SciELO - Scientific Electronic Library Online

 número44"Missing in action": the significance of bodies in African bereavement rituals índice de autoresíndice de assuntospesquisa de artigos
Home Pagelista alfabética de periódicos  

Serviços Personalizados



Links relacionados

  • Em processo de indexaçãoCitado por Google
  • Em processo de indexaçãoSimilares em Google


Psychology in Society

versão On-line ISSN 2309-8708
versão impressa ISSN 1015-6046

Psychol. Soc.  no.44 Durban Jan. 2013


Using Appreciative Inquiry (AI) to evaluate an education support ngo in Soweto



Angelo Fynn*

University of South Africa Pretoria




The purpose of this paper is to discuss the application of the Appreciative Inquiry (AI) approach while evaluating an emerging Non Governmental Organisation (NGO) in Soweto, South Africa. The AI approach was originally introduced as an organisational development tool intended to shift the focus from deficit-based evaluation to an appreciative stance that considered the organisation when it functioned at its peak. The use of this approach was motivated by the belief that it would provide emerging NGOs with an acknowledged method of implementing their development agendas in the face of donor or governmental organisational resistance. While the AI approach intends to provide a broad platform for stakeholder interaction, the experience of this study was that it requires a significant amount of political resources to include stakeholders from outside the organisation to participate actively. Our experience of the AI approach was that it seems to be difficult to implement in an environment where there are low levels of trust and cohesion among the stakeholders. The paper concludes with suggestions for researchers, especially novice researchers, entering the field with the intention of utilising AI in the Soweto context.

Keywords: Appreciative Inquiry (AI), research process, South African education, non governmental organisations




In his Nobel prize acceptance speech Richard Feynman (1965) highlighted that: "We have a habit in writing articles published in scientific journals to make the work as finished as possible, to cover all the tracks, to not worry about the blind alleys or to describe how you had the wrong idea first, and so on. So there isn't any place to publish, in a dignified manner, what you actually did in order to get to do the work, although, there has been in these days, some interest in this kind of thing". The thrust of Feynman's speech discussed how the positive bias in publication often neglects the mistakes and flaws in research processes to present a dignified, complete study at the expense of sharing the learning experience inherent in the flawed research process followed (Feynman, 1965). This paper should be viewed as a reflection on the Appreciative Inquiry (AI) approach and its use in the South African context written primarily from the perspective of the researchers and, while it doesn't examine the empirical data from the study on which it is based, it uses the study as the context in which to critically reflect on the meta-process and application of AI.

This paper discusses the dynamics experienced during the application of the AI approach to evaluate an emerging Non Governmental Organisation (NGO) in Soweto, a township in the Gauteng province of South Africa. The secondary purpose of this paper is to discuss the mechanics of the AI approach within the context of the research undertaken and to reflect on the dynamics of conducting participant directed research in the highly contentious arena of South African education.

The AI approach was originally introduced as an organisational development tool that sought to shift the focus from deficit-based evaluation to an appreciative stance that considered the organisation at its peak of performance. This approach assumes that the identity and functioning of an organisation is a social construction based on the dominant shared language within the organisation (Bushe, 2007). By creating a shift in the language, the approach seeks to generate positive change that is embedded within the culture of the organisation which increases the chances of sustainable change. In simpler terms, AI posits that the way we think and speak about the organisation shapes the way we act and defines the focus of the organisation, ultimately defining the nature of the organisation (Cooperrider & Whitney, 2005).

The study this paper is based on adapted the AI approach for use within the South African education context with the aim of examining the utility of AI as a cost-effective evaluation technique that is based on the principle of empowered participation. The focus of this paper is on the dynamics, challenges and achievements experienced during the use of the AI method to evaluate an emerging NGO in Soweto. In discussing the utility of AI in this context we will briefly describe the context, discuss the basic principles underpinning AI, describe the ideal method for using AI and, finally, discuss the use of AI as an evaluation tool in Soweto.



Cooperrider and Whitney (2005), define AI as the cooperative, co-evolutionary search for the best in people, their organisations and the world around them. It involves systematic discovery of what gives life to an organisation, community or social programme when it is most effective and capable in economic, ecological and human terms (Cooperrider & Whitney, 2005). AI is an approach that explicitly, but not exclusively, focuses on the positive within a community, social programme, organisation and the individuals within these entities (Bushe, 2007). An explicit focus on the positive entails exploring those aspects of the organisation, community or programme that work and are valued by individuals within the organisation, community or programme (Datta, 2003; Cooperrider & Avital, 2004; Datta, 2007). Coghlan, Preskill & Catsambas (2003: 5) see AI as a process that "inquires into, identifies, and further develops" the best of what is in organisations in order to create a better future. The approach works on the assumption that by focusing on the positive and collectively constructing ways of building on what works contributes to strength-based development for the programme, organisation or community (hereafter collectively referred to as the programme). The AI approach is not a once-off measure that instantly corrects the deficits within the programme but rather, a process that incorporates the entire organisation in what is termed a whole system event (Reed, 2007). In simpler terms this means that the act of inquiry or evaluation in the AI process is not a once-off event within a closed system. The full AI process continues long after the culmination of the inquiry summit and ideally embeds within the culture, practice, planning and collective thinking of the programme (Bushe & Kassam, 2005; Bushe, 2007; Dunlap, 2008). Appreciative Inquiry is also described as "social construction in action" McNamee (in Reed, 2007: viii). This statement describes the central role ascribed to the function of social construction within the approach and seeks to demonstrate how constructions of reality make certain actions possible and others prohibited (Burr, 1995; Gergen, 1999; Terre Blanche & Durrheim, 1999; Bradley & Morss, 2002). AI, based on underpinning Constructionist principles, proposes that the most important aspect of social life is what people do together because in their joint actions they create a world that values certain beliefs and practices (McNamee, 2003). In other words the meaning given to particular representations of people, objects and phenomena informs our experience of these people, objects and phenomena which are then shared through a shared language that aids the construction of reality (van Sant, 1989; Gergen, Gergen & Barrett, 2004; Valsiner, 2006). The centrality of this assumption to AI is what drives the need for inclusivity in the process (Lewis & van Tiem, 2004). By excluding a participant group, the research team lose a portion of the reality that constitutes the social programme. This principle of inclusivity describes why AI is cooperative and co-evolutionary (Cooperider & Srivastva, 1987; Cooperrider, Whitney & Stavros, 2003; Jacobsgaard, 2003). As each participant contributes what they value about the organisation, the collective story of the organisation evolves to accommodate each contribution to its history and everyday life.

