complete leisure to fully-committed social action and the mechanisms that keep the movement together are elastic. At present these consist of rituals, external paraphernalia and a well-run organizational structure. Local groups take care of their own funding, and contribute to the running costs of regional and international headquarters.

Fourth, there is yet another danger to "cookie-cutter" replications which remain faithful in their form and content to the original "model" programme. They may succeed and even be locally supported, especially if they appeal to a well-resourced leadership, but do they work to the benefit of their target group? Sustaining such a model may even become counterproductive as it could absorb all available resources and discourage the promotion of other, more appropriate models. Everything else becomes less attractive, not to be emulated. For example, it is a familiar sight in derelict or deprived areas to see first class community centres or sporting grounds that resemble the facilities available in better endowed places. These exact replicas have often been established by benefactors who, in tandem with local leaders, want "the best of the best" for youth. They are usually the show pieces reserved for visitors and the media but they may reach only a fraction of the youth living in the neighbourhood. It is also likely that the services offered are appropriated by the least disadvantaged youth. For financial and psychological reasons, the existence of such a service could foreclose any other form of assistance to deprived youth in the area.

Finally, in the current trend towards "planned" replication, there is an undue stress on technical and organizational aspects at the cost of human and social aspects. Social reality is inherently complex and does not submit in a predictable manner to externally imposed interventions, no matter how well planned or technically sound they may be. At the heart of all social processes are people and they are also the intended beneficiaries of social programmes. Unless the human aspect is given due consideration in the design and dissemination of programmes, the effort is likely to fail. According to Cernea (1991:7) "the neglect of social dimensions in intervention-caused development always takes revenge on the outcome". (27) In the world of development practice and literature, there is an increasing awareness of the need to "put people first" in the planning of intervention programmes (Cernea 1991; Chambers 1993; Korten & Klauss 1990). A change is called for in the conventional approach to planning, which is dominated by technical factors and administrative details. The rationale for this change is sought not just on ethical and humanitarian grounds but is rooted in the belief that this is essential for assuring the effectiveness of programmes.

The Summer Training and Education Programme (STEP) - a USA remedial training programme for 14 and 15 year old poor urban youth over two summers - provides a very good illustration of a planned replication where more importance was given to form over content. (28) STEP has been hailed as a model of staged replication and is seen as an unqualified success as far as the replication process is concerned. However, long-term evaluations show that the programme had little or no impact on the youth it aimed to serve once they had left the programme. This outcome would not have been so surprising to the sponsors of the programme if the complex social dynamics surrounding issues such as poverty, urban deprivation, unemployment and teenage pregnancy had been acknowledged and incorporated into the programme at the outset. Few astute social observers would have believed that a short intervention like a summer programme could rid youth of multiple, structural disadvantages.

II.4 The Contextualist Approach: An Appraisal

The contextualist approach recognizes the uniqueness of each particular setting, thus precluding the wholesale cloning of models and practices from one context to another. Primacy is given to addressing local needs, adapting to local environments and acknowledging the validity of local knowledge. The relationship between the giver and receiver is viewed as equal and non-hierarchical and each exchange is a potential opportunity for mutual learning. The very term "dissemination" is seen to carry the connotation of a dependent or passive receiver; as are the words "target audience", "consumers" and "takers". Instead, notions of "partnership" and "convergence" should govern all exchanges.

In a parallel discussion on the merits and demerits of centralized versus decentralized diffusion, Rogers and Marcus (1983) note that a centralized approach is preferred only when highly-technical expertise is required. Contrasting the two strategies they observe that decentralization invites local control, stimulates staying power, promotes peer-to-peer diffusion and horizontal networks, encourages local experimentation by local non-experts, is problem centred and demand-driven and has a higher degree of adaptation. Centralization is dependent on highly-trained experts, is top down, draws on research for innovation, manifests a low degree of adaptation and is supply driven.

It is obvious that the contextual approach is more suited to the transfer of components and principles and not to the actual replication of a project or programme. The key words frequently used are: indirect, dispersed, inadvertent, spontaneous, less measurable, and less geographically bounded (Chambers 1993). Not surprisingly, no framework or blue-print is prescribed for implementing this strategy and there is room for using indirect means for achieving wider impact and coverage. Thus, activities such as lobbying, influencing policy, advocacy, training and networking are deemed to have equal, if not more, significant results than direct dissemination.

