Cover Image
close this bookMulti-Grade Teaching - A review of research and practice - Education research paper No. 12 (DFID, 1994, 63 p.)
close this folderChapter 2 - Lessons from developing countries
View the document(introduction...)
View the documentZambia: Teacher education and support for multi-grade schools
View the documentPeru: Indigenous schools
View the documentSri Lanka: Approaching multi-grade via multi-level teaching
View the documentThe impact system of mass primary education
View the documentConclusion

(introduction...)

Five innovations in multi-grade teaching over the past two decades in developing countries are presented in this chapter. The first two experiences, from Zambia and Colombia, have addressed the multi-grade reality of rural primary schools directly. The Zambian experience is less than ten years old and relatively small-scale. The Colombian experience spans three decades and is large-scale. Both have involved external agencies and support from the government mainstream. The third experience, from Peru, describes the multi-grade reality found among indigenous communities and the recent involvement of NGOs in teacher education programmes oriented towards multi-grade teaching. The fourth, from Sri Lanka, illustrates how some solutions to the problems faced by the multi-grade teacher are being offered on a small scale through the recognition that even mono-grade classes contain very wide differences in achievement and that a single grade may be conceived as a multi-grade context. The final experience draws on innovations in rural primary education in several countries. Project Impact began as a radical reform of primary education in Indonesia and the Philippines and expanded subsequently to Liberia, Malaysia, Bangladesh and Jamaica. Although not oriented exclusively to the problems of multi-grade classes, many of the features of the innovation offered solutions to them.

Zambia: Teacher education and support for multi-grade schools

The 1984 education census returns in Zambia indicated that 800 of the 3008 primary schools were "ungraded" (i.e. 26%), meaning that they offered only four grades of primary schools in a full cycle of seven grades. These ungraded schools had fewer teachers than grades, small enrolments in each grade and were located in remote and sparsely populated areas.

Much of rural Zambia is sparsely populated. In 1990 the population density was 11 persons sq. km. In 1980 this figure was even lower - 7.5 persons sq. km. Figures for India and China in 1990, by contrast, are 260 and 118 respectively. Although the population growth rate of 4% between 1980 and 1990 is among the highest in Africa, the rate in some areas is low and declining, with implications for the pattern of provision of education and other social services.

While a majority of children attending primary school in urban areas completes seven years of primary education, the percentages are smaller in rural areas where enrolment in an ungraded or incomplete school represents a terminal educational experience. A particular problem faced in many of these incomplete schools is the small enrolment of children in each grade. The problem is essentially one of resources - rural communities in sparsely populated areas too impoverished to contribute to the physical development of a school, and the high costs of providing teachers to teach each grade separately. Two solutions have been adopted (Lungwangwa 1989). The first, tried out in SO of the 800 schools, is a biennial intake in which students enrol in grade 1 only every second year. The second is the use of multi-grade teaching in which two or more grades are taught simultaneously.

Multi-grade teaching was introduced to a number of Zambian primary schools in rural areas in the mid 1980s as part of a consciously formulated "project" supported by the Ministry of General Education and Culture and the Swedish International Development Authority. It started from a very small base in four schools in Mkushi district and was extended to a further 40 schools in 1986 and 50 in 1987. It was argued that multi-grade teaching would enable small schools with low enrolments in each grade to upgrade themselves to complete grade 7 primary schools without requiring additional classrooms and teachers.

In-service training courses in multi-grade teaching were developed and mounted by the Malcolm Moffat Teachers' Training College (MMTTC). Others "inputs" to the project consisted of language and maths texts and exercise books, follow-up seminars, inspection, evaluation and the incorporation of multi-grade teaching as part of the regular preservice teacher training programme.

Approaches to curriculum organisation and teaching appear to have been promoted by the MMTTC included

the common timetable option: where all children learn the same subject in a given timetable period, but each grade group follows its own work, according to its own work programme and grade level

the subject stagger option: subjects are staggered on the timetable so that grade groups learn different subjects in the same period. Subjects which require high teacher-pupil contact are matched with those requiring little

the subject grouping option: subjects are presented to all grade groups together at the same time. Some subjects eg music, art, religious knowledge and social studies lend themselves well to this option.

An evaluation of the four pilot schools and the College's involvement in teacher's training was undertaken by Lungwangwa (1989). The evaluation addressed

the extent to which the multi-grade system had enabled all children in a school catchment area have access to the full primary level programme

the costs of this form of provision

the role of the teacher and the nature of his pedagogical activities in a multi-grade school

the impact of multi-grade teaching on the cognitive achievement of students

the impact of this form of teaching in promoting independent and self-directed habits of study

the internal efficiency of multi-grade schools

the impact of multi-grade teaching on the participation and performance of girls

the perceptions of multi-grade teaching held by participants, parents of students and the wider local community

the impact that the introduction of this system has had on the attitudes and mores of the local community

(Lungwangwa 1989: 13-14)

The College

The Malcolm Moffat Teacher Training College (MMTTC) is the Zambian-designated institution for the formal in- and pre-service training of multi-grade teachers. Keen support was lent to the idea of multi-grade teaching by the principal, the vice-principal and the multi-grade "coordinator", the first two persons having had some prior exposure to it in Australia in the 1970s. Despite the obvious enthusiasm of these key individuals a number of problems have been encountered in the implementation of the training programmes. Lungwangwa's (1989) review notes that

(i) By 1988 the multi-grade teacher education course had not been well integrated into the mainstream programme for teacher preparation.

Since most members of staff had themselves no experience of multi-grade there was anxiety and a general lack of enthusiasm for it

(ii) Multi-grade training has enjoyed the status only of an extra-curricula activity since its inception and has been time-tabled to occur during the "last few days before teaching practice commences"

(iii) Because of its lack of formal status students do not take it seriously (and)...see it as a filler not as an important component of their training... there are no examinations in multi-grade teaching... (and it)... is resented because it is considered to be a preparation to teach in the remotest parts of the country, a situation they would like to avoid at any cost

(iv) The four lecturers involved in the multi-grade programme feel "overstretched" because their teaching loads are already full. They feel that the absence of special remunerations and formal training in multi-grade affects their recognition and status

(v) A scarcity of resources in the schools restricts what the trainees can do during their multi-grade teaching practice. "Independent learning" is seen by staff to form the basis of multi-grade teaching and this, in turn, rests on the availability of learning resources.

Despite these perceived constraints, however, college staff believe that multi-grade teaching can have some positive outcomes. It can enhance independent learning, it encourages teachers to adopt pupil-centred approaches to teaching, it facilitates revision of materials covered in earlier grades, it increases pupil interaction and contributes to the country's objective of universalising basic education.

