The survey of farmers with fish ponds in the North Western Province was designed:
Thus, the survey had definite and limited objectives. Additional or other objectives could be set for a survey of this kind, in particular to establish the following:
This report starts with a description of the general status and the context of fish farming in Northwestern Province. It goes on to describe survey executon, analyzes the data obtained, and presents conclusions and recommendations.
The North Western Province borders on Angola to the west, Zaire to the North, the Copperbelt Province of Zambia to the east and the Western Province to the south. It covers a territory of 125 826 km2 and in 1980 had a population of 302 688. Administratively the Province is divided into six districts: Mwinilunga, Kasempa, Solwezi, Kabompo, Zambezi and Chizera. The first three were visited by the survey teams, and are briefly described below.
The Northwestern Province has received a heavy influx of Angolan and Zairian refugees. In 1987 some 23 000 were settled, a large part in the Mwinilunga District. The Province has received donor and government assistance to help settle refugees and to combat malnutrition among them and the Zambian population. A part of this aid has been used to promote the culture of fish in ponds.
The population density in the Province is 2.4 per km2, the lowest in Zambia (1981). Of the population 87% live in rural areas where the main occupation is subsistence farming.
Farmers spread their risk amongst several crops and animals as well as off-farm activities. Only a minor part of incomes is obtained in the form of cash. Farming differs somewhat from district to district. In the Kasempa and Solwezi districts the farming system is based on sorghum while in Mwinilunga district it is cassava based. In all three districts shifting cultivation is practised by the majority of farmers.
Until recently, there was no organized marketing of the major crops -- sorghum, millet and cassava -- in the province. These crops were produced -- and are still produced -- mainly to serve subsistence needs, with only a small and occasional surplus traded or bartered in local markets or converted into beer. Crops are now gradually brought into the formal marketing system administered by the North Western Province Co-operative Union. Seasonal credit to qualified farmers to acquire inputs is provided by the Agricultural Finance Company (AFC).
In Kasempa and Solwezi districts the most important single source of cash is home-brewed sorghum/millet beer, sold locally. Other sources in order of importance are game meat, wild honey, chickens and beans. Occasionally goats are sold in Solwezi, and fish by those living near rivers and streams. Off-farm activities are at least as important for generating cash as farming activities.
In Mwinilunga district small amounts of beans, onions, cassava and chickens are sold for cash. Labour is rarely hired for cash, more often in return for beer or game meat. In addition, there are several off-farm activities to generate cash income; wild honey (sometimes brewed into beer and sold), hunting, fishing, and handicrafts.
There are a few “semi-commercial” farmers in the three districts. Apart from their subsistence production they produce special crops for the markets -- such as maize, Irish potatoes and pineapples -- also using purchased inputs. About 3% of the households in Mwinilunga, Kasempa and Solwezi rear cattle for sale.
Full-blown commercial farming is found only in a few cases and is insignificant in relation to the areas and numbers of individuals involved.
Migration to find employment for longer or shorter periods takes place from all districts in the Province. Most migrants go to the Copperbelt Province.
Mwinilunga district, the one most thoroughly covered in this survey, is characterized by a gently undulating topography with plenty of valleys, each with perennial streams and rivers, much of it still virgin land. The conditions for fish farming are excellent in the district, as well as in the rest of the Province. The population density is low. Land and water are in abundance and there is little, if any competition amongst the different types of land use (fish farming as opposed to agriculture, irrigation, grazing, settlement, etc.)
By the beginning of the 1980s the total number of earthen ponds was somewhere between 100 and 200. Tilapia of mixed origin were cultured. The number grew rapidly during the 1980s in response to aid projects intended mainly to alleviate the plight of refugees.
The records available with the Department of Fisheries and the ICARA show that by the end of 1988 there were in the Province 1 418 farmers who between them had 2 625 ponds with a combined surface area of 127 hectares. The details are given in Table 1.1 below.
Table 1.1 Farmers with fish ponds: North Western Province
|No of farmers||No of ponds||Area (ha)|
|Mwinilunga district||1 195||2 189||108.6|
|Province (Total)||1 418||2 625||127.0|
Source: Department of Fisheries (Preliminary, not official figures)
(i) The “ICARA” fish farmers
During the early '80s, the UNHCR funded a fish culture project to combat malnutrition among peasant farmers, especially refugees from Angola and Zaire. By 1984, 800 ponds had been completed.
