4. MAIN FINDINGS OF SURVEY 1: FISH CONSUMPTION AND FISH EATING ATTITUDES

This chapter details the main findings of Survey 1. Results from the 1985/86 Household and Income Expenditure Survey (HIES) are used as a comparison for the general socio-economic data collected as part of the survey.

4.1 Description of Household Heads

Two-thirds of the sampled households were female headed (214 households), of which 52% were de-jure female headed households and the remainder de-facto female headed (household head is absent) households. This is higher than the national estimates given in the HIES which show that female-headed households account for around 45% of the total number of households. The bias in the result may be because only occupied houses were interviewed as well as because it was the start of the ploughing season.

The percentage distribution of households by type of head at each site is set out in Table 3. For dam numbers 2 and 3, the low percentage of female- headed households relative to the average for the whole sample could be attributed to the small sample size.

Table 3 Household Heads by Type

The average age of household heads is set out in Table 4. In general male heads were older than female heads, a trend also apparent in the HIES estimates.

Table 4 Average Age of Household Heads (years)

The marital status of these household heads is set out in Table 5. On average, approximately 66% of all households heads were married, 24% were single and 11% were widowed.

Table 5 Marital Status of Household Heads by Dam

The main economic activity of most household heads fell into three main categories: salaried employment (31%), livestock farming (23%) and crop production (17%). Just under 8% of household heads were too old to be economically active, 7% did nothing and the same percentage were involved with a self-help scheme, such as irrigated horticultural projects located in Gampudi, Letlhakane and Semarule. Just over 5% were self-employed, and 8% were either looking for work or involved in occasional work.

Of the 104 household heads who were absent, 56% were engaged in salaried employment in the Republic of South Africa, 22% were employed elsewhere in Botswana, 19% were at their lands and just under 2% were at the cattle post.

4.2 Description of Respondents

An overwhelming majority of respondents were female (86%). This may be because they were found at home during the time of the interview. The ploughing season had just commenced and most male household members were at their lands. The average age of respondents was 41.5 years but the standard deviation is large. Table 6 shows the breakdown by each dam.

Table 6 Average Age of Respondents

All respondents were related to the household head in some way. 44% of respondents were the spouse of the household head, 34% were the household head, and 17% were either the son or daughter of the household head. The remaining 5% of the respondents were close relatives.

4.3. Household Composition and Size

The average household size by each surveyed area is given in Table 7. These figures are not directly comparable to averages obtained from the HIES, because the HIES figures excluded absentee household members. A comparison has therefore not been made. However, apart from households interviewed at Gampudi, the average household size in each area differs very little as does the standard deviation. At Gampudi, small average household size may be attributable to the small sample as well as the higher than average age of household heads as most of those interviewed were all participants of a self-help project mainly comprised of elderly people.

Table 7 Average Household Size

The aggregated frequency distribution by number of persons by household, including absentee members is given in Figure 1, shows that the number of persons per household cluster between 5 and 10 persons with 9 persons being the most frequent size.

Figure 1: Household Size by Numbers of Persons in Household

4.4 Age of Household Members

The survey covered 2662 persons, of which 56% were 16 years old or over. The aggregated adult/child ratio (numbers of children per adult) including both present and absent members of the household was 0.8 and for present members 1.1. For the purposes of the survey, adults were classified as persons of 16 years or over. Table 8 sets out these ratios by each area based on the total number of persons (present and absent) as well as the for de jure and de facto female headed households. The figures indicate that de facto female-headed households tend to have a higher number of children per adult.

Table 8: Adult/Child Ratio by Type of Household and Area

4.5 Main Sources of Income for Households

Households were asked what their main and secondary sources of cash and income were. This latter question had the objective of finding out about other, perhaps less regular, sources of cash and/or subsistence available to the household. However, the results suggest that the responses may have been inaccurate as difficulties were experienced in phrasing the questions correctly in order to obtain the information. It is likely that households did not always mention subsistence activities or activities which did not generate any cash or income. For example, some households did not mention crop production although, according to results from other surveys as well as discussions with key informants, it is likely that some of their members were engaged in subsistence crop production. A likely situation which emerged from the survey results is a situation where the household was made up of a women and 2–3 children and an absent household head who was working in South Africa. The respondent, the wife, often stated that she did nothing and depended solely on remittances from her husband.

With regard to remittances from household members working in South Africa or elsewhere in Botswana, the answers vary a great deal. When wages from employment are clearly the main source of cash/income for the household, they were recorded. However, where absent members were stated as being engaged in wage employment, remittances were often not given as a source of income. This is a flaw in the interview schedule, and therefore the responses have been ‘cleaned’ so that where absent household members are in wage employment, but remittances have not been mentioned, an assumption has been made that the household does receive remittances from these members. This assumption has been recorded separately under the heading remittances. These results should be treated with extreme caution, not only because they may be an invalid assumption but also because the amount and the regularity of such remittances would vary enormously.

Tables 9 and 10 show the sampled households' main sources and secondary sources of income respectively.

Table 9 Main Sources of Cash by Number of Responses

For households at all dams, livestock, crop production and wages from South Africa and Botswana (with the exception of Dam 6) were the most important sources of income. At the main villages of Kanye (Dam 1), Moshupa (Dam 5) and Molepolole (Dam 7) salaried employment is also an important source of income. Letlhakane, Gakgatla and Semarule are all lands areas and therefore crop production is the most commonly reported main source of income.

Table 10 Secondary Sources of Income/Subsistence of Responses

According to the results of the survey, remittances and crop production form an important secondary sources of income. Livestock as a source of income (cattle and small stock) was only stated by a small number of respondents (6.5%) but this is probably because they are considered as a source of ready cash as numerous studies have indicated.

Just over one-third of the respondents stated that their households had no secondary sources of income. As previously mentioned, this may be because they did not consider crop production or any other activity for household use only as a source of income. These latter activities, could include, for example, the motselo which can provide an important source of income for women. This is a system whereby a group of women periodically take turns to hold a party where people can buy food and drink, the proceeds going to the person that held the party.

The number of people responding that they brew beer as a source of income may be underestimated because questions regarding sale of beer were not asked and other surveys¹ have indicated that sale of beer (including chibuku and tinned beer) as opposed to brewing only, is an important source of income for women.

¹ Gulbrandsen, O Agro-Pastoral Production and Communal Land-Use: A Socio-Economic Study of the Bangwaketse, Rural Sociology Unit, Ministry of Agriculture 1980 Haggblade, S. The Shebeen Queen or Sorghum Beer in Botswana: The Impact of Factory Brews on a Cottage Industry, University Microfilms International, Ann Arbor, Michigan USA 1984

Finally, occasional work, such as farm labouring or occasional village work was another stated secondary source of income for a small percentage of respondents.

4.6 Distance to the dam

The average distance between interviewed households and the dam by each surveyed area is given in Table 11. While at the larger villages of Kanye and Molepolole (Dams 1 and 7) households were interviewed located at varying proximity to the dam to enable more representative coverage, most households at the lands areas of Letlhakane and Gakgatla were further away from the dam than at the other sites.

Table 11 Average Distance Between Household and Dam

4.7 Consumption of Fish

The survey revealed that 23% (78) of all households ate fresh fish, and 80% (280) of all households ate fish of some kind. These findings are described in detail in the following Section.