Organisational development practitioners, such as David Cooperrider, were concerned with the performance and sustained improvement of programmes in their contexts (Coghlan, Preskill & Catsambas, 2003; Bushe, 2007). Cooperrider noticed that by asking questions about what people valued in their organisation they spoke in an unrestricted manner that provided greater insight into the assumptions and beliefs that underlie everyday practice (Cooperrider & Whitney, 2005). By focusing on the positive aspects of the programme, the assumption is that positive development of the programme will be built on what works in the organisation. This is described as the positive-focused development model (Cooperrider & Whitney, 1998). Positive focused development is described as the process through which the strengths of a programme are exposed and the circumstances in which these strengths are displayed are understood and a plan of action is developed to build on these strengths (Reed, 2007). This can be contrasted with the deficit-focused model of development which is the core of more orthodox developmental strategies (Patton, 2003). The deficit-focused model of development begins with the assumption that the programme has a problem that needs to be fixed or that the programme is not working as well as it should be (Patton, 2003; Boyd & Bright, 2007). This approach may imply that the programme is unsatisfactory, inadequate or underperforming without taking into consideration the context in which the programme operates (Patton, 2003). To borrow from Community Psychology theories, what may seem like inadequate programme performance or poor programme implementation may simply be organic adaptation to the needs and circumstances of the individuals that the programme services (Nelson & Prilleltensky, 2005).

Boyd and Bright (2007) discuss the contrast between these approaches on a continuum which is subject to what they term normative forces. In this depiction the underlying assumption is that all programmes will strive to revert to a state of normal which is neither positively nor negatively deviant (see Figure 1). Within this continuum the deficit-focused approaches form part of the dynamics of reaction and restoration which are enacted to fix problems within the programme in order to shift it away from negative deviance to ordinary (Boyd & Bright, 2007). On the other hand positive-focused approaches, such as AI, seek to move the programme from ordinary to positive deviance through the dynamics of pro-action and extension that aim to elevate strengths within the programme (Boyd & Bright, 2007). As mentioned, both of these processes are subject to what is termed normative momentum, which is the drive toward the normal (Bushe & Kassam, 2005; Grant & Humphries, 2006). The implication is that unless a positive change is accepted systematically and integrated into the everyday life of a programme, the programme will inevitably revert back to a state of normality where the exceptional remains the exception. What AI strives to do is to shift the entire system so that what is extraordinary or positively deviant, becomes the new standard for ordinary. The distinction between AI and other positive-driven approaches is that AI is not just about the positive (Bushe, 2007). According to Bushe (2007), the focus of AI is generativity, which aims to give new ways of viewing, understanding and constructing social structures and institutions to provide new options for action.



Appreciative Inquiry process

AI is a process; therefore, AI practitioners do not simply cherry-pick research techniques to suit the desired results. Rather, AI practitioners follow a clearly delineated process of inquiry that guides the implementation of the principles and assumptions of the approach. The AI process places the power of collective storytelling at the centre of its theory and practice and therefore leans more toward the qualitative end of the research continuum. When participants are asked for their input, they are not limited to a pre-defined assessment of what is valuable to the organisation but are allowed to express their experience of the organisation from within their worldview (Cooperrider, Whitney & Stavros, 2003). This allows the study to adopt a reflexive design that incorporates the input from participants organically and in real time so that the programme receives feedback that is immediately relevant to its current situation (Patton, 2003).

The typical implementation process for the AI process is through the Appreciative Inquiry summit (Watkins & Mohr, 2001; Bushe & Kassam, 2005). The summit is a whole-system event where all the stakeholders of a programme are gathered over a period of four days to work through the various phases of the AI process (Bushe & Kassam, 2005). By placing all stakeholders in the same room the process of building relationships that are geared toward achieving the new objectives of the programme can begin. As mentioned previously, AI is not about asking positive questions but rather about asking generative questions aimed at reframing the reality of individuals within the programme (Gergen, Gergen & Barrett, 2004; Bushe, 2007). The aim of these generative questions is to generate novelty and surprise among stakeholders about the programme they all know about by sharing stories about the programme they may not have been aware of. These questions are also aimed at engaging participants at more than just the intellectual level and seek to also engage participants by aligning the goals and objectives of the programme with the passions of the stakeholders to generate greater buy-in (Bushe, 2007). The aims of generative questions are depicted in Figure 2 and are not placed in any particular order as the various objectives transcend the entire AI process.



The summit is structured around the 4-D process that provides the overarching framework for all interactions that take place during the summit.

The 4-D process

The 4-D process (see Figure 3) is the most often used process in the AI approach (Coghlan, Preskill & Catsambas, 2003). The 4-D process or cycle has four distinct stages that are named Discovery, Dream, Design and Destiny/Delivery.




The Discovery phase of the 4-D process aims to appreciate what gives life to the programme (Reed, 2007). This is the opening step in the process and is often participants' first exposure to the approach. In this phase the aim is to discover what stakeholders value about the programme and what they perceive as the strengths of the programme.

The Dream phase of the 4-D process aims to encourage participants to envision what might be (Cooperrider & Whitney, 2005). The Dream phase is where participants work together to develop ideas of what the future might or could be. Participants are encouraged to think creatively and to let go of the constraints the programme currently faces. The rationale behind dreaming big is that it provides a long-term goal to collectively strive toward. This phase builds on the positive aspects of the programme identified during the Discovery phase.

