Answer to Question 1
B
Answer to Question 2
The primary objective when using regression analysis for descriptive purposes is to measure the relationship between the dependent and independent variables. There are several statistics that are useful for this purpose. The first is the correlation coefficient. The closer this measure is to positive 1.0 or minus 1.0, the stronger is the linear relationship. The R-square value measures the percentage of variation in the dependent variable that is explained by the independent variable. Values close to 1.0 (or 100 percent) indicate that a strong linear relationship exists.
However, the primary statistics of interest are the regression coefficients. The intercept value (b0 ) can be interpreted to be the average value of y when x = 0.0. Note: This interpretation is valid only when the data used to develop the regression equation can legitimately have values of the x variable equal to zero. If not, then the intercept has no particular interpretation. The regression slope coefficient (b1 ) is usually of prime importance in a descriptive analysis. We interpret it to mean the average change in y for a one-unit change in x. We are particularly interested in the sign on the slope coefficient and the actual value of the coefficient too. We remember that the value of b1 is a point estimate and, thus, subject to sampling error. Therefore, we would likely develop a confidence interval estimate for the true population regression slope coefficient. If this interval does not cross zero, we say that there is a significant linear relationship.