5.1 In the three pilot programmes the REFLECT approach proved to be both more effective at teaching people to read and write and more effective at linking literacy to wider development.

5.2 Of those adults who initially enrolled in REFLECT circles 65% in El Salvador, 60% in Bangladesh and 68% in Uganda, achieved basic literacy over a one year period. This compared to 43%, 26% and 22% in the respective control groups [and a typical 25% according to Abadzi (1994)]. REFLECT was particularly effective with women (and in Bangladesh specifically with younger women in the 15-19 age group). Participants in REFLECT circles remained well motivated and dropped out in much lower numbers than those in the control groups. There were positive signs that the participants are developing literate habits but it is too early to evaluate fully the extent to which literacy skills have been permanently consolidated.

5.3 In respect of empowerment the three evaluations identified the following major outcomes:

• Participants in all three pilots spoke of self realisation as one of the major benefits of the REFLECT circles. Most spoke of better self esteem and the increased ability to analyse and solve problems as well as articulate ideas. Furthering their knowledge of the local environment (agriculture, health, income generation and survival skills) helped this process of self realisation, which was also reflected by improved relations within the community (and within the household).

• Increased participation in community organisations was a concrete outcome of the REFLECT circles in Uganda and El Salvador. Most strikingly, 61% of learners in El Salvador reported that they had now assumed formal positions of responsibility in community organisations which they did not hold before the REFLECT literacy programme (eg chair, secretary or treasurer on the Community Council, Cooperative Directorate, Credit Committee, PTA, health committee, women's group or church group).

• The discussions in the literacy classes often led to community level actions to improve local conditions. These actions ranged from the economic sphere (constructing grain-stores, diversifying crops, cooperative buying or selling) to community projects (small infrastructure such as re-grading access roads, school repairs, water pipes); from the environmental sphere (terracing, organic fertilisers, tree nurseries, tree planting) to the health sphere (digging a tubewell, building latrines, clearing rubbish, cleaning stagnant water). The key factor in achieving the implementation of these actions was felt to be that the learners had independently arrived at decisions to do something through their own analysis - they felt a local ownership of the problems and of the possible solutions.

• The evaluations also revealed that the REFLECT circles had a positive influence on people's resource management at an individual or household level. Women in Bangladesh repeatedly spoke of the value of calendars and matrices to strengthen their analytical skills, enabling them to plan better, develop more effective coping strategies (eg bulk buying and storing goods) and have more control over decisions regarding loan use (which was previously dominated by men). In Uganda there were what appeared to be the beginnings of significant attitudinal changes seen in relation to child spacing, polygamy and traditional cultural practices which can undermine food security.

• The REFLECT pilots appeared to have had a positive initial impact on gender roles and relations in Uganda and Bangladesh. In Uganda learners and facilitators reported that many men have taken on domestic work, such as carrying water and fetching fuel wood, previously carried out by women. Women are now more vocal and more involved in key household and community decisions. In Bangladesh women attributed their growing involvement in household decision making to the REFLECT circle. However, in El Salvador, where the organisations and individuals involved in the pilot lacked basic gender awareness, there was no significant impact on gender roles, revealing that much depends on how the methodology is interpreted and applied.

• The evaluations revealed that the REFLECT circles had a positive impact on health awareness, typified by the comment of one woman from Bangladesh: "We learnt something of health before but it was not very practical and felt like a lot of rules. With making maps it was a lot more helpful and we understand things a lot better." This was translated into concrete actions in many communities, particularly involving latrine building and more effective disposal of waste.

• In respect of children's education the most dramatic results were seen in Uganda. Government schools fed by REFLECT parents have experienced a 22% increase in enrolment; and parents in over one-third of the REFLECT classes have started their own NFE centre for primary age children. A more modest increase in school attendance was registered in the other pilots.

5.4 The REFLECT approach proved to be low cost and cost effective in Bangladesh (£12 per learner) and Uganda (£11 per learner), in both cases cheaper than an equivalent primer-based programme. In El Salvador, the only pilot programme to use volunteer teachers, the costs were surprisingly higher (£34 per learner) owing to the small scale of the programme in a country where costs are high. In a REFLECT programme resources are shifted from printing to training, which makes the REFLECT approach generally cheaper than a primer-based approach at a time of high printing costs.

5.5 A process of methodological learning has taken place through the pilot programmes so that the REFLECT approach is now stronger. Certain core recommendations are made. For example: facilitators should normally have at least 6th grade primary education in order to teach other adults; visual cards should be made much more simple than in the pilots; a broad range of approaches to reading and writing integrated with the graphics should be stressed (avoiding the use of key words throughout) and training for facilitators should be ongoing. Most of these observations are relevant to making any adult literacy programme effective. The essence of the REFLECT method has been compiled into a REFLECT Mother Manual available from ACTIONAID.

5.6 Literacy does not empower people. The control groups showed very few signs of having changed peoples lives. It seems that many of the past claims about the benefits of literacy are bogus. Literacy in itself probably does not empower and does not bring benefits in respect of health, productivity or population growth.

However, this is not to say that literacy can never bring huge benefits. This research has shown that the REFLECT methodology has brought quite dramatic benefits in the three pilot projects. This appears to be because the REFLECT approach involves two parallel and interweaving processes: a literacy process and an empowering process. The literacy gives people practical skills which will help in the empowerment process (eg as they assume positions of responsibility in community organisations) and the empowerment process in turn creates uses for literacy in people's everyday lives. This mutual consolidation and reinforcement is the essence of why it makes sense to fuse the two processes. Literacy programmes, then, can be very empowering if the literacy process is interwoven with other processes through a well structured participatory methodology.

Literacy programmes in the past (especially since Freire) have tried to fuse the two processes and some have succeeded, with remarkable results. However, most have failed because they have fallen into believing that either literacy in itself is sufficient (so they have ignored other processes and focused on the product) or they have assumed that empowerment in itself is enough (but have in practice tried to "indoctrinate" people into new ideologies). REFLECT holds these two processes in an effective balance and helps them to build on each other.

5.7 There are many unanswered questions that remain. How flexible will the REFLECT approach prove to be? Will it work in urban areas, with refugees, with adolescents, within a government programme? Will it work on a large scale or will the participatory essence be lost? Will people who have learnt in the original pilots retain their skills in the longer term? Three things are needed:

• ongoing evaluation of the original pilot programmes and of new REFLECT experiences;
• a capacity to train others and promote best practice so that methodological learning is continuous.
• the continual experimentation and scaling up of the approach in different contexts.

ACTIONAID is planning to address all three of these in the coming three years.