Hypothesis
Ho: Socio-economic characteristics of the rural farmers have no significant effect on their income generated from agriculture.
Data Analysis
The specific objective was analyzed using multiple regression analysis. The model specifications are as follows;
Multiple regression analysis.
Y = (FX1, X2, X3, X4, X4, X5, X6)
Implicit form
Y = a0+ a1 + a1 + a2 + a2 + a3 + a3 + a4 + a4 + a5 + a5+a6+a6+et
Explicit form
Where;
Y = Level of income generated form Agriculture produce
X1 = Age (yrs)
X2 = Educational level
X3 = Farm size
X4 = Farm Experience
X5 = Farm Enterprise
X6 = Family size
a0 = Constant
a1-a6 = Coefficient.
et = Error term
Regression analysis of the relationship between the socio-economic characteristics of the rural household and their income generated from agriculture in Abakaliki L.G.A.
The result of the analysis is presented in table below. The income generated by rural household form agricultural activities where regressed against the exploratory variable (Socio-economic characteristics of the households) which are age(X1), Educational level (X2) farm size (X3), farming Experience (X4) farming enterprise (X5) and household size (x6) including the stochastic error term.
|
Variable |
Variable name |
Regression coefficient |
Std error |
T-value |
Sign |
|
Consistent |
– |
1.757 |
0.727 |
2.416 |
0.21 |
|
X1 |
Age |
0.112 |
0.074 |
-1.505 |
0.142 |
|
X2 |
Educational level |
0.007 |
0.229 |
0.030 |
0.976 |
|
X3 |
Farm size |
0.056 |
0.076 |
0.736 |
0.467 |
|
X4 |
Farm Enterprise |
0.373 |
0.184 |
2.026 |
0.051 |
|
X5 |
Farming experience |
0.023 |
0.065 |
-0.401 |
0.691 |
|
X6 |
Family size |
0.313 |
0.114 |
-2.747 |
0.010 |
R2 (square) = 0.873
Adjusted R2 = 0.848
Standard Error of the estimates = 0.73636
Durbin-Watson = 1.849
Source: Computed data, 2013
A multiple regression was utilized to analyse the relationship between the socio-economic characteristics of the rural household and their income generated from agriculture. In the analysis of the relationship between the socio-economic characteristics of the rural household and their income generated form agriculture, the computed coefficient of multiple determination (R2) 0.873 showed that 87.3% of total vacation in the dependent variable (income generated from agriculture) was caused by the combined effect of the the explanatory variables.
The high value of the adjusted R20.848 shows a good fit in the model specific
ation. This indicates that the proportion of changes accounted by the explanatory (independent) variables is high. Thus the exploratory variables help to predict the income generated from agriculture. The significant variables in the model are farm size and farming experience from table above, it sowed that variable, Age(X1) with a positive coefficient was not statistically significant. This showed that coefficient of Age had a positive relationship with the income generated from agriculture. Thus the apriori expectation not met since the result showed that increase in the age of farmer increase output. This finding refute that of Anele (2010) who claimed that the higher the age of farmer, the lower the farm output and thus income generated form such. This could due to lack of strength by some aged men who take agriculture as their lucrative business.
It was further discovered that the coefficient of educational level was positively signed related to the dependent variable (Y) and was not statistically significant. This is implies that the level of education of rural household increases their income generated from agriculture in the study area. This is justified by Babatunde et al (2007). Who claimed that educational level of rural households allows for flexibility in decision making hereby affording them the opportunity to take risk through engagement in multiple agricultural enterprises which can improve their aggregate income.
On the other hand, farm size (X3) had a positive coefficient and is significant. This mean that as farm size increase the income generated form agriculture increases. Thus, the higher the size of farm, the higher the income generated from Agriculture. This result conforms to the apriori expectation.
Farm enterprise (X4) also bore a positive coefficient and was not statistically significant. This conforms to the a priori expectation. It showed that the higher the farm enterprise and the level of experience in farm operations by rural households the higher the income generated from agriculture in the study area.
The result of regression analysis showed that the coefficient of farming experience complied with the apriori experience of the study. It shows that the higher the is a positive effect of farming experience on the level of income generated form agriculture produce in the study area.
House hold size (X6) which bore a positive coefficient did comply with the a priori expectation. This revealed that the size of households in the study area determines to a great extent the income generated from agriculture. This findings is in line with the claims of Okunmadewa (2001). According to him household size is a significant factor which contributes to increase in farm income.
Y = 1.757 + 0.112 + 0.007 + 0.056 + 0.373
(0.727) (0.074) (0.229) (0.076) (0.184)
(0.056) (0.114)
Hypothesis Testing
Decision Rule
If the F-cal is greater than the F-tab reject null hypothesis otherwise accept the alternative hypothesis.
F-cal = R2(N-K)
1-R2(K-1)
where;
R2 = Coefficient determination
N = Number of variable
F-Cal ?
F-cal = 0.873(40.6)
1-0.873(6-1)
F-cal = 29.682
0.635
F-cal = 46.74
F-tab = 2.10
Since the f-cal is greater the f-tab, null hypothesis was rejected while alternative hypothesis was rejected while alternative hypothesis.
Be the first to comment