The contextualist approach is usually developed as part of a broader strategy which also includes other, more direct means of replication and going-to-scale. (29) Each individual situation should determine the choice of strategy to be followed; in certain cases a combination of approaches may well be the most feasible course of action. While "expansion" and "addition" are seen as obvious means of increasing impact and coverage, it is felt that indirect means of replication should be given due recognition as they can often have superior results. Advocates of the contextual school rarely approach dissemination and replication strategies in isolation but place them in the wider framework of development theory and discussions on NGO management, impact and efficiency. (30)

There are several merits to the contextualist approach - more particularly, its sensitivity to the local level, the importance that is given to local knowledge and to need-driven demands for information, and its acceptance of the relevance of direct and indirect means of increasing impact - to make it an attractive component of any replication strategy. It is empowering, ensures local control and encourages self-generated learning. At first glance, it would also appear to contain all the elements required for developing an appropriate strategy for replicating policies, programmes and practices - the starting point of this paper. A closer look at how this approach works in practice reveals some weaknesses.

First, notwithstanding its strengths, critics of the contextualist approach would deny it the label of a strategy since there are few rules governing its implementation. The replication effort is informal and dispersed and there are few well-defined criteria for evaluating its success. The premise of the uniqueness of each situation can sometimes be taken to an extreme. If all commonality is denied, room can be left open for unnecessary re-inventions of the wheel, with each local agency expending time and resources to find new solutions to problems that are not unique. It is not rare for small organizations to become self-serving and inward-looking with little or no contact with other like-minded actors. This precludes them from coalition-building and from joining forces in the interests of a common cause. A major avenue for increasing the impact of the work of the local sector could then be lost.

Second, the objective of increasing impact and coverage is also not served well if there is an undue focus on processes at the cost of outcomes. Some proponents of this approach would go so far as to reject the notion of planned, step-wise change, especially in community-based work. According to Smale (1993:16), "most people need to reinvent their own wheels and want to use them in their own way". Consequently, there cannot be a blue-print for community-based practice as "there are no destinations, only journeys". However, it would be difficult to deny that journeys could and should be undertaken with some sense of destination in mind. If the aim of social programmes is to mitigate the effects of social disadvantage, the interests of the target group will not be served well by a strategy which leaves so much to chance.

Third, an element of wishful thinking and romanticization can be detected in the notion of decentralization of diffusion efforts. It is necessary to bear in mind the critiques developed in the discourse on decentralization where devolution of authority is sharply distinguished from democratization. It is emphasized that the focus should not be just on shifting the responsibility for financial allocations and decision-making but equally on understanding the nature of local power structures and political processes which get "empowered" as a result of decentralization. Writing more than a decade ago about the "myth of decentralization", Bryant and White (1982) warn that it can invite corruption, internecine warfare and take-over by local elites. (31)

Finally, it is important to ask the question: who assesses local needs and how are they legitimized? All too often, this is done by an outside agent - the "animator" - or by the representative of a donor agency. Needs, priorities and areas for intervention are often defined on the basis of a short exposure or a superficial knowledge of the field. Local hierarchies, power structures and disagreements may not be reflected in the message that is taken back. For example, undue emphasis may be given to the voices of the most vocal and visible youth, who may not be the most vulnerable and marginalized.


The universalist and contextualist approaches both have their relative merits and in real situations they can often be combined, bringing out their individual strengths and mitigating their weaknesses. The actual form of these combinations will vary from one context to the next and will be informed by such considerations as the impact on the target group, the numbers reached, and the sustainability of the interventions. No matter what approach is used, accumulated experience - in research, policy and practice - shows that replication is a complicated, costly and time-consuming process. There are no easy solutions to it and no short cuts. However, it is useful to look at the various trends in social programming, policy and practice that offer promise - including those found in development experience -and also to explore the long research tradition in related social science disciplines where parallel discussions are held on the subjects of knowledge transfer and use. The integration of inputs from these distinct field of practice and research yield some considerations that need to be kept in mind when discussing replication in the social sector.

III.1 Commonality vs Specificity

The problems facing most countries of the world have an element of commonality. Korten (1990) mentions, by way of illustration, over twenty development problems that do not recognize North-South or East-West distinctions. Some of the needs that are increasingly shared in common are:

    • reducing chronic unemployment;
    • controlling drug trafficking and abuse;
    • managing population growth and distribution;
    • reducing teenage pregnancy;
    • providing housing for the homeless;
    • making credit available for micro-economic activities;
    • reducing hunger, illiteracy and infant mortality among difficult to reach populations;
    • treating AIDS victims and controlling the spread of the disease;
    • meeting the needs for bilingual education;
    • facilitating reconciliation among racial, religious and ethnic groups;
    • resettling refugees;
    • ensuring the preservation of human rights; and
    • increasing citizen awareness of global development issues.
These global risk factors have a direct bearing on the well-being of families, particularly those who already live under stress caused by poverty, discrimination, or disabilities. But families in the richer countries are also sensitive to these risk factors. A recent study reveals that 15-30% of children in the OECD countries could be considered "at risk" for the same reasons as in developing countries (Evans 1995).

