the estimated coecient will be biased.

1

. For each scenario below, determine whether the estimated coefficient will be biased. If there will be a bias, determine the sign of the bias. In every scenario, the researcher is running a regression with SAT score as the dependent variable and a dummy variable for whether a person takes an SAT prep course as the independent variable to test whether taking a prep course is associated with a higher SAT score. The dummy variable is equal to one if a person takes a prep course and zero otherwise. Treat each scenario separately (in other words, when answering one part, ignore any of the omitted variables mentioned in the other parts). There may be multiple correct answers for a part, the key is to be able to explain the economic intuition behind your chosen answer.

(a) Individuals who take prep courses are more likely to have parents that will buy them other test prep materials which help improve scores.

(b) Individuals who take prep courses do so because they tend to do poorly on standardized tests and need extra help.

(c) All students sign up for prep courses but due to space limitations, only a random subset of students are admitted into the prep course.

(d) All students sign up for prep courses but due to space limitations, only those students willing to pay the most are admitted into the prep course.