Open Research Online The role of OER localisation in building a knowledge partnership for development: insights from the TESSA and TESS-India teacher education projects

Open educational resources (OER) continue to support the needs of educators and learners globally. However, it is clear that to maximise their potential more focus is needed on reuse and repurposing. Accordingly, adapting OER for local contexts remains one of the greatest challenges of the open education movement, with little written about how to support communities of users to adapt materials. This paper emerges from the ongoing debate around education quality in low income countries (LICs), taking as its focus two OER projects led by the Open University –TESSA and TESS-India. These projects have collaboratively developed core banks of OER for teacher education that respond to regional and national priorities and pedagogies. In this paper we explore how the projects have supported localisation of the OER and how processes of OER localisation can contribute to more equal knowledge partnerships in the pursuit of education quality.


Introduction
As ever more open education resources (OER) are produced with the aim of widening access to learning in international contexts, debates around the localisation of OER have been increasingly voiced (e.g. West, Taylor & Teemant, 2011). It is generally agreed that sharing OER across continents is not just a matter of distributing resources to those who need them on a "one size fits all" basis-"whereby the rich north would push these resources at the south without thought of reciprocity" (Glennie, Harley, Butcher & van Wyk, 2012, p. v). Bateman, Lane and Moon (2012, p. 3) observe a tendency for the OER Movement to be seen as (and see itself as) "benevolent, developed country 'providers' of OER' as distinct from 'passive, developing country 'users' of them," while Miyagawa (2005) warns that by ignoring such concerns we may see a global information society resembling "a map of the world in the 16th century composed of those that colonize and those that are colonized." Unsurprisingly, an outcome of the 2012 UNESCO World OER Congress in Paris was the suggestion that OER producers need to give more attention to reuse and repurposing, yet adapting OER for local contexts remains one of the greatest challenges of the open education movement (Wolfenden & Buckler, 2012) and very little has been written about how to support communities of users to adapt materials.
Open Praxis, vol. 6 issue 3, July-September 2014, pp. 221-233 Our paper is situated in this literature, and also in the ongoing debate around the quality of education in low income countries (LICs). Access and enrolment have been a key focus of government strategies to meet the international Education for All (EFA) targets and Millennium Development Goals for education: primary enrolment in Sub-Saharan Africa increased five times faster between 1990faster between -2005faster between than between 1975faster between -1990faster between (UNESCO, 2010. However, a decline in pupil achievement has been reported across expanding systems in LICs, and targets for increasing the number of teachers and improving the quality of teaching are gaining momentum: "quality" education is increasingly understood in relation to its appropriateness and relevance to learners (UNESCO, 2005;Tikly & Barrett, 2011;Buckler, 2012).
With all of the potential of OER it is essential that the pursuit of global standards of education quality connect with national policies and local needs. It could be argued then that for OER to be truly valuable within an EFA agenda they need to be truly open (in terms of both licence, and access) for adaptation. This paper explores the work of two projects which are working to maximise access to and appropriateness of the OER they create. It describes and debates the localisation processes of the Teacher Education in Sub-Saharan Africa (TESSA) and Teacher Education through School-based Support in India (TESS-India) programmes-two international collaborations devel oping materials for teachers and teacher educators. It aims to extend the global conversation about the localisation of OER through adaptation and repurposing with the aim of contributing to an emerging framework for localisation to ensure more equitable and sustainable OER development and use.