Further, the causes of these problems, and their manifestations, do not run parallel to each other, but are closely interdependent. Being commonly rooted, these problems often appear in clusters. Studies from the United States confirm that risk behaviours are interrelated in children and youth: nearly 50% of American youth are involved in two or more of the four categories of risk behaviours that have been identified as: drug and alcohol use and abuse; unsafe sex, teenage pregnancy, and teenage parenting; school failure, underachievement and dropout, and; delinquency, crime and violence.(Lerner 1995).

This connectedness also manifests itself internationally. Events in one country could have an immediate impact on the lives of people living at the other end of the globe and vice versa. Communication and exchange of values and ideas occurs all the time and with considerable speed and few remain untouched. Ling (1989) notes, for example, that "lifestyle illusions have become the new communicable disease - transmitted through the information media...They are initiated as fast as communications speed information from one country to another". Common problems call for joint action and learning.

In recognizing the commonalities of social issues, the underlying specificities of each local situation should not be ignored. Seeking common solutions to shared problems could well imply a negation of underlying contextual differences and a stifling of need-based local action. Care needs to be exercised not to impose "universal" solutions to problems which are intrinsically local and to avoid coopting local initiatives into donor-determined, homogenized approaches. In fact, a sensitive balance has to be struck between accepting what is universal or global while recognizing and protecting what is valuable at the local level. This issue strikes at the very heart of the discussions on how to disseminate and replicate social policies and practices across cultures and borders.

III.2 What Makes Programmes Work?

A first step to understanding why programmes work is to uncover the principles and processes underlying "good practice". Most reviews of successful programmes are mainly descriptive in nature, they seldom go further than offering evidence that the project has a positive effect on the target group and should, therefore, be supported or emulated. Analytical studies that reveal why programmes work, under what conditions, and how are rare. Without this understanding, the dissemination of projects, or of their elements, could degenerate into a form of blind cloning or become a matter of intuition.

A number of principles have already been referred to in the text as they appear to be essential to most successful programmes. They include empowerment of users, recognition of cultural diversity and local needs, promotion of holistic development, and parental involvement. However, mechanical adherence to these principles will not automatically lead to positive development; their meaning and function should be continually analyzed.

It is important to bear in mind that knowledge and practice are not stable - they have to be reviewed all the time. Questions have to be posed continually and in each different context as the answers will vary accordingly. Understanding the issues underlying wider principles is, therefore, necessary to counter the mechanical application of outdated practice. It will also ensure that programmes and policies are not static but remain responsive to changing needs.

III.3 Giving Validity to all Knowledge

A review of critical writings in the field of knowledge use and transfer reveal important guidelines for replication in the social sector. These are of particular relevance in situations where government departments or large funding agencies take the lead in dissemination efforts. (32) The most fundamental conclusion of these critiques is that knowledge is not objective or value-free; it is identified with the groups that create it and it serves to further their interests while disregarding those of others. In order to ensure that dissemination does not become a way to exert power and control over small, local organizations, care should be taken not to treat "knowledge users" as empty receptacles with no mechanisms for their own knowledge creation. This requires giving validity to all kinds of knowledge - be it research or practitioner knowledge. Similarly, in order to be truly effective, knowledge should not be imposed from outside but should be owned or internalized by users. Ideally, two-way information sharing, rather than knowledge emanating from a single source, would be one way to avoid this situation. At the practical level, it would be more effective to present users with a range of programme and policy options rather than promoting one particular prototype. This would allow them to make comparisons and to select and combine elements to suit their particular environment.

III.4 Networking as a Tool for Dissemination

Outcome-oriented networking of networks could be yet another way of disseminating good practice. The participants should belong to vertically and horizontally linked structures, connecting public and private organizations. These networks should be multi-nodal and comprise autonomous subsystems. There should not be a tightly-structured chain of command or communication. The participants should have the capacity to act and learn without being forced to do so and they should have the potential for voluntary and collective action. Most importantly, they could form coalitions of smaller NGOs, or even GOs, who could act together to make an impact.

