The relationship between public debt and economic growth in

RESEARCH RESULTS 
Regression analysis 
In order to how advances from commercial banks , government stock , treasury bills , overdraft 
at the central bank of Kenya and treasury bonds (independent variables) contributes to Economic 
Growth (dependent variable), multiple linear regression analysis. The study conducted a multiple 
regression analysis. The findings were as shown in the table 1 below: 
Table 1: Model Summary 
Model Summary
Model 
R 
R Square 
Adjusted R Square 
Std. Error of the 
Estimate 
1 
.992
a
.985 
.962 
38651.09391 
a. Predictors: (Constant), External Debt, Advances from Commercial banks , Government Stock 
, Treasury Bills , Overdraft at the Central Bank of Kenya , Treasury Bonds
Coefficient of determination explains the percentage of variation in the dependent variable that is 
explained by the independent variables. It is used to explain the extent to which changes in the 

---

dependent variable can be explained by the change in the independent variables. 
From the analysis, the independent variables (commercial banks, government stock, treasury 
bills, overdraft at the central bank of Kenya and treasury bonds and public debts) in this study 
contributed to 98.5% of the variation in economic growth as explained by adjusted R
2
of 0.985.
The study conducted an Analysis of Variance, in order to test the significance of the model. The 
findings were as shown below: 
Table 2: ANOVA
b
 
Model 
Sum of Squares 
df 
Mean Square 
F 
Sig. 
1 
Regression 
3.872E11 
6 
6.453E10 
43.193 
.001
a
Residual 
5.976E9 
4 
1.494E9 
Total 
3.931E11 
10 
a. Predictors: (Constant), External Debt, Advances from Commercial banks, Government Stock, 
Treasury Bills, Overdraft at the Central Bank of Kenya, Treasury Bonds
b. Dependent Variable: Economic Growth measures by GDP 
From the ANOVAs results, the probability value of .001
a
was obtained implying that the 
regression model was significant in predicting the relationship between Economic Growth and 
the predictor variables as it was less than α=0.05.