The Schools

The four pilot schools were visited and evaluated in line with the points noted above. We present here just two of the four cases, contrasting in several respects.

The Mwape primary school

Mwape Primary School was established by Jesuit missionaries in 1945. Between 1945 and 1963 children who completed the grade 4 and who wished to continue their primary education enrolled in grade 5 at the Chingombe mission boarding school, some 75 km away. Because of the distance and the annual boarding fees few students continued. In 1964 the Mbosha school was established at a closer location and it was estimated that about 5 children proceeded to grade 5 each year between 1964 and 1984. However few of these remained beyond the end of grade 5 because of the boarding costs, and between 1970 and 1984 only three students who began their education at Mwape succeeded in graduating from grade 12.

The introduction of multi-grade teaching has resulted in increased enrolments, although migration out of the area and the greater attractiveness of another primary school (where children receive free uniforms) has led to increases smaller than might have been expected. By 1988 45 boys and 43 girls were enrolled, compared with 17 boys and 16 girls in 1983. In 1988 the school served 10 villages.

There are two teachers. One teaches grades 1 and 2 in the morning and 3 and 4 in the afternoon. The other teaches grades 5, 6 and 7 as a combined class. The teachers expressed concern over lack of resources and workload. Requests from them for a third teacher had gone unheeded. Because formal lessons were taught during both the morning and afternoon, creative work, practical skills and production unit activity have been displaced. Despite these constraints the teachers were pleased to report that of eight students who sat the grade 7 examination in 1987 four had qualified for grade 8. The school-community relationship was not particularly strong and no contribution had been made by the parents towards the rehabilitation of school buildings

Kalombe Primary School

The Kalombe school experience has been a little different. The school, which currently serves four villages, was established later than Mwape, in 1965. Between then and 1984 those children who proceeded to grade 5 attended a school 19 km away. The weekly boarding fees prevented most children from proceeding and the Kalombe school was regarded as a "dead end" by most parents. After the introduction of multi-grade teaching, enrolment increased rapidly, from 123 in 1985 to 204 in 1988. However the numbers in each grade are now large enough to justify a mono-grade structure. One of the reasons for the increased enrolment has been in-migration, due mainly to a resettlement scheme. Parents expressed the view that the provision of upper primary grade schooling, made possible through the multi-grade teaching, had been one of the attractions of settling in the area.

There are three teachers in the school, two of whom were trained teachers and had specialised in multi-grade teaching. However they recalled that their multi-grade training consisted of two weeks practice in 1984 which, though valuable, was inadequate. They felt a need for more in-service training in multi-grade teaching and felt that the concept of multi-grade teaching should be introduced to all teachers as part of the normal pre-service training. The system of multi-grade teaching had been implemented "vigorously" and the recommended methods had been tried out. Teachers felt that the 'common timetable' and 'subject stagger' approaches had been the most useful, partly because of the large numbers of students involved. Unfortunately the volume of learning resources had not kept pace with increased enrolments, making the idea of "independent" learning difficult to implement. Increased enrolments were also creating pressure on desk and seat space. In general the teachers felt that the present enrolment justified their school being upgraded to a mono-grade 'complete' primary school. Its continued classification as a multi-grade institution, they felt, led to too great a workload for the teachers.

Teachers felt that with multi-grade teaching students were better prepared for self learning after they had left school. It was felt that multi-grade teaching contributed greatly to the mastery and enduring impact of basic skills, an interesting perception which could warrant further investigation. The progression rate from grade 7 to 8, of 12.5% in 1987 was of a level similar to the national average.

In contrast to Mwape the school-community relationship is very strong. A very strong parent-teacher's association has built a shelter for grades 1 and 2 and a brick house for the third teacher. The PTA maintains the classroom block to a very high standard and is raising money for another classroom. At the same time the community is aware of the pressures under which the three teachers work and cited instances of low proficiency of children in reading. They are aware that if more teachers were provided they would be expected to build more teachers' houses and raise money for additional classrooms. This prospect was viewed positively.

Mwape and Kalombe are just two of the schools studied and their experiences are rather different. The increases in enrolment in both schools have been impressive, though the average class size of each of the seven grades in Mwape remained too small to develop a mono-grade teaching structure. In Kalombe the increases in enrolment justified a switch to mono-grade teaching. Mwape is experiencing a degree of out-migration, with students attending another school where the incentives are higher and parents are shifting their homes in search of better farming lands. Kalombe, by contrast, has benefitted from the in-migration of children and families generated through a resettlement scheme. Perhaps it is this difference in orientation, the former "out" and the latter "in", which has contributed to the very different levels of support offered the respective schools by the community - rather low in Mwape and high in Kalombe. Teachers in both schools were enthusiastic about multi-grade teaching, perceived that it had learning benefits and were keen to receive further training. At the same time all teachers felt that multi-grade teaching created a heavy workload and was compromised by a failure of resources to keep pace with increases in enrolment. In Kalombe the numbers were now such that they could justify a changeover to a mono-grade system, the dominant system in Zambia.

Colombia: Escuela Nueva

In rural Colombia students receive, on average, 1.7 years of schooling, compared with 3.8 years in urban areas (Colbert, Chiappe and Arboleda 1993). In 1985 the transition rate of students from first to second grade was just 45 per cent in rural areas, with repetition rates in these two grades averaging 20%. Compared with schools in urban areas the quality of rural education has been characterised by a greater use of passive pedagogy, the use of inappropriate urban-biased curricula, lack of educational materials, rigidity of calendar, and a lack of community involvement. Underpinning all of these is a multi-grade reality of school organisation insufficiently supported by teacher training and materials. A number of efforts to address these problems have been made over the years, beginning in the 1960s with the implementation of the unitary school, an idea promoted by UNESCO in 1961. In 1967 the Colombian government decreed a unitary school system of one-teacher schools in sparsely populated rural areas. Between 1967 and 1974 a number of approaches were adopted, but the diversification of approach to the problems of the rural school, "each responding to different aspects of the problem", led to a lack of consensus on strategy and "universalisation fell short" (Colbert, Chiappe and Arboleda 1993).

In 1975 Escuela Nueva - the New School programme - was organised in an attempt to address the problems of rural education which persisted in spite of the unitary school approach. A number of accounts of the development of this programme are available (eg Colbert, Chiappe and Arboleda 1993, Colbert and Arboleda 1989, Colbert 1987 and Colbert and Mogollon 1977). This account is based on Colbert et al 1993.

By 1992 Escuela Nueva included 17,000 schools. It provides

active instruction, a stronger relationship between the school and the community, and a flexible promotion mechanism adapted to the lifestyle of the rural child. It comprises four main components - curriculum, training, administration and community relations.