Thematic and theoretical context
Both TESSA and TESS-India were conceived to help address major problems of quality and quantity of teacher education materials in contexts with enormous numbers of unqualified teachers and insufficient capacity to train new and existing teachers. In Bihar state in northern India, for example, 45 per cent of teachers in schools do not have the minimum qualification (MHRD, 2013) and 75 per cent of teacher education institutions did not conduct any training between 2007-2010(UNICEF, 2010. The Sudanese government raised the minimum qualification for teachers to a degree in the 1990s and by 2002 fewer than 10 per cent of teachers had been upgraded (Wolfenden & Buckler, 2013). Sub-Saharan Africa will need an additional five million teachers by 2030 (UNESCO, 2013).
In addition, existing materials for teacher education reportedly fail to align with national policies or curriculum frameworks (NCFTE, 2009), and fail to integrate national ideas around pedagogy with subject knowledge and the realities of teachers' practice (MHRD, 2012;Buckler, 2012). Calls for teachers and teacher educators to be given the opportunity to play a more active role in the development of learning materials are common across the world, but especially in LICs where the gap between national goals and local possibilities is often greater (GoI, 2012;MoE, 2008). TESSA and TESS-India intended to harness the affordances of OER to support national priorities, enrich the pedagogic toolkit of teachers, ensure these toolkits are deeply relevant at the local level by involving teachers and teacher educators in their development and embed teachers' learning in their own contexts, both material and symbolic (Wenger, 1998). Richter and McPherson (2012, p. 202-204) suggest that "OER will be of value. . . only if they are genuinely reusable or at least fully adaptable," but local relevance has tended to be an after-thought within the OER movement. UNESCO's definition of an OER is: the open provision of educational resources, enabled by information and communication technologies for consultation, use and adaptation by a community of users for non-commercial purposes' (UNESCO, 2002); yet adaptation is not mentioned in the minutes from a UNESCO OER discussion forum for higher education (Albright, 2005). A recent survey of individual and institutional readiness for OER in India does not cover adaptation of materials (Harishankar, 2012).
In this paper we situate OER localisation within two development paradigms. First, the notion of "knowledge for development" driven by the World Bank (2011) and others: Obamba (2013, p. 127) indicates a "clear shift" towards the conflation of development with learning or knowledge, and McArthur and Sachs (2009) suggest that a knowledge production paradigm is increasingly emphasised in contemporary development theory and practice. Alongside this we consider the "partnership for development" paradigm popularised by policy papers such as Department for International Development's (DFID, 2005) Partnerships for Poverty Reduction document and the older, but still referenced, OECD (1997) approach to development that emphasises collaboration and contextual embeddedness. We suggest that OER have the potential to straddle these to comprise a distinct paradigm of knowledge partnerships for development, but only if more consideration is given to their localisation potential. We carried out a small-scale study into the localisation processes of TESSA and TESS-India and, here, we consider how analysis of these processes can provide insights into this developing paradigm.

The Teacher Education in Sub-Saharan Africa (TESSA) Programme
TESSA is an OER project based at the Open University (UK) but representing a consortium of teacher education institutions from nine original member countries in Sub-Saharan Africa (Ghana, Nigeria, Kenya, South Africa, Sudan, Rwanda, Tanzania, Uganda and Zambia). The programme is funded through grants from a range of philanthropic trusts and government funds. Between 2006 and 2009 TESSA academics collaboratively created a bank of 75 Pan-African OER study units for teacher development in five subject areas (literacy, numeracy, science, social science and life skills/ arts); all units followed the same template of activities, case studies and resources. Materials were developed through workshops and followed up with virtual working. The resources are conceptualised as a "professional learning and strategy toolkit" and are characterised by a focus on school-based development supporting teachers to "interrogate and expand their repertoire of practice" (Wolfenden & Buckler, 2012, p. 3) and aim to shift the dominant frame of learning within a classroom. The programme also supported a localisation and translation process as well as providing support for integration of the materials into existing programmes, or the creation of new programmes depending on the needs of each institution (see www.tessafrica.net).

