On 11 September 2013, Professor Elaine Unterhalter gave the BAICE Presidential Address at the 12th UKFIET International Conference on Education and Development: Education and Development Post-2015: Reflecting, Reviewing and Re-visioning at the University of Oxford, UK
The text of Professor Unterhalter’s Address, Walking backwards into the future: A comparative perspective on education and the post 2015 framework, can be found below and can be downloaded here. This is an Author’s Original Manuscript of an article submitted for consideration in Compare [the exclusive right to publish residing with Taylor & Francis, copyright BAICE].
Once the published Presidential Address is available in Compare, you will find the link here. In the meantime, please enjoy Professor Unterhalter’s Author’s Original version here on the BAICE website.
Walking backwards into the future: A comparative perspective on education and the post 2015 framework
BAICE President, Professor Elaine Unterhalter, Institute of Education, University of London
12th UKFIET International Conference on Education and Development: Education and Development Post 2015: Reflecting, Reviewing, Re-visioning
University of Oxford, 10-12 September, 2013
ABSTRACT: Good education is central to many contemporary visions for a post 2015 framework, given prominence in popular consultation and official policy pronouncements. This places considerable responsibility on our network of academics and practitioners in commenting on how we can support this process. This BAICE Presidential Address reviews what special perspective a comparative approach brings to visions for the future. It comments on differences between the confident assumptions made about the world by the architects of the MDGs and EFA, and those we now draw on in which uncertainty plays a key role. It argues that our view of the present and the recent past are a crucial resource to use in considering the future. But the nature of this comparative process needs clarifying. Descriptive and prescriptive approaches to comparison, often isolating a single variable, are different to approaches which are more reflexive and pay attention to uncertainty and provisionality. In highlighting the scope of each approach to comparative education this lecture argues for greater clarity regarding what the grounds for comparison are in addressing some of the challenges of post 2015. Reflections are presented from a number of research projects I have contributed to over the last five years looking at the implementation of the gender, poverty, and education MDGs and aspects of EFA in different sites, ranging from multilateral organisations to local NGOs in a number of African countries. These suggest a number of concrete strategies for future research, policy development and approaches to measurement and practice, which acknowledge, and go with the grain of uncertainty. In this way, the Presidential Address attempts to draw out some of the potential of reflexive comparative education.
Thanks to Caroline for the introduction and to BAICE for the honour of appointing me their President for the past year, which entailed the invitation to deliver this lecture. My thanks also to my students, former students, colleagues, friends, and family who have been willing to discuss some of the ideas in this talk over the past few weeks. It is a little daunting to do this, not least given the importance of this conference and after Amina’s thought provoking address, but here goes…
A good education is emerging as the number one priority for the post 2015 world for the nearly 1 million people around the world who have voted in the UN/civil society poll regarding the world we want. (The blue boxes at the top of this bar chart indicate this) .As specialists in this field the responsibilities of reflecting, revisiting and revisioning education for the world we want are considerable. What special perspective does a comparative approach bring? The argument I want to make is that much of the work on the MDGs and EFA, which focuses on enrolment, progression and attainment takes too narrow a remit for why we do comparative education. Although much useful traction on the problem of children out of school or learning little within can be gained from looking at a limited range of issues, important insights are lost. These include why distribution is patterned in particular ways, the nature of the relationships between people, ideas and social formations, how and why particular people do ordinary and extraordinary things, what normative compass we take, and why. Crucially such a limited remit disorientates us as to why we might do comparison at all. In this lecture I want to reflect on what some of these gains and losses are and then argue for a distinct approach to the comparative process, which I consider is especially useful to a post 2015 framework.
I must admit at the outset to being unable to come down off the fence in relation to the MDGs and EFA. I see these global frameworks as offering opportunities, but also difficulties. The MDGs and EFA have put together a significant global collaboration to expand and improve the provision of education. Much has been gained and learned in this process. But persistent critiques of the MDGs, include, firstly, that in specifying goals, targets and indicators there is a separation out of concerns with education, health, and poverty so that each is approached down its own track with little thought to contributions from the others; sometimes this approach has generated contradictions between MDGs . I am not sure the High Level Panel (HLP) report on post 2015 in identifying principles, such as ‘no child left behind’ but not processes for sustaining the dynamic between goals, has addressed this question adequately . A second critique is that the approach to development entailed by the MDGs focuses on results and sets aside engagement with larger ethical questions concerned for, example with the nature of social justice or global obligation. Fukuda Parr and Hulme(2009) see this as entailing a rejection of equity for crossing minimum lines of sufficiency. The HLP introduces equity translated through metrics of distribution, rather than principles or practices. This may be adequate, but why it is adequate, and with what provisos requires careful consideration. It is a technical response to a political problem, which may address some, but not all of the critiques of a lack of focus on equity.