The programme assumes that the rural schools involved in the programme are multi-grade and that innovations in the curriculum and teacher training need to be organised with this type of school in mind. Student self-instruction, flexible promotion, learning centres and teacher training are central to the multi-grade strategy.

Self Instruction Study Guides, Flexible Promotion and Learning Centres

The self instruction study guides are developed for children from grades 2 to 5 in four basic curriculum areas (natural science, mathematics, social studies and language). The guides adopt a method which promotes active learning, cognitive skills, discussion, group decision-making and the development of application skills within the local environment. The guides contain sequenced objectives and activities. Because the student follows the work at his/her own pace the schools operate a system of "flexible promotion". Hence students do not repeat grades. They are promoted to the next grade of work when they have mastered the present objectives and activities.

The study guides reflect both the national curriculum and regional and local adaptations. The national material are developed and printed centrally. The regional and local adaptations are developed by teachers during training courses and are produced using simple technology. The printing of the core study guides is done nationally. The study guides are used by groups of two to three children at a time and facilitate the work of the teachers required to work with several grades in the same classroom. Conventional textbooks tend not to facilitate self instruction.

Learning Activity Centres in each school complement the study guides. Materials to be used in the four basic curriculum areas are housed within the centre and students are guided to specific activities and observations based on these materials by the study guides. School libraries complement the study guides and the learning centres and contain reference material, dictionaries, textbooks and children's literature. The cost is low - a library of 100 volumes costs US$225.

In-service teacher training is an integral part of the New School strategy. Each teacher attends three in-service workshops over a period of one year, with a series of follow-up workshops thereafter.

The first workshop initiates the teacher in the basic concepts and methods of the programme, the purpose of involving students in the organisation of the school, the use of learning centres and group work in the organisation of learning, and the mobilisation of community resources for the development of the school. All these objectives are written up as self-study units in a teacher's training manual and the teachers follow these, engaging in active learning, in exactly the same manner as children will follow their guides in the classroom.

The second workshop is on the use and adaptation of children's study guides. This workshop takes place only after the school has been reorganised and the community mobilised. During the workshop teachers study the children's materials and learn how to use them for multi-grade teaching and "flexible promotion". Sets of materials for the children's use are delivered to the teachers during the workshop. The third workshop focuses on the role of the school library as a complement to the study guides and learning centre. The teachers receive the books for the library at the end of the workshop.

The follow-up workshops are organised monthly to "exchange ideas, analyse problems and discuss results". Over time these local non-formal workshops became formalised into "microcentres", described as

a participatory experience where teachers could evaluate, create, enrich their own experiences, innovate, criticize, analyse and carry out projects for the improvement of the school and the community

(Colbert et al 1993: 59)

Demonstration schools also play an important role in training. During the initiation workshops teachers visit a school which is implementing the curriculum approach effectively and which is operating as an effective community centre. Both the micro-centres and demonstration schools maintain a horizontal training network and are regarded as a "decentralised, in-service, low-cost mechanism to maintain quality in the process of going to scale" (p 59).

Stages in Going to Scale

Since 1975 three stages of development of the programme have been observed. These have been described as 'learning to be effective', 'learning to be efficient' and 'learning to expand'.

stage 1: learning to be effective: this stage occurred between 1975 and 1978 when the programme was implemented in 500 schools in three regions. During this stage materials for teachers and students were designed, administrative and financial arrangements put in place, administrators and teachers trained, delivery systems organised, materials reproduced and distributed, the programme implemented and initial evaluation conducted. The Agency for International Development (AID) provided financial support.

stage 2: learning to be efficient: this stage occurred between 1979 and 1986 when the programme was extended to 8000 schools. Training courses which had been developed during the first phase were replicated at the national level and the teacher's manual and children's study guides reproduced. Revised versions of the training courses and self study materials were developed and a core team established within the Ministry of Education. During this stage the Colombian government adopted the Escuela Nueva strategy to universalise rural primary schooling throughout the country. A variety of sources - government, the Interamerican Development Bank, the Coffee Grower's Federation, the Foundation for Higher Education and the World Bank - provided financial support.

stage 3: learning to expand: the third and current phase began in 1987 and was planned to include 27,000 schools by 1992. By this stage the Escuela Nueva movement was no longer a programme; it was now the declared official policy of government embodied in a national plan. New forms of organisational capacity were built at national, department and school cluster level.

Programme Evaluation

Colbert et al (1993) report the results of the evaluations which have been conducted on the programme to date. Rodriguez (1978) suggested during the first stage that there was no difference in the levels of creativity of children in multi-compared with mono-grade rural schools, but the self esteem of both boys and girls was higher. More recently Rojas and Castillo (1988) report that a majority of teachers believe that the New School is superior to other types of traditional rural school. Students in New Schools performed better on tests of socio-civic behaviour, self esteem and some subjects in some grades.

In short, it has been suggested that the New School system responds successfully to the needs of the rural child in Colombia because

it offers a multi-grade approach that permits provision of complete primary schooling where incomplete schooling exists

one or two teachers can handle five grades in the same school

it involves administrative agents and communities as well as children and teachers

the learning strategy adopted encourages active, creative, participatory and responsible learning

through their participation in the school government children learn civic and democratic behaviour

children learn at their own pace using self instructional materials

there is no grade repetition: promotion to the next objective or grade is progressive and flexible. Children can study at school and at home. They can continue to help their parents at home while studying

materials are affordable - one set is shared among three children and each set lasts several years. The content of the materials reflects a national curriculum and can also include regional and local adaptation

teachers are facilitators: they guide and orient learning. Teacher workshops employ a pedagogy similar to the one they will use in their classrooms

the inservice training of the teachers is local, replicable and permanent

(adapted from Colbert et al 1993)

We return to the results of the evaluation in Chapter 3.

Peru: Indigenous schools

Thirty nine percent of primary schools in Peru are one-teacher schools (Tovar 1989). One-teacher schools occur in rural areas, predominantly in the Andean and Amazon regions. References to multi-grade teaching and multi-grade schools occur very infrequently in the Peruvian educational literature. Where they do appear reference is usually made to the prevalence of one-teacher schools over multi-teacher schools, rather than multi-grade teaching as opposed to mono-grade teaching. In rural areas multi-grade teaching is the norm for most schools, be they multi-teacher schools or one-teacher schools. Figures for the number of teachers per primary school nationally are 2.2 and for the Amazon region 1.5 (Chirif 1991:47).

While the provision of primary schooling for children in rural areas has increased over the 1980s, this has been achieved through an increase in the number of one- and two-teacher schools. In the case of the one-teacher school, a teacher can be faced with an age range from ¾ up to 15/16 years, divided into nursery grade (initial) and six grades.