The Design phase aims to determine what will be and brings participants closer to the real world of the programme (Watkins & Mohr, 2001). The Design phase is where participants work together to craft plans for the future by developing provocative propositions, which are statements about what the programme wants to achieve (Reed, 2007). These provocative statements are essentially collectively designed values and objectives put forward by the stakeholders of the programme and represent a commitment to achieving them and should be set out as unequivocal ambitions or statements of intent with no caveats or conditions (Reed, 2007).

The Destiny or Delivery phase entails planning what will be and is the phase where the energy moves towards implementation planning (Preskill & Catsambas, 2006). This phase requires participants to work out specific steps that need to happen for the programme to realise the commitments made in the previous phase. This stage draws on all of the previous stages, especially on the Discovery phase which highlights past successes that can contribute to the future success of the programme.

Criticisms of Appreciative Inquiry

The primary criticism of Appreciative Inquiry (AI) by evaluators is that it is a simple bait and switch strategy employed by researchers to alleviate the anxiety and tension that is associated with an evaluation of a programme (Patton, 2003). Proponents of this stance argue that AI is used as a method of deceiving stakeholders into participating in a typical evaluation. Evaluation is a highly political activity and as such power dynamics play a significant role in the process. Evaluation also has the negative connotations of being judged and assessed, often with the implication that non-performance or non-compliance will be penalised (Patton, 2003). The AI approach has been criticised for being co-opted as a strategy for engaging participants who would not participate in a typical evaluation by emphasising the positive, appreciative nature of the approach without ensuring that the principles of the approach are applied during implementation (Cooperrider, Whitney & Stavros, 2003).

AI has been criticised for its unbalanced, uncritical almost biased focus on the positive (Patton, 2003; Reed, 2007). When applied uncritically, the application of the emphasis on the positive can be seen to discourage constructive criticism. Criticisms and weaknesses emerge as part of the narrative in which the positive is highlighted (Reed, 2007). The crucial difference is that when the weaknesses are brought to the fore as part of a positive narrative they do not dominate the narrative and provide the stakeholders with a positive foundation on which future endeavours can be built. Encountering negative feedback or criticism within an environment that emphasises an appreciative approach can facilitate easier integration of this negative feedback for individuals (Watkins & Mohr, 2001).

The approach has also been criticised for the delay between the intervention, often in the form of the AI summit and the observable effects of the intervention (Reed, 2007). This is due to the fact that AI aims to bring about sustained change within a programme and places the responsibility for this change in the hands of the stakeholders. The implication of this is that stakeholders may work on a deadline that does not correspond with the schedule of those conducting the impact assessment. A second reason is that each programme is unique and change within these organisations will take place at varying rates (Reed, 2007). In addition the changes within these organisations may not be to the explicit system in terms of the procedures, policies, protocols and results but rather to the implicit system that consists of the relations, interactions and perceptions of the individuals within the programme. The latter would be hard to discern if an evaluator was not explicitly searching for these changes within the programme.



The section starts with a brief overview of the South African education system and of the role of NGOs in the system, with a particular focus on the NGO that was evaluated. It then follows with a narrative of the process followed that will be integrated by a reflection on the process.

The South African education system is in crisis; with low matric pass rates, high dropout rates, teacher strikes, rising pregnancy rates among teenaged learners, and assaults by learners on educators and other learners (Sabates et al, 2010; Williams, 2010). These issues are embedded within the wider social systems and are symptomatic of a social structure that fails to sufficiently meet the needs of individuals, social groups and communities (Jansen & Taylor, 2003; Ahmed & Sayed, 2009; Motala, 2009). The education system is not only faced with external challenges but also struggles with immediate internal challenges such as a lack of infrastructure that leads to overcrowding in class rooms and insufficient learning materials (Sabates et al, 2010; Williams, 2010). High attrition rates due to voluntary resignations to emigrate to escape the high crime rates or move to higher paying work, as well as due to death as a result of the HIV/AIDS pandemic and the education legacy of apartheid have also led to a large number of under-qualified teachers being placed within the system (Molteno, 1984; Demombynes & Ozler, 2005; Arends, 2007; Hammet, 2007).

Habib (2003) places the state-civil society relationship in three distinct blocs, each with a different set of relationships with the state. On the one end of the spectrum he places informal NGOs, mainly working in and with marginalised communities, who have no relationship with the state; in the second position there are those organisations that have entered into partnerships with or are sub-contracted to the state; lastly there are those organisations that actively challenge and oppose what Habib (2003) describes as the implementation of neo-liberalism by the state (Habib, 2003: 8). He describes this as the sum of state policies aimed at the "liberalisation of the financial and trade markets, deregulation of the economy and the privatisation of state assets" (Habib, 2003: 8) which have led to the largely negative consequences of the realisation of state deficit targets at the cost of higher unemployment, poverty and inequality in the country. Within this context the twin roles of NGOs are to continue to act as monitors for the public good and play a role in enabling poor and marginalised communities to survive within the context of modern, post-liberalisation South Africa; and to act as both advisor and critic to the state in their role of activist for their beneficiary group (Reitzes & Friedman, 2001; Habib, 2003).

NGOs play a variety of roles in relation to education service delivery (Rose, 2009). The types of interventions typically implemented by NGOs in the education sector can be divided into two distinct but interlinked categories. The first category is aimed at school improvement at various levels and includes interventions aimed at improving the infrastructure, school management practices, capacity building for teaching staff, curriculum reform and policy development. NGOs are also involved in lobbying and advocating for educational transformation by working individually or through networks to participate in policy dialogue with government (Miller-Grandvaux, Welmond & Wolf, 2002).