Networking also has its drawbacks: network meetings are not low-cost. They can easily degenerate into talking shops, or turn into elite groups, excluding others and monopolizing the debate. These unproductive dynamics can be avoided by encouraging the participation of groups or sub-networks. This can be done by identifying specific needs and problems among the network partners; by setting goal-oriented agendas; and by facilitating and monitoring progress. Sustained, effective and locally rooted dissemination is most likely to take place through outcome-directed networking. Replication through networking is not likely to evolve spontaneously or from the bottom up; guidance and direction by a centralized force is usually needed, not only to initiate but also to supervise and sustain the process. Traditionally, government agencies and grant-making organizations assume this role. Some of the problems associated with this could be avoided if NGO coalitions were also to take on these functions.


This paper has not sought to produce a list of recommendations. These abound in the literature emanating from the non-profit sector and display a general tendency to reduce the many dimensions of this issue into a set of how-to-do guidelines. It is only when the processes of going to scale are appreciated in their full complexity that meaningful policy can be drawn up. The search for simplicity has been a main stumbling block in the formulation of this policy. Some of the complexities underlying the question of programme replication have been touched upon in this text.

The progress of replication efforts is often measured in terms of criteria such as the effect on the participants, the numbers reached; the spread of project sites over the country, region or world; the volume of services extended, of the institution or of its staff. When these indicators meet expectations, the programme is judged to be successful. Seldom, if ever, are advances gauged against overall needs in a given country or region. Thus, while an organization is expanding its exemplary programmes the impact on the total target population may remain insignificant. It is hard to judge the relevance of dissemination programmes that are carried out without a clear understanding of the total picture. There is a need for a system that would monitor this situation. At the most basic level, such a monitor would gather data on the target population, their needs, what programmes are offered for them, how many participate and who and how many are left out, or require special attention. (33) Additionally, the monitor would feed back information and demonstrate the effectiveness of dissemination programmes to all participants. A self-monitoring system used by the participants would be an important force to increase the outreach of services.

The expansion of programme coverage is often the outcome of an intricate interplay between donor and recipient agencies. The relationship between donors and recipients is, by definition, unequal. Barring rare occasions where popular NGOs are courted by the entire donor community, recipients are dependent for their income and livelihood on donors and will, in varying degrees, cut their cloth according to the donors' whims and wishes. In spite of the often encountered expressions of "genuine partnership" and "critical dialogue" it is the donor who, in the ultimate analysis, calls the shots. As most donors and development agencies are identified with the West, there is also the danger of imposing models and practices developed in the West at the cost of existing and effective local customs and approaches. Dissemination efforts spearheaded by prestigious persons such as first ladies, celebrities or royalty should also be looked at with circumspection, particularly when they take a guiding role with respect to the direction and content of the programme. (34) Throughout this text, various pointers have been suggested on how to professionalize the relationship between donor and recipient. Early involvement in project design and planning by all stake-holders, working through transparent networks, NGO capacity building, and evaluations are some of the most prominent tools to this end. The philosophical underpinnings of the code of conduct for donors have been worked out extensively by Pantin (1979,1983), who summarizes this philosophy in two words: respectful listening.

A main argument running through this paper is that effective and sustainable programmes reaching out to large numbers are few and far between and take a long time to establish. Take the example of programmes for children and youth. In 1975, a survey showed that life for youth in the inner cities of North America had only worsened after a decade of intensive debates and federal, state and municipal involvement (Goldman and Dotson 1975). Now, twenty years later, the situation is not better and has, in fact, deteriorated further. (35) This has happened, or has been allowed to happen, against the back cloth of efforts to expand programmes that "work"; the availability of tried and tested programme and policy scenarios; and the existence of vast skills and knowledge on how to implement these. The situation of children in other countries is often not much different. (36) Given this track record, it behooves politicians, researchers, donors and policy makers to adopt an attitude of profound modesty, or even wariness, about their proposals and about any future scenarios they set in motion. It may well be that an informed uncertainty proves to be a better compass to go by in the debate on replication than a set of over-confident, but ill-informed recommendations.


Archie, M. et al (1993) Partners in Action: Environmental Social Marketing and Environmental Education, Washington, Academy of Educational Development.

Backer, Thomas E. (1992) "Getting Management Knowledge Used", Journal of Management Inquiry, Vol. 1, No. 1, 39-43.

Bendahmane, Diane B. (1994) "The Quiet Revolution: Child Survival Comes of Age" Grassroots Development, Vol. 18, 2-12.

Biemann, Betsy and Tomlinson, John (1992) Community-Based Development Experience: Can it be replicated from South to North and so how? New York, The Synergos Institute.