The Teacher Education through School-Based Support in India (TESS-India) Programme
TESS-India, also based at the Open University (UK), is a DFID funded initiative that drew on the success of TESSA, but was not an attempt to replicate the process in a different geographical context: the aims, purpose and process were determined through collaboration with the main stakeholder, India's Ministry of Human Resource and Development (MHRD), and the materials were developed to align with the pedagogic strategies outlined in the National Curriculum Framework (NCF) (2005) and National Curriculum Framework for Teacher Education (NCFTE) (2009). TESS-India OER materials have been developed collaboratively with Indian teachers and teacher educators and consist of teacher development units (TDUs) in elementary and secondary maths, science and English as well as leadership development units (LDUs) for teachers and school leaders (see www.tess-india.edu.in).
For both projects, the broad distribution goals necessitate localisation of resources to meet end users' diverse linguistic, cultural and pedagogic needs. Both TESSA and TESS-India adopted a Open Praxis, vol. 6 issue 3, July-September 2014, pp. 221-233 two-tier model of localisation. In line with OER practice, use of Creative Commons licensing for all resources will allow for adaptation by the end users: teacher educators, headteachers and teachers. However, an earlier stage of supported localisation is embedded within the production process via a series of workshops in which local academics work together to version resources in terms of language, imagery and cultural references. Participants then take a portion of the materials away with them to complete. It is on this initial phase of supported localisation that this paper focuses.

Methodological approach of the study
Initially, the aim of this research was to focus on the outcomes of the TESS-India and TESSA materials (i.e. the specific changes made to the localised materials). We hoped to develop a better understanding of these changes through a better understanding of the perspectives and processes underpinning the localisation process. However, changes to the time-frame of the TESS-India project following the appointment of a new academic director, and a re-framing of the OER meant that the localisation process as it is reported here will not be replicated across the project and (at the time of data-collection) there were too few localised OER to analyse robustly. Our research questions, therefore, became focused around the process of localisation itself, not just as a means to an end (in which the "end" is a localised OER) but as a site of knowledge partnership and knowledge production.
We adopted a small-scale, qualitative and reflexive approach to understanding the localisation processes of these two projects. Our research was small-scale because we wanted to capture activities at the two early TESS-India localisation workshops in a way that might inform the re-framing exercise. It was qualitative, with an intention to "make vivid what had been obscure" (Eisner, 2001, p. 136) through an exploration of how participants in the localisation processes of the two projects "made sense" of OER localisation and how they "made sense" of their experiences as OER localisers (Merriam, 2009, p. 13). The research was reflexive because it was necessary for us to acknowledge our positionality as members of the Open University UK whose roles are affiliated with the TESSA and TESS-India projects (although not directly involved in the localisation process) as well as acknowledging that TESSA is an established project with several widely reported successful outcomes (e.g. Harley & Barasa, 2012), while TESS-India is a new project and was in a state of transition when the study was carried out. Our methodological approach was selected to "meaningfully shape rather than distort" (Smith, 2000) perspectives from across the two projects in a way that would provide insights into the ongoing development of TESS-India and other OER projects. Table 1 gives a summary of the data upon which our research is based. Data from the TESS-India project was collected through participant observation at two localisation workshops involving localisers from the three states of Bihar, Madhya Pradesh and Uttar Pradesh. This allowed for detailed examination of the ways in which those tasked with localising the resources worked together to identify aspects of the adaptation and the support required for this task. Observation was carried out by the Hindi-speaking author in order to capture the details and nuances that were difficult for the other authors to interpret through a translator. Additional data was collected in the form of workshop reports and interviews with participants including two facilitators and three localisers. The TESS-India data was analysed alongside a retrospective analysis of TESSA adaptation documents and interviews conducted between 2009-2010 with two facilitators and eight participants (four Ghanaian, two Kenyan and two Sudanese) from three TESSA localisation workshops. A thematic approach to analysis was undertaken that drew on Chase's (1995;2003) guidelines for coding narrative data. Themes were sought in particular that responded to the research questions around conceptualisations of OER and how knowledge is developed and shared, although several new themes emerged from the data, relating particularly to logistical issues around the process of localisation. Highlights from the analysis are discussed below.

Developing communities of practice for localising OER: two case studies
This section outlines the process of localisation as it occurred in the two projects. It is not intended as a "how to" guide. Indeed, localisation of OER must necessarily be a process designed with the very specific needs and skills of the end-users in mind. However, a practice in which OER developers and / or users explicitly describe processes of adaptation, including the choices and justifications of these choices involved, could be of tremendous value to the OER community. This section aims to contribute to this limited literature.