The three blocs of NGO described above should not be viewed as stark and immutable. An organisation can have a cooperative relationship with government on one issue while acting as a staunch critic of the same on another (Habib, 2003). Townsend et al (2004: 872) further refine this point by describing NGOs as a "fluid web of interrelations" that can simultaneously act as an "arm of government" and make possible the "insurrection of subjugated knowledge". Townsend et al (2004) also point out that even the most complaint NGOs may be creating spaces of resistance or carving out different spaces and visions from that of donor or government organisations while those organisations who actively question development agendas may need to comply with certain requirements for the sake of organisational survival. Of particular relevance to this paper is the practice of subtle subversion of the donor's development agenda by appearing to comply with accountability mechanisms while implementing the organisations' own development agenda and approach - which is more suited to the context - in practice (Townsend, Porter & Mawdsley, 2004). Townsend et al (2004: 877) refer to this practice as working subversively within the "master-servant relationship" and striving hard to adopt the donor language and style with the long-term view of creating space in which to promote the NGOs own development ideas.

The second category of NGO in education service delivery is direct education provision which focuses on providing education to children traditionally excluded from the education system (Rose, 2009). In other words, this form of intervention by NGOs places emphasis on providing education to hard-to-reach populations of children who, for reasons concerning ethnicity, citizenship status, socio-economic status or disability, are not accommodated by the formal education system (Rose, 2009; Ahmed & Sayed, 2009). The NGOs in this category can be said to provide non-formal or alternative education measures. In this role, NGOs are described as providing Non-Formal education. Non-Formal education is a contentious issue to define but can be described, at its most basic, as any organised and sustained activities that take place both within and outside educational institutions and that caters to persons of all ages with a focus on holistic education that has differing durations and may or may not have certification of learning achieved (UNESCO, 1997: 47).

The NGO that was evaluated as part of this study, falls within the latter category of providing non-formal education support to high schools in Soweto that are classified as underperforming by the Department of Basic Education. The long-term aim of this organisation is to develop a formal partnership through which it can inform government delivery of education based on the experience gained through its practice in communities where the delivery of education is problematic and inconsistent (Matoane & Fynn, 2010). It also aims to challenge those policies and practises that are seen to sustain structural inequalities in education within the schools in which it intervenes (Matoane & Fynn, 2010). The organisation intervenes in a context where there are high levels of poverty and unemployment. As such, some learners do not have access to basic resources such as school uniforms, stationery and textbooks, which hampers their ability to learn (Matoane & Fynn, 2010). Within the context of Habib's (2003) blocs, this organisation can be positioned as both partner and opponent to state providers of education.

The organisation has put in place a programme of support for learners of township schools. The programme of support is described as a holistic, empowerment focused programme aimed at preparing learners for a successful transition from school to the world of work or tertiary education (Matoane & Fynn, 2010). The support programme consists of three aspects, namely, financial, mentoring and psycho-social development workshops.

The organisation provides financial support to the learners selected into the programme. The financial support of the learners aims to provide the learners with the basic materials needed to ensure that teaching and learning takes place (Matoane & Fynn, 2010). This includes purchasing school uniforms, stationery, textbooks and additional study material.

The second component of the programme is the mentoring aspect. The organisation aims to provide learners with both academic and psycho-social support. Both of these aspects are provided by placing each learner with a mentor (Matoane & Fynn, 2010). The mentorship component of the support is designed to assist learners to deal with the aspects of their curriculum that they find challenging (Matoane & Fynn, 2010). By creating a one-on-one environment for learning, the programme ensures that each student can learn at an optimal pace. The mentors are young professionals from a range of fields that include Engineering and Financial Specialists (Matoane & Fynn, 2010). The mentors offer their services on a voluntary basis. The mentors act as role models for the students to model attitudes, behaviours and skills that the learner will need to be successful in life (Matoane & Fynn, 2010).

The psychological, academic and social development of the learners is the core objective of the organisation and as such the learners are also required to attend a series of workshops aimed at personal growth and development (Matoane & Fynn, 2010). The workshop topics include career orientation, self-awareness, communication, goal setting and being a teenager (Matoane & Fynn, 2010). This is referred to as the learner empowerment component of the programme. This component is aimed at providing the learners with a safe space to learn and practice life skills that are deemed crucial to their success in their schooling career, as well as in their post-schooling career (Matoane & Fynn, 2010). The workshops emphasise experiential learning and place the learners in situations where they can practice the skills taught to them in these workshops (Matoane & Fynn, 2010). As such, the learners are often required to present what they've learnt in the classrooms to their peers (Matoane & Fynn, 2010).

At the time of the study in 2010, the NGO had been operating for a year and, due to increased interest from the community, was in the position to expand its sphere of operation to a second school in the same district. The board of directors determined that it would be necessary to have a clear sense of how the organisation was performing before committing to the increased workload that the second school implied. After multiple consultations with tertiary institutions, the board opted to try the AI approach. The rationale for this choice was that the organisation was still defining its vision, mission and method, the large number of stakeholders in the organisation required an inclusive and participative approach to ensure an accurate reflection of stakeholder's perceptions and, lastly, the board wanted an approach that would encourage reflection on the practical, everyday solutions to issues facing the organisation with the aim of using these lessons to improve services with the inclusion of the second school.

A secondary consideration for the use of the AI approach was to find a method of ascertaining what works in a context dominated by discourses of neglect, underperformance and a history of adversarial relationships between the stakeholders involved in the context (Matoane & Fynn, 2010). The belief was that the AI method would allow for an exploration into how the successes, if any, were achieved despite the negative context described above, without avoiding the reality of the situation in which the organisation operates.

The data sources for this paper are primarily from the field notes kept by the research team during the process. The field notes were compiled with the view to developing an easy to use guide for the organisation to use in subsequent evaluations. The primary focus of the original study on which this paper is based, was to evaluate the intervention approach of the organisation and therefore the focus of participant feedback was primarily directed toward that purpose. The discussion of the process will primarily be in the third person with a narrative approach to discussing the process.



The remainder of this paper describes an attempt to apply AI in an NGO active in the South African education sector. While reading through this section, it is important to note that the study on which this paper is based is viewed as the habitat in which these processes were experienced. The source of the reflections described below originates from the researchers' emersion in the context and experience of the research process. The focus is therefore not on empirical data obtained from participants but from the reflective praxis on which the research approach was based.

