A single village was selected in each of the two regions. The criteria for selecting the two villages were (i) It should have a relatively high number of semi-intensive fish farmers4 and (ii) The distances from village to market and village to extension office should be different for the two villages. Subira village (11 km west of Songea) in Songea Urban District, was selected from Ruvuma region; and Ihanda Village (11 km southwest of Vwawa) in Mbozi district was chosen from Mbeya region.
Three weeks were spent in Subira village and just over two in Ihanda village. The methods used to identify adoption and viability criteria for semi-intensive fish farming included interviews with key informants, Participatory Rural Appraisal (PRA) with fish farmers and semi-structured interviews with semi-intensive fish farmers. Checklists were developed on the basis of other studies carried out by ALCOM5. The information obtained from these as well as from the PRAs are attached in Appendices 1 to 5. (Appendices 6 & 7 contain interviews with key informants and with semi-intensive fish farmers. These are available at ALCOM for those who are interested). Furthermore, information on the extension service was obtained through meetings with Fisheries Officers at the district and regional levels. Finally, general information on the status of fish farming as well as baseline data from the two regions were acquired through existing literature6.
Interviews with key informants were carried out to get a picture of agriculture in general and fish farming in particular in the villages. They also served as a check on information obtained through interviews with semi-intensive fish farmers. In both villages the key informants were selected on the basis of consultations with the village chairman and the village secretary, each place yielding three interviews. In Subira, a female primary school teacher and pond owner, a fish farmer who was an opinion leader on fish farming and the village chairman were interviewed. In Ihanda, the interviewees included a male primary school head teacher and fish farmer, a village elder and former pond owner, and three men who constituted a farming cooperative which also had fish farming among its activities.
The PRA sessions were carried out among 9–10 fish farmers (PRA participants) in Subira village and 6–8 fish farmers (PRA participants) in Ihanda village. They were selected to represent the more intensive fish farmers and fish farmers who had a comprehensive knowledge of fish farming in the village. Not all participants were present throughout all three days of the PRA. In Subira, one (out of two active) female fish farmers was among the PRA members. In Ihanda there are no female fish farmers.
5 E.g. van der Mheen-Sluijer (1991).
6 Seki and Maly (1993) and Msuya (1992).
The subjects explored through the PRA sessions covered:
a historical background with emphasis on fish farming;
a village map displaying social and physical features;
a seasonal diagram presenting the agricultural tasks throughout the year for each crop linked with an activity profile on amount of labour involved by each type of labourer (man, woman, child and hired labour). It further included seasonal distribution of rainfall, incomes and expenditures;
a ranking of definitions for a successful fish farmer;
a ranking of criteria leading to successful fish farming;
Venn Diagrams showing the importance of various institutions and persons for a fish farmer.
Semi-structured interviews were carried out with semi-intensive fish farmers (respondents) in order to obtain in-depth information on their fish farming practices. Six fish farmers were selected in each village. The more intensive fish farmers in the village were identified on the basis of information provided by three individual informants7 (not the key informants). In most cases they were individual farmers. However, in Subira, one so-called “respondent” actually referred to a couple, each of whom owned a pond but carried out farming activities together. In Ihanda, one so-called respondent actually referred to a farm cooperative with three farmers in the process of splitting up farm activities on an individual basis with fish farming still being a cooperative activity. It turned out that in each village one respondent had also been interviewed as a key informant.
Cross-checking of information from the different methods applied was an ongoing process throughout the study.
In order to reduce bias in responses, no fisheries officers or village officials were present throughout the PRA and the interviews with the villagers.