Besides their rural location, a number of other factors influence the multi-grade school. These include

an absence of teacher training. One-and two-teacher schools, located in the most remote rural areas, are considered very low-prestige schools and are allocated the lowest qualified teachers. In the Andean Department of Apurimac, more than 85% of teachers in one-teacher schools have no teaching qualifications (Zu 1989); in the Tambo region in the Central Rainforest, one-teacher schools are staffed by local teachers who have only some years of primary schooling plus a few summer vacation courses. Here class sizes reach 75 students (Heise 1987).

a lack of resources. Rural multi-grade schools are very poorly equipped, not only in terms of the fabric of the school itself but in terms of text books and other educational materials, which in some schools do not exist at all. This poses serious constraints for the teacher who becomes entirely dependent on the blackboard, on which it may be barely possible to write.

cultural and linguistic diversity. In both the Amazon and Andean regions, which have large indigenous rural populations, there is a high cultural and linguistic diversity. Consequently, multi-grade teachers are faced with the task of teaching a monocultural and monolingual Spanish curriculum to indigenous children, who, in many cases, are completely monolingual and monocultural. The cultural and linguistic backgrounds of the teachers are often distinct from those of their students.

This is the situation which exists in many areas and was heavily criticise in southeastern Peru by a wide range of institutions and NGOs working in education

The lack of teacher training hinders the use of adequate teaching methodologies, which becomes an acute problem when the teacher is working in a one-teacher school where the student population utilise a diversity of mother tongues depending on their ethnic affiliation

(CAAAP 1992: 13).

Faced with this situation, a multi-grade teacher will often divide the students into two groups. The first, grades 1 and 2, comprise the monolingual mother tongue students and the second, grades 3 to 6, comprise students with at least some understanding of Spanish. Given that each grade contains students who are repeating the grade and students new to the grade there may be considerable variation in ability within each grade.

Teacher Education

Teacher training colleges (Institutos Superiores Pedagos) show little concern for training new teachers to cope with multi-grade classes. This reflects the orientation of teacher training in Peru towards teaching in well equipped urban schools where there is one teacher per grade. Not only is this type of school in the minority but within the education system it is the most prestigious and the most lucrative. The most highly qualified and experienced teachers aim for an urban posting in a Mestizo area where, like themselves, the students are Spanish speaking.

In contrast, the least prestigious schools rural, indigenous one-teacher schools - receive a high percentage of teachers with no training and many such schools find it difficult to get teachers at all. Mestizo teachers arrive with no methodological orientation for multi-grade teaching and no ability to communicate with their mono-lingual students. This situation is linked to the general abandonment and lack of concern for rural regions, in particular those with a high indigenous population. This abandonment encompasses education in general and indigenous education in particular.

Nevertheless, there is a growing concern for improving the quality of primary school education for indigenous peoples in Peru, which has taken the form of a Ministry of Education directive instructing all Instituto Superiores Pedagos to offer training courses for indigenous teachers in intercultural bilingual education. The new courses, however, are designed to encourage teachers to adapt the content of the existing 6 grade curriculum, with its rigid system of grade promotion through formal exams. It does not encourage the development of new styles of pedagogy and classroom organisation which would enhance the task of the multi-grade teacher. The training on these new courses is primarily concerned with developing a satisfactory methodology for teaching Spanish as a second language.

The national curriculum as it stands today provides the possibility for 'unitary' teaching. That is, the curriculum is presented in terms of textbooks for each grade comprising lesson plans and structured in such a way that certain subjects or themes appear through all grades with appropriate levels of ability. However a study in indigenous Amazon schools has shown that teachers appear to lack the ability to organise the official curriculum in terms of thematic units. They work instead with each grade independently and may be dealing with several different topics with several different grades simultaneously (see GaschI>et al. 1987).

The following example from a primary school in the Arakmbut community of San Jose presents a contemporary multi-grade teaching reality in an indigenous area. The primary school is run by two lay-missionary Spanish speaking teachers with no formal teaching qualifications and illustrates the labour intensify of teaching common in many multi-grade schools. The two teachers had divided the students into one class of Initial, Grade 1 and Grade 2, while the other comprised Grades 3, 4, 5 and 6.

The day began with maths and the lesson proceeded in a pattern familiar to the students: Grades 3 and 4 children watched and waited while the teacher wrote fractions on the blackboard for both grades to do together and a sentence explaining what had to be done: "order the fractions according to increasing number, then into decreasing number". The teacher read out this instruction, reminded them of similar work they had being doing the previous day, and then left them to copy everything on the board, including the written instructions, into their exercise books. The teacher then turned her attention to Grades 5 and 6. Once they had been given their work she walked back across the classroom to the group copying the fractions and began to call them to the board one by one to complete fractions. The other children watched and copied so that at the completion of the lesson all the children had perfect answers.

In the second classroom in San Jose the Grade 2 had settled down to copying and completing 2 X table multiplications from the blackboard which kept them occupied for an hour. Meanwhile, Grade 1 worked with the teacher from a colourful poster of a huge grand piano, repeating words which were printed below in Spanish. While the teacher focused most of her attention on this language lesson, she was at the same time overseeing the work of the four Initial Grade students who were restlessly copying over and over again the four words which she had written into their jotters in cursive script.

(Aikman 1994, Appendix C).

So, despite the possibility of "unitary" teaching prescribed by the national curriculum, the teachers continue to organise their work as though they were teaching mono-grade. They move continually from one discrete group of students studying one subject to another group, often engaged on a completely different lesson.

NGO teacher training programmes

Where schools have no supervision and local authority and training college staff never visit, personnel from educational institutions and training colleges have no idea of what it is like to teach in a school in an indigenous community (Heise 1987). Moreover, they have little experience of poorly equipped multi-grade classrooms, as their experience pertains mostly to the relatively privileged urban settings surrounding the training colleges. An exception to this situation, however, can be found in some of the NGO teacher training programmes set up specifically for training indigenous, and bilingual intercultural teachers in the Amazon region. These include programmes such as the Programme for Intercultural Bilingual Education in the Alto Napo River; the Bicultural and Bilingual Experimental Education Project for the Ashaninka of the River Tambo run by the Amazon Centre for Anthropology and Applied Practice (CAAAP); and the Association for the Development of the Peruvian Rainforest (AIDESEP) and the Instituto Superior Pedago de Loreto (ISPL) programme for training indigenous teachers in intercultural bilingual education.

