ABSTRACT
Linear regression is highly similar to correlation in that both deal with the relationship between two variables (X and Y). As regression (prediction) employs the correlation coefficient to predict Y from X, it is clear that the accuracy of the prediction depends on the magnitude of the correlation coefficient. Linear regression, that is, predicting Y from X, consists of finding the best fitting line that comes closest to all the points on a scatter plot formed by the X (READ) and Y (GPA) variables. The most commonly used description of the line of best fit is the prediction line that minimizes the total error of prediction, that is, the sum of the squared errors of prediction. Based on this description, the line of best fit is often called the least-squares regression line. A measure of the strength of the computed equation is R-square, sometimes called the coefficient of determination.
