The intercept turns out to be a particular predicted value. The intercept in the regression line for the Statistical Knowledge Quiz scores

is 10.9. If pre-SKQ = 0, then we would predict post-SKQ = 10.9 + .42(0) = 10.9. In other words, the intercept is the predicted value for the vertical axis variable when the horizontal axis variable equals 0. Sometimes this special predicted value is of interest and sometimes not. In this case, it is not interesting because a score of 0 on the twenty-item true-false Quiz is a very unlikely value. The actual pretest scores ranged from 8 to 14, which does not come close to including 0. A general rule is that the regression line should not be used to make predictions outside the range of the data from which it was derived.