The discussion of the research process will follow a chronological narrative based on a synthesis of the field notes taken by the researchers. The research team consisted of two academics in Psychology at the University of South Africa and the director of the NGO. The academics acted as the primary researchers who brought the knowledge of the AI process and research process while the director of the NGO was the context expert with the resources to negotiate access to participants. The director also played the role of the critical examiner who checked the interpretation of the research data against the context in which she worked every day and who disseminated the findings of the research to the stakeholders as part of the member checking process.

This section is divided into the initial entry into the organisation and research context, implementing the AI summit, and reflection on the research process.

Initial entry into the organisation and research context

The organisation approached the Department of Psychology at UNISA to assist them in designing and implementing an evaluation of their intervention approach. The board of the organisation felt that by using a tertiary education institution to evaluate their activities, they would lend credibility to their approach through an independent evaluation. They approached the Psychology Department because of the strong emphasis on psycho-social development in the programme and the community setting in which they operated.

The need of the organisation was to evaluate the intervention method a year into the existence of the programme and to include all the stakeholder groups in this process. The NGO intervenes in a school setting and identified the learners in the programme, the parents of these learners, educators in the school, the school governing body, the school management team, the board of the NGO, the mentors and auxiliary volunteers in the organisation as stakeholder groups to be consulted. The researchers then introduced the board to the principles and process of AI and the collective decision was made to use this approach for the evaluation.

As mentioned in an earlier section, the AI process revolves around the AI summit where all the stakeholders are assembled to collectively discover the moments of peak performance in the organisation and to dream up possible futures based on these moments. Convening this summit proved to be incredibly difficult in the organisation being evaluated. After three months of negotiation with the various stakeholder groups and aborted attempts to convene the summit with all the relevant, or an acceptable representation of, the stakeholders in the organisation the decision was made by the board of the organisation to abandon the AI summit. The primary reason for the failure to convene the AI summit was that the organisation did not have the material or political resources to gather the stakeholders in the same space for a period of three to four days. To convene the AI summit would require that parents and mentors take leave from their employment and would also entail disrupting the school schedule to include the educators and learners in the process. While these issues were foreseen as stumbling blocks in the process from the outset, the board and researchers hoped that the stakeholder groups would be able to cooperate to ensure that the summit could take place.

The board did not, however, want to abandon the AI approach to the evaluation. They still appreciated the underlying philosophy of inclusivity and the focus on peak performance. After a brainstorming session between the researchers and the board of the organisation, the decision was made to run a series of smaller AI summits and to ask a percentage of participants from each summit to attend the subsequent summit. In this way we hoped to bridge the communication gap between stakeholder groups and to carry over the energy from each summit while keeping the researchers on the periphery of the change process.

With the aid of the executive director, who was the primary point of contact for all stakeholder groups, the dates and times were negotiated for the series of summits. An interesting feature of these negotiations was the clearly defined time boundaries. The board and school were adamant that schooling hours, including extra lessons were non-negotiable and could not be infringed upon. The parents and mentors were clear that work hours and religious days, specifically Saturday evenings and Sundays, were considered as untouchable. Another unforeseen difficulty in securing a date for the summit was that many of the parents were single parents who did not have any form of assistance in running their households. For these participants, Saturdays were used to manage the household and prepare for the week ahead. The impact of these rigid and limited time boundaries was that we were allowed three hours per summit for each stakeholder group. This is significantly less than the suggested four-day summit proposed by AI practitioners and placed pressure on the research process to provide sufficient time for engagement of the participants and data gathering.

When considering the limitations imposed on us by the participatory nature of the approach and we debated whether the degree of control the participants held strengthened or hampered the research process. In exercising the power inherent in their role as key informants, the participants strictly limited the researchers to the time they were willing to allocate to the evaluation regardless of whether we could perform the evaluation in this time frame. We saw this as a negative effect of the participatory nature of the AI process. However, we may have had an inexplicably low rate of participation had we followed a research approach that did not allow for the open discussion of seemingly mundane and irrelevant aspects of the participants' lives which held such an important space for the individuals participating in the study.

To maximise the use of time available the summits were designed to focus on the Discover and Dream phases of the AI cycle. This focus, we hoped, would generate positive conversations about those moments when the organisation was seen to perform at its peak and provide multiple perspectives of what lends strength to the organisation. The idea was to follow up these summits with planning sessions focusing on the Design and Destiny phases using smaller groups of volunteers from the various stakeholder groups.

We requested that the stakeholders organise themselves into the groups they believed would be most appropriate to achieving the goal of sharing peak experiences of the organisation. The decision was taken to group the mentors with the learners, parents with the school governing body and the board with the school management team.

Implementing the AI summit

After confirming the times and dates for the various summits, the national teachers union embarked on nationwide industrial action, closing most schools (Cohen, 2010). The school in which the study was based was affected by the industrial action and we could not conduct the study on the school premises. Due to the industrial action, the educators withdrew from the study as a group for fear of being victimised by the parents and school governing body. The withdrawal of the educator stakeholder group had a significant impact on the research process. AI is driven by the need for inclusivity (van Sant, 1989; Gergen, Gergen & Barrett, 2004; Valsiner, 2006) and acknowledges that each stakeholder group holds a unique view of what is important and valuable for the organisation (Lewis & van Tiem, 2004). The concern with the withdrawal of the educator group was that we lost the only view of what made the organisation work or not work within the classroom setting. Not only did we lose the perspective of the educators, but we were also concerned about how this would affect the perspective of the learners.

Postponing the study was discussed with the board, but was rejected. The inclusion of the new school required an evaluation of the intervention approach to ensure that the successes experienced thus far could be replicated. Consequently, the dates for the summits were changed and the participants were taken to a location where they could safely participate in the study. The time allocated for each summit remained the same as described above, as did the composition of each summit. We conducted three separate summits with volunteers from each summit giving feedback to the various stakeholder groups on what was discussed and how the summit was conducted. Through these volunteers the ethos of each summit was transmitted or communicated from the perspective of the participants and formed the foundation on which the next summit would be developed.

