|Getting Books to School Pupils in Africa - Education Research Paper No. 26 (DFID, 1998, 134 p.)|
|Chapter five - Book box libraries: Mozambique|
Choice of methodology and rationale
For gathering data on the 'mobile' libraries, the methodology used in the study, because of the problems which have been identified already, was basically a series of interviews with the person responsible for the project and the use of the few pieces of archive material and reports which had not been destroyed.
For the 'portable' libraries, the methodology was similar. A series of interviews were held with the executive secretary of the NGO Progresso and a number of its reports and archive documents studied.
Methods used in data collection and analysis
Statistics, published annual reports and archive material from the institutions responsible for the book box libraries were used collect background information. Written reports and other documentation were a particularly important source of information on the CODE project.
Based on the general picture which was given to this researcher by the initial interviews and after a first analysis of some of the archive reports and documents, the bulk of the data was collected through structured interviews with the people in charge of the projects. These interviews were designed to take account of the data necessary for presenting the case studies but which did not figure in the documentation used or where a reading of the same gave rise to doubts over its interpretation.
Interviews with a sample of teachers or students to gather data on satisfaction and impact did not prove possible. In the case of the MINED project, the situation of instability and mobility among the population (in this case of teachers and heads of schools) caused by the war, together with the destruction of the archives relating to the project, meant that any sampling procedure would be impracticable.
An intermediate evaluation of the CODE project was conducted at the end of 1993, based on a sample of schools in Cabo Delgado. It did not use probability sampling, which is typical in statistics, but rather 'convenience sampling', which is characteristic of qualitative methods, namely selecting elements which will supply information which is pertinent to an in-depth study. The people responsible for the sampling chose two districts in the Province which were among the first where the project was implemented or where there was ease of access. By using the processes of interviews and direct observation, they visited a total of 10 schools in the two districts and interviewed about 25 pupils; they also interviewed teachers, the people responsible for the libraries, executives from the Provincial Education Office in Cabo Delgado and the district authorities for Pemba and Montepuez who were involved in education. The data collected in this evaluation is used in the present case study.
In respect of the MINED project, the biggest problems lay in the fact that the project archives had all been destroyed (a result of changes to the structures and management organs of the Ministry); the huge level of mobility among heads and teachers of schools throughout the whole country (mainly as a result of the war) meant that there was no possibility of interviewing any of the teachers or heads who had taken part in the pilot project.
Time and expense contributed to the difficulties encountered in collecting data on the CODE project. The lack of information in circulation in Mozambique on this project meant that it was only at the end of November 1997 that this researcher learnt of the existence of the project. The school teaching year in Mozambique is from February to November, which meant no data from the field could be collected until the following year, outside the time limits of the research. In addition, the costs of travelling and staying in Cabo Delgado (a province in the extreme north of the country), amounted to US$1,200, also outside the research, in terms of the funding available.