Birman, Beatrice F. and Kaufman, Martin K. (1991) A Knowledge Market Model of Dissemination and Use in Special Education, Office of Special Education Programs, U.S. Department of Education.

Bryant, C, and White, L. (1982) Managing Development in the Third World, New York, Westview Press.

Cernea, Michael M. (1991) Using Knowledge from Social Science in Development Projects, Washington DC: The World Bank.

Chambers, Robert (1993) Challenging the Professions: Frontiers for Rural Development, London, Intermediate Technology Publications.

Chambers, Robert (1992) "Spreading and Self-Improving: a Strategy for Scaling-up", in : Michael Edwards and David Hulme (eds.) 1992.

Clark, John, (1991) Democratizing Development The Role of Voluntary Organizations, West Hartford, Kumarian Press.

Clinton, Hillary Rodham, (1995) It Takes a Village: and other lessons children teach us, New York, Simon Schuster.

Conservation Company (1993) Principles and Lessons Learned Regarding Social Innovation Transfer, The Conversation Company.

Conservation Company & Public/Private Ventures (1993) Building from Strength: Replication as a Strategy for Expanding Social Programs that Work, Philadelphia, Replication and Program Services Inc.

Council on Foundations (1993) Evaluations for Foundations, San Francisco, Jossey-Bass.

De Lone, Richard H. (1995) Replication: A Strategy to Improve the Delivery of Education and Job Training Programs, Philadelphia, Public/Private Ventures, (2nd printing).

Dichter, Thomas W. (1989) Findings 89: NGOs and the Replication Trap, Connecticut, TechnoServe.

Dillinger, William (1994) Decentralization and its implications for Urban Service Delivery, Discussion Paper Nr 16, Washington, World Bank.

Edwards, Michael and David Hulme (eds) (1992) Making a Difference: NGOs and Development in a Changing World, London, Earthscan Publications Ltd.

Evans, Peter (1995) "Children and Youth `at risk'" in OECD Our Children at Risk.

Garvin, Alexander (1995) The American City, What Works, What Doesn’t, New York, MacGraw Hill.

Glaser, E.M., Abelson, H., & Garrison, K.N. (1983) Putting Knowledge to Use, San Francisco, Jossey Bass.

Goldman, D. and Dotson J.L. (1975) "Ten Years On", Newsweek.

Grant, J.P. (1989,1990) The State of the World's Children Report, New York, UNICEF.

Higgins, J. (1978) The Poverty Business, London, Basil Blackwell.

Huberman, Michael (1994-95) "Research Utilization: The State of the Art", Knowledge and Policy: The International Journal of Knowledge Transfer and Utilization, Vol. 7, No. 4.

International Initiative (undated) Rebuilding the Family from Within: A different Way of Working, Warford (UK), International Initiative.

International Youth Foundation (1991 - 1995) Annual Reports, Battle Creek.

Klein, Susan S., and Gwaltney, M.K. (1991b) A Research-Based Framework for a National Education Dissemination System, US Department of Education. Washington, D.C.

Klein, Susan S., and Gwaltney, M.K. (1991a) "Charting the Education Dissemination System", Knowledge: Creation, Diffusion, Utilization, Vol 12, no 3. March, 241-265.

Korten, D. and R. Klaus (eds), (1990) People Centred Development: Contributions Towards and Theory and Planning Frameworks", Kumarian Press.

Korten, David (1990) Getting to the 21st Century, Voluntary Action and the Global Agenda, West Hartford, Kumarian Press.

Kozol, Jonathan, (1995) Amazing Grace, The lives of Children and the Conscience of a Nation New York, Crown Publishers.

Lerner, Richard M. (1995) America's Youth in Crisis: Challenges and Options for Programs and Policies, Thousand Oaks, California, Sage Publications.

Ling, C.S.J. (1989) "New Communicable Diseases: a Communicative Challenge", Health Communication, Vol I, nr 4, 253-260.

Mannes, Marc & Steven G. Meilleur (1989) How to Market Human Service Innovations: A Guide to the Development and Dissemination of Social Technologies, Albuquerque, American Indian Law Center, Inc.

Mott Foundation 1990 Replication: Sowing Seeds of Hope, 1990 Annual Report, Mott Foundation, Flint MI.

Myers, Robert (1992) The Twelve who Survive: Strengthening Programmes of Early Childhood Development in the Third World, London & New York, Routledge.

National Association of Social Workers (1993) Effective Strategies for Increasing Social Program Replication/Adaptation, Washington, NASW.

