Linear Regression Analysis
H0 0 (Annual wage is not influenced by years of education.)
H1 0 (Annual wage is influenced by years of education.)
The selected level of significance, is 0.5. The selected test is hypothesis test for zero slope.
Table 1 shows the data for employees annual wage and years of education. Both years of education and annual wage are interval (ratio) scale variables. Figure 1, shows the scatter plot of employees current annual wage against their years of education. Since, an ellipse can be visualized therefore, assumption of linearity is met. There appears a positive linear relationship between years of education and annual wage.
Table 2 shows the results of regression analysis for employees annual wage against their numbers of years of education. The p-value (.001) is less than the selected level of significance of 0.05, thus, null hypothesis H0 is rejected.
Years of education significantly influences annual wage, .41,t(98) 4.43,p .001.Years of education also explains a significant proportion of variance in annual wage,R2 .17,F(1, 98) 19.58,p .001. The effect as measured by coefficient of determination (R2) is small.
The regression equation is given by
Wage (in Dollars) -699.95 2477.09(Years of Education)
The slope coefficient suggests that every years of education increases employees wage by about 2,477. The intercept coefficient is not relevant in this context.
In conclusion, employees years of education significantly influences their annual wages, however, there is a small effect of years of education on annual wage.