A repeated measures study was conducted to examine the effects of two different store displays for a household product on sales in four successive time periods. Eight stores were randomly selected and four were assigned at random to each display. The sales data are shown below.

1. The experimenter obtained the residuals for repeated measure model (i.e. split-plots design model) and plot them again the fitted values. He also prepared a normal probability plot of the residuals. What do you conclude about the appropriateness of the model based on these plots? Specifically, please comment on the assumptions of the normality and homoscedasticity.