Part One – Correlation
Read Lecture Ten. Lecture Ten introduces the idea that different variables may move together—sometimes due to causation and at other times due to an unknown influence. An example involves the perfect (+1.0) correlation between annual number of rum barrels imported into the New England region of the U.S. between the years 1790 and 1820 and the number of churches built each of those years (citation lost). Discuss this correlation: What does it tell us? Does rum drinking cause church building? Does church building cause rum drinking? Or what else could it tell us? If this correlation shows a cause and effect relationship, what drives what? If not, why does it exist? What could this correlation be used for? (This should be started on Day 1.)
Part Two – Linear Regression
Read Lecture Eleven. Lecture Eleven provides information showing a strong positive correlation and a significant linear regression existed between the individual’s salary and midpoint (used as a substitute for grade). This is not an unexpected outcome in a company. How useful are these in understanding what drives salary differences? Why? What examples of a linear regression might be useful in your personal or professional lives? Why? (This should be started on Day 3.)
Part Three – Multiple Regression
Read Lecture Twelve. In Lecture Twelve, a multiple-regression equation was developed that showed the factors that influenced a person’s salary and—almost as important—factors that did not influence salary. How do we interpret a multiple-regression equation? Pick one of the factors—whether statistically significant or not—used in the analysis, and describe its impact on salary, what the coefficient is and what it means, what its significance is, and whether you expected this outcome or not. (This should be completed by Day 5.)
Your responses should be separated in the initial post, addressing each part individually, similar to what you see here.