Nilsson, Kjell and Sunesson, Sune (1993) "Conflict or Control: Research Utilization Strategies as Power Techniques", Knowledge and Policy: The International Journal of Knowledge Transfer and Utilization, vol. 6, No. 2, 23-36.

OECD (1995) Our Children at Risk, Centre for Education Research and Innovation, Paris, OECD.

Oster, Sharon M., (1992) "Non-Profit Organizations as Franchise Operations", Nonprofit Management & Leadership, Vol 2, no 3, Spring.

Paisly William J. and Mathilda Butler eds. (1983) Knowledge Utilization Systems in Education, Beverly Hills, Sage Publications.

Pantin, Gerald (1983) The Servol Village, SERVOL, Kingston

    Pantin, Gerald (1979) A Mole Cricket Called SERVOL: the Early Years of a Community Project in the West Indies, High/Scope Press, Ypsilanti.

    Pittman, Karen Johnson (1995) Premises, Principles, Practises: Defining the Why, What, and How of Promoting Youth Development through Organizational Practise, Centre for Youth Development and Policy Research, Academy for Educational Development, January, Washington D.C.

Pottier, Johan (1993) Practising Development Social Science Perspectives, London, Routledge.

Public/Private Ventures (1990) Replication: Some Policy and Program Considerations, Philadelphia, Public/Private Ventures.

Rogers, E.M. and Marcus, J.E. (1983) "Advances in Diffusion Theory", in: Paisley William J. and Mathilda Butler eds. (1983) Knowledge Utilization Systems in Education, Beverly Hills, Sage Publications.

Rogers, E.M. (1983) Diffusion of Innovations, New York, Free Press.

Rogers, E.M (1976) "New Product Adoption and Diffusion", Journal of Consumer Research, Vol.2, 290-301.

Rothman, Jack and Edwin J. Thomas (eds.) (1994) Intervention Research: Design and Development for Human Service, New York, The Haworth Press.

RPS (1994) Building from Strength: Replication as a Strategy for Expanding Social Programs that Work, Philadelphia, Replication and Program Services, Inc.

Schorr, Lisbeth B. and Schorr, Daniel (1989) Within our Reach: Breaking the Cycle of Disadvantage, New York, Anchor Books.

Smale, Gerald G. (1993) "The Nature of Innovation and Community-Based Practise", in: E.F. Martinez-Brawley, Transferring Technology in the Social Services, Washington, NASW Press.

Van Oudenhoven, Nico (1989) Child Development and Community Response, UNESCO Notes, Comments, nr 81, Paris, UNESCO.

Van Tilborg, L, and Riemersma, F. (1995) Haalbaarheidsonderzoek Jeugdmonitor 0-25 jaar, Eindrapport (Feasibility Study Youth Monitor,0-25 years, Final report), The Hague, Sardes.

Walker, Gary and Villela-Velez, Frances (1992) Anatomy of a Demonstration: The Summer Training and Education Program (STEP) From Pilot Through Replication and Postprogram Impacts, Philadelphia, Public/Private Ventures.

Watkins, John (1994) " A Postmodern Critical Theory of Research Use", Knowledge and Policy: The International Journal of Knowledge Transfer and Utilization, Winter, Vol 7, no 4, 13-33.

Zuzovsky, Ruth (1994) "Utilization of Research Findings: Matter of Research Tradition" Knowledge and Policy: The International Journal of Knowledge Transfer and Utilization, Winter, Vol 7, no 4, 78-93.


1. We would like to thank the International Youth Foundation, which commissioned an earlier version of this study and Renee Pittin, for her comments on the draft.

2. The following descriptors are useful in accessing the subject area: acceptance, adaption, additionality, adoption, application, assimilation, communication, coverage, diffusion, dissemination, distribution, exchange, expansion, extension, flow, going-to-scale (also scaling up and upscaling), growth, innovation, multiplication, new knowledge research, new practice research, new products research, reception, replication, retrieval, spread, transfer, transmission, utilization, and also, frequently, planning.

3. Much of the current debate on going-to-scale turns out to be a repetition of earlier, readily available information, without reference to lessons learnt in the first round of replication efforts more than a decade ago.

4. There could be little doubt or disagreement about the desirability of mass immunization, or of its intrinsic welfare-raising contribution. It would also be fair to argue that beneficiaries could not really be expected to develop such a product at the micro-level. The focus then falls on developing the best version of the product, of an efficient delivery mechanism and of ensuring the acceptance of the product by the beneficiaries. Large scale replication systems are thereby necessary.