Both critiques implicitly invoke a particular form of boundary setting. The charge of working in differentiated areas of social policy identifies boundaries which run vertically deploying modes of professional knowledge and practice associated with particular administrative forms with regard to health, education, or poverty reduction. In education it has had the effect of separating out a focus on universal primary education (UPE), from concerns with other phases, such as secondary or higher education, and of privileging enrolment over quality. The charge of providing education, food, income or health care only up to a bare minimum, completing a primary cycle of schooling or earning $1a day, invokes horizontal boundaries, in which particular kinds of ideas about sufficiency (set , in my view, at too low a level) and the location of responsibility are professionally and administratively endorsed, rather than more widely debated and engaged. These processes of setting horizontal and vertical boundaries in education and social policy have effects. They have galvanised much useful action, as you all know well. But in pointing much policy, planning and practice in a particular direction, they empty out the space in which we can think about opportunities, outcomes, rights, capabilities, people and relationships. Indeed, by suggesting the main format of comparison is the league table, they close off an investigation as to what our reasons for making comparisons are. They provide particular frameworks which generate answers, but do not enrich our capacity to address some of the pressing problems we confront. I am going to argue that a different kind of comparison, that I term reflexive comparison, is a useful complementary approach.
The historian, political theorist and commentator on South East Asian area studies, Benedict Anderson, refers to the spectre of comparison. By this he means a giddying imaginary feat where one attempts to hold together a collection of impossible opposites: the strange and the familiar, constructed ideas about a place, a nation and a region and its real time dynamics, the explanatory envelope of a theory and the messy detail of real events. I find this a particularly generative idea because it suggests the traces the past leaves on the present are an important resource for trying to make sense of confusing times. This resonates for me, partly because I have an academic background as a historian, and partly because my formative educational experiences were in South Africa, where everything about where and how we learned was saturated with historical struggles over race, class and gender relationships. This is a photograph of the South African past, taken when I was at school but not of my school. It is, using Anderson’s phrase, a spectre of comparison that is always with me. The black freelance photographer, Ernest Cole, took the photo at a school in Johannesburg, only a few miles geographically, but socially a world away from my suburban school.
Thinking about Anderson, who like many comparativists, raises questions about the process, and, returning constantly, like so many South Africans, to concerns about inequalities, a central question for me is what particular kinds of comparisons are useful now. In trying to answer this and develop a notion concerning a reflexive historical perspective, I have been taken by what I understand is the linguistic form of the Aymara language, spoken in the Andes, which alerts us that we walk backwards into the future. It is the past and present that are in front of us and the future that is behind us. We go forward looking at what we are doing in the present and possibly what we’ve done in the past. Thus we need to make reflexive comparisons to go carefully.
Inevitably, when we look to the future what we see is a distorted version of the present. The High Level Panel report on Post 2015 does this explicitly. It identifies 6 features of the present that will accelerate and contribute to building the vision of peace, justice and prosperity they wish for the post 2015 world. Their list for the future comprises a doubling of global output, with significant growth in developing countries. It entails a refocused development co-operation as many of today’s low income countries will be middle income. Thus development co-operation can be refocused. Enormous demographic expansion, increasing international migration, and a shift to the major portion of the world’s population to cities will generate demographic dividends. The widespread use of affordable technologies will enhance information and communication flows. Interestingly for us, these assumptions do not include comments on education or on gender and other equalities. I think they highlight how uncertain we still are about what we have achieved.
