APPENDIX 1
Methodology

1. The study area

Four areas in Mwinilunga District were selected as study area and for interviews with farmers. They were Ikelenge, Nyakaseya, Mwiniyilamba and Kanongesha. Eight chiefs in the district have fish farming activities.

2. Interviews with fish farmers/visits to fish ponds

In all, 23 of the planned 24 structured and informal interviews were completed. In 12 households the farmers' wives were also interviewed. Including the interviewed farmers, a total of about 40 farmers were met for brief discussions, and about as many fish farms were visited.

For the interviews, questionnaires were developed dealing with pond construction, pond management, resource allocations, harvesting methods and disposal of harvested fish. These questionnaires served to structure the informally held interviews but were not strictly followed.

3. Selection of fish farmers for interviews

In the ALCOM fish farmer survey, 56 fish farmers were interviewed in the four areas under study. These farmers had been randomly selected so as to form a rough 10% sample of the fish farmers in the area. At the stage of selecting farmers for this study, the information provided by the fish farmer survey was not sufficient to group the farmers according to the farming system they practised. Instead, the 56 farmers were divided according to age and origin of fish ponds, assuming these to be factors influencing production levels and management practices.

As a result, four groups emerged :-

1. ICARA farmers with subsidized ponds only, established in 1986 or later.
2. “Status quo” farmers with subsidized ponds only established 1985 or earlier.
3. Expanding with subsidies (farmers with subsidized ponds).
4. Expanding with own means (after initially subsidized ponds, farmers expand using own labour and/or money).

From each group, six farmers were selected for interviews and four back-ups using the ballot.

In the course of the study it became obvious that the information on which this division was founded, was insufficient. Farmers often gave contradictory information as to the origin of pond construction, and present and past means.

After the study the 23 farmers interviewed were grouped as follows :-

A comparison between the four groups intended was not meaningful given the small number in the first two groups. The two expanding groups were instead compared with the three “non-expanding” farmers from the first two groups.

4. Interviews with Department of Fisheries (DoF) staff

All eight Fish Scouts in the district were interviewed using a formal questionnaire (Appendix II).

General discussions were held with the Provincial Fisheries Officers in Solwezi and the staff at the fisheries station in Mwinilunga.

5. Interviews with key informants

People indirectly involved in fish farming, or who otherwise could provide information of interest for the study were informally interviewed. They were agriculture extension workers, nutritionists, missionaries, fish wardens, primary school teachers and representatives of multipurpose co-operatives and fish farmer committees. Besides, staff at the Mutanda agricultural research station were interviewed in their capacity as farming systems researchers.

6. Collection of harvest results

The harvest results for all individual ponds (not only those of the interviewed farmers) that were cropped within the period 1st June 1988 to 1st June 1989, were collected from the Fish Scouts for chiefs Ikelenge, Nyakaseya, Mwininyilamba and Kanongesha. Besides, results from the previous harvest for these ponds were collected in order to assess the duration of the average production period and the productivity. This information was obtained from records kept by Fish Scouts in the study area.

7. Analysis of the relationship between fish farming and farming systems

After the interviews, when more data was available, the 23 farmers were divided according to the farming systems they practised.

The Farming Systems Research Programme carried out by the Adaptive Research Planning Team (ARPT) provided the farming systems classifications which made up the framework of the study. According to ARPT, two major farming systems were present in Mwinilunga District. They were:-

• the cassava-based traditional subsistence shifting cultivation system
• the small-scale, semi-commercial system.

Of the interviewed farmers, 13 belonged to the semi-commercial group and 10 to the subsistence-oriented group.

These groups were compared with respect to pond management, harvesting methods and disposal of harvested fish. Information from interviews with Fish Scouts and key informants was incorporated into this comparison.