Agriculture Reference
In-Depth Information
various groups' relationships with other institutions. The research was done
with the participation of a number of teams within the departments of
Agricultural Research and Extension and the Ministry of Agriculture (both
the National Early Warning Unit and the Harare Urban Extension teams).
Both work in the areas of UA, food security and livelihoods research.
A review of the available literature on UA in general and in Harare in
particular was carried out, using both published and unpublished documents.
Six weeks of observation and unstructured interviews with city authorities
and households involved in UA were a starting point for the research.
Study sites
Harare Metropolitan Province is divided into four districts: Harare Rural,
Chitungwiza, Epworth and Harare Urban. Harare Urban had the largest
proportion (76 per cent) of the population in the 2002 national census. The
present study took place in the Harare Urban district, which is made up of 48
wards spread over 27 suburbs. In this district, on-plot agriculture exists and
was provided for in low-density suburbs such as Borrowdale. Meanwhile, off-
plot agriculture is found across the city in suburbs of varying densities, with
varying degrees of intensity and tenure regimes. Peri-urban agriculture (UPA)
exists in areas surrounding the city and some of the lands are smallholdings
zoned for agricultural purposes.
A total of 16 suburbs where most of UA was assumed to be taking place
were selected for the study. Generally speaking, high-income residents tend to
inhabit low-density areas, medium-income residents live in suburbs of all
density types and low-income residents mostly inhabit high-density suburbs.
The distribution of the 372 households among the selected suburbs is shown
in Table 4.1.
A household survey which included a consent form was used to collect
information from sampled households. It collected information on
demographics, assets, support to UA, UA-related activities and formal and
informal employment. It also collected information on coping strategies
related to income, expenditures and consumption.
Survey sampling and sample size
The study used a two-stage sampling strategy, where the first stage was the
ward or suburb. A list of these was obtained from the Central Statistics Oce,
representing wards used for the 2002 population census. These were
deliberately chosen on the basis of residents' participation in UA. The second
stage for sampling was households, chosen randomly by selecting people who
were in their fields during the time of the interview.
The minimum sample size required was calculated using the formula
(Equations 4.1, 4.2) for large samples given by Poate and Daplyn (1993):
n
¼
z
2
c
2
x
2
ð
4
:
1
Þ