In 2000 the architects of the MDGs looked forward to now, and believed now would work like then, with rational flows of aid linked to national planning, growing economies with a private sector not a major player in education, imminent agreement on climate change and sustainable development. For a range of reasons, partly to do with the Millennium moment, they were less reflexive than they should have been. In some things and for some places they were right, and in others they were wrong. On how you get children to go to school, I think they were right for those children who were out of school because they had to pay fees, or schools were too far away. But they did not know in 2000 why other children were out of school or attending irregularly, and they did not take the time to ask how particular kinds of social relationships structured these inequalities. The assumption yielded results and the reduction of the numbers of children out of school are impressive. UNESCO estimated the number of primary school aged children out of school in 1999 at 108million. By 2010 this had fallen to 61 million. In 1999, it was estimated only 75% of primary school children in developing countries remained to the last grade, but by 2009 this had increased to 81%. . Through the actions of governments, supported by the Education for All (EfA) movement, and the mobilizing dynamic of the MDGs, which was largely about directing money to primary schools, more children have had the opportunity to enrol and progress through school than ever before. Partly the MDG planners were right, because they had the money and influence to be right. But their vision of the future was also incorrect, partly because they did not know enough and partly because they did not take time to ask. One of the things they did not know very much about was gender, even though gender equality and women’s empowerment was an MDG goal, gender parity was an MDG indicator, and gender equality in education is one of the EFA goals. In these frameworks gender is separated out, as a particular field of expertise, and targets are to be met by crossing a minimal line of sufficiency, in the MDG case equal amounts of provision for girls and boys, ie gender parity. But this descriptive meaning of gender as a noun in no way addresses the complex relationships of power, history, emotions or reflection that the scholarship on gender alerts us is particularly salient. The frameworks we have had have thus focussed on comparing gender parity in enrolments, and have blocked off richer engagements with the meaning of gender and how and why we should compare.
A second area they did not take time to inquire into concerned the content of schooling. For the architects of the MDGs, although less so the drafters of the Dakar framework, increasing enrolment was more important than improving quality. However, for both no attention was given to equality except in a broad brush stroke concern with gender parity. I surmise that thinking the future might look like the present, the content of education was considered primarily a matter of national deliberation or market forces. The architects of the MDGs and EFA, having lived through the 1990s, with the optimism generated by the end of the Cold War, did not anticipate our hard times where destructive environmental events are not occasional disasters, but almost monthly occurrences, where the hope of the Arab spring has turned to confusion, confrontation and terrible war, where political leadership has not been forthcoming to support the poor through a global economic crisis, and where, despite the highest levels of access to education and information ever, many young people cannot get decent jobs or anticipate better futures. This suggests to me the content of education is of crucial importance. Our capacity to understand these processes, and think through responses, requires dialogues and understanding of very difficult issues. As a minimum we all need in depth teaching about sustainability, equality and solidarity.
The MDGs were in part successful, because, by design, they set out to disregard complexity, the traces left by history and established comparison largely as a technique of information control, comparing one item of information about each country with another, say net enrolment rates. Their focus on results based management suggested complexity was irrelevant. But they were also unsuccessful because they paid insufficient attention to the complexity of education, the way in which the activity entails human ideas, aspirations, needs, emotions, and relationships, real people and their lived experiences all resonating with histories. Even using its own framework the MDGs were problematic partly because they neglected seeing education as a continuum between levels. The stress on UPE meant little consideration was given regarding how enough teachers would be trained to deliver UPE, where trained birth attendants for MDG 5 would come from and how, with only primary education, women would have enough status to stand for parliament as required by one of the indicators for MDG3. When I look at the illustrative goals in the HLP report, I do not think that this disarticulation has been addressed. The disarticulation relates partly to how the goals might work together, but also, crucially what kinds of education, and to what levels, for whom, where, are needed to make the goals work.
Thus the MDGs and EFA, were exercises in eliciting comparisons between countries and regions in relation to a limited range of areas of intervention. These limits were both fruitful, but also restrictive. In 2005 I wrote a number of articles about the MDGs, which ended with the hope that the simplicity of the goals and indicators would generate more work on complexity. That is by engaging with the limited notion of achieving gender parity in education, equal numbers of girls and boys enrolled or passing each level, more people would be drawn to discuss the nature of gender equality and women’s rights.