Towards the end of the research process, some of the educators agreed to meet the researchers in a closed session to discuss their experiences of the programme but did not assent to giving feedback to the other stakeholder groups. This was an interesting, albeit frustrating, turn of events that provided insight into the complexities of the dynamics of state-NGO relations. Further reflections on this point are in the next section of this paper.

The process for each summit was the same: each summit began with a description of the objectives of the study followed by a detailed discussion of the AI process and how it should be implemented to make the summit effective. The summit then moved to the Discovery phase which opened with the statement to: "think about a peak experience you had in the organisation. Describe the moment to us and explain why this moment was so important to you." The open ended nature of the question allowed participants to reflect on their involvement in the organisation and to draw what they found most valuable from this.

The Dream phase of the summit opened with the proposition that participants picture what they would like to see or experience in the organisation in the next three years. They were asked to abandon all consideration of limited resources or buy-in and to simply "dream big". This proposition opened the imagination of the participants to the alternative growth possibilities open to the organisation despite the immediate context and its limitations (Bushe, 2007).

We believed that AI, like most process-driven approaches, requires a significant amount of time and a degree of flexibility during the implementation of the approach to ensure that all participants are engaged and actively participating (Coghlan, Preskill & Catsambas, 2003). What we found in practice was that, once we had outlined how the approach works and discussed how the principles applied to the current summit, participants easily took control of the summit and directed the inquiry into those areas of the organisation in which they were most invested. The appreciative slant to the inquiry offset the tendency to blame and facilitated open, honest discussion about difficulties that the organisation experienced.

Our experience of this approach was that, once the opening questions were asked and the ground rules for interaction were set, the participants were comfortable to discuss their experiences and ideas with little prompting or probing from the researchers. In an attempt to keep the focus on the positive but avoid censorship of the negative, we proposed a "parking lot" where issues that were contentious and held the organisation in a negative cycle were written down and displayed in highly visible locations throughout the venue. These issues were periodically re-visited during the summit to see whether any of the lessons learnt from the positive experiences could be applied toward solving them.

The implementation of the Design phase primarily happened with the board. The information from the various summits was integrated into the plan for the organisation developed by the board as they revised their vision and mission of the organisation. While a working solution, this approach does not include the broad stakeholder interaction envisioned in the AI approach and may hold negative implications for the buy-in required to implement the revised plan as the organisation continues to grow.

Reflection on the research process

The purpose of this paper is not to argue for or against the use of AI in the South African context but to contribute to the body of knowledge on evaluative techniques that are based on the principles of empowerment and equal participation. Reflections on the research process consistently raised the question of whether the study described above is an example of bad implementation of the AI approach or whether it questions the utility of the AI approach for emerging NGOs in Soweto.

While the AI approach purports to provide a broad platform for stakeholder interaction, the experience of this study is that it requires a significant amount of political resources to include stakeholders from outside the organisation to participate actively. Our experience of the AI approach is that it seems to be difficult to implement in an environment where there are low levels of trust and cohesion among the stakeholders. A second negative reflection arose from the time required to conduct the AI summit.

The NGO used in this study simply did not have the capacity to stop its operations for four days. Nor could it request the same for the stakeholders, especially the educators and learners, to do the same. Considering that the summit is the central space in the AI process, this is a critical problem. While a workaround was attempted by hosting a series of summits with smaller groups, it could be argued that this is contrary to the system-wide shift that AI aims to facilitate (Cooperrider & Avital, 2004). Perhaps this research study could have benefited by drawing more heavily on Participatory Action Research by allowing for multiple research cycles that have a consultation process that is more drawn out to allow the research team time to embed themselves within the context as partners rather than external resources (Boyd & Bright, 2007).

The research process was punctuated by points of resistance by participants which provided an interesting insight into the dynamics of providing educational support within this particular context; in particular the resistance of the educators to participating in the process. From immersion in the context it appears to be a crucial issue to highlight the resistance of the educators as resistance to the state and not necessarily to the NGO alone. Based on interactions with stakeholders such as the school principal and immersion in the environment it appears that the resistance was driven by the perception that the NGO was co-opted by the state structures. At this point the researchers reflected on whether we were witnessing the "bait and switch effect" described by Patton (2003), which posits that the emphasis on the positive allows AI to be co-opted as a strategy to engage participants who would not otherwise have participated in the evaluation process (Cooperrider, Whitney, & Stavros, 2003). This effect describes how the positive focus of the AI process may be used to ease concerns around the negative preconceptions of the research process (Patton, 2003). Within the context of this study, the effect here possibly took the form of inviting the educators to actively participate in a process from which they were -seemingly -excluded. On closer examination, however, the educator participation was not as active agents in executing the intervention programme but as key informants into how the programme should be executed as others. When considering that the AI process is participant centred (Bushe, 2007), with an emphasis on participatory action principles (Boyd & Bright, 2007); it is particularly ironic that the research process discussed here inadvertently excluded one of the most crucial stakeholder groups.

That said, once the summits were convened, the open-ended approach of AI facilitated an environment that allowed open and honest communication between stakeholders at different levels of the organisation. The focus on what was appreciated and what works led to conversations about those aspects of the organisation that were not functioning optimally could be improved or adapted. This focus allowed participants to move beyond the unit of analysis focus that typifies evaluations to an approach that takes cognisance of various interests, interpersonal dynamics, organisational and community level politics that shape how an organisation (dys)functions (Patton, 2003). Especially when focusing on the effectiveness government supported programmes, it is important to understand how bureaucracies and the power dynamics between larger and smaller ecologies of knowledge affect not only the nature of the programme but the manner in which it is implemented (Patton, 2003; Townsend, Porter & Mawdsley, 2004).