The TESSA localisation process
For TESSA, a consortium of institutions from nine countries, localisation took place initially through regional workshops lasting two or three days, followed up by almost a year of materials development. Prior to this, workshops were held with TESSA coordinators from all institutions in order to "develop collective understandings of the factors to be considered when adapting OER for use in a particular environment" (Wolfenden, 2008, p. 2). A collective decision was made that 40 per cent of each Pan-African TESSA study unit would be open for adaptation in the supported process; it was intended that this would ensure that the materials "spoke to experiences of teachers in a particular context whilst retaining the integrity and internal consistency of the OER" (Wolfenden & Buckler, 2012).
TESSA institution coordinators were responsible for the recruitment of staff to undertake the localisation process, often drawing from the pool of original TESSA authors: "they knew what kind of people they wanted, and the ones they chose had already demonstrated skills but more importantly commitment to the idea of OER: there was much less drop-out of localisers than there were authors" (TESSA localisation facilitator, 2010). A minimum of two localisers for each subject area was suggested for each country although due to staff-commitments Nigeria was the only country which could provide a full cohort. Institutions were paid by the project and given autonomy on how the resources were spent-usually on staff-buyout; few localisers were paid directly for their work on the materials as it was built into their institutional duties. A TESSA localisation handbook was provided to all participants, and at some of the workshops trips to local schools were included to act as a basis for discussion and as case studies for testing out ideas regarding what changes would be appropriate. Workshops included presentations about the original TESSA materials and the principles behind them and "mock" localisation activities in groups, but the majority of the workshop was for the localisers to work on their materials, supported by the facilitators and each other. Depending on the number of localisers involved, support required and adaptations necessary, the first draft of localisation tended to be completed at the workshop. Over a period of several months, subsequent drafts (often up to five) were completed, edited and developed electronically with the exception of Sudan where localisers adapted the materials by hand and delivered hard copies (some travelling several hundred miles) to the TESSA coordinator at the Open University of Sudan. While the Ghanaian materials remained in English, the Kenyan materials were translated into Kiswahili and the Sudanese into Arabic after the localisation process.

The TESS-India localisation process
TESS-India localisation was also launched at workshops; one in Madhya Pradesh and a combined workshop for Uttar Pradesh and Bihar held in Lucknow (UP). This stage of localisation focused on three subsets of the TESS-India TDUs: elementary English, elementary science and secondary maths. These were the first subject areas to be completed and consisted of 45 TDUs.
As TESS-India is managed through the national and local governments in India (rather than teacher education institutions, as in TESSA), the State Council for Educational Research and Training (SCERT) released a call for expressions of interest for localisation-related roles to statelevel resource centres, teacher education institutions and university departments of education. Interested participants submitted a CV and were invited to an informal interview with the State Representative for TESS-India and the State Localisation Manager (SLM). For each state, two State Localisation Experts (SLEs) were chosen for each subject area. The SLMs were paid through an external agency and the SLEs were given a stipend for attending the workshop then a set amount per unit localised.
The first workshop in UP lasted three days but the programme of the workshop in MP was reduced to two days to account for an unexpected public holiday. Feedback from participants in UP led to the development of a Localisation Handbook created for participants at the second workshop, as well as the provision of a Hindi-English dictionary. The TDUs were translated into Hindi prior to the workshops which were carried out in Hindi with assistance from interpreters.
Localisation was primarily completed in hard-copy. SLMs were responsible for writing the changes into a MS Word document before sending electronically for critical reading, translation checks and quality assurance.

