Academic Integrity: tutoring, explanations, and feedback — we don’t complete graded work or submit on a student’s behalf.

28. Suppose we want to estimate the effects of alcohol consumption (alcohol) on

ID: 2925147 • Letter: 2

Question

28. Suppose we want to estimate the effects of alcohol consumption (alcohol) on college grade point average (colGPA). In addition to collecting information on grade point average and alcohol usage, we also obtain attendance information (say, percentage of lectures attended, called attend). A standardized test score (SAT) and high school GPA (hsGPA) are also available. (2 points) Propose at least two different ways to measure "alcohol consumption" (describe your variables and be very specific, talk about the units) (2 points) Write down the econometric model including variables and coefficients (general case, level-level) a) b)

Explanation / Answer

a)

Yes, we should include attendance information along with alcohol consumption as explanatory variables in a multiple regression model. This is because attendance is likely to be correlated with alcohol consumption in sense that higher is attendance, lower time will be spent on alcohol consumption and therefore lower will be the alcohol consumption. So, if we exclude attendance from the regression, beta alcohol will give gross effect of alcohol consumption on college GPA and therefore our results would be biased whereas if we include attendance in the multiple regression model, the beta alcohol will give net effect of alcohol consumption on college GPA and therefore the results would be robust.

This implies that we must include attendance information along with alcohol consumption as explanatory variables in a multiple regression model.

b)

Although SAT and hsGPA are likely to be correlated with alcohol consumption but there is no ground to say that SAT and hsGPA both will affect college GPA. So including or excluding SAT and hsGPA would make no difference to the “beta alcohol”.

However, if we assume that SAT and hsGPA both will affect College GPA, then SAT and hsGPA should be included as explanatory variables in the multiple regression model.

Hire Me For All Your Tutoring Needs
Integrity-first tutoring: clear explanations, guidance, and feedback.
Drop an Email at
drjack9650@gmail.com
Chat Now And Get Quote