International Journal of Horticulture, 2016, Vol.6, No.3, 1-7
3
Materials and Methods
This study was carried out in Osun State, Nigeria. Osun State is one of the six States that make up the South West
Geo-political zone in the South-western part of Nigeria. The State experiences approximately eight months
(March to October) of bimodal rainfall and four months (November to February) of dry season each year, the
average annual rainfall ranges from 1,125mm in derived savannah to 1,475mm in the rain-forest belt. The mean
annual temperature ranges from 27.2°C in the month of June to 39.0°C in December. The area is mainly agrarian.
Food crops grown in the area include maize (
Zea mays
), yam (
Dioscorea spp.
), cassava (
Manihot esculenta
),
cocoyam (
Colocasia spp
.), rice (
Oryza sativa
) and vegetable (
Amaranthus spp
.). The permanent crops include
cocoa (
Theobroma cacao
), kolanut (
Cola nitida
) and oil palm (
Elaeis guinensis
). These crops are usually mixed or
intercropped. The State is made up of 30 Local Governments Areas (LGAs) which are distributed into three
agro-ecological zones by the State’s Agricultural Development Program (OSSADEP) namely; Ife-Ijesa
(rainforest), Osogbo and Iwo zones (derived savannah), with six administrative zones.
The forest zone is characterized by vegetation with rainfall averaging 1,475 mm perannum. The soils of the area
are mainly derived from undifferentiated basement complex rocks with pebble beds and to a lesser extent coarse
granite parent material. Thetopography is mainly undulating plains, dissected plains and patches of nearly level to
gently undulating plains. Arable crops cultivated include maize (
Zea mays
), cassava, okra (
Hibiscus esculentus
),
and chilli pepper (
Capsicum
sp.). The derived savannah zone is located at the northern part of the state and it is
characterized by savannah vegetation. The topography is mainly gently undulating plains with undulating
dissected plains. Annual rainfall is about 1,125 mm per annum. The soils of the area are derived mainly from
quartzite and basement complex rocks. The main crops cultivated are maize, cassava, okra, soybean (
Glycine max
)
and cowpea (
Vigna unguiculata
).
The selection of respondent farmers was multistage. In the first stage, two of the agro-ecological zones in the State;
Osogbo and Iwo were purposively selected based on land use, the zones are noted for arable crop farming being
derived savannah zone. Second stage involves selection of four Local Government Areas (LGAs) each from the
two agro-ecological zones; three villages from each of the four LGAs, five farmers from each village were
selected using systematic random selection at each sampling stage. A total of 120 respondents were interviewed.
Both primary and secondary data were used for the study. Primary data was collected on the socio-economic
status of the farmers, price and quantity of inputs and outputs, farm size, with the use of well-structured
questionnaire. Secondary data were obtained from records and documents on recommended agronomic practices
on fertilizer usage from Osun State Agricultural Development Programs (OSSADEP) and Ministry of Agriculture
and Natural Resources.
Descriptive statistics was used to determine the socio-economic characteristics of smallholder farmers in arable
crop production. Logistic regression analysis was used to determine factors influencing the use of fertilizer and
the major constraints to its use among smallholder arable crop farmers in the State respectively. The logistic
regression model derives its name from the logistic probability function (Gujarati, 2004) expresses the
quantitative dependent variable. A structural relationship is specified, the quantity of fertilizer used, ranging from
zero (i.e. non use of fertilizer) to different levels of fertilizer use, was regressed on the socio-economic factors,
and other independent variables such as level of output of the farmer, farmers family size, farmers’ income from
other farm enterprise, size of farmland, price of fertilizer which is the principal variable as quantity demanded of a
commodity is proportional with its price
ceteris paribus.
Following Amanze et al
.
(2010), since the dependent
variable in this study is both qualitative (i.e. probability of use) and quantitative, coefficient of multiple
determination was not used in deciding the level of significance rather Chi squared (χ
2
) was used. (Ohajianya et al
.
,
2007).
The model specified equation is stated below:
Ln Y = Ln (P/1-P) …………………………………………………. (i)