These programmes multi-grade teaching as an integral and important characteristic of the teaching situation in rural schools. Because these programmes have been set up specifically for indigenous teachers, one- and two-teachers schools, lack of teaching materials, lack of formal training, absence of supervision and distance from urban centres become central foci of the programmes. Multi-grade indigenous community schools are not considered remote and low prestige. Furthermore, solutions to the neglect in these schools and communities are sought at the local level and in the schools themselves.

The AIDESEP/ISPL Teacher Education Programme

The AIDESEP/ISPL programme has been designed by and for indigenous Amazon teachers working in communities with the same linguistic and ethnic background. An important part of the training course is the production of a new primary curriculum by the trainee designed to suit the indigenous communities and their particular situation. This includes a new pedagogical approach to multi-grade teaching, methodology and classroom practice. It is in multi-grade classrooms in small one or two-teacher schools that the trainee will be most free to introduce and trial his/her new curriculum during the final three years of this training course. Over this period the trainee will carry out teaching practice with on-site support and supervision from a local team of trainers comprising educators, anthropologists, linguists and community members. Unlike other training courses the NGO programmes emphasise the importance of the college coming to the trainee rather an overwhelming emphasis on in-college work.

The new curricula which students on the AIDESEP/ISPL course are each developing focus on problems and issues which are significant for the indigenous community and avoid 'fragmenting the indigenous reality and view of the world by dividing knowledge into discrete subjects' in a manner alien to the society (see Trapnell 1990; 1991). Similarly, they aim to avoid fragmenting students into discrete autonomous grades which do not reflect the way children learn within their indigenous society and makes poor use of both student and teacher ability and potential.

Though there is still very little direct reference to multi-grade teaching in the reports and articles concerning these new NGO programmes, they break from the centralised graded and highly authoritarian primary teaching methods and rigid curriculum propagated by the Ministry of Education. This break suggests that in the future multi-grade teaching may be given more explicit recognition in teacher training in both NGO and Ministry of Education programmes, and that multi-grade classrooms may be recognised as a new potentially rich learning context rather than an unavoidable setback.

Sri Lanka: Approaching multi-grade via multi-level teaching

Sri Lanka has a highly developed system of education and enrolment in the primary cycle is near universal. However several types of social and economic disadvantage remain and multi-grade schools are associated with some of these. Multi-grade schools are most likely to be found in remote rural areas where access is difficult, population sparse and the living conditions for teachers unattractive. They are to be found in both the Sinhala-speaking and Tamil-speaking rural areas. In some types of disadvantaged areas, for example, the Tamil-medium schools in plantation areas, shortages of qualified teachers mean that multi-grade teaching is the norm. The government policy of a minimum of 3 teachers per school, however small, is sometimes not implemented. Even then, with a five-grade national curriculum some grades will be treated as multi-grade. Although the percentage of one-teacher schools has decreased from 5.6% in 1986 to just 2% in 1991, the percentage of schools where there are multi-grade classes will be far higher estimated to be 24% in 1986 (Abhayadeva 1989). Many small schools cover more than the five grades of primary, sometimes six, sometimes eight. A UNICEF report, written in 1987 commented

such schools generally cater to disadvantaged populations in rural areas whose earnings are low and unstable... In addition to paucity of teachers, such schools have inadequate physical facilities, equipment, books and are neglected by the education system in regard to maintenance, repair and supervision. They remain in a state of suspended animation. They are charged with providing education to over a quarter of million children. Since there are only one or two teachers per school of five grades, no teaching occurs for many children over large portions of the year

(Ratnaike 1987)

The national primary school curriculum in Sri Lanka is oriented towards mono-grade schools. A number of changes have been introduced recently to the content of the grade 1 and 2 curriculum, in recognition of the low levels of basic number and language skills with which many children in disadvantaged areas enter primary school. However, these changes continue to be framed within the assumption that primary schools are organised on mono-grade lines. The organisational realities facing the multi-grade teacher are not readily addressed in the formal guidance of the primary school syllabus, teachers' guides, textbooks, pre-service and in-service training. The following notes from the author's field diary illustrates the problem

an unannounced visit to a tea estate school. The school has 163 children enrolled in grades 1 to 6 and two female teachers. The principal teacher has nine years experience and followed her formal training through the distance mode. She lives in the estate. The other teacher travels daily by bus from a small town ten miles away. The bus is infrequent and usually late. Already it is 8.30 am and the school has been in session officially for one hour. Grade 1 children sit quietly outside the one-roomed school under a shed-like construction with no roof The morning sun forces them to take shelter on a lower ten-ace under a roughly constructed thatched enclosure. None of the grades has been set any work and the children sit expectantly, their unopened books in neat piles in front of them. The enrolment register has been checked. This morning one hundred children have turned up for school, an attendance rate of just 61%. Although we wish to spend fifteen minutes or so with the teacher discussing a recent inservice training which she attended recently, we suggest that the students be set some work before we do so. The teacher moves quickly around the grade groups crammed inside the single classroom and grade 1 outside, giving instructions, opening books, writing exercises on the blackboard. Three male monitors, apparently self-appointed, dart around the six groups, distributing verbal punishment here, physical punishment there. A grade 3 girl takes an envelope out of her satchel. It contains small picture cards of animals and flowers. Two boys snatch one each, a squabble ensues, the monitor intervenes, peace is restored and they await more attention from the teacher. The grade 2 children have been set language work. One child reads out one or two words from the set book, the others repeat in unison. All the grade 2 children have a language book, distributed as part of the government's free text book scheme. Although all participate in the chant, and although all are holding a copy of the relevant book, not all are reading the words. It is enough to hold the book and repeat the chant.

The male kangani stands outside looking in. He is an elderly man paid to shepherd the children one and half miles from the estate line-room where they live, through the tea fields to the school, and back home each day. He is paid Rs 250 (£4) a month by the parents for this work. His stick is poised, ready to intervene should the classroom become too unruly.

By 9.00 the bus arrived and the second teacher joins us. We discuss various activities and suggestions made during the recent in-service seminar on health. Both teachers had also attended an in-service training course for teaching methods in the lower primary grades. One two hour lecture had been devoted to multi-grade teaching. What could they remember from it? They remembered a discussion about introducing a topic common to all grades. For example the topic could be fruits. After introducing the topic different grades of children could be asked to engage in different activities based on this common theme For example grade 1 children could talk in small groups about the fruits they eat; grade 2 could write out single names of fruit and compare fruits in different ways; grade 3 could compare fruits for their vitamin value.

Although it was clear that the in-service seminars had generated ideas in the minds of the teachers it was more difficult to judge whether these were being transacted at the classroom level. The teachers had made a large number of learning and teaching aids at the in-service training sessions and were able to produce these from the store cupboard to show us. But none was in use during the one and a half hours of our visit.