In September 1984, a proposal for expansion of the project was accepted by the US Government's Bureau for Refugee Programmes. It was set up as a 33-month US$969 000 project (September 1984 – June 1987). The Zambian Catholic Secretariat implemented the project, with administrative support from the International Catholic Migration Commission (ICMC, Geneva). This is usually referred to as the ICARA project.
From July 1, 1987, the Department of Fisheries in the Ministry of Agriculture and Water Development, Zambia, became the implementing agency for a one-year “consolidation” project known as ICARA-II. Financial and administrative backstopping was provided by the US Government through the ICMC. The project's long-term objective was to develop a system of integrated fish farming to counter malnutrition and provide cash income, for refugees and Zambians alike, living in Mwinilunga district and the Meheba Refugee Settlement Scheme.
The main project components were training of fish farmers, loans and extension advice for the construction, stocking and cropping of ponds, inputs and extension services for five crops for five agricultural components -- ducts, rabbits, field crops, vegetable crops and fruit trees.
Although a project basically intended to assist refugees, using fish farming both as a developmental trigger and as a substitute for food relief, ICARA did not aim exclusively at refugees. The participation of Zambians in the project was encouraged, so that the local community accepted the idea of subsidies directed primarily at refugees. The project's target ratio was 60% refugees and 40% Zambians.
Farmers received from ICARA Kwacha 1 000 to construct two ponds, each of 600m2. This applied to farmers selected by ICARA. The project funded their participation in fish farming residential courses at the Mwekera fish farm in the Copperbelt Province. Farmers who did not take part in training courses, but nevertheless approached the project for assistance, received Kwacha 100 for each 200m2 of pond area, with an upper limit of Kwacha 300. Between 300 and 400 farmers receiving subsidies participated in training courses.
Under the ICARA II project, 1 067 farmers received direct subsidies leading to the construction of 1 441 ponds with a combined surface area of 70 ha. They are located in the Mwinilunga District and in the Maheba Refugee Settlement in Solwezi district (see Table 1.2 below).
Table 1.2: Farmers and ponds (1985–88) Subsidized directly by ICARA
|No of farmers||No of ponds||Area (ha)|
|Mwinilunga district||886||1 259||60.40|
|Solwezi district (Maheba R. Settlement)||181||182||10.45|
|Province (Total)||1 067||1 441||70.85|
Source: ICARA Final Report
Note: The actual number of ICARA ponds in operation by mid-1988 is not known.
The number of farmers benefiting from ICARA is however, larger than the number which actually received subsidies. In Mwinilunga District too, those farmers who were previously established through the UNHCR project or had constructed ponds on their own, were serviced by the well developed ICARA/DOF extension network. Adding the farmers in Maheba Refugee Settlement, a total of 1 376 farmers (97% of all farmers with fish ponds in the province) benefited directly or indirectly from the ICARA project.
No doubt the ICARA project has had a great impact on fish farming in general in the North Western Province, particularly in the Mwinilunga District. Extension services were well developed and effectively provided to the farmers through the DoF staff. The project helped train the extension staff, posted them throughout the district and made them mobile by providing them with motorcycles. Farmers were visited regularly and assisted with nets at the time of major harvests. Fish scouts kept records of major harvest and farmers' progress in general. Fingerlings were provided from DoF ponds in Mwinilunga and Solwezi and brought to the farmers by ICARA vehicles.
However, as the demand for fingerlings exceeded the supply at the DoF fish farm, additional fingerlings had to be provided by established farmers on a nearby lake. Some successful farmers received additional training and were assigned as fish wardens for their neighbourhoods, receiving a token salary. At several locations, groups of farmers had formed fish farming committees, assisted by the fish wardens. When the ICARA project terminated in mid-1988, the entire extension network (already partly DoF) and project equipment was handed over to the DoF.
To sum up, the impact of ICARA was quick, massive, concentrated and to a great extent immediately successful, judged by the number of farmers involved and ponds constructed. In a very short period (three years) a body of knowledge on how to grow tilapia in ponds was formed and rooted in the area.
