True or False? Correlation always implies causation. If a correlation is negative, then as it becomes even more negative, r^2 increases. If the units used to measure the X variable change (like from inches to centimeters), but the same data are analyzed, then the value of r will not change. As |r| increases, the average deviation of data from the predicted value (according to the best-fit regression line) increases. The best-fit regression line to predict Y when you know X will always go through the point Z_x = 0 and Z_y = 0. If a positive correlation exists between X and Y, and the range of X is then greatly restricted, |r| must increase. If a positive correlation exists between X and Y, and a new data point is added whose Z_x = 3 and Z_y = 0, the correlation will decrease. If no correlation exists between X and Y, and a new data point is added whose Z_x = 2.5 and Z_y = 2.5, r will increase. If a negative but imperfect correlation exists between X and Y, and a new data point is added whose Z_x = 2.5 and Z_y = 2.5, |r| will decrease. In correlation, reverse-scoring the X variable by multiplying each z score by -1 (so that originally high scores on X are now low, and originally low scores on X are now high) will not change the value of r^2.