Suppose you want to estimate the effect of class attendance on student performance using the simple model = Bo + Biattrate + u sper f where sperf is student performance and attrate is attendance rate. (i) Is attrate endogenous in this model? Come up with an unobserved variable that is plausibly correlated with u and attrate. (ii) Let dist be the distance from a student's living quarters to campus. Explain how dist could potentially be correlated with u. (iii) Maintain that dist is uncorrelated with u despite your answer to part (ii) i.e. it is exogenous. Now, what condition must dist satisfy in order to be a valid IV for attrate? Discuss why this condition might hold.