In 2013, the areas in which global frameworks have and have not realised this promise appear more evident. My research on the GEGPRI project, a comparative, collaborative study of how meanings of gender, education and poverty were understood and negotiated in a range of sites where policy relating to MDG 1, 2 and 3 were being implemented. These sites of vertical comparison from global to local, and horizontal comparison between state and non –state provision are shown on the slides. The findings from the study suggest that the complex forms of the political economy of the global policy framework, the construct of the nation in South Africa and Kenya, or affiliations around a movement like EfA, contains the simple directives associated with the MDGs and indicates what emphasis will be given to these allegedly simple steers to policy. Thus, in some sites, the framing of the MDG around gender parity limited work on gender equality issues, and that the lack of attention to building and supporting women’s movements constrained the implementation of policies to address violence against women and support and address the multiple sites in which gender and other inequalities persist
Reflecting on the experience of researching the implementation of EFA and the MDGs in GEGPRI, the question that emerges is not just what we found, but what its implications are, particularly how can a comparative approach help to take forward the question regarding what kind of education we want for a post 2015 world? I think particular insight in response to this question can come from understanding the grounds of comparison, considering historically why people have undertaken comparisons. A simile, the comparative figure of speech, works by saying one thing is like another. But it is accepted that this is not a literal comparison. When we say a woman is brave like a lion, we do not mean she is literally like a lion, or even that her bravery looks like the actions a lion makes. What we mean is that our idea about her bravery gains depth by being compared with ideas of strength and nobility that the symbolic figure of the lion conveys. It is this loose and evocative sense of comparison that I think is helpful for us in contemplating a comparative perspective on the relationship of complexity and simplicity, and the connection from the present to the future. Our academic reasons for doing comparisons in this way in a time of uncertainty are different, to, say the ways in which third world nationalist leaders did comparisons in the 1960’s to build the solidarity of an anti-colonial struggle, or the ways in which global corporations monetize a particular form of education and compare returns in different locations to maximise profits.
There are two features of what I want to call an evocative or reflexive mode of comparison that I want to draw out and explain why I think this is a particularly useful form of comparative education for the present. The first is that the grounds of comparison apply only to certain features, a part, not a whole. The second is that the special perspective comparison brings is the introduction of a relational dynamic which asks how and why these features are similar and different, what our responses to this are and why they matter. Comparison is thus not immediately evident, but its significance is made reflexively, which is important given the uncertainty of the times. Let me give some examples to illustrate this from recent research projects. The GEGPRI project set out to investigate engagements with MDG 1,2 and 3 in a range of different settings from the global to the local. Using a standard vocabulary of comparative analysis our finding were very similar to others on the implementation of EfA and the MDGs, noting how global frameworks, fail to attend sufficiently to national or local contexts. Some extracts from the data on the slide show this clear support in national ministries and lack of understanding and sense of distance in a school and rural NGO. Camfield et al discussing the MDgs have termed this the contrast between approaches based on measurement and those based on voice. However, another finding from the project showed that it was important to distinguish different locations where global frameworks were put into practice. The MDGs were a familiar framework to particular groups of people, those, for example, working in global organisations, at senior levels in national Ministries of Education, and in NGOs engaged with the global agenda. They all talked the same language and shared the EfA or MDg framework. This was unsurprising, as was the general lack of knowledge or engagement with this framework by teachers or Ngo workers in very rural contexts. The extracts from interviews I and others in the research team carried out which highlight how close and distant different groups of practitioners felt from the MDG project. This is quite familiar terrain
But what was new, that we had not expected, was that more junior staff in national ministries, officials at the provincial level, and workers in a range of NGOs had only a very partial understanding of the MDG or EfA framework and supplemented this with confusion about the meaning of gender, essentialized and stigmatized descriptions of the poor, in which blame was a major trope. These extracts from the data exemplify this. On a more positive note we documented a middle level official, in this case in South Africa, who had studied a gender course at university, helping some school girls who were trying to bring a complaint against a teacher for sexual harassment. From the interview notes what emerged was her concern to support and protect them, buy them air time for their phones to keep in touch with each other, accompany them to the hearing and make sure they were sustained through the harrowing business of giving evidence.
The nature of this middle group, the history of its different constituents, its key importance in connecting top-down and bottom up processes, whether and why it shares interests or concerns with measurement and/or with voice, and how it is regarded by constituencies who implement global policy at the top and critique it at the bottom, all seem to me extremely pertinent questions for any post 2015 framework. It is this group that will deliver a significant component of the practice of education post 2015 and we need to understand them better. Up till now we have looked at this group as a node on a system concerned with decentralisation or efficiency or teachers’ professional skill. We have not seen this as a peopled group, cross cut with historical, political and social relationships articulating capabilities. It is this that I think particularly needs addressing. But these questions about the middle are not questions that would arise without considering comparison as relational and evocative, rather than descriptive and prescriptive.