5. In the case of high-yielding variety seeds as well, the model of external development of the product, followed by mass diffusion would appear to be justified. However, it became quickly apparent that the benefits of such technical change were very unevenly distributed within the target population, with those whose need was the greatest benefiting the least. The externally generated product was defined exclusively in terms of positive attributes but experience has revealed negative dimensions, most prominently in the form of the high degree of dependence on chemical fertilisers, and the consequent environmentally damaging effects. Contrary to the case of immunization programmes, the impact of this intervention was not self-contained and neutral with respect to other social outcomes. Finally, partly as a result of the eventual discovery of these problematic side-effects, there has been a certain grudging recognition of the values of the original farming technologies, and new approaches attempt to build on the essential features of these endogenous systems.

6. See, for example, Myers (1992), NASW (1993), Pittman (1995), Schorr (1989). More recently, the American First Lady, Hillary Rodham Clinton (1995) also contended that "we do not lack the information, but rather the will to do what is best".

7. See, for example, Myers (1992), NASW (1993) and Schorr (1989). The International Youth Foundation (IYF), whose explicit mandate it is to "replicate good practice", operates YOUTHNET, a data bank that contains a growing selection of international "exemplary" projects.

8. There is little evidence in the literature, however, to suggest that planned or staged replication of project prototypes is an any way less costly than starting a new programme. On the contrary, most research shows that it can be an expensive process requiring vast human and financial resources. Bieman and Tomlinson (1992) note the high costs involved in any exchange of practices and ideas within a North-South context.

9. This trend is particularly strong in the United States, as witnessed by the fact that many foundations have taken up "going-to-scale" as a major topic in their programming. See for example, Birman and Kaufman (1991); Council of Foundations (1993); Mott Foundation (1990); Paisly et al (1983); and International Youth Foundation (1991-1995).

10. See, for example, Cernea (1991), Chambers (1993), Edwards and Hulme (1992) and Pottier (1993).

11. There is a tendency among donor agency staff to undertake new initiatives and to respond much more enthusiastically to innovations rather than to more of the same.

12. See, inter alia, Backer (1993); Chambers (1993); Conservation Company (1993); Conservation Company & Public/Private Ventures (1993); Dichter (1989); de Lone (1995); Edwards & Hulme (1992); Mott Foundation (1990); NASW (1993); Public/Private Ventures (1990); RPS (1994); Rothman & Edwin (1994).

13. Garvin (1995) suggests that many urban problems could be resolved by applying solutions found elsewhere. If there is a problem, he argues, go and look where they have found the answer.

14. With respect to the area of children and youth, a review of in-depth as well as meta-studies shows that most successful programmes meet a certain combination of criteria. Briefly, they focus on children; promote positive growth; are preventative; allow for optimal participation by the children, parents and communities; are contextual and respond to local needs; have a positive bias towards vulnerable groups, especially poor children and girls; are horizontally and vertically embedded in organizational structures; and are low-cost. They are also well managed; conduct regular evaluations and offer training opportunities to their staff (Grant 1990; International Initiative 1991a,b; International Youth Foundation 1991-5; Van Oudenhoven 1989).

15. The first four paths to replication have been identified by Replication and Program Services, Inc. (RPS) on the basis of their survey of US practice (RPS 1994). RPS also provide a listing and descriptions of US-based youth programmes which fall into each of the four categories.

16. The main proponents of this approach are: Backer (1992a,b); Mott Foundation (1990); Conservation Company (1993a,b); de Lone (1990); Public/Private Ventures (1990); RPS (1994).

17. An important study on replication, which was conducted under the auspices of some US agencies, concluded with the following three recommendations: a how-to publication which would pull together relevant information in a manual; establishment of a replication resource group designed to "serve the interests and needs of private and corporate philanthropy, concerned with the most cost-effective use of program development, demonstration project and replication strategies"; and a national fund for programme replication (RPS 1994).

18. Recognizing the importance of such an agent, private organizations in the USA have established the specialised agency Replication and Program Services, Inc (RPS). Based in Philadelphia, this new office provides support to foundations and private voluntary organizations in disseminating their work.

19. See Archie 1993; Backer 1992; Mannes & Meilleur 1989; Oster 1992; RPS 1994.

20. The role of the "champion" is seen by many as crucial for dissemination efforts. The argument runs that somebody is needed who believes in disseminating programme outcomes, who is committed to it, is internally motivated, who can push and move things, and has the skills, endurance and personality to carry on and to convince others to follow. However, as these champions are not always easily found, it is often recommended that an external, professional "replication agent" be appointed to guide the dissemination. What they would lack in personal qualities would be made up by their expertise, professional interest, and external incentives.