Technology and time
The collected data revealed some challenges to localising OER that were common to both projects. The most prominent of these is time: all of the participants interviewed felt that the workshops were too short, and all felt that the deadlines for returning versioned materials to the project were too tight. This is partly related to the unavailability and unreliability of technology in the partner countries-a Kenyan TESSA versioner, for example, was sharing one computer with five other colleagues and the Nigerian institute was disconnected from the Internet for several weeks at a time. Technological competence-including the additional complexity of typing in Hindi (a skill none of the localisers could demonstrate)-also worked as a brake in the TESS-India workshops: "Firstly, the localisers were working with hard copies, scribbling on them. Printing and typing took a lot of time. Reading handwriting was difficult. This is a logistical problem. . . most localisers do not know typing in an electronic format. They're not familiar with computers. Only one person at Lucknow was working on a laptop" (TESS-India localisation facilitator A, 2014) Time pressures were also partly to do with participants' familiarity with the concept of OER; at each workshop far more time was required for induction than anticipated. Only one of the eight localisers interviewed from TESSA had worked with OER before, so familiarity of OER was a stated criteria for participation in TESS-India localisation. At the time of TESS-India localisation, the OER movement was far more active in India (Kumar, 2009) than it had been four years pre viously in Sub-Saharan Africa. However, while SLEs had used them in their work, their experience of developing or adapting them was perceived by the facilitators to be limited: "Most of the people who are SLEs were textbook writers for SCERTs-so they write the state curriculum. They knew of OER but many didn't really understand what they were. They've not really been exposed to OER writing, for example the language or the style" (TESS-India localisation facilitator B, 2014) Unfamiliarity, in itself, is not necessarily problematic but a key area for future working identified in both projects was the need to build in additional sessions for orientation and practice, as well as follow-up and support.
"I'd give people much more time to familiarise themselves with the underlying philosophy of TESSA, all of it really, the teaching and learning, interaction, distance learning, OER, maybe that would take three days, maybe five days, and then there would be time to learn, and also to get the work done in that incredibly supportive environment that just wasn't as possible later at a distance. We assumed they could work at an OU pace, which of course was completely unreasonable because we'd been thinking about these underlying concepts for years and we expected them to pick them up in a morning" (TESSA localisation facilitator A, 2010)

Cultural differences and the "right" kinds of experts
A key tenet of both TESSA and TESS-India is the idea that bringing together subject experts and experts in teacher education materials production, OER creation and experts in national education policy and systems will contribute towards a high-quality, nationally appropriate yet locally relevant product. The data suggests that the way "expert" has been defined across the projects highlights cultural differences between the different stakeholders which can influence collaborative working around OER adaptation.
For TESS-India, these issues appear to have their roots in the original OER writing process and in the selection, management and support of the subject experts recruited for this purpose (a key reason behind the re-framing exercise): "When I've seen the authors and the subject leads together I've seen very warm, very productive, very collegial relationships which is fantastic-what it should be about-but there were also a few difficulties [and] pedagogic and practical issues that were never really sorted out. . . Several authors entered into a great dialogue and there was a big mutual benefit, but others felt affronted by any challenge or change. . . and there was this sense that both sides tried their hardest to avoid conflict. This happened at the localisation workshops too, and it is visible in the materials which barely seemed to change" (TESS-India localisation facilitator A, 2014) Open Praxis, vol. 6 issue 3, July-September 2014, pp. 221-233 In both projects, responsibility for the selection of participants for versioning and localisation was given to those working in-country. In TESSA, institution coordinators tended to recruit authors of the TESSA materials who had shown particular commitment to the ideals of OER: "so they knew the kind of people that they wanted and if they'd had problems with writers in the first instance they knew how to select better this time. In fact I think people did better selecting for versioning that they did for writing, there weren't as many people drop out and I think their understanding of what TESSA was about had grown" (TESSA localisation facilitator A, 2010) An apparent advantage of the TESSA model in this respect was the direct relationship the programme had with institutions; coordinators could instruct their staff-who were active in teacher education and who already worked collegially-to contribute to the process as part of their professional development duties. In TESS-India, recruitment was managed by the project office in Delhi in collaboration with State Councils of Educational Research and Training (SCERTs): Many of the TESS-India localisers were considered to be subject experts who had extensive experience of writing textbooks for use in the Indian education system. This experience was both an asset and an obstacle and their ideas around their "autonomy and expertise" needed to be navigated during the workshops: explaining how an OER differed from a textbook, and how the focus of the materials was pedagogy and strategies-rather than subject knowledge-was a key challenge for the facilitators in both projects: "There were some very intelligent people there, people with PhDs, who really wanted to engage more with the topic than the technique. . . we didn't want the TDUs to focus too much on content, on the topic itself, it was the methods that needed to be the priority. . . we don't need to duplicate the textbooks, but a lot of people honed in on the subject matter, that's what they wanted to deal with because that's familiar territory-that's familiar to them. The other stuff-the OER stuff-that's very unfamiliar" (TESS-India localisation facilitator A, 2014) "The mathematicians, they're just very focused and maths is maths and you teach it like this, rather than thinking what's behind the mathematics, or how you make the context relevant to children" (TESSA localisation facilitator B, 2010) This desire to want to focus on strengthening teachers' subject knowledge is understandable. In India, for example, the Teacher Eligibility Test (TET)-introduced in 2011 and a pass in which is a condition of employment-is reporting pass-rates as low as 1% in some states (ToI, 2013); it is the professional purpose of these senior curriculum developers to address this deficit in teachers' subject knowledge. However these experts were so senior many had little exposure to the real and on-the-ground experiences of teachers in their state. Some appeared to have a negative perception of teachers and questioned the value of the OER for many: "Most of the teachers are clueless about how to teach. . . For the willing teachers, who are ready to experiment. . .this will be very beneficial. Frankly speaking. . .there are two types of teachers. First are the ones who have got no interest in teaching and just teach for the sake of it. There will be no impact on those. The other type are the willing types. These teachers will use the TDUs most and will adapt them as well" (TESS-India localiser C, 2014) This perspective is especially interesting as it resonates with Indian literature-both policy and academic-around the "explicit positioning" by teachers of some learners as "uneducable" (Saigal, 2012(Saigal, : p. 1011Menon, Chennat & Gunjan, 2010;GoI, 2012). In contrast, the selection of expert teacher educators for the role of localisation does not resonate with claims in the National Curriculum Framework for Teacher Education (NCFTE, 2009) that the "larger academic debates on equity, gender and community' do not enter the 'day-to-day discourse of teacher educators' and that teacher education happens in contexts which are 'severed from ground realities as well as the aims of education they espouse" (p. 10).

