Question

Part A: Imagine a researcher thinks that using caffeine before a test increases test scores. If they had the consent of all the students in a 120-person class to participate in a study, a good experimental design would follow which protocol?

	a.	Collect information after the test from the students about the caffeine they consumed the morning of the test. Regress that against their test scores.
	b.	Ask the whole class to drink16 ounces of coffee the morning of the test, but randomly vary the ratio of caffeinated and decaffeinated coffee in their cups. Regress the caffeine ratio on test-scores
	c.	Get test scores of top scoring students in the class after the test, and then ask how much caffeine they had before the test. Regress the answers on the best test scores.

Part B: In a regression equation, the coefficient for the constant (aka intercept) represents,

	a.	The estimated value of the dependent variable when all model variables are at their minimum observed value in the data
	b.	The estimated value of the dependent variable when all model variables are equal to zero
	c.	The estimated value of the dependent variable when all model variables are at their mean value
	d.	The minimum observed value of the dependent variable in the data.

Answer 1