Teachers may be encouraged at in-service seminars to experiment with multi-grade teaching methods in their classes. However this expectation is paradoxical. Those teachers who find themselves in schools which force them to adopt multi-grade methods are not only teaching in the most disadvantaged economic and social conditions, but they are themselves the most disadvantaged teachers in terms of education, level of training, status and, often, social background. Professional support from the central curriculum and administration authorities is based on an assumption of mono-grade organisation Curricula are developed with an image of a relatively well-educated and trained urban teacher in mind. The adaptation of national curriculum materials, the reorganisation and structuring of timetables and groups, the creative use of space and the management of time are challenges which the lesser educated and lesser trained person is expected to confront and master. The more educated and trained teacher simply works within the norm and is not expected to make major intellectual adaptations.

In recent years, two developments, one stemming from the National Institute of Education, and the other from the Ministry of Education, have acknowledged the pedagogic value of multi-grade organisation However, their approach has been via the needs of mono-grade teachers and the recognition that there are multiple levels of ability within single grades.

National Institute of Education: the development of multi-grade and multi-level teaching strategies

One approach to multi-grade teaching is to link it with the ideas of multi-ability or multilevel teaching. The general idea is that, even within a grade in the dominant mono-grade structure, there are wide differences in competency in the basic skill areas of language and mathematics. In a useful paper on the development of multi-grade and multilevel teaching strategies developed at the National Institute of Education, Abhayadeva (1989) underlines the point that the multi-grade teaching strategy should be

a feature even in regular situations with a teacher per grade... (and that) a single grade with multi-levels could be conceived as operating in a multi-grade context

(Abhayadeva 1989)

Abhayadeva supports her case with reference to data on language and mathematics competency at entry and exit from grade 1. Using data from a large-scale survey of competencies she constructs a distribution of competency in writing movement coordination and concepts of quantity for a class of 40 students. Competency levels are assessed at five levels of mastery. Table 3 suggests that there is a wide distribution in both skills among those who enter grade 1. Although the range of variation reduces a little as children reach the end of their first grade of schooling, in the sense that larger numbers approach mastery on this particular skill, the variation remains marked. In principle the majority of children who reach the end of grade 1 should have achieved mastery or close to mastery in these skills. In practice less than half are doing so in the basic concepts of quantity.

Abhayadeva (1989) suggests that these data have several implications for organising or grouping children:

for example there are 32 (24+8) halfway to mastery at entry to year I and at exit... and there are 30 (10+20) at close to mastery level and another 16 (4+12) who have mastered (writing movement). According to the data those at halfway to mastery would be struggling to copy shapes of letters while those who have mastered would copy (or write) letters with ease.

Competency levels in quantity show that for more than half... learning activities should be geared at a lower level... a multi-grade organisation would be beneficial... some of these students would need more than one year, 18-24 months or perhaps even longer to attain competency in a given grade specific curriculum. Sri Lanka has approached almost full enrolment at Year 1. However grade repetition is found at each grade level with year 2 having the second highest rate of repetition... The cumulative effect of learning problems mostly caused by the necessity to adhere to a grade-wise curriculum is revealed by continuing repetition rates which has reached its highest in primary for year 4. Grade repetition and the accompanying sense of failure leads to early school dropouts. It becomes difficult to maintain the momentum reached by achieving full enrolment at school entry.

Perhaps a non graded approach where all children need not necessarily be transferred to the next grade at the end of the year, will help certain children to overcome some of the learning problems encountered due to competency level... a flexible approach which reaches out to different and parallel competency levels in adjacent grades would require reorganisation of the graded structure at least in the first two or three years of the primary school

Table 3: Distribution of competency in writing and mathematics, grade 1, Sri Lanka

Competency level

Writing movement

Quantity


at entry

end of Gr. 1

at entry

end of Gr. 1

not started mastery

0

0

1

0

started mastery

2

0

16

5

halfway to mastery

24

8

18

18

close to mastery

10

20

4

8

mastery

4

12

3

9

It is clear from this that a multi-grade approach is being advocated for children at the beginning of the primary cycle. In other words, teachers in mono-grade schools are being urged to reorganise their work along multi-grade lines, rather than the reverse.

The practical steps which have been taken to date to effect some of these ideas include the establishment of a pilot project in 20 schools carried out by the Primary Education Project of the National Institute of Education in schools under the UNICEF-assisted programme for quality development of primary education, the long term objectives of which include exploring the possibility of using multi-grade teaching in "normal" classrooms and in large mono-grade schools and incorporating the experiences in a guide for teachers, and to introduce multi-grade teaching as a component of pre-service and in-service education.

The Ministry of Education's Plantation Sector Education Development Programme (PSEDP): Self-Study materials and Graded Learning

Since 1987 the Ministry of Education in Sri Lanka has run a programme of support to primary education in the tea and rubber estates. Until fairly recently estate schools had been managed and "owned" by plantation companies. They were connected with the National Ministry of Education through a grant-in-aid system in which schools were subject to an annual inspection and award of grant based on academic achievement. During the 1970s these schools and their teachers began to be "taken-over" by the state and incorporated fully into the state system. In 1984 there were 558 estate schools, with a total of 63,389 students and 1,148 permanent teachers. Thus the average student-teacher ratio was 55: 1. A large proportion of the schools had only one teacher.

A programme initiated in 1987 was designed to upgrade the estate schools, most of which covered the first six grades. The objectives were to increase enrolment and to improve the quality of education. Although the majority of schools are in fact multi-grade, the student population in many of the estate catchment areas warrants a mono-grade structure were universal enrolment is to be achieved. Hence priority attention was given to stimulating enrolment and attendance among the school-age population, stimulating the supply of teachers to teach in the Tamil medium and upgrading teacher performance through inservice training and on-site support from teacher educators. To have started from the multi-grade reality would have been unwise, for it would have distracted attention away from the more fundamental problems facing children in the estate community large numbers of children dropping out from primary, too few teachers and too many untrained teachers. Gradually, and as some of those problems are being met, more attention is being given under the programme to appropriate pedagogy for groups of 40-50 children, (i) through the development of self-study materials and (ii) the encouragement of a graded and individualised approach to the learning of reading. Both strategies can support multi-grade teaching as well as multi-ability teaching within a mono-grade structure.

Self study materials

In 1993 PSEDP embarked on a programme of development of self study materials in the Tamil language for use by students in years 3-5 of the primary cycle. The idea for this programme arose out of concerns expressed by the teachers and teacher educators of PSEDP

shortage of teachers, lack of additional learning material and variation in comprehending ability of the learners, especially in the primary cycle, are a few of the reasons of slow achievement in language and mathematics. Also due to these factors slow learners or low ability groups were generally left behind and at the other extreme the leaning needs of fast learners or high ability groups are not catered for. This situation leads to the necessity of identifying and developing learning material which would make learning interesting and encourage the learner to face challenges in learning.