Quality, control and openness
Falconer, McGill, Littlejohn and Boursinou assert that Belief in quality is a significant driver for OER initiatives, but the issue of scale-able ways of assuring quality in a context where all (in principle) can contribute has not been resolved, and the question of whether quality transfers unambiguously from one context to another is seldom surfaced (Falconer et al., 2013, p. 4).
The two-tier TESSA and TESS-India localisation processes, with their quite directive initial phase of resource adaptation, are intended to offer a way of ensuring that the changes that are needed to meet local needs actually do take place during the production process, while also allowing for further localisation by teachers and teacher-educators once they have bought in to the resources' use within their own practice.
Our data suggests that there is an interesting relationship between control, quality and openness-particularly at this interim stage of supported localisation where the intended result is an OER that stands up as an exemplar of quality in terms of its alignment with national policies and the pedagogical approach embedded in these. While in both projects localisation was intended from the start, TESSA managed this process more tightly by determining sections of the materials that could be localised, and sections that couldn't.
"Versioning was always on the cards. I think that concept was in [the Director's] head very clearly. What TESSA was trying to do was to have a template that gave a structure and a form so when people used it they knew there was a case study, they knew there was an activity, but that activity is related to them in their context and their particular issues and things that they have to deal with. But, underneath all of that there was an approach to teaching and learning that is consistent and is interactive and helping teachers to have a better impact in the classroom" (TESSA localisation facilitator A, 2010) Within this structured template and guided activity, practitioners appeared to approach the project in a task-oriented way: "You see, the objectives have been decided beforehand. So I was trying always to keep these objectives, not to distract from these objectives" (TESSA localiser, Sudan, 2010) The TESSA template perhaps prevented some localisers from fully engaging with the concept of being a "partner" in the process. Yet in TESS-India, the absence of such structured support-where localisers in the workshops had more autonomy to make changes-led to even fewer changes being made to the materials: "Just changing a few place names, addresses and sticks to stones is quite superficial I think. I would like to see more relating to State-specific objectives, you know, like if they want to see more assessment done in the classroom or if they want more attention paid to low achievers that sort of thing. But as far as I've seen they haven't. Really, I'd like more radical localisation rather than safe localisation but there's a reluctance, a deference that gets in the way" (TESS-India localisation facilitator B, 2014)