(ii) “NON-ICARA” farmers
In Solwezi and Kasempa districts, the extension service rendered by DoF without ICARA assistance is poor. Lack of public support was a common complaint among the 49 fish farmers recorded by the DoF for these two districts. The survey team found this to be a most effective constraint on expansion of fish culture in the two districts. The fish scout posted in Kasempa for example, had no means of transport whatsoever. Fish farmers interviewed by the survey team said they had not seen the extension service for several years. Moreover, the survey team came to know about fish farmers with fish ponds, who were not known to the local DoF staff. The total number of such “anonymous” farmers is not known.
DoF records indicated that 1372 farmers had fish ponds in the North Western Province. This large number of farmers posed a problem. Could a representative sample be surveyed in the time and with the resources available to the survey team? Were farmers subsidized by ICARA more homogeneous than others? Could they therefore be covered through a relatively smaller sample? If so, there would be more time to interview non-ICARA farmers. Should any particular categories of farmers (e.g. women headed households) be over - represented in the sample?
It was agreed that the survey should be allowed to reveal whether farmers subsidized by ICARA are more homogeneous in their fish farming habits or social and cultural characteristics than those who have not received ICARA subsidies. Also, it was agreed that no special group should be sought out for interviews.
It was agreed furthermore that a target of 130 farmers, about 10% of those having ponds in the province, would provide a realistic picture of the condition of practising and former farmers, and that this number of interviews could be accomplished. These 130 farmers were identified through random selection. In addition, 50 “back-up” farmers were identified. They would be interviewed in case those first selected were not available. “Potential” farmers were to be approached in the course of the survey; the target number was 75. Thus it was intended to interview a total of just 200 farmers.
It was also agreed that the survey staff would strive to meet certain conditions for the interviews.
Each farmer should ideally be interviewed in the presence of both an ALCOM aquaculturist and a person trained in social sciences. The fish ponds should preferably be inspected by the aquaculturist.
To the extent possible, an effort should be made to avoid during the interviews the presence of anyone who had advised (or were supposed to do so) the farmers on fish culture matters, so that farmers did not give biased information about harvests and pond management.
Fish farmers should also ideally be kept aside from neighbours, friends and particularly other fish farmers when interviewed, so that they felt quite free to express attitudes and provide answers on the basis of the results from their activities.
Before starting an interview, the enumerator should carefully explain the purpose of the survey, and the type of questions to be asked. The farmer should be given the opportunity to ask questions before and after the interview.
It proved impossible to meet all these conditions during all interviews. Although working close together during the entire survey, it was sometimes not possible for the socio-economist and the aquaculturist to be both present during interviews, nor for the aquaculturist to visit fish ponds. In about 35 cases DoF staff were present during interviews.
The third and fourth criteria posed no problem. Explaining the purpose of the survey turned out to be important. Upon arrival the survey team was often identified by the farmers as ICARA staff or, as ICARA was about to leave the area, as representatives of other donors with similar aims. The survey team took great effort to clarify the purpose of the survey and the independent character of the survey team in relation to the ICARA project and the DoF.
The survey was executed in three visits during April – June 1988.
During the first visit the survey was conducted in Solwezi district, which includes the Maheba Refugee Settlement, and in parts of Mwinilunga district. The first interview was carried out on the 12th of April, the last on the 21st of the same month.
The survey team included the following Lusaka - based personnel: Ms. Lilian Chinkumbi and Ms. Nelly Mazingaliwa, social science graduates from the University of Zambia, and ALCOM staff members Rolf Larsson, socio-economist (APO) and Karl-Otto Wahlstrom, aquaculturist (APO). The following personnel from the province participated: Mr. B. Chakunguka, ICARA counterpart manager (DoF), local fish scouts and locally recruited interpreters.
Three teams were formed: Ms Chinkumbi and a fish scout (interpreter and aquaculturist); Ms. Mazingliwa, Mr. Wahlstrom and a locally recruited interpreter (contracted from a nearby school); and, Mr. Chakunguka (aquaculturist), Mr. Larsson and a locally recruited interpreter. The three teams had two vehicles, one of which was provided by the Department of Fisheries in Solwezi.