21. The franchising of programmes, products, names, and logos is common practice in the field of social programmes in the USA. More than half of the top one hundred charitable non-profits, e.g. American Red Cross, YMCA, and Scouting groups, are franchising organizations (Oster 1992). They transfer to franchisees the exclusive right to use their "trademark" or sell certain products, usually in a particular territory, in return for a payment. The franchiser provides assistance and exerts control over certain aspects of the operation. Profits, losses and liabilities are borne locally. Similar practices are now becoming common in other countries as well.

22. According to Rogers (1976) research and evaluation studies of replication have also been heavily "source biased" as they have mainly been commissioned by sponsors.

23. What, for example, is the commitment needed to sell a McDonald hamburger?

24. It is not surprising that most of the deliberations on replication take place at the behest of organizations that wield power to implement decisions.

25. See various information leaflets from Scouting Nederland and WOSM/WSF Geneva.

26. The many local and national differences became manifest, almost to the point of mutual non-recognition, at the triennial Jamboree in The Netherlands in 1995. The Japanese scouts were appalled to see Dutch scouts kissing each other. The Dutch were, in turn, stunned by the military look of the Japanese, while many Muslim scouts were not allowed to have any companionship whatsoever with the other sex. Uniforms varied equally widely.

27. A re-evaluation of 25 World Bank financed projects shows that 13 of these were unsustainable, not for financial reasons, but because socio-cultural factors had been neglected at the stage of project formulation and implementation (Cernea 1991).

28. For a comprehensive description of the replication process involved in the STEP programme see Walker & Villela-Velez (1992).

29. See Cernea (1991); Chambers (1993); Clark (1991); Edwards & Hulme (1992); Korten & Klaus (1990).

30. A recent British publication, exploring the different ways in which NGOs should increase their impact lists three strategies for dissemination: additive -implying an increase in the size of the programme or organization; multiplicative - where impact is achieved through deliberate influence, networking, policy and legal reform, or training; and diffusive - where spread is informal and spontaneous. No preference is expressed for any one strategy as each would be effective in a different circumstance and no clear-cut criteria are provided for implementing the various strategies (Edwards & Hulme 1992). It is interesting to note that this volume on replication makes no reference to the parallel discussion taking place among US organizations.

31. The discussion about the merits and demerits of centralization and decentralization, and components of these multi-layered processes still attracts attention. Dillinger (1994), overseeing the international scene and writing for the World Bank, states that the objectives of decentralization are only tangentially related to administrative performance. He believes that governments are not genuinely interested in delegating responsibilities aimed at improving service delivery. At best they agree to a disorderly and reluctant series of concessions, primarily to maintain political stability. Recent developments in inter-sectoral networking may merge the advantages of both centralized and decentralized strategies.

32. See Huberman 1994-95; Nilsson & Sunesson 1993; Watkins 1994.

33. See Van Tilborg & Riemersma (1995) and Zuzovsky (1994) for detailed descriptions of the role of monitors.

34. The involvement of well-know people in programme promotion has become a regular feature on the development scene. Their use is obvious: they help to attract attention, open up doors and rally people around a good cause. It is argued here, however, that involvement should be restricted to promotion and not to guiding the direction and content of programmes.

35. Kozol (1995), writing about children in urban ghettos reports that child poverty in USA has reached its highest level since 1964. "People have become tired of shouting", he notes.

36. Evidently, much has gone wrong and several attempts have been made to reflect on these poor outcomes and on what could be learnt from the past. See, for example, Grant (1989) and Klein and Gwaltney (1991a,b). The implicit suggestion in their work is that if certain missing areas were given proper attention, things would go better. It is not certain if this optimism, which is expressed at many fora, is warranted.

About the authors

Nico van Oudenhoven and Rekha Wazir, psychologist and sociologist, are co-founders of International Child Development Initiative (ICDI), Hooglandse Kerkgracht 17, 2312 HS Leiden, The Netherlands. ICDI is an international development support agency specializing in programmes, policy and research for marginalized children and youth. The authors have written extensively on these issues and also on the topic of development management. Their recent publications include «Partnership - a development Strategy for Children» (1997) and «Child Sexual Abuse: What can Governments do?» (1997).

The opinions expressed in this series are those of the authors and do not necessarily reflect the views of UNESCO.

To MOST Clearing House Homepage

To see the document in English
Pour voir le document en Français
Para ver el documento en