There is scope for further examination of whether this approach is useful in those contexts where the gap between the funding organisation and the NGO is difficult to bridge without significant loss of power for the NGO. This is particularly pertinent when considering the work of Townsend et al (2004) and their description of the common practice of how NGOs subtly subvert donor agendas by appearing compliant to accountability mechanisms while implementing a more relevant approach tailored to their beneficiaries. Despite the challenges faced, the AI approach does allow for stakeholders in an organisation to define their theory of change in a context that recognises the legitimacy of their knowledge without sacrificing the accountability held sacred by donors and government agencies (Gergen, Gergen & Barrett, 2004). Hopefully this approach, when implemented appropriately, can provide a space to formally subvert the "master-servant relationship" between donor and government organisations by being an acknowledged means of communicating not only stakeholder needs but also stakeholder development ideas (Townsend et al, 2004: 877).



This paper reflected on the implementation of a method that could provide an acknowledged space in which emerging NGOs could utilise to develop and describe their development agenda. The AI approach, with its focus on stakeholder participation, could act as a mechanism through which emerging NGOs can comply with the traditional understandings of evaluation practice while challenging the development agendas of donor and government organisations and the implementation of the same through deficit focused evaluation practices.

Simultaneously, this paper questioned the utility of the AI approach with reference to the research team's experience of the research process. Of particular interest was the resistance to the research process by the educator stakeholder group and how this was a manifestation of their inadvertent exclusion from the intervention process. This was particularly ironic when considering the inclusive nature of the AI process and was reflective of what Patton (2003) referred to as the bait and switch effect.

Lastly, there is scope for further inquiry into whether the challenges experienced in this study are present in other contexts or whether this study is an example of how not to implement AI in the South African context.


The author would like to acknowledge Dr Elizabeth Archer for her role in this paper.



Ahmed, R & Sayed, Y (2009) Promoting access and enhancing education opportunities? The case of "no-fees schools" in South Africa. Compare, 39(2), 203-218.         [ Links ]

Alkin, M C & Christie, C A (2004, January 23) An evaluation theory tree. Available at Chapter 2.pdf. Date retrieved: 23 March 2011.         [ Links ]

Arends, F (2007) Teacher education and development in South Africa: A research and development programme. Pretoria: Centre for Evaluation and Assessment, University of Pretoria.         [ Links ]

Boyd, N M & Bright, D S (2007) Appreciative Inquiry as a mode of Action Research for Community Psychology. Journal of Community Psychology, 35(8), 1019-1036.         [ Links ]

Bradley, B S & Morss, J R (2002) Social Construction in a world at risk: Toward a Psychology of experience. Theory & Psychology, 12(4), 509-531.         [ Links ]

Burr, V (1995) An introduction to Social Constructionism. New York: Routledge.         [ Links ]

Bushe, G R (2007) Appreciative Inquiry is not (just) about the positive. OD Practitioner, 30-35.         [ Links ]

Bushe, G & Kassam, A (2005) When is Appreciative Inquiry transformational? A meta-case analysis. The Journal of Applied Behavioral Science, 41(2), 161-181.         [ Links ]

Coghlan, A T, Preskill, H & Catsambas, T T (2003, Winter) An overview of Appreciative Inquiry in evaluation. New Directions in Evaluation, 5-22.         [ Links ]

Cohen, M (2010) South African teachers' strike shuts schools, compounds educational crisis. Available at Date retrieved: 31 December 2012.         [ Links ]

Cooperider, D & Srivastva, S (1987) Appreciative Inquiry in organisational life, in Pasmore, W A & Woodman, R W (eds) Research in Organisational Change and Development, 1 (129-169). Greenwich: JAI Press.         [ Links ]

Cooperrider, D L (1990) Positive image, positive action: The affirmative basis of organising. Available at Date retrieved: 24 April 2011.         [ Links ]

Cooperrider, D L & Avital, M (2004) Advances in Appreciative Inquiry - Volume 1: Constructive discourse and human organisation. London: Elsevier.         [ Links ]

Cooperrider, D L & Whitney, D (1998) A positive revolution in change: Appreciative Inquiry. Available at Date retrieved: 27 April 2011.         [ Links ]

Cooperrider, D L & Whitney, D (2005) A positive revolution in change: Appreciative Inquiry, in Cooperrider, D L, Sorensen, P F, Whitney, D & Yaeger, T F (eds) Appreciative Inquiry: Rethinking human organisation toward a positive theory of change. Champaign: Stipes.         [ Links ]

Cooperrider, D L, Barrett, F & Srivastva, S (1995) Social Construction and Appreciative Inquiry: A journey in organisational theory, in Hosking, D, Dachler, H P & Gergen, K (eds) Management and Organisation: Relational alternatives to individualism (157-200). London: Ashgate Publishing.         [ Links ]

Cooperrider, D L, Whitney, D & Stavros, J M (2003) Appreciative Inquiry handbook. San Francisco: Berrett-Koehler.         [ Links ]

Cross, M, Mungadi, R & Rouhani, S (2002) From policy to practice curriculum reform in South African education. Compararative Education, 171-187.         [ Links ]

Datta, L (2003, Spring) Important questions, intriguing method, incomplete answers. New Directions for Evaluation, 37-46.         [ Links ]

Datta, L (2007, Spring) Looking at the evidence: What variations in practice might indicate. New Directions for Evaluation, 35-54.         [ Links ]

Demombynes, G & Ozler, B (2005) Crime and local inequality in South Africa. Journal of Development Economics, 76, 265-292.         [ Links ]

Dunlap, C A (2008) Effective evaluation through Appreciative Inquiry. Performance Improvement, 47(2), 23-29.         [ Links ]

Eshach, H (2007) Bridging in-school and out-of-school learning: Formal, non-formal, and informal education. Journal of Science Education and Technology, 16(2), 171-190.         [ Links ]