(PSEDP workshop report 1993)

The development of the materials involved a number of well planned steps and a methodical built-in evaluation of the materials before their mass production. The steps were as follows:

Step 1: A 5-day workshop was held at a school, organised by a teacher educator from the Ministry of Education and a resource person from the National Institute of Education. The other resource persons were the teachers themselves who identified problem learning areas in language and maths, designed preliminary materials, tried them out in the school in which the workshop was being held, revision and grading of achievement level.

Step 2: The materials were then tried with a large sample of children in plantation sector schools. An assessment format for the try-out was devised, experienced teachers trained to conduct the try-out and data collected is analysed by and with the teachers who have conducted the try-out. The analysis was conducted in terms of the percentage of children who gained correct answers, plus a listing of the different types of error made those who gained incorrect answers.

Step 3: On the basis of the analysis the materials are revised by the teacher educators and teachers. After revision the materials are typeset, mass produced and laminated for durability of use. Several hundred self study "cards" have been produced to date.

A Graded Approach to the Learning of Reading

The second example within the PSEDP is the development of a graded approach to reading. A teacher educator involved in the scheme described the idea and the follow-through

We had been talking for some time about reading skills... All the teacher supervisors in the group agreed that the reading ability was low. When we discussed the problem with the teachers they always said, simply, "children are not interested, they do not have the ability". Teachers never seem to acknowledge that they can improve their practice.

We decided to send formats to every school and asked them to do compulsory evaluation of reading... each child was graded A-E. We specified the criteria. We then asked the teachers to use the Tamil reader and other story books. We discovered, after the first assessment, that there were some children in the upper classes who were extremely weak in their reading. The teachers had not been giving them any remedial activity. We suggested that they should ask these children to go back to the year 1 and year 2 reader.

When asked whether she thought that the testing per se was responsible for the improvement in reading standards or whether the intervention comprised a number of elements the teacher educator explained

As supervisors we showed an interest in the reading and showed the teachers how to do their own evaluation using the criteria. We started the evaluation in June (1993). Some teachers have done the evaluation three times, others four times by now (February 1994). We believe that if change is coming about it is because of several things. We are helping the teachers become aware of the individual difference in levels of reading skill. We think that the testing itself may be having a motivating effect on the students and the teacher. We are suggesting that story books as well as the official book be used with the children. We are supplying schools with extra reading material through our mobile library. And of course we ourselves are showing an interest in the reading abilities of the children. All of these things are happening at the same time. We cannot say which is the most important factor. But we are using the teachers' assessments of the levels judged against the criteria.

Although the teacher-educator and her colleagues requested all the schools in the areas for which they were responsible to do the compulsory testing, they decided to "study in a systematic way the improvement in 10 schools, selected randomly".

Some of our results are as follows: In school I there have been three testings. The percentage with A grade increased from 0% to 21%; the percentage with E grade decreased from 35% to 7%. In a second school there were five year 5 children who could not read at all. We started them off with the year 1 book. After 6 months two of the five got their promotion to year 6; they were able to read at the year 5 level. The other three did not reach the level and will repeat the year... but they will catch up. In another school we found that when we tested in March there no D's and quite a few A's, but when we retested in June, we used a different book but of about the same level, not the set Tamil reader. This time there were no A's. We felt that the children had been memorising the set book and were unfamiliar with reading anything that was outside the set book. We recommended the use of story books as well as the set book. We are helping them find the extra books through our mobile library.

This is a grassroots example of teachers and teacher educators working together to identify differences in the pace at which children in the same grades are learning. Traditionally, reading is taught through whole grade groups reading out aloud from the same text; the set text is often the only book available in the classroom and the home. The identification of individual differences in reading has been an eye-opener for many teachers, especially those who are untrained. The realisation that additional reading materials can be developed at low or no cost and that supplementary story books can be made available on loan is transforming the work of the teacher and the learning experience of the student.

Although neither the self study materials nor the reading project arose primarily out of a need to find solutions to the problems facing the multi-grade teacher, clearly the materials and approaches being developed are appropriate for both the multi-grade and the mono-grade teacher.

The impact system of mass primary education

During the early 1970s the Regional Centre for Educational Innovation and Technology (INNOTECH), based in Quezon City in the Philippines, devised a radical approach to rural education. Known originally as the "no more schools" concept, it proposed to replace schools, textbooks, teachers and grades with learning centres, self-instructional materials, peer, tutor and community support and instructors responsible for the management of learning among groups as large as 150-200 students. Over time the concept became better known as Instructional Management by Parents, Community and Teachers (IMPACT). The innovation began in the Philippines and Indonesia, spreading subsequently to Malaysia (INSPIRE), Jamaica (PRIMER), Liberia (IEL) and Bangladesh (IMPACT).

The rationale for the "no more schools" concept and the components of a new system for the delivery of education are presented by Cummings (1986). The original rationale was based on the educational conditions facing much of South-east Asia in the early 1970s

one-half of rural children in Southeast Asia do not complete more than 4-5 years of school and, due to a projected rapid population growth rate, this situation is likely to worsen. Shortage of classrooms, considered one of the major causes of the low educational attainments of rural youth, is unlikely to improve because overstrained national education budgets will not be able to fund many new places at existing unit costs. The inflexibility of conventional school schedules, which causes children to miss lessons, fall behind, and eventually dropout, is another reason for low educational attainment. Thus to improve rural education, a new delivery system needs to be devised with a more flexible schedule and lower student costs. Since teachers make up 80-90% of unit costs in conventional schools, such costs can be reduced by increasing the student-teacher ratio and supplementing teacher supervision with assistance from students, parents, and community resources. Self instruction, relying on programmed instructional materials, can be another means to reduce costs and, moreover, enable greater flexibility in scheduling individual learning.

(Cummings 1986: 6)

Although there is no explicit reference to multi-grade teaching in the rationale, many of the schools which were subsequently to participate in the innovative system were multi-grade schools with fewer teachers than grades. The greatest early success of the project was experienced in central Kalimantan and Sabah, island provinces of Indonesia and Malaysia, where one-teacher primary schools were common and where teachers experienced difficulty in using the conventional texts and materials designed with the mono-grade primary school in mind. As we shall see below the components of the delivery system could be adapted to the multi-grade school.

