2.2. Study Design and Sampling

This study was carried out as a cross-sectional study in five subcounties of Nyeri County between 24^th and 29^th of August 2015. The questionnaire was organized as a semistructured questionnaire, and after being prepared, it was pretested on a selected number of farmers; a final version that incorporated the results from the pretest was prepared and used for all subsequent interviews. The information gathered from the fish farmers was done through a face-to-face interview by the researcher with the help of enumerators (12 persons in total involved).

The interviews were done with respondents who were either the owners of the fish farms (n = 42), farm managers (n = 21), caretaker (n = 72), or pond attendant (n = 17) and the remaining having combined responsibilities (owner/caretaker n = 12) or serving different functions in the local society (n = 42). The questionnaire focused on fish farm details including type of pond (earthen, earthen with linen, wooden-linen, and concrete), source of water (flowing or stagnant), and age and species of fish and reported mortalities, management practices including pond fertilization, pond drainage and treatment, pond water testing, source of fingerlings, sharing of fingerlings, sharing of water sources, sharing and treatment of nets, and disposal of fish waste. For fish health, we focused on indicators of clinical signs of disease, i.e., feed uptake, poor growth, stationary fish, abnormal swimming behavior, skin lesions, abnormal skin color, swollen abdomen, and floating of fish in the ponds and scored as absent (0) or present (1). Farm management practices included ownership and usage of net, methods of net cleaning, fishpond types, sources of water, flow of water, pond fertilization, and disposal of fish waste after processing. The above parameters were chosen to provide information about disease occurrence and association with management practices. The interviews were combined with on-farm visits to the farms.

The subcounties studied were Kieni East, Kieni West, Nyeri Central, Othaya, and Tetu. A list of fish farmers who owned at least one pond was obtained from the Director of Fisheries in the county, and this list formed the sampling frame. Samples were purposively drawn from all the subcounties. In this study, households were the sampling units used. For each subcounty the following farmers were interviewed: Kieni East (n = 51), Kieni West (n = 30), Nyeri Central (n = 37), Othaya (n = 41), and Tetu (n = 49).