Feynman, R P (1965) Richard P Feynman - Nobel Lecture: The Development of the Space-Time View of Quantum Electrodynamics. Available at prizes/physics/laureates/1965/feynman-lecture.html. Date retrieved: 14 May 2013.         [ Links ]

Fitzgerald, S P, Murrell, K L & Miller, M G (2003) Appreciative Inquiry: Accentuating the positive. Business Strategy Review, 14, 5-7.         [ Links ]

Gergen, K (1999) An invitation to Social Construction. California: Sage.         [ Links ]

Gergen, M M, Gergen, K J & Barrett, F (2004) Appreciative Inquiry as dialogue: Generative and transformative, in Cooperrider, D L & Avital, M (eds) Advances in Appreciative Inquiry -Volume 1: Constructive discourse and human organization. Oxford: Elsevier.         [ Links ]

Grant, S & Humphries, M (2006) Critical evaluation of Appreciative Inquiry: Bridging an apparent paradox. Action Research, 4(4), 401-418.         [ Links ]

Gray, D E (2005) Doing Research in the real world. Thousand Oaks: Sage.         [ Links ]

Habib, A (2003) State-Civil Society relations in Post-Apartheid South Africa, in Daniel, J, Habib, A & Southall, R (eds) State of the Nation: South Africa, 2003-2004 (227-241). Cape Town: HSRC Press.         [ Links ]

Hammet, D (2007) Cuban intervention in South African health care service provision. Journal of Southern African Studies, 33(1), 63-81.         [ Links ]

Jacobsgaard, M (2003, Winter) Using Appreciative Inquiry to evaluate project activities of a nongovernmental organisation supporting victims of trauma in Sri Lanka. New Directions for Evaluation, 53-62.         [ Links ]

Jansen, J & Taylor, N (2003) Educational change in South Africa 1994 - 2003: Case studies in large-scale education reform. Education Reform and Management Publication Series, 2(1), 1-47.         [ Links ]

Lewis, J & van Tiem, D (2004) Appreciative Inquiry: A view of a glass half full. Performance Improvement, 43, 19-24.         [ Links ]

Liebowitz, B (2000) Education for democracy: Some challenges facing education in South Africa. Emerging Democracies, Citizenship and Human Rights Education. Washington: Educational Resources Information Center (ERIC).         [ Links ]

Matoane, M & Fynn, A (2010) First steps: The first annual report of the ERF. Johannesburg: The Educhange and Research Foundation.         [ Links ]

McNamee, S (2003, Winter) Appreciative Evaluation within a conflicted educational context. New Directions for Evaluation, 23-40.         [ Links ]

Miller-Grandvaux, Y, Welmond, M & Wolf, J (2002) Evolving partnerships: The role of NGOs in basic education in Africa. Washington DC: US Agency for International Development, Bureau of Africa, Office of Sustainable Development.         [ Links ]

Molteno, F (1984) The historical foundations of the schooling of Black South Africans, in Kallaway, P (ed) Aparthied and education. Johannesburg: Ravan Press.         [ Links ]

Moore, M (2008) Appreciative Inquiry: The why? The what? The how? Practice Development in Health Care, 7(4), 214-220.         [ Links ]

Motala, S (2009) Privatising public schooling in post-apartheid South Africa - Equity considerations. Compare, 39(2), 185-202.         [ Links ]

Nelson, G & Prilleltensky, I (2005) Community psychology: In pursuit of liberation and well-being. London: Palgrave MacMillan.         [ Links ]

Patton, M Q (2003, Winter) Inquiry into Appreciative Evaluation. New Directions for Evaluation, 83-98.         [ Links ]

Reed, J (2007) Appreciative Inquiry: Research for change. Thousand Oaks: Sage.         [ Links ]

Rogers, P J & Fraser, D (2003, Winter) Appreciating Appreciative Inquiry. New Directions in Evaluation, 75-83.         [ Links ]

Rose, P (2009) NGO provision of basic education: Alternative or complimentary service delivery to support access to the excluded?. Compare, 39(2), 220-233.         [ Links ]

Sabates, R, Akyeampong, K, Westbrook, J & Hunt, F (2010) School dropout: Patterns, causes, changes and policies. Vienna: Paper commissioned for the EFA Global Monitoring Report 2011, The Hidden Crisis: Armed Confilct and Education.         [ Links ]

South African Department of Education (1998) Further Education and Training Act No. 98 of 1998. Cape Town: South African Government Gazette.         [ Links ]

South African Department of Education (1995) The White Paper on Education and Training. Cape Town: South African Government Gazette.         [ Links ]

Terre Blanche, M & Durrheim, K (1999) Social constructionist methods, in Terre Blanche, M & Durrheim, K (eds) Research in practice. Cape Town: UCT Press.         [ Links ]

Terre Blanche, M & Kelly, K (1999) Interpretive methods, in Terre Blanche, M & Durrheim, K (eds) Research in practice. Cape Town: UCT Press.         [ Links ]

UNESCO (1997) International standard classification of education. Washington DC: UNESCO.         [ Links ]

Valsiner, J (2006) Dangerous curves in knowledge construction within Psychology: Fragmentation of methodology. Theory & Psychology, 16(5), 597-612.         [ Links ]

van Sant, J (1989). Qualitative analysis in development evaluations. Evaluation Review, 13(3), 257-272.         [ Links ]

Watkins, J & Mohr, B (2001) Appreciative Inquiry: Change at the speed of imagination. San Francisco: Jossey-Bass/Pfeiffer.         [ Links ]

Whitney, D & Trosten-Bloom, A (2003) The power of Appreciative Inquiry. San Francisco: Berrett-Koehler.         [ Links ]

Williams, S E (2010, October) Exploring the viability of school-based support for vulnerable children: A case study of two township schools in Johannesburg. CREATE Pathways to Access Research Monograph No. 46, 1-36.         [ Links ]



* Corresponding author:

Creative Commons License Todo o conteúdo deste periódico, exceto onde está identificado, está licenciado sob uma Licença Creative Commons