Some conclusions: building knowledge partnerships in OER localisation
TESSA and TESS-India both have a strong commitment to knowledge partnerships and this supported stage of localisation is intended to demonstrate-through both process and product-this Open Praxis, vol. 6 issue 3, July-September 2014, pp. 221-233 commitment. However, the data suggests that the notion of knowledge partnerships is more pre valent in the minds of the localisation facilitators than in some of the participants selected to contribute to localisation activities. While the literature around OER implies that barriers to knowledge partnership are cross-cultural (Miyagawa, 2005), our study of these two projects suggests that these barriers appear to be embedded in the hierarchical structures of institutions.
This appears to be the case whether participants were involved as part of their departmental duties (as in TESSA) or because of financial incentives (as in TESS-India). The enthusiasm participants expressed about the projects and their reported pride in their involvement, combined with a reluctance to challenge the writing and opinions of "experts" and a disinclination to disrupt hierarchical notions about who owns (and should own) knowledge and who should share it suggests that the supported environments for localisation created by the projects only shifted them into the "medium engagement" step of Joanna Wild's stairway model of educators' engagement with OER (see Figure 1, Wild, 2012 andPegler, Fitzgerald, Hardy, Waller, Manista &Wild, 2012). In this model, low engagement involves educators using and sharing resources with no adaptation, medium engagement involves educators integrating OER into core teaching materials and "tweaking" them to meet their own needs, and high engagement involves producing and sharing OER and becoming an advocate for OER use.
This was, of course, the very aim of the workshops and TESSA materials have gone on to be used in (and adapted further for) teacher education programmes across the continent. However, our data suggests that capacity building around the highest level may have increased commitment and enhanced engagement at the middle level. Both projects adopted an approach that can be mapped against a traditional ascent up the ladder; capacity building was limited to induction and "tweaking." Perhaps if capacity building is focused around high-level engagement, including the production of OER themselves, a localiser may be able to temporarily "climb down the ladder" in order to create a richer product for the project and develop their own skills to create new high quality OER. "What I think is [the workshops] are the stepping stones and what I would like is for people to be writing more materials, new materials based on the experience with these that have been written. . . but I don't know if we did enough for that" (TESSA localisation facilitator A, 2010) "So what we should have done is see these people far more clearly as the people that are going to take this forwards and outwards, potentially, they are the writers of new OER, they could make a difference, they could take two OER and stick them together, they could do something really different. So getting

Figure 1: Wild's OER Engagement Ladder
them to move from a faithful re-version into a more radical change would be something to pursue and we could have gone on that journey with them. And that would have been a really nice way to have gone about capacity building. . . but we haven't described it as that, we described it as a process where people sign off another product. I think we missed a trick there" (TESS-India localisation facilitator A, 2014) Getting beyond a low-engagement, piecemeal use of OER is important to the resources' potential being fully realised within individual OER initiatives, but also to the sustainability of the OER movement itself. However, we suggest that this is only possible if more attention is accorded to issues of user access, skills and confidence to imagine and realise localisation as well as their role and status within the education system. In supported localisation it appears important to aim to develop a community of practice by enabling different forms of reification at the organisational level in order to develop a collective sense of group identity (as OER developers) and purpose (Wenger, 1998). Bateman et al. (2012) argue that the "promise of OER does not reside solely in the resources themselves, but also in developing the conceptual framework and methodological approaches that organise, manage and ascribe meaning to them in a variety of educational environments." We suggest too that it is not only the end product of an OER that needs to be contextualised, but also the frameworks and processes that lead to and support its contextualisation.