A fellow student conducts a study on the Serial Position Effect. He gives subjects a list of words and then tests their recall. He hypothesizes that words early in a list and
words at the end of a list will be remembered better than words in the middle of the
list. He calculates a regression line using word position as the predictor variable and
recall as the response variable. Then, he calculates a confidence interval to test the
Null Hypothesis about the slope of the regression line. To his surprise, he fails to
reject the Null (see figure below). He does not understand how this can be, because
he knows from the research literature that there is a very strong relationship between
word position and recall. What assumption was not satisfied in order for your fellow
student to calculate the regression line?