At the start of each day a plan was made specifying how the two vehicles were to bring the three teams to the selected farmers. On the whole there were no major difficulties in locating farmers for random selection. In a few cases the teams had to interview “back-up” farmers as the ones selected to be the first “target” were not available. In all, 56 interviews (42 practising, 5 former and 9 potential farmers) were carried out during the trip.
The second trip started on the 4th of May. The last interview was carried out on the 21st of the same month. During the second trip a total of 86 farmers were interviewed (67 practising, 6 former and 12 potential farmers).
For this trip only one vehicle was available. This meant that only two teams could participate. They consisted of Mr. Larsson and Mr. Wahlstrom, each assisted by a locally recruited interpreter. The local interpreters, all of them school teachers, were most helpful in locating farmers. They provided excellent assistance.
The farmers selected were approached to the extent possible, given the constraints of distance and means of transport (only one vehicle). It was, however, not always possible to strictly follow the list of sampled farmers. Sometimes, neither the first selected farmer, nor the back-up was available. Instead of spending time driving, often on bad roads, looking for the next farmer on the list, any other farmers nearby (either with ponds or the intention to dig them) were located and interviewed. Similarly, if one survey team was active interviewing, the other used the occasion to interview some other fish farmer or potential farmer who was present. Thus, in effect, during the second trip, the farmers interviewed were a combination of randomly selected fish farmers according to the original sampling procedure, and of farmers living close by.
In the Mwinilunga District the survey teams could not visit and measure about 25% of the ponds, because time was short. However, all those ponds are of the standard 600m2 size, and managed by farmers who had received ICARA subsidies.
It turned out to be difficult to locate former farmers. Very few appeared “spontaneously” in the sample. When this became obvious the district fish scouts were asked to prepare lists of farmers known to have abandoned their ponds. From these lists, the teams selected some at random for interviews. In the Mwinilunga Boma area, three women fish farmers were added to the randomly selected sample.
Scrutinizing the returns from the first two trips, it was realized that 97% of the farmers interviewed had received support from ICARA. If no more farmers were interviewed it would be impossible to make any meaningful comparisons between “ICARA” and “non-ICARA” farmers.
A third visit was therefore made to the area, following procedures developed during the second trip and with the same teams. It started on the 6th of June. The last interview was carried out four days later. A total of 14 farmers were interviewed in the Kasempa and Solwezi District (13 practising farmers and one ex-farmer).
Table 1.3 below gives an indication of the extent to which the survey plan could be followed. It does this by showing the number and location of farmers the teams were scheduled to interview (in accordance with the random sampling procedures) and the farmers actually interviewed.
Table 1.3: Farmers sampled and interviewed
(P & F)1
|Mwinilunga Boma and surroundings||17||10||0||4|
1 P. & F. stands for Practising and Former
Thus, on the one hand the initial target of interviewing 130 practising and former farmers was achieved, as 134 were interviewed. On the other hand, time did not permit as many interviews with potential farmers as intended.
The survey questionnaires were designed by ALCOM. They had been tested in a previous survey of the Northern Province and subsequently modified.
The reports referred to are
Fish farmers in rural communities: Evaluation of questionnaires and survey routines used during a pilot survey in the Northwestern Province of Zambia. By U. N. Wijkstrom and H. Aase. FI: GCP/INT/436/SWE.4. Rome, 1988.
Fish farmers in rural communities: Results of a socio-economic pilot survey in Northern Province of Zambia. By Ulf N. Wijkstrom and Hans Aase. GCP/INT/436/SWE.5. Lusaka, Zambia, 1989
The questionnaires are designed to elicit information which describes the farmers' current situation in terms of:
To establish the identity of the “fish farmer” the respondent is subsequently questioned about: (i) his motives, and the extent to which they are circumscribed or constrained by social or cultural factors; (ii) his social and economic situation; and, (iii) allocation of resources amongst fish farming, other farming, and non-farming activities.
These data when analyzed against the “current situation” information, will indicate the degree to which individuals with different backgrounds are: (i) more or less constrained in their management of fish ponds (ii) likely to continue or abandon the activity. It also indicates how far pond management practices might influence success in fish farming.
The data obtained during the survey were entered on a data base management programme operated on a personal computer. The same programme was used to analyze the data.