IMPACT System Components and Principles

Although the details of delivery programmes were developed in the field and were to take on a different form in different settings there were some common elements. The original "no more schools" concept envisaged a delivery system based around personnel, instructional materials and instructional organisation

Personnel

- in place of the conventional teacher, an instructional supervisor able to manage up to 200 primary students

- community members, enlisted on a voluntary basis, to provide instruction in particular life skills

- primary-school graduates, provided modest pay, to give courses in reading and other academic subjects

- parents to take responsibility for motivating their children and monitoring their progress

Instructional material

- the use of modular instructional materials, with many of these materials being self-instructional to allow children to proceed at their own pace

- instructional radio programme sessions to supplement the written material

Instructional organisation

- the primary mode of learning was to be sell: paced, individual instruction under the guidance of tutors and the instructional supervisor. This mode could be supplemented, where appropriate, with group sessions

- a simplification of organisational procedures with no specific age required for entry to the community centre, few set class periods during the day, no prescribed schedule for completing modules, and no individual grades maintained other than a record of completed modules

Over time, and as delivery systems were worked out in detail in different settings, several principles of the IMPACT system began to emerge. These are described by Respati and Mante (1983:9-14), two of those intimately involved in the development of the idea in Indonesia and the Philippines. They are presented in a slightly adopted form below.

The subject of education are children of primary school age

the Impact system is open in character. It does not close off the opportunity to the children to study although they may have dropped out of primary school. All children of primary school age can effectively follow the primary school curriculum from the start until they finish. School dropouts do not stop learning. They can go on studying till they finish and gain the elementary school certificate.

Learning materials are based on the current primary school curriculum

learning materials are based on the approved curriculum of the school system. The fact remains that the students of IMPACT schools are still governed by the requirements of the system such as the successful passing of official examinations

The essence of education is the learning process

Education in the IMPACT system, through the intermediary of modules will encourage children to learn by themselves. There will be a minimum of exhortation. Instead, the process will throw the child right into situations that will require him to learn by himself

Learning can take place anywhere

learning (takes place) anywhere - not only in the classroom. The attitude that dropouts inevitably will cease to learn, or that graduation terminates the learning process is contrary to this principle and, therefore, must be changed. The school building is merely the centre of learning from where guidance, material and resources may be derived. In the countryside where the dropouts and adults may have easy access, learning posts are established.

Multiple Entry and Exit

The principle of multiple entry and exit will help solve the problem of dropouts by meeting them half-way, by literally allowing students to enter when they wish or when it is most proper for them, to leave at the most opportune time and obtain the elementary education certificate.

Progress based on mastery and individual speed

Children will not be forced to proceed at a pace beyond their capacity and readiness. On the other hand they will not be hindered when they are ready and capable. This principle recognises the concept of individual differences quite realistically. The children learn by themselves and pacing is individual. The basis of progress is mastery.

Education is a socialising process and provides leadership training

Children of today are citizens of tomorrow. Group learning is a primary mode of learning. Older ones act as tutors or "programmed teachers" to the younger ones. In peer groups children assist each other.

Education is the responsibility of parents, the community and the government

Teachers are not the only source of education. The acceptance of the responsibility of education by parents, the community and the government is crucial. In the conventional school system, the participation of the community is chiefly financial. In the IMPACT system, community participation is more substantial and technical. Parents are encouraged to monitor their children's progress, to assist them through tutoring in their studies if they are capable.

The teacher as the manager of the learning process

The teacher's duty is to direct and manage the learning process. The teacher will no longer do much direct "teaching". She will be expected to exercise a different role - that of managing all the resources of the school, all the sources of education and ensure that the children is benefited maximally - hence the designation "Instructional Supervisor".

(Respati and Mante 1983: 9-14)

Although these components and rationale Underpinned the early "no more schools" concept, the specific objectives and components of innovations varied from place to place. Cumming's summary (1986:19) suggests that the objective of improving education quality was common to all six projects, whereas that of lowering the unit cost of schooling was common to the Philippines, Liberia and Bangladesh, but not Indonesia, Malaysia or Jamaica. The shift in the teacher's role to instructional supervisor and manager of the students learning through self-instructional modules characterised the programmes in the Philippines, Indonesia, Liberia and Bangladesh but not in Malaysia and Jamaica. In Jamaica and Malaysia greater emphasis was placed on the development of instructional guides and aides. All the countries divided class groups into small groups but they varied in the extent to which they encouraged cross-age and peer tutoring, programmed learning and differential pacing.

In terms of the challenges facing those who work with the multi-grade teacher two features of the IMPACT experience stand out. The first is the quality of materials, which was particularly high in the Malaysian and Liberian cases. Of the Malaysian materials Cummings (1986:85) writes

a teacher with a minimum of preparation could successfully fill virtually every classroom minute with interesting and attractive activities. The teaching guide tells the teacher what to prepare before class; how, in the first few minutes, to recall old material; and how, in the next 20-30 minutes, to present new material. For the remainder of the period, the teacher is able to lead slow learners through a special drill with attractive instructional aids while the other students work on self-instructional worksheets. The instructional kit also provides periodic tests.

The second is the preparedness of the teachers and their support team to innovate in school organisation and classroom management (Cummings 1986: 86).

In the Indonesian experiment, one is especially impressed with several of the innovations affecting classroom management In the lower grades in all the experiments, students tended to proceed at more or less the same pace under the management of their teacher. However, in the upper grades, as the students turned to self-instructional modules two problems emerged: boredom and differential pace, the first innovation developed was peer-group learning, wherein three to six students at the same grade level would form a group to study modules together... modules were revised to assign distinct roles to a peer-group leader... the position of group leader could rotate among the members. While these innovations alleviated boredom, differential pacing remained a problem As the Indonesian project team became more familiar with the principles of mastery of learning, they began to introduce pre- and post-tests for each module and to insist that no group could move on to a new module until all members had achieved 90% on the post test... fast learners manifested a more helpful attitude towards those in difficulty, and the overall cooperative spirit in classrooms was considerably enhanced.

Conclusion

These five experiences of multi-grade teaching in developing countries have a number of common themes. All address educational problems in disadvantaged rural settings with low populations. All have involved teacher training in the techniques of multi-grade teaching at the local level. Some have succeeded in having multi-grade recognised by government as a legitimate area of enquiry for teacher educators and teacher trainees at national level. Others rely on the support of NGOs and teachers self help groups. The issue of cost has not been a dominant theme in any of the examples. The multi-grade strategy has involved a number of components besides teacher training. The design, reproduction and distribution of large quantities of self-study materials to support individual, peer and small group learning; a system of evaluating learning progress and achievement; and forms of internal school and class organisation which establish routines for students independently of the teacher appear to be among the characteristics of effective multi-grade teaching and learning.