PRESENTATION OF REGRESSION RESULTS AND ANALYSIS GOTTEN FROM THE IMPACT OF FOREIGN AID ON POVERTY IN NIGERIA

The method of Ordinary Least Squares (OLS) was used for running the regression. The estimates of the regression results will be subjected to various tests using the empirical findings provided by the results, a choice analysis will be made so as to come out with robust policy suggestions.

Variable	Coefficient	Std. Error	t-Statistic	Prob.
C	7.452407	1.063112	7.009992	0.0000
LOG(ODA)	0.212939	0.087926	2.421798	0.0256
LOG(GR)	-0.129545	0.062113	-2.085636	0.0507
LOG(PGR)	-4.986452	0.834317	-5.976690	0.0000
LOG(CAP)	0.041455	0.032670	1.268911	0.2198
LOG(GRGDP)	-0.004304	0.022743	-0.189229	0.8519

R-squared	0.936304
Adjusted R-squared	0.919541
F-statistic	55.85799
Durbin-Watson stat	1.253807

tα/2 2.069

F0.05 2.53

ANALYSIS OF RESULTS

From the table presented above, we arrive at the results

P = b₀+ b₁ODA + b₂GR + b₃PGR + b₄CAP + b₅GRGDP+Ut

Then,

LOG(P)= 7.452 + 0.2129ODA - 0.1295GR - 4.9864PGR + 0.0415CAP- 0.0043GRGDP + ut

The coefficient of L(ODA) = 0.2129, which shows that the elasticity of P with respect of ODA. Thus a percent change in ODA while other variables are held constant would lead to 0.02129% increase in P.

The coefficient of L(GR) = 0.1295, measures the relative change in the mean value of P for a percentage change in grants received, excluding technical assistance. It is negative relationship and it signifies that increase in Grants reduce poverty. Therefore over the period studied (1980-2008), 1% change in grants, excluding technical assistance reduces poverty on the average by 0.1295% holding other variables constant.

On the part of population growth rate PGR = -4.9864 and it measures the relative relationship between PGR and P. the population in poverty decrease by 4.9864% for a unit increase in the population growth rate holding other variables constant.

The coefficient of L(CAP) = 0.0415, measures relative change in the mean value of P for change in capital expenditure. There is a positive relationship and thus means that 1% change in capital expenditure increases poverty by 0.011% while keeping other variables constant.

The coefficient of GRGDP = -0.0043, which measure the elasticity of P with respect to growth rate of gross domestic product. It has a negative relationship implying that a unit increase in GRGDP reduces population in poverty by 0.0043% holding other variables constant.

EVALUATION BASED ON THE STATISTICAL CRITERIA (FIRST ORDER TEST)

Coefficient of Determination R²:

The R² of a multiple regression measures the degree of association of the independent variables taken together for percentage of total variation in LOG(P). It is the goodness of fit and has value of 0.936304 and this implies that 93% of the variation in the dependent variables is explained by the variation of the independent variable. While 7% of the variation in the dependent variables is explained by other variation which are captured by the error term.

Student T-Test

It is used to test for the statistical significance of individual estimated parameter.

Hypothesis:

H₀: b₁ = 0 (b₁ is not statistically significant)

H₁: b₁ ≠ 0 (b₁ is statistically significant)

Decision rule:

Reject H₀, if tcal > tα/2 at 5% level of significance where tcal = computed values of t and tα/2 = tabulated value. The value of t at tα/2 significance with degree of freedom = n – k.

The t_0.025 = 2.069

Variable	Parameter	t-Statistic	t- tabulated	Conclusion
C	β₀	7.009992	2.069	Statistically Significant
LOG(ODA)	β₁	2.421798	2.069	Statistically Significant
LOG(GR)	β₂	-2.085636	2.069	Statistically Significant
LOG(PGR)	β₃	-5.976690	2.069	Statistically Significant
LOG(CAP)	β₄	1.268911	2.069	Statistically insignificant
LOG(GRGDP)	β₅	-0.189229	2.069	Statistically insignificant

THE F-STATISTICS TEST

It is used to test for the joint influence of the explanatory variables on the dependent variable.

Hypothesis

H₀ = b₁ = b₂ = b₃ = b₄ = b₅ = 0 (all the slope co-efficient estimate are simultaneously zero)

H₁ = b₁ ≠ b₂ ≠ b₃ ≠ b₄ ≠ b₅ ≠ 0 (all the slope co-efficient estimate are not simultaneously zero)

Decision Rule

If F-cal > F-tab, reject the null hypothesis and conclude that the regression plane is statistically significant. Otherwise accept the null hypothesis.

With degree of freedom (K-1) (n-k); d.f = (6-1), (29- 6) = (6,23)

Thus, F_0.05= 2.53 and Fcal = 55.85799

Conclusion

Since Fcal > F0.05 i.e (55.86 > 2.53) we reject the null hypothesis (H₀) and accept the alternative hypothesis. We therefore conclude that our regression model is quite significant.

DURBIN WATSON (DW):

It is used to test for the presence of autocorrelation (serial correlation). The computed DW is 1.254. At 5% level of significance with 5 explanatory variables and 29 observations, the tabulated DW for dL and du are 1.050 and 1.840 respectively. The value of DW lies between the lower and upper limit. Therefore, we conclude that there is inconclusive evidence regarding the presence or absence of positive first order serial correlation.

EVALUATION OF RESEARCH HYPOTHESIS

The hypothesis can be evaluated by considering the result of the model. From the result, t-statistics of the values shows that capital expenditure and growth rate of GDP have no significant impact on the poverty level in the country, while the other variables including foreign aid (ODA and Grants) have significant impact on poverty.

Furthermore, the F-test could be used to evaluate the research hypothesis and thus given the F-statistics and F-tabulated we observe that our model is significant and this depict that, there is a relationship between poverty and foreign aid.

i. For the first hypothesis, we reject the null hypothesis and conclude that there is relationship between foreign aid and poverty in Nigeria.

ii. For the second hypothesis, reject the null hypothesis and conclude that there is significant impact of foreign aid on poverty in Nigeria.

POLICY IMPLICATION OF THE FINDINGS

So far, the research findings have been critically analysed and at this junction the economic implications and policy implications can be laid bare.

The statistical significance of net official development assistance (ODA) indicates that under the assumption that other factors remain constant, the increase in ODA does not reduce poverty in Nigeria. This could be attributed to the level of corruption in the country, institutional failure administrative incompetence and lots more such that 1% increases in ODA would cause the number of people living in poverty to increase by 0.2129%

On the other hand, the grants received excluding technical assistance show statistical significance and reduces poverty by when increase by 1% if other factors remain constant. This could be attributed tot eh specificity of purpose which accompanies most grants and as such there is much monitoring of the grants. These grants are often aimed at augmenting the efforts of the government on the standard of living of its people and also aid economic growth.

READ MORE ON FOREIGN AID

· SUMMARY, CONCLUSION AND RECOMMENDATION OF THE IMPACT OF FOREIGN AID ON POVERTY IN NIGERIA