Here is a second example. The TEGINT project was a collaboration I worked on with ActionAid, local implementing partners and co-researchers in Tanzania and Nigeria working to enhance poor girls’ participation to claim rights at school. Its rationale was framed by the MDGs and EFA and the kinds of partnerships it established between government officials and NGOs, north –south dialogue, research , policy and practice exemplified many of the best features of the EFA movement. The project drew on knowledge circulating in the EFA community of how to change practice and entailed establishing girls’ clubs work on gender with teachers and school management committees and some rich dialogues about how to use research data in relation to project implementation. Because the contexts in Tanzania and Nigeria were different, one reading of the research findings emphasised the importance of understanding context to interpret the indicators we developed linked to gender profiles and empowerment. But the project highlighted an area that was not prefigured in the MDGs or EFA, but which has gained prominence in discussion of a post 2015 framework, notably the extent of violence against women and girls, and some of the silences and difficulties in working with schools to change this, which figured in the empowerment indicator we developed for the project endline research. But this work raises wider questions. While the HLP illustrative goal on gender includes a candidate indicator to eliminate all forms of violence against women, this begs many questions. What are the grounds for comparison of different forms of violence , how do different histories form similarities and differences regarding violence, what kind of knowledge do we have about the extent of violence, its interpretation, and how it does and does not link with poverty and other inequalities? To counter violence, what kind of education work do we need and how can this be addressed in different kinds of societies? Why is this an important area of women’s rights activism, and how does this work, often undertaken by civil society, connect with education, generally the realm of the state and political economy? These seem to me important questions for the future, touching crucially on the grounds for comparison between the public and the private that a reflexive, evocative comparative education attentive to histories could undertake.
My third example relates to the problem of intersecting inequalities and the inter-relationship of particular kinds of structures of discrimination. The MDGs and EfA used gender, and a particularly condensed notation for measuring this, as a shorthand for inequality. Important work has been done on identifying other sites of inequality, associated for example with the nature and form of work, ethnicity, disability, or class . But holding particular kinds of inequality, like poverty or gender or rurality, separate from each other does not allow one to see how they interconnect, except in some analysis of intensification, as these figures from the latest GMR on skill levels, suggest. These show that in countries where a majority of rural people do not complete lower secondary school, young women from these regions comprise the largest proportion of those who lack foundation skills. In Benin, Cameroon, Liberia and Sierra Leone, about 85% of young women living in rural areas have not completed 9 years of schooling, compared with 70% of young men. But what how do we take this information further? Understanding inequality only in terms of position on a particular line or series of lines closer or further away from a desired point, does not allow you to understand where the exclusions associated for example with gender, poverty, and rurality may overlap, why, what and where the structures of exclusion are, how histories have shaped this, in what dimensions they are different, and what feelings and actions are elicited in response. In the TEGINT and GEGPRI projects one of the findings was that children in rural areas, who did not see other children at school, and had absolute, rather than relative experiences of exclusion, found it more difficult to articulate their demands. It pointed to the need for projects to give poor children more exposure to other children’s lives, to build their confidence to claim rights. The power relations associated with a system of classification of inequalities used in EfA and the MDGs has generated a tendency to work primarily with ideas based on inequality defined by falling below particular lines, rather than to think more complexly about structure, agency and context in addressing inequality. This yields a social policy in which professionals, who confront these problems in day-to-day work, have inadequate knowledge resources or processes for gathering information or reflexively engaging with complex inequalities. The missing resources are partly financial, in that there are not budgets of time or money to attend to making connections and this requires government action. But in addition they are conceptual, so that the ways in which ideas connect in the direction of social justice is not modelled at the international or national level and replicated downward to province, community or school, and there are no mechanisms to take good practice from local settings outward or upward. In addition no space or analytical authority is given for reflection on everyday practice and how this can help with reviewing processes of classification and framing. A more reflexive comparative education can, in my view, assist with this connection.
What are the implications for the future of applying this reflexive, evocative approach to comparison, drawing on a complex understanding of the past in the present, constantly questioning the grounds for comparison? I want to suggest three and link these with some concrete strategies for research and work on policy and practice.
Firstly, this approach to comparison highlights the importance of bringing together methodologies that allow one to not only answer questions relating to what, but also how and why. We increasingly have the capacity to collect and analyze big data. This is likely to grow in the post 2015 world. But only asking questions of distribution or association, even with multi-dimensional metrics, without critically examining why this might be important to know, what the comparison does and does not tell us, how long these trends have existed, how and why they have changed, is to miss too much. We need to develop better understandings of processes and how these link to outcomes, and what the implications of actions might be, rather than miss many of the opportunities big data offers, through rigid conceptions of acceptable research design or approaches to interpretation. Therefore an important strategy entails research teams working collaboratively using quantitative and qualitative methods to complement each other.
Secondly, the approach indicates the importance of researching relational dynamics – how do the top and bottom speak to each other, what role is played by structure and agency in the middle, particularly for that crucial group of classic middle ground professionals – teachers and local administrators. How and why do different levels of education articulate and what does this tell us about practice and development? How do we make practical knowledge about the intersection of inequalities and how do we support critical reflection on this? How do the life histories and assumptions of this group help us understand their practice? If the post 2015 framework is not just about crossing particular kinds of lines of minimum adequacy but about fostering more complex engagements for equality and addressing poverty, building these kinds of relational engagements into policies, programmes, or measurement frameworks seems crucial. Therefore a second strategy entails developing approaches to enquiry and practice that do not overlook groups in the middle, the new and very fragile middle classes in developing countries, the people who work on say poverty and gender at the same time, approaches to measurement that help us capture multidimensionality, including various forms of inequality in education.
Third, the approach tries to introduce provisionality and uncertainty together with an invitation to be explicit about a normative fame. The work of many friends and colleagues who have looked into deepening ideas about the right to education, particularly through consideration of what the notion of capability adds, has been particularly generative. An important dimension of this works with, rather than against the grain of uncertainty. This is a strength, because it emphasises the need for critical reflexive, historically aware deliberation in assessing rights and capabilities. For example, the notion of capabilities adds to rights a sense of human diversity, and expands the contexts for realisiation. These entail not just public spaces or the responsibilities of states, but a range of private spaces as well, including the family, civil society, the market, or a faith community. Capabilities signal opportunities that require material and non-material resources. A third strategy entails giving more, not less time to, many of the questions associated with a discussions of rights and capabiltiies. Some of these are associated with education, culture, the relation of private and public, heterogenous age groups and the generations who are unborn. What does history tell us and what do we want to tell history about our normative strategies? A key area where I think reflexive normative discussion has a great deal to contribute concerns the current stand-off relating to proponents and critics of public and private provision of schooling for the poor. The history of the application of the Ruggie principles on business and human rights to protect, respect, and remedy in relation to the actions of trans-national corporations and the demand that remedy is rigorously monitored by civil society and trade unions suggest to me a useful way forward to evaluate some interventions of the private sector in education. Our normative range from the UDHR through CEDAW to CRC is broad, and any post 2015 framework needs to complement, not displace it.
The argument I have made is that reflexive, evocative comparison is important in these times of uncertainty, partly because it compels us to ask deeper questions about the past and the present and consider some of the implications of the provisional nature of knowledge. Recent trends in comparative and international education have either gone for high levels of certainty on a narrow band of issues, or very detailed contextualized accounts of a particular process at the top, the bottom or in transfer between places. My suggestion is that academic networks such as ours pay reflexive attention to how to connect these two trends, thinking about people and ideas in the middle, and giving further consideration to their histories. Secondly, we need to think more about the reasons, both normative and political for comparison, and hence the kind of education we have reason to believe we want. In giving more attention to these two processes, I consider we may be able to walk more thoughtfully backwards to the future.
As I was writing this talk I was saddened by the death of the poet Seamus Heaney, who explored evocatively many ideas about the traces of the past and what we hope for the future. He was, among his many attributes, an anti-apartheid activist. He was translating the Cure at Troy in 1991 at the time Nelson Mandel was released from prison. This is my last slide.
History says, Don’t hope
On this side of the grave,
But then, once in a lifetime
The longed-for tidal wave
Of justice can rise up
And hope and history rhyme.
 UNESCO, 2012, EfA Global Monitoring Report Paris: UNESCO, 355
 UNESCO, 2012, EfA Global Monitoring